From mailman-owner@ietf.org  Fri Mar  1 06:33:20 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA10936
	for <ieprep-archive@odin.ietf.org>; Fri, 1 Mar 2002 06:33:20 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id GAA03874
	for <ieprep-web-archive@ietf.org>; Fri, 1 Mar 2002 06:33:22 -0500 (EST)
Date: Fri, 1 Mar 2002 06:33:22 -0500 (EST)
Message-Id: <200203011133.GAA03874@optimus.ietf.org>
From: mailman-owner@ietf.org
Subject: ietf.org mailing list memberships reminder
To: ieprep-web-archive@ietf.org
X-No-Archive: yes
Precedence: bulk
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Session Initiation Protocol <sip.ietf.org>

This is a reminder, sent out once a month, about your ietf.org mailing
list memberships.  It includes your subscription info and how to use
it to change it or unsubscribe from a list.

You can visit the URLs to change your membership status or
configuration, including unsubscribing, setting digest-style delivery
or disabling delivery altogether (e.g., for a vacation), and so on.

In addition to the URL interfaces, you can also use email to make such
changes.  For more info, send a message to the '-request' address of
the list (for example, ieprep-request@ietf.org) containing just the
word 'help' in the message body, and an email message will be sent to
you with instructions.

If you have questions, problems, comments, etc, send them to
mailman-owner@ietf.org.  Thanks!

Passwords for ieprep-web-archive@ietf.org:

List                                     Password // URL
----                                     --------  
ieprep@ietf.org                          fPSM      
https://www1.ietf.org/mailman/options/ieprep/ieprep-web-archive@ietf.org


From ieprep-admin@ietf.org  Thu Mar  7 19:40:54 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA04874
	for <ieprep-archive@odin.ietf.org>; Thu, 7 Mar 2002 19:40:54 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA09089;
	Thu, 7 Mar 2002 19:39:23 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA09057
	for <ieprep@optimus.ietf.org>; Thu, 7 Mar 2002 19:39:21 -0500 (EST)
Received: from sj-msg-core-1.cisco.com (sj-msg-core-1.cisco.com [171.71.163.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA04825
	for <ieprep@ietf.org>; Thu, 7 Mar 2002 19:39:17 -0500 (EST)
Received: from nisser.cisco.com (nisser.cisco.com [171.71.176.85])
	by sj-msg-core-1.cisco.com (8.11.3/8.9.1) with ESMTP id g280cod23638;
	Thu, 7 Mar 2002 16:38:50 -0800 (PST)
Received: from SKEAGY-W2K1.cisco.com (sjc-vpn2-752.cisco.com [10.21.114.240]) by nisser.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with ESMTP id QAA15951; Thu, 7 Mar 2002 16:38:48 -0800 (PST)
Message-Id: <4.3.2.7.2.20020307162602.01ec6af0@malone.cisco.com>
X-Sender: skeagy@malone.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 07 Mar 2002 16:38:48 -0800
To: Fred Baker <fred@cisco.com>
From: Scott Keagy <skeagy@cisco.com>
Cc: ieprep@ietf.org
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Ieprep] Fwd: 53rd IETF - Bridge MIB WG (bridge MIB) cancelled
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

Fred,

I'm in process on other issues related to emergency services, but this is 
one that I just noticed and it is important that someone speaks up for it 
(bridge MIB) at the IETF mtg (which I will miss while tending my newborn baby).

Right now, Cisco ER can query only Cisco switches because of 
inconsistencies in the bridge MIB between different vendors. Whenever VLANs 
are present, there is no consistent way to present info via SNMP about 
which mac addresses appear on which switchport. This is a solution that 
desperately needs to be resolved as a starting point of locating callers in 
an ethernet campus environment. Otherwise, any application developer needs 
to write separate code to query every vendor's ethernet switches.

With no changes here, all standard location schemes will have to be based 
on IP address, which will only provide location granularity to the size 
covered by an IP subnet (which may easily span buildings - and technically 
cities or continents- in the presence of VLANs).

Hope this helps as an immediate action for ieprep.

Regards,
Scott


>To: IETF-Announce:;
>From: agenda@ietf.org
>Subject: 53rd IETF - Bridge MIB WG (brigde) cancelled
>Date: Thu, 07 Mar 2002 11:27:27 -0500
>Sender: dinaras@cnri.reston.va.us
>
>
>Bridge MIB WG (bridge) scheduled on Thursday, March 21 at 0900-1130
>has been CANCELLED.


Regards,

Scott Keagy, CCIE# 3985
Product Manager, Enterprise Voice and Video Business Unit
Cisco Systems, Inc.
tel:+1-408--853-0170
mailto:skeagy@cisco.com


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar  8 03:10:54 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA25002
	for <ieprep-archive@odin.ietf.org>; Fri, 8 Mar 2002 03:10:54 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id DAA09384;
	Fri, 8 Mar 2002 03:09:33 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id DAA09348
	for <ieprep@optimus.ietf.org>; Fri, 8 Mar 2002 03:09:31 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA24947
	for <ieprep@ietf.org>; Fri, 8 Mar 2002 03:09:28 -0500 (EST)
Received: from FRED-W2K6.cisco.com (dhcp-128-107-163-27.cisco.com [128.107.163.27])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2888xh19679;
	Fri, 8 Mar 2002 00:08:59 -0800 (PST)
Message-Id: <4.3.2.7.2.20020308000448.01c552d8@mira-sjcm-2.cisco.com>
X-Sender: fred@mira-sjcm-2.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 08 Mar 2002 00:06:32 -0800
To: Scott Keagy <skeagy@cisco.com>
From: Fred Baker <fred@cisco.com>
Cc: ieprep@ietf.org
In-Reply-To: <4.3.2.7.2.20020307162602.01ec6af0@malone.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Ieprep] Re: Fwd: 53rd IETF - Bridge MIB WG (bridge MIB) cancelled
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 04:38 PM 3/7/2002, Scott Keagy wrote:
>Right now, Cisco ER can query only Cisco switches because of 
>inconsistencies in the bridge MIB between different vendors. Whenever 
>VLANs are present, there is no consistent way to present info via SNMP 
>about which mac addresses appear on which switchport. This is a solution 
>that desperately needs to be resolved as a starting point of locating 
>callers in an ethernet campus environment. Otherwise, any application 
>developer needs to write separate code to query every vendor's ethernet 
>switches.

I could have sworn that there was some work in bridgemib to resolve that; 
in any event, that's a function for the bridgemib working group. It is 
their job to make management be consistent for LANs and VLANs.

I'm not sure what ieprep can really do in that area beyond note that it is 
an issue. What are you looking for?


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar  8 13:19:28 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA21957
	for <ieprep-archive@odin.ietf.org>; Fri, 8 Mar 2002 13:19:28 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA22292;
	Fri, 8 Mar 2002 13:19:06 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA22262
	for <ieprep@optimus.ietf.org>; Fri, 8 Mar 2002 13:19:04 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA21918
	for <ieprep@ietf.org>; Fri, 8 Mar 2002 13:19:02 -0500 (EST)
Received: from nisser.cisco.com (nisser.cisco.com [171.71.176.85])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g28IIWh10771;
	Fri, 8 Mar 2002 10:18:33 -0800 (PST)
Received: from SKEAGY-W2K1.cisco.com (sjc-vpn1-5.cisco.com [10.21.96.5]) by nisser.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with ESMTP id KAA07352; Fri, 8 Mar 2002 10:18:32 -0800 (PST)
Message-Id: <4.3.2.7.2.20020308095710.01623f88@malone.cisco.com>
X-Sender: skeagy@malone.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 08 Mar 2002 10:18:31 -0800
To: Fred Baker <fred@cisco.com>
From: Scott Keagy <skeagy@cisco.com>
Cc: ieprep@ietf.org
In-Reply-To: <4.3.2.7.2.20020308000448.01c552d8@mira-sjcm-2.cisco.com>
References: <4.3.2.7.2.20020307162602.01ec6af0@malone.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Ieprep] Re: Fwd: 53rd IETF - Bridge MIB WG (bridge MIB) cancelled
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

I'd just like to see this feature requirement (bridgemib) documented by 
ieprep to affect the prioritization of initiatives by the bridgemib working 
group.... To your point that there was some work to resolve the problem, I 
just found the following: http://ietf.org/rfc/rfc2674.txt

I'm checking with our developers now to see if this meets our needs (seems 
to me like it does).

Regards,
Scott



At 12:06 AM 3/8/2002 -0800, Fred Baker wrote:
>At 04:38 PM 3/7/2002, Scott Keagy wrote:
>>Right now, Cisco ER can query only Cisco switches because of 
>>inconsistencies in the bridge MIB between different vendors. Whenever 
>>VLANs are present, there is no consistent way to present info via SNMP 
>>about which mac addresses appear on which switchport. This is a solution 
>>that desperately needs to be resolved as a starting point of locating 
>>callers in an ethernet campus environment. Otherwise, any application 
>>developer needs to write separate code to query every vendor's ethernet 
>>switches.
>
>I could have sworn that there was some work in bridgemib to resolve that; 
>in any event, that's a function for the bridgemib working group. It is 
>their job to make management be consistent for LANs and VLANs.
>
>I'm not sure what ieprep can really do in that area beyond note that it is 
>an issue. What are you looking for?


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar  8 15:25:12 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01664
	for <ieprep-archive@odin.ietf.org>; Fri, 8 Mar 2002 15:25:12 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA03677;
	Fri, 8 Mar 2002 15:24:58 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA03644
	for <ieprep@optimus.ietf.org>; Fri, 8 Mar 2002 15:24:56 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01650
	for <ieprep@ietf.org>; Fri, 8 Mar 2002 15:24:52 -0500 (EST)
Received: from FRED-W2K6.cisco.com (dhcp-128-107-163-27.cisco.com [128.107.163.27])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g28KOOh18252;
	Fri, 8 Mar 2002 12:24:24 -0800 (PST)
Message-Id: <4.3.2.7.2.20020308122335.029d0220@mira-sjcm-2.cisco.com>
X-Sender: fred@mira-sjcm-2.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 08 Mar 2002 12:23:57 -0800
To: Scott Keagy <skeagy@cisco.com>
From: Fred Baker <fred@cisco.com>
Cc: ieprep@ietf.org
In-Reply-To: <4.3.2.7.2.20020308095710.01623f88@malone.cisco.com>
References: <4.3.2.7.2.20020308000448.01c552d8@mira-sjcm-2.cisco.com>
 <4.3.2.7.2.20020307162602.01ec6af0@malone.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Ieprep] Re: Fwd: 53rd IETF - Bridge MIB WG (bridge MIB) cancelled
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 10:18 AM 3/8/2002, Scott Keagy wrote:
>I'm checking with our developers now to see if this meets our needs (seems 
>to me like it does).

if it does, then we can release the concern, correct?


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 11 12:35:10 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA16746
	for <ieprep-archive@odin.ietf.org>; Mon, 11 Mar 2002 12:35:10 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA03317;
	Mon, 11 Mar 2002 12:34:30 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA03285
	for <ieprep@optimus.ietf.org>; Mon, 11 Mar 2002 12:34:28 -0500 (EST)
Received: from imo-d02.mx.aol.com (imo-d02.mx.aol.com [205.188.157.34])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA16702
	for <ieprep@ietf.org>; Mon, 11 Mar 2002 12:34:23 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d02.mx.aol.com (mail_out_v32.5.) id l.106.e86937c (4561)
	 for <ieprep@ietf.org>; Mon, 11 Mar 2002 12:33:52 -0500 (EST)
Message-ID: <106.e86937c.29be447f@aol.com>
Date: Mon, 11 Mar 2002 12:33:51 EST
To: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_106.e86937c.29be447f_boundary"
X-Mailer: AOL 6.0 for Windows US sub 353
Subject: [Ieprep] Comments on draft-carlberg-ieprep-framework
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--part1_106.e86937c.29be447f_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

To all,

Some comments on details in the draft:

4.1.1 Current SIP (RFC2543) and its revision define only four levels. There 
is no "other-priority". (There was a new value proposed in an earlier draft, 
but this has been removed.)

The draft referenced for [15] has expired and does not represent current 
thinking. The information is now included in the new Resource-priority header 
defined in draft-polk-sipping-resource-00, February 2002. It would probably 
be advisable to add a new "namespace" for IEPS to that draft for the values: 
"authorized-emergency", "general-emergency", "high", and "normal". If not 
included in the draft, they would be added later through IANA.

At this point, it would be good to consider whether the possibility should be 
included for multiple levels of "authorized". (Note that the parallel work in 
SG16 on H.460.4 is leaning towards including such a capability.)

4.1.2 Diff-Serv

Since it is unlikely that a new class with 5 drop precedence values would be 
approved, the alternative is to add 2 values to one of the current classes, 
resulting in 5 ascending levels. The problem with either of these plans 
concerns the potential use of the existing 4 classes. As depicted in 
draft-baker-ieprep-requirements, an example of their use could be the 
following:

EF for voice, with signaling of bandwidth
AF4 for video, with signaling of bandwidth
AF3 for voice/video signaling
AF2 for transaction/ERP traffic
AF1 for traffic that will potentially move large volumes

In this example, there would need to be the 5 drop precedence levels for EF, 
as well as for each AF class, since eventually there should be a need to 
apply precedence type services to other traffic types besides voice. Clearly, 
there are not enough available DSCP code points.

The description at the end of the third paragraph relates the 5 drop 
precedence levels of a possible new class with the 3 levels of the existing 
ones. There is a confusion, first of all, in the numbers. RFC 2597 defines 
1=low drop precedence and 3=high. That means the emergency traffic would use 
the low numbers (1 or 2) to get a lower probability of packet loss. Secondly, 
the values assigned within one class can't be compared with those of another 
class. They are completely disjoint sets of values, since each class 
represents difference queues or traffic types. (As RFC 2597 states: "Packets 
in one AF class MUST be forwarded independently from packets in another AF 
class, i.e., a DS node MUST NOT aggregate two or more AF classes together.")

If a new class were defined with 5 levels, then, for other non-emergency 
traffic, "the lowest drop precedence they are assigned to is 3" and 
"Emergency traffic would be set to 2 or 1".

As described in draft-pierce-assured-service, the preferential treatment for 
Assured Service traffic needs to be provided by a combination of call setup 
priority and packet forwarding priority. Since the most important capability 
is provided by the call setup priority, the presently defined 3 drop 
precedence levels per class may be sufficient for packet forwarding 
preferential treatment, as shown in Annex A of that draft.

Mike Pierce
Artel


--part1_106.e86937c.29be447f_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>To all,
<BR>
<BR>Some comments on details in the draft:
<BR>
<BR>4.1.1 Current SIP (RFC2543) and its revision define only four levels. There is no "other-priority". (There was a new value proposed in an earlier draft, but this has been removed.)
<BR>
<BR>The draft referenced for [15] has expired and does not represent current thinking. The information is now included in the new Resource-priority header defined in draft-polk-sipping-resource-00, February 2002. It would probably be advisable to add a new "namespace" for IEPS to that draft for the values: "authorized-emergency", "general-emergency", "high", and "normal". If not included in the draft, they would be added later through IANA.
<BR>
<BR>At this point, it would be good to consider whether the possibility should be included for multiple levels of "authorized". (Note that the parallel work in SG16 on H.460.4 is leaning towards including such a capability.)
<BR>
<BR>4.1.2 Diff-Serv
<BR>
<BR>Since it is unlikely that a new class with 5 drop precedence values would be approved, the alternative is to add 2 values to one of the current classes, resulting in 5 ascending levels. The problem with either of these plans concerns the potential use of the existing 4 classes. As depicted in draft-baker-ieprep-requirements, an example of their use could be the following:
<BR>
<BR>EF for voice, with signaling of bandwidth
<BR>AF4 for video, with signaling of bandwidth
<BR>AF3 for voice/video signaling
<BR>AF2 for transaction/ERP traffic
<BR>AF1 for traffic that will potentially move large volumes
<BR>
<BR>In this example, there would need to be the 5 drop precedence levels for EF, as well as for each AF class, since eventually there should be a need to apply precedence type services to other traffic types besides voice. Clearly, there are not enough available DSCP code points.
<BR>
<BR>The description at the end of the third paragraph relates the 5 drop precedence levels of a possible new class with the 3 levels of the existing ones. There is a confusion, first of all, in the numbers. RFC 2597 defines 1=low drop precedence and 3=high. That means the emergency traffic would use the low numbers (1 or 2) to get a lower probability of packet loss. Secondly, the values assigned within one class can't be compared with those of another class. They are completely disjoint sets of values, since each class represents difference queues or traffic types. (As RFC 2597 states: "Packets in one AF class MUST be forwarded independently from packets in another AF class, i.e., a DS node MUST NOT aggregate two or more AF classes together.")
<BR>
<BR>If a new class were defined with 5 levels, then, for other non-emergency traffic, "the <U>lowest</U> drop precedence they are assigned to is 3" and "Emergency traffic would be set to 2 or 1".
<BR>
<BR>As described in draft-pierce-assured-service, the preferential treatment for Assured Service traffic needs to be provided by a combination of call setup priority and packet forwarding priority. Since the most important capability is provided by the call setup priority, the presently defined 3 drop precedence levels per class may be sufficient for packet forwarding preferential treatment, as shown in Annex A of that draft.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_106.e86937c.29be447f_boundary--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 11 19:25:23 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA00552
	for <ieprep-archive@odin.ietf.org>; Mon, 11 Mar 2002 19:25:22 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA25212;
	Mon, 11 Mar 2002 19:21:32 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA25174
	for <ieprep@optimus.ietf.org>; Mon, 11 Mar 2002 19:21:30 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA00408;
	Mon, 11 Mar 2002 19:21:26 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2C0Ksh03202;
	Mon, 11 Mar 2002 16:20:54 -0800 (PST)
Message-Id: <4.3.2.7.2.20020311155247.0c4a3be8@mira-sjcm-2.cisco.com>
X-Sender: fred@mira-sjcm-2.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Mon, 11 Mar 2002 16:20:36 -0800
To: agenda@ietf.org, ieprep@ietf.org
From: Fred Baker <fred@cisco.com>
Cc: Scott_Bradner <sob@harvard.edu>
In-Reply-To: <200203111536.KAA11853@ietf.org>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Ieprep] 53rd IETF - Internet Emergency Preparedness
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

The agenda for IEPREP is as follows. We have two hours, and need to cut to 
the chase.

10 minutes: overview and agenda bashing
15 minutes: Kimberly King, GETS Overview
20 minutes: Ken Carlberg, Framework for IEPREP Telephony
15 minutes: James Polk: Ongoing MLPP discussions in SIP
60 minutes: discussion and assignment of tasks

These three presentations are primarily designed to make sure we are all on 
the same page. Kimberly's talk is to ensure we understand the key points of 
the service we are interacting with on the telephone system side. Ken's 
talk is to discuss issues in one draft which I believe represents a working 
group outcome. James will discuss separate but related work in SIP that we 
need to be cognizant of.

Hal Folts has also asked for some time to present his white paper, but 
frankly I think attendees can read it (and the other posted drafts), and 
that the concepts were presented and discussed in the BOFs. It seems better 
to me to focus on the task at hand. Please do read them; the documents 
present a useful overview.

Per our charter, we have two deliverables; we can also produce other 
documents if they don't conflict with getting these out. The two documents 
are:

    Best Current Practice:
         IETF Recommendations for the Emergency Telecommunications Service 
using
         existing protocols - what can be done with existing protocols and what
         can not be done.

    Informational:
         Requirements for Internet Emergency Preparedness in the Internet.
         Framework for Supporting Internet Emergency Preparedness in IP 
Telephony.

I expect Ken Carlberg's draft to become the "Framework" document, and we 
may also choose to publish Brown's Security draft. As to the other two 
posted drafts, text from them may fit into the above, or they may become 
the above; exactly what needs to be written and who is to write it are 
among the major decisions before the house.

Posted drafts are:
    http://www.ietf.org/internet-drafts/draft-baker-ieprep-requirements-00.txt 

    http://www.ietf.org/internet-drafts/draft-brown-ieprep-sec-00.txt 

    http://www.ietf.org/internet-drafts/draft-carlberg-ieprep-framework-00.txt
    http://www.ietf.org/internet-drafts/draft-folts-ieprep-white-paper-00.txt


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 11 19:51:49 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA01030
	for <ieprep-archive@odin.ietf.org>; Mon, 11 Mar 2002 19:51:49 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA26659;
	Mon, 11 Mar 2002 19:51:24 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA26625
	for <ieprep@optimus.ietf.org>; Mon, 11 Mar 2002 19:51:22 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA01024
	for <ieprep@ietf.org>; Mon, 11 Mar 2002 19:51:19 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2C0onh26237;
	Mon, 11 Mar 2002 16:50:49 -0800 (PST)
Message-Id: <4.3.2.7.2.20020311163603.0256ab98@mira-sjcm-2.cisco.com>
X-Sender: fred@mira-sjcm-2.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Mon, 11 Mar 2002 16:49:16 -0800
To: Mpierce1@aol.com
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
Cc: ieprep@ietf.org
In-Reply-To: <106.e86937c.29be447f@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 09:33 AM 3/11/2002, Mpierce1@aol.com wrote:
>At this point, it would be good to consider whether the possibility should 
>be included for multiple levels of "authorized". (Note that the parallel 
>work in SG16 on H.460.4 is leaning towards including such a capability.)

are these "levels of authorization" or "kinds of authorization"? My 
experience with levels is that the discussion is always about winners and 
losers: "my stuff is more important than his stuff, so I think there needs 
to be another level."

What makes sense to me in SIP is a simple statement: "I assert that this 
call is an emergency call". If you are not authorized to make that 
assertion, the correct policy is to ignore the statement. If you are 
authorized, the network will have a policy to apply such as "permit this 
call even if we are well into the overload zone" or "drop a random call in 
order to complete this one." The network should determine the policy, not 
the standard.

>4.1.2 Diff-Serv
>
>[snip] In this example, there would need to be the 5 drop precedence 
>levels for EF, as well as for each AF class, since eventually there should 
>be a need to apply precedence type services to other traffic types besides 
>voice. Clearly, there are not enough available DSCP code points.

I think you have a very different understanding of diffserv than I do. When 
you talk about a drop precedence for EF (which doesn't have a drop 
precedence), what are you talking about - dropping packets, or dropping calls?

Recall that in GETS, you don't get differing qualities of circuits. You get 
differing probabilities of getting a circuit, but once you have a circuit, 
it is just a circuit and works exactly like any other circuit. In IP 
telephony, you might have varying call preferences, but once you decide to 
carry voice at all, EF is the service that carries voice. If you're trying 
to model GETS in IP Telephony, dropping packets makes no sense. Dropping 
calls might make sense.

So it's not in the least obvious where diff-serv fits here. Please advise?

>As described in draft-pierce-assured-service...

is that draft-pierce-sipping-assured-service-01.txt?

BTW, in view of RFC 2597 "Assured Forwarding PHB Group", which is often 
called "Assured Service", I'd personally appreciate it if you could find 
another name. My head snaps every time I hear the name :^)


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 12 03:59:50 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA18156
	for <ieprep-archive@odin.ietf.org>; Tue, 12 Mar 2002 03:59:50 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id DAA28146;
	Tue, 12 Mar 2002 03:59:07 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id DAA28113
	for <ieprep@optimus.ietf.org>; Tue, 12 Mar 2002 03:59:06 -0500 (EST)
Received: from rbhub100.chamb.disa.mil (rbhub100.chamb.disa.mil [209.22.120.23])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA18147;
	Tue, 12 Mar 2002 03:59:02 -0500 (EST)
Received: by rbhub100.chamb.disa.mil with Internet Mail Service (5.5.2653.19)
	id <GWRLL6ZQ>; Tue, 12 Mar 2002 03:59:04 -0500
Message-ID: <AFBFF033E351D511999900204804EE54031758@rbmail104.chamb.disa.mil>
From: "Folts, Harold" <foltsh@ncs.gov>
To: "'Fred Baker '" <fred@cisco.com>, "'agenda@ietf.org '" <agenda@ietf.org>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Cc: "'Scott_Bradner '" <sob@harvard.edu>
Subject: RE: [Ieprep] 53rd IETF - Internet Emergency Preparedness
Date: Tue, 12 Mar 2002 03:59:03 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

 Fred,

Of course participants can read the white paper ID as it has been discussed
in the past. However the current ID is a revised version from
draft-folts-ieps-white-paper-00.txt. Anyway, will this ID be the basis for
the Information "requirements for ieprep"? If so, I would be interested in
view of what ajdustments might be necessary?

Hal

-----Original Message-----
From: Fred Baker
To: agenda@ietf.org; ieprep@ietf.org
Cc: Scott_Bradner
Sent: 3/11/02 7:20 PM
Subject: [Ieprep] 53rd IETF - Internet Emergency Preparedness

The agenda for IEPREP is as follows. We have two hours, and need to cut
to 
the chase.

10 minutes: overview and agenda bashing
15 minutes: Kimberly King, GETS Overview
20 minutes: Ken Carlberg, Framework for IEPREP Telephony
15 minutes: James Polk: Ongoing MLPP discussions in SIP
60 minutes: discussion and assignment of tasks

These three presentations are primarily designed to make sure we are all
on 
the same page. Kimberly's talk is to ensure we understand the key points
of 
the service we are interacting with on the telephone system side. Ken's 
talk is to discuss issues in one draft which I believe represents a
working 
group outcome. James will discuss separate but related work in SIP that
we 
need to be cognizant of.

Hal Folts has also asked for some time to present his white paper, but 
frankly I think attendees can read it (and the other posted drafts), and

that the concepts were presented and discussed in the BOFs. It seems
better 
to me to focus on the task at hand. Please do read them; the documents 
present a useful overview.

Per our charter, we have two deliverables; we can also produce other 
documents if they don't conflict with getting these out. The two
documents 
are:

    Best Current Practice:
         IETF Recommendations for the Emergency Telecommunications
Service 
using
         existing protocols - what can be done with existing protocols
and what
         can not be done.

    Informational:
         Requirements for Internet Emergency Preparedness in the
Internet.
         Framework for Supporting Internet Emergency Preparedness in IP 
Telephony.

I expect Ken Carlberg's draft to become the "Framework" document, and we

may also choose to publish Brown's Security draft. As to the other two 
posted drafts, text from them may fit into the above, or they may become

the above; exactly what needs to be written and who is to write it are 
among the major decisions before the house.

Posted drafts are:
 
http://www.ietf.org/internet-drafts/draft-baker-ieprep-requirements-00.t
xt 

    http://www.ietf.org/internet-drafts/draft-brown-ieprep-sec-00.txt 

 
http://www.ietf.org/internet-drafts/draft-carlberg-ieprep-framework-00.t
xt
 
http://www.ietf.org/internet-drafts/draft-folts-ieprep-white-paper-00.tx
t


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 12 07:40:53 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA20342
	for <ieprep-archive@odin.ietf.org>; Tue, 12 Mar 2002 07:40:53 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id HAA07896;
	Tue, 12 Mar 2002 07:40:41 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id HAA07847
	for <ieprep@optimus.ietf.org>; Tue, 12 Mar 2002 07:40:39 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id HAA20317
	for <ieprep@ietf.org>; Tue, 12 Mar 2002 07:40:35 -0500 (EST)
Received: from posh.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.21632-0@bells.cs.ucl.ac.uk>; Tue, 12 Mar 2002 12:40:31 +0000
X-Mailer: exmh version 2.5 07/13/2001 with version: MH 6.8.3 #25[UCI]
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-reply-to: Your message of "Mon, 11 Mar 2002 12:33:51 EST." <106.e86937c.29be447f@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Tue, 12 Mar 2002 12:40:29 +0000
Message-ID: <832.1015936829@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

hi,

> 4.1.1 Current SIP (RFC2543) and its revision define only four levels. There 
> is no "other-priority". (There was a new value proposed in an earlier draft, 
> but this has been removed.)

thanks.  this will be updated accordingly.

> The draft referenced for [15] has expired and does not represent current 
> thinking. 

well, technically the old draft hasn't expired yet.  but the point
is taken that a new draft, under a new name, exists.  however,...

> The information is now included in the new Resource-priority header 
> defined in draft-polk-sipping-resource-00, February 2002. It would probably 

...this newer draft wasn't available when the Framework draft was 
submitted.  James and I will probably sit down and hash out some 
of the finer details next week.

> be advisable to add a new "namespace" for IEPS to that draft for the values: 
> "authorized-emergency", "general-emergency", "high", and "normal". If not 
> included in the draft, they would be added later through IANA.

the current Framework draft only recomends a single value be defined for
a new NameSpace to be associated with IEPS, and I prefer not to go
beyond this single value.  one reason is that this single value would
correlate nicely to a one-to-one association with the existing single
NS/EP code point in the ISUP/IAM message used for GETS in the U.S. PSTN.

the other reason is that the absence of any other value implies that the
default mode is an INVITE without the Resource Header value -- which is
what we have today (and which reflects the installed base of existing
implementations).  i'll add wording to this effect in the next version
of the draft to clarify this position.

> At this point, it would be good to consider whether the possibility should 
> be included for multiple levels of "authorized". (Note that the parallel 
> work in SG16 on H.460.4 is leaning towards including such a capability.)

given the reasons above, i'd rather stay away from multiple levels for
IEPS unless there is a very strong compelling reason not to.  the effort
in SG16 involves the introduction of a field that has room for expansion.
the Resource Header for SIP by James and Henning also allows for
expansion, albeit it in a different form.


> 4.1.2 Diff-Serv
> 
> Since it is unlikely that a new class with 5 drop precedence values would be 
> approved, the alternative is to add 2 values to one of the current classes, 
> resulting in 5 ascending levels. 

your first contention is possible.  but i don't have much confidence that
your latter proposal would ever come to pass because it involves changes 
in specifying the existing classes of what AF, not to mention the domino
effect with existing implementations.  but i'll defer to Fred on his
view about the possibility of redefining existing classes of AF.

> The problem with either of these plans 
> concerns the potential use of the existing 4 classes. As depicted in 
> draft-baker-ieprep-requirements, an example of their use could be the 
> following:
>
> EF for voice, with signaling of bandwidth
> AF4 for video, with signaling of bandwidth
> AF3 for voice/video signaling
> AF2 for transaction/ERP traffic
> AF1 for traffic that will potentially move large volumes
>
> In this example, there would need to be the 5 drop precedence levels for EF, 
> as well as for each AF class, since eventually there should be a need to 
> apply precedence type services to other traffic types besides voice. 
> Clearly, there are not enough available DSCP code points.
>
> The description at the end of the third paragraph relates the 5 drop 
> precedence levels of a possible new class with the 3 levels of the existing 
> ones. 

ok, it seems you're mixing together two different and separate
"candidate recomendations" put forth in the draft.  i used the word
"candidate" as way of indicating ideas to be considered beyond what we
have now.  i didn't mean that they should be combined.  i'll add some
wording to make this more clear.

note, for those that haven't read this section, I also state that one
could use the existing mix of AF & EF code points to accomplish our
goals in the near term.  however, i think down the line there may be a
need to go beyond what we currently have (and the reasons for this are
in the draft, so as not to repeat myself here) -- hence, the candidate
recommendations.


> There is a confusion, first of all, in the numbers. RFC 2597 defines 
> 1=low drop precedence and 3=high. That means the emergency traffic would use 
> the low numbers (1 or 2) to get a lower probability of packet loss. 
> Secondly, 
> the values assigned within one class can't be compared with those of another 
> class. They are completely disjoint sets of values, since each class 
> represents difference queues or traffic types. (As RFC 2597 states: "Packets 
> in one AF class MUST be forwarded independently from packets in another AF 
> class, i.e., a DS node MUST NOT aggregate two or more AF classes together.")
>
> If a new class were defined with 5 levels, then, for other non-emergency 
> traffic, "the lowest drop precedence they are assigned to is 3" and 
> "Emergency traffic would be set to 2 or 1".

your statement above is correct: a different class is disjoint.  if a
single new class for AF were defined, one could potentially have the
following:

  AFx                               Ken's New Class
  ---                               ---------------
  High Drop Precedence = 3      ->  3 = High Drop Precedence
  Mid   "      "       = 2      ->  2 = Mid   "        " 
  Low   "      "       = 1      ->  1 = Low   "        "
                                ->  4 = Lower  Drop Precedence
                                ->  5 = Lowest Drop Precedence

there would be three levels in the new class; each correlating to the
same numeric value order as the current set of classes defined in AF.
there doesn't need to be a continuing ascending order of values for
adding types of precedence.

again, its just a candidate proposal and not to be treated as the only
one to consider.

> As described in draft-pierce-assured-service, the preferential treatment for 
> Assured Service traffic needs to be provided by a combination of call setup 
> priority and packet forwarding priority. Since the most important capability 
> is provided by the call setup priority, the presently defined 3 drop 
> precedence levels per class may be sufficient for packet forwarding 
> preferential treatment, as shown in Annex A of that draft.

agreed.  the Framework already draft states that for the time being we
can use the existing DSCP values.  i would add that one shouldn't get
too attached to always relying on a combination of EF and AF in cases
where diff-serv is supported because there may be providers that only
offer AF and Best Effort, and nothing else.  my hunch is that AT&T, who
recently announced support for diff-serv, only supports AF.

thanks for the feedback,

-ken




_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 12 11:09:32 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25775
	for <ieprep-archive@odin.ietf.org>; Tue, 12 Mar 2002 11:09:32 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA21434;
	Tue, 12 Mar 2002 11:09:10 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA21354
	for <ieprep@optimus.ietf.org>; Tue, 12 Mar 2002 11:09:06 -0500 (EST)
Received: from imo-m01.mx.aol.com (imo-m01.mx.aol.com [64.12.136.4])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25770
	for <ieprep@ietf.org>; Tue, 12 Mar 2002 11:09:02 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m01.mx.aol.com (mail_out_v32.5.) id i.7e.241d5b47 (4397);
	Tue, 12 Mar 2002 11:08:24 -0500 (EST)
Message-ID: <7e.241d5b47.29bf81f8@aol.com>
Date: Tue, 12 Mar 2002 11:08:24 EST
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
To: K.Carlberg@cs.ucl.ac.uk
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_7e.241d5b47.29bf81f8_boundary"
X-Mailer: AOL 6.0 for Windows US sub 353
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--part1_7e.241d5b47.29bf81f8_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 3/12/02 7:40:54 AM Eastern Standard Time, 
K.Carlberg@cs.ucl.ac.uk writes:


> hi,
> 
> > be advisable to add a new "namespace" for IEPS to that draft for the 
> values: 
> > "authorized-emergency", "general-emergency", "high", and "normal". If not 
> > included in the draft, they would be added later through IANA.
> 
> the current Framework draft only recomends a single value be defined for
> a new NameSpace to be associated with IEPS, and I prefer not to go
> beyond this single value.  one reason is that this single value would
> correlate nicely to a one-to-one association with the existing single
> NS/EP code point in the ISUP/IAM message used for GETS in the U.S. PSTN.
> 
I thought your draft was proposing two values: "authorized-emergency" and 
"general-emergency".

> he other reason is that the absence of any other value implies that the
> default mode is an INVITE without the Resource Header value -- which is
> what we have today (and which reflects the installed base of existing
> implementations).  i'll add wording to this effect in the next version
> 

But the absence of a value has to default to something. Text is being added 
to the Resource-priority draft to indicate that one of the defined values 
needs to be specified as being the default. So, in addition to the defined 
levels, there needs to be a "Normal" level defined, even if it is never 
signaled.


>>  At this point, it would be good to consider whether the possibility should 
> > be included for multiple levels of "authorized". (Note that the parallel 
> > work in SG16 on H.460.4 is leaning towards including such a capability.)
> 
> given the reasons above, i'd rather stay away from multiple levels for
> IEPS unless there is a very strong compelling reason not to.  the effort
> in SG16 involves the introduction of a field that has room for expansion.
> the Resource Header for SIP by James and Henning also allows for
> 

The effort in SG16 seems to be allowing for expansion, but it isn't clear to 
me yet from their drafts how such expansion would occur.
Right now, the Resource-header draft is not allowing expansion within a 
defined namespace.
I thought previous discussions on this subject identified the need to 
identify various users for possible multiple levels. The MLPP model for the 
similar service uses 5 levels: Routine, and 4 "emergency" levels.


>>  4.1.2 Diff-Serv
> > 
> > Since it is unlikely that a new class with 5 drop precedence values would 
> be 
> > approved, the alternative is to add 2 values to one of the current 
> classes, 
> > resulting in 5 ascending levels. 
> 
> your first contention is possible.  but i don't have much confidence that
> your latter proposal would ever come to pass because it involves changes 
> in specifying the existing classes of what AF, not to mention the domino
> effect with existing implementations.  but i'll defer to Fred on his
> 


Yes, I suspect what you said is true - changes to existing AF definitions, 
even if simply two new drop precedence levels for one of the classes ("very 
low" and "very high"?) might be very difficult. But my comment (as in 
draft-pierce-sipping-assured-service) is that 5 levels is probably unecessary.


> There is a confusion, first of all, in the numbers. RFC 2597 defines 
> 
> > 1=low drop precedence and 3=high. That means the emergency traffic would 
> use 
> > the low numbers (1 or 2) to get a lower probability of packet loss. 
> > Secondly, 
> > the values assigned within one class can't be compared with those of 
> another 
> > class. They are completely disjoint sets of values, since each class 
> > represents difference queues or traffic types. (As RFC 2597 states: 
> "Packets 
> > in one AF class MUST be forwarded independently from packets in another 
> AF 
> > class, i.e., a DS node MUST NOT aggregate two or more AF classes 
> together.")
> >
> > If a new class were defined with 5 levels, then, for other non-emergency 
> > traffic, "the lowest drop precedence they are assigned to is 3" and 
> > "Emergency traffic would be set to 2 or 1".
> 
> your statement above is correct: a different class is disjoint.  if a
> single new class for AF were defined, one could potentially have the
> following:
> 
>   AFx                               Ken's New Class
>   ---                               ---------------
>   High Drop Precedence = 3      ->  3 = High Drop Precedence
>   Mid   "      "       = 2      ->  2 = Mid   "        " 
>   Low   "      "       = 1      ->  1 = Low   "        "
>                                 ->  4 = Lower  Drop Precedence
>                                 ->  5 = Lowest Drop Precedence
> 
> there would be three levels in the new class; each correlating to the
> same numeric value order as the current set of classes defined in AF.
> there doesn't need to be a continuing ascending order of values for
> 

Yes, that's possible, I just dodn't think that a non-consecutive numering 
would be a good idea. Hard for my brain to deal with, anyway. How about 4 = 
higher and 5 = highest.
> 
> Mike Pierce



--part1_7e.241d5b47.29bf81f8_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 3/12/02 7:40:54 AM Eastern Standard Time, K.Carlberg@cs.ucl.ac.uk writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">hi,
<BR>
<BR>&gt; be advisable to add a new "namespace" for IEPS to that draft for the values: 
<BR>&gt; "authorized-emergency", "general-emergency", "high", and "normal". If not 
<BR>&gt; included in the draft, they would be added later through IANA.
<BR>
<BR>the current Framework draft only recomends a single value be defined for
<BR>a new NameSpace to be associated with IEPS, and I prefer not to go
<BR>beyond this single value. &nbsp;one reason is that this single value would
<BR>correlate nicely to a one-to-one association with the existing single
<BR>NS/EP code point in the ISUP/IAM message used for GETS in the U.S. PSTN.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">I thought your draft was proposing two values: "authorized-emergency" and "general-emergency".
<BR>
<BR>t<BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">he other reason is that the absence of any other value implies that the
<BR>default mode is an INVITE without the Resource Header value -- which is
<BR>what we have today (and which reflects the installed base of existing
<BR>implementations). &nbsp;i'll add wording to this effect in the next version
<BR>of the draft to clarify this position.</BLOCKQUOTE>
<BR>
<BR>But the absence of a value has to default to something. Text is being added to the Resource-priority draft to indicate that one of the defined values needs to be specified as being the default. So, in addition to the defined levels, there needs to be a "Normal" level defined, even if it is never signaled.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">&gt;<BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px"> At this point, it would be good to consider whether the possibility should 
<BR>&gt; be included for multiple levels of "authorized". (Note that the parallel 
<BR>&gt; work in SG16 on H.460.4 is leaning towards including such a capability.)
<BR>
<BR>given the reasons above, i'd rather stay away from multiple levels for
<BR>IEPS unless there is a very strong compelling reason not to. &nbsp;the effort
<BR>in SG16 involves the introduction of a field that has room for expansion.
<BR>the Resource Header for SIP by James and Henning also allows for
<BR>expansion, albeit it in a different form.</BLOCKQUOTE>
<BR>
<BR>The effort in SG16 seems to be allowing for expansion, but it isn't clear to me yet from their drafts how such expansion would occur.
<BR>Right now, the Resource-header draft is not allowing expansion within a defined namespace.
<BR>I thought previous discussions on this subject identified the need to identify various users for possible multiple levels. The MLPP model for the similar service uses 5 levels: Routine, and 4 "emergency" levels.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">&gt;<BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px"> 4.1.2 Diff-Serv
<BR>&gt; 
<BR>&gt; Since it is unlikely that a new class with 5 drop precedence values would be 
<BR>&gt; approved, the alternative is to add 2 values to one of the current classes, 
<BR>&gt; resulting in 5 ascending levels. 
<BR>
<BR>your first contention is possible. &nbsp;but i don't have much confidence that
<BR>your latter proposal would ever come to pass because it involves changes 
<BR>in specifying the existing classes of what AF, not to mention the domino
<BR>effect with existing implementations. &nbsp;but i'll defer to Fred on his
<BR>view about the possibility of redefining existing classes of AF.</BLOCKQUOTE>
<BR>
<BR>
<BR>Yes, I suspect what you said is true - changes to existing AF definitions, even if simply two new drop precedence levels for one of the classes ("very low" and "very high"?) might be very difficult. But my comment (as in draft-pierce-sipping-assured-service) is that 5 levels is probably unecessary.
<BR>
<BR>
<BR>&gt; There is a confusion, first of all, in the numbers. RFC 2597 defines </FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0"><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>&gt; 1=low drop precedence and 3=high. That means the emergency traffic would use 
<BR>&gt; the low numbers (1 or 2) to get a lower probability of packet loss. 
<BR>&gt; Secondly, 
<BR>&gt; the values assigned within one class can't be compared with those of another 
<BR>&gt; class. They are completely disjoint sets of values, since each class 
<BR>&gt; represents difference queues or traffic types. (As RFC 2597 states: "Packets 
<BR>&gt; in one AF class MUST be forwarded independently from packets in another AF 
<BR>&gt; class, i.e., a DS node MUST NOT aggregate two or more AF classes together.")
<BR>&gt;
<BR>&gt; If a new class were defined with 5 levels, then, for other non-emergency 
<BR>&gt; traffic, "the lowest drop precedence they are assigned to is 3" and 
<BR>&gt; "Emergency traffic would be set to 2 or 1".
<BR>
<BR>your statement above is correct: a different class is disjoint. &nbsp;if a
<BR>single new class for AF were defined, one could potentially have the
<BR>following:
<BR>
<BR> &nbsp;AFx &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Ken's New Class
<BR> &nbsp;--- &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;---------------
<BR> &nbsp;High Drop Precedence = 3 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;-&gt; &nbsp;3 = High Drop Precedence
<BR> &nbsp;Mid &nbsp;&nbsp;" &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;" &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;= 2 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;-&gt; &nbsp;2 = Mid &nbsp;&nbsp;" &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;" 
<BR> &nbsp;Low &nbsp;&nbsp;" &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;" &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;= 1 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;-&gt; &nbsp;1 = Low &nbsp;&nbsp;" &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;"
<BR> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;-&gt; &nbsp;4 = Lower &nbsp;Drop Precedence
<BR> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;-&gt; &nbsp;5 = Lowest Drop Precedence
<BR>
<BR>there would be three levels in the new class; each correlating to the
<BR>same numeric value order as the current set of classes defined in AF.
<BR>there doesn't need to be a continuing ascending order of values for
<BR>adding types of precedence.</BLOCKQUOTE>
<BR>
<BR>Yes, that's possible, I just dodn't think that a non-consecutive numering would be a good idea. Hard for my brain to deal with, anyway. How about 4 = higher and 5 = highest.
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR></BLOCKQUOTE>Mike Pierce
<BR>
<BR></FONT></HTML>

--part1_7e.241d5b47.29bf81f8_boundary--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 12 11:09:35 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25787
	for <ieprep-archive@odin.ietf.org>; Tue, 12 Mar 2002 11:09:35 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA21399;
	Tue, 12 Mar 2002 11:09:08 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA21337
	for <ieprep@optimus.ietf.org>; Tue, 12 Mar 2002 11:09:05 -0500 (EST)
Received: from imo-r10.mx.aol.com (imo-r10.mx.aol.com [152.163.225.106])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25768
	for <ieprep@ietf.org>; Tue, 12 Mar 2002 11:09:01 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r10.mx.aol.com (mail_out_v32.5.) id c.151.a4b3478 (4397);
	Tue, 12 Mar 2002 11:08:20 -0500 (EST)
Message-ID: <151.a4b3478.29bf81f3@aol.com>
Date: Tue, 12 Mar 2002 11:08:19 EST
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
To: fred@cisco.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_151.a4b3478.29bf81f3_boundary"
X-Mailer: AOL 6.0 for Windows US sub 353
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--part1_151.a4b3478.29bf81f3_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 3/11/02 7:51:13 PM Eastern Standard Time, fred@cisco.com 
writes:


> At 09:33 AM 3/11/2002, Mpierce1@aol.com wrote:
> >At this point, it would be good to consider whether the possibility should 
> >be included for multiple levels of "authorized". (Note that the parallel 
> >work in SG16 on H.460.4 is leaning towards including such a capability.)
> 
> are these "levels of authorization" or "kinds of authorization"? My 
> experience with levels is that the discussion is always about winners and 
> losers: "my stuff is more important than his stuff, so I think there needs 
> 

The idea is that there could be many different agencies/persons/functions 
that need emergency access - FEMA, police, medical, Red Cross, network 
administrators, etc. It's hard to image that all would be treated the same. 
If there is only one possible level to apply to all, it would be very 
difficult to decide whether or not a particular group would get the 
authorization, e.g., is the Red Cross one that needs this? Providing a way to 
distinguish would help. Whoever administers the authorization to use 
emergency levels in a network would decide who gets which level. Within 
"emergency" there needs to be 3 or 4 levels. (Remember, In MLPP, or "Assured 
Service", one could think of the 5 levels of precedence as consisting of a 
"Routine" level, and then four levels of "emergency".)

> 
> What makes sense to me in SIP is a simple statement: "I assert that this 
> call is an emergency call". If you are not authorized to make that 
> assertion, the correct policy is to ignore the statement. If you are 
> authorized, the network will have a policy to apply such as "permit this 
> call even if we are well into the overload zone" or "drop a random call in 
> order to complete this one." The network should determine the policy, not 
> 

Not sure of your point. The standard should allow a way to designate 
sufficient levels of importance, in order for a network to apply whatever its 
policy is. I can certainly envision that a desired network policy is to 
control access based on the relative level of the emergency call and the 
authorization of the person making the attempt. The MLPP model is that each 
user is authorized to use up to a certain level, and may signal any level up 
to that one in a call setup. The question is: how many distinct levels are 
needed?
> 
> >4.1.2 Diff-Serv
> >
> >[snip] In this example, there would need to be the 5 drop precedence 
> >levels for EF, as well as for each AF class, since eventually there should 
> >be a need to apply precedence type services to other traffic types besides 
> >voice. Clearly, there are not enough available DSCP code points.
> 
> I think you have a very different understanding of diffserv than I do. When 
> you talk about a drop precedence for EF (which doesn't have a drop 
> precedence), what are you talking about - dropping packets, or dropping 
> 

The current draft shows "EF for voice, with signalling of bandwidth". 
Therefore it whould need to have multiple levels of drop precedence. This 
refers to dropping individual packets, since that is where the marking is. 
Actually, I think that EF only for the signaling and one of the AF classes 
for the voice packets would make more sense, as I suggested in Annex A of 
draft-pierce-sipping-assured-service.

> 
> Recall that in GETS, you don't get differing qualities of circuits. You get 
> differing probabilities of getting a circuit, but once you have a circuit, 
> it is just a circuit and works exactly like any other circuit. In IP 
> telephony, you might have varying call preferences, but once you decide to 
> carry voice at all, EF is the service that carries voice. If you're trying 
> to model GETS in IP Telephony, dropping packets makes no sense. Dropping 
> 

If GETS resulted in different routing, it might mean different circuit 
quality (better or worse). Yes, it was "just a circuit", so there were 
limited things that could be done to play with quality. IP provides other 
options.

Yes,dropping calls might make sense, but I recognize the dislike for 
"preemption" and understand that it isn't allowed some places. The point I'm 
making is that some resources can be stolen for the more important call by 
causing the less important call to drop more packets, thereby stealing some 
of its QoS. The result, when the QoS of the less important call gets 
intolerably bad, is for the parties to give up and disconnect, freeing up 
more resources. Meanwhile, the establishment of new calls is being controlled 
based on precedence (importance), so their reattempt is blocked.
> 
> 

Diff-serv AF provides the drop precedence for what I described above, without 
any change. Is there anything else that could do this?

> 
> >As described in draft-pierce-assured-service...
> 
> 

Certainly. My fast typing often gets ahead of by slow brain.

> 
> BTW, in view of RFC 2597 "Assured Forwarding PHB Group", which is often 
> called "Assured Service", I'd personally appreciate it if you could find 
> another name. My head snaps every time I hear the name :^)
> 
> I'm certainly sensitive to using a term for two things. Is the term 
> "Assured Service" used in specs, or just in everyday speech? We'll consider 
> other names.

Mike


--part1_151.a4b3478.29bf81f3_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 3/11/02 7:51:13 PM Eastern Standard Time, fred@cisco.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">At 09:33 AM 3/11/2002, Mpierce1@aol.com wrote:
<BR>&gt;At this point, it would be good to consider whether the possibility should 
<BR>&gt;be included for multiple levels of "authorized". (Note that the parallel 
<BR>&gt;work in SG16 on H.460.4 is leaning towards including such a capability.)
<BR>
<BR>are these "levels of authorization" or "kinds of authorization"? My 
<BR>experience with levels is that the discussion is always about winners and 
<BR>losers: "my stuff is more important than his stuff, so I think there needs 
<BR>to be another level."</BLOCKQUOTE>
<BR>
<BR>The idea is that there could be many different agencies/persons/functions that need emergency access - FEMA, police, medical, Red Cross, network administrators, etc. It's hard to image that all would be treated the same. If there is only one possible level to apply to all, it would be very difficult to decide whether or not a particular group would get the authorization, e.g., is the Red Cross one that needs this? Providing a way to distinguish would help. Whoever administers the authorization to use emergency levels in a network would decide who gets which level. Within "emergency" there needs to be 3 or 4 levels. (Remember, In MLPP, or "Assured Service", one could think of the 5 levels of precedence as consisting of a "Routine" level, and then four levels of "emergency".)
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>What makes sense to me in SIP is a simple statement: "I assert that this 
<BR>call is an emergency call". If you are not authorized to make that 
<BR>assertion, the correct policy is to ignore the statement. If you are 
<BR>authorized, the network will have a policy to apply such as "permit this 
<BR>call even if we are well into the overload zone" or "drop a random call in 
<BR>order to complete this one." The network should determine the policy, not 
<BR>the standard.</BLOCKQUOTE>
<BR>
<BR>Not sure of your point. The standard should allow a way to designate sufficient levels of importance, in order for a network to apply whatever its policy is. I can certainly envision that a desired network policy is to control access based on the relative level of the emergency call and the authorization of the person making the attempt. The MLPP model is that each user is authorized to use up to a certain level, and may signal any level up to that one in a call setup. The question is: how many distinct levels are needed?
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>&gt;4.1.2 Diff-Serv
<BR>&gt;
<BR>&gt;[snip] In this example, there would need to be the 5 drop precedence 
<BR>&gt;levels for EF, as well as for each AF class, since eventually there should 
<BR>&gt;be a need to apply precedence type services to other traffic types besides 
<BR>&gt;voice. Clearly, there are not enough available DSCP code points.
<BR>
<BR>I think you have a very different understanding of diffserv than I do. When 
<BR>you talk about a drop precedence for EF (which doesn't have a drop 
<BR>precedence), what are you talking about - dropping packets, or dropping calls?</BLOCKQUOTE>
<BR>
<BR>The current draft shows "EF for voice, with signalling of bandwidth". Therefore it whould need to have multiple levels of drop precedence. This refers to dropping individual packets, since that is where the marking is. Actually, I think that EF only for the signaling and one of the AF classes for the voice packets would make more sense, as I suggested in Annex A of draft-pierce-sipping-assured-service.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>Recall that in GETS, you don't get differing qualities of circuits. You get 
<BR>differing probabilities of getting a circuit, but once you have a circuit, 
<BR>it is just a circuit and works exactly like any other circuit. In IP 
<BR>telephony, you might have varying call preferences, but once you decide to 
<BR>carry voice at all, EF is the service that carries voice. If you're trying 
<BR>to model GETS in IP Telephony, dropping packets makes no sense. Dropping 
<BR>calls might make sense.</BLOCKQUOTE>
<BR>
<BR>If GETS resulted in different routing, it might mean different circuit quality (better or worse). Yes, it was "just a circuit", so there were limited things that could be done to play with quality. IP provides other options.
<BR>
<BR>Yes,dropping calls might make sense, but I recognize the dislike for "preemption" and understand that it isn't allowed some places. The point I'm making is that some resources can be stolen for the more important call by causing the less important call to drop more packets, thereby stealing some of its QoS. The result, when the QoS of the less important call gets intolerably bad, is for the parties to give up and disconnect, freeing up more resources. Meanwhile, the establishment of new calls is being controlled based on precedence (importance), so their reattempt is blocked.
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>So it's not in the least obvious where diff-serv fits here. Please advise?</BLOCKQUOTE>
<BR>
<BR>Diff-serv AF provides the drop precedence for what I described above, without any change. Is there anything else that could do this?
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>&gt;As described in draft-pierce-assured-service...
<BR>
<BR>is that draft-pierce-sipping-assured-service-01.txt?</BLOCKQUOTE>
<BR>
<BR>Certainly. My fast typing often gets ahead of by slow brain.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>BTW, in view of RFC 2597 "Assured Forwarding PHB Group", which is often 
<BR>called "Assured Service", I'd personally appreciate it if you could find 
<BR>another name. My head snaps every time I hear the name :^)
<BR>
<BR>I'm certainly sensitive to using a term for two things. Is the term "Assured Service" used in specs, or just in everyday speech? We'll consider other names.</FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>Mike
<BR></FONT></HTML>

--part1_151.a4b3478.29bf81f3_boundary--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 12 11:59:14 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA27415
	for <ieprep-archive@odin.ietf.org>; Tue, 12 Mar 2002 11:59:13 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA24406;
	Tue, 12 Mar 2002 11:58:54 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA24369
	for <ieprep@optimus.ietf.org>; Tue, 12 Mar 2002 11:58:52 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id LAA27368
	for <ieprep@ietf.org>; Tue, 12 Mar 2002 11:58:46 -0500 (EST)
Received: from posh.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.14990-0@bells.cs.ucl.ac.uk>; Tue, 12 Mar 2002 16:58:39 +0000
X-Mailer: exmh version 2.5 07/13/2001 with version: MH 6.8.3 #25[UCI]
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-reply-to: Your message of "Tue, 12 Mar 2002 11:08:24 EST." <7e.241d5b47.29bf81f8@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Tue, 12 Mar 2002 16:58:37 +0000
Message-ID: <1242.1015952317@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


> I thought your draft was proposing two values: "authorized-emergency" 
> and "general-emergency". 

yes, originally i did.

> But the absence of a value has to default to something.

correct.  and it seems that it will be helpful to add text about this
in the Framework draft.

> Text is being
> added to the Resource-priority draft to indicate that one of the defined 
> values needs to be specified as being the default. So, in addition to
> the defined levels, there needs to be a "Normal" level defined, even if
> it is never signaled. 

my assumption was that the text you speak of was in relation to the 
example DSN NameSpace.  Otherwise, the Resource Header draft is 
implicitly requiring that any defined namespace must have at least 2 
values.  i'd prefer not to see that constraint in there.

> The effort in SG16 seems to be allowing for expansion, but it isn't clear
> to me yet from their drafts how such expansion would occur. 

a fair assessment.  i don't know how ITU documents are structured.
perhaps one needs either an appendix that defines the format of the
values stored in the 'emergency' related field, or a subsequent document
needs to be written with the explanation.  i'll ask the fellows putting
it together.

> Right now, the Resource-header draft is not allowing expansion within a 
> defined namespace. 

this is probably a question to James and Henning, but couldn't one
define a new NameSpace?  like MLPPv2, or extended-MLPP?

> I thought previous discussions on this subject identified the need to
> identify various users for possible multiple levels. The MLPP model for
> the similar service uses 5 levels: Routine, and 4 "emergency" levels.

the 5 levels that have been discussed are strictly for MLPP.  I may have
introduced confusion by trying to have IEPS be symbiotic with MLPP.  so
I may go back and add some clarity in that regard to the Framework draft.

> Yes, I suspect what you said is true - changes to existing AF definitions,
> even if simply two new drop precedence levels for one of the classes 
> ("very low" and "very high"?) might be very difficult. But my comment 
> (as in draft-pierce-sipping-assured-service) is that 5 levels is probably 
> unecessary. 

ok, i think i understand.  it could be the existing 3 + 1 additional one
for IEPS.  but in the end, its mostly a trial balloon of an idea.

> Yes, that's possible, I just dodn't think that a non-consecutive numering
> would be a good idea. Hard for my brain to deal with, anyway. How about 
> 4 = higher and 5 = highest. 

something to consider.

cheers,

-ken



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 12 14:55:44 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA04197
	for <ieprep-archive@odin.ietf.org>; Tue, 12 Mar 2002 14:55:44 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA05558;
	Tue, 12 Mar 2002 14:54:43 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA05526
	for <ieprep@optimus.ietf.org>; Tue, 12 Mar 2002 14:54:41 -0500 (EST)
Received: from imo-d06.mx.aol.com (imo-d06.mx.aol.com [205.188.157.38])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA04151
	for <ieprep@ietf.org>; Tue, 12 Mar 2002 14:54:37 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d06.mx.aol.com (mail_out_v32.5.) id i.28.23639eb1 (30970);
	Tue, 12 Mar 2002 14:53:55 -0500 (EST)
Message-ID: <28.23639eb1.29bfb6d3@aol.com>
Date: Tue, 12 Mar 2002 14:53:55 EST
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
To: K.Carlberg@cs.ucl.ac.uk
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_28.23639eb1.29bfb6d3_boundary"
X-Mailer: AOL 6.0 for Windows US sub 353
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--part1_28.23639eb1.29bfb6d3_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 3/12/02 11:58:59 AM Eastern Standard Time, 
K.Carlberg@cs.ucl.ac.uk writes:


> 
> > I thought your draft was proposing two values: "authorized-emergency" 
> > and "general-emergency". 
> 
> 

Is there a new draft after version -00 from Feb 15? It proposes two values, 
doesn't it?

> 
> > But the absence of a value has to default to something.
> 
> correct.  and it seems that it will be helpful to add text about this
> in the Framework draft.
> 
> > Text is being
> > added to the Resource-priority draft to indicate that one of the defined 
> > values needs to be specified as being the default. So, in addition to
> > the defined levels, there needs to be a "Normal" level defined, even if
> > it is never signaled. 
> 
> my assumption was that the text you speak of was in relation to the 
> example DSN NameSpace.  Otherwise, the Resource Header draft is 
> implicitly requiring that any defined namespace must have at least 2 
> values.  

In a response from Henning on Mar 7 to my following comment he wrote:

> [MAP] No, I don't think it is academic. It is worth considering. The result 
is that
> an added item would be needed in resource-priority to designate which level
> is the default for each namespace and to include the action it says above. I
> don't know that this would be a "well-defined" value. It would be different
> in each domain (namespace). For the dsn namespace, it would be "Routine"; 
for
> the q735 namespace, it would be "4".

[Henning] I've added a sentence in each about the default value.

> 
> > Right now, the Resource-header draft is not allowing expansion within a 
> > defined namespace. 
> 
> this is probably a question to James and Henning, but couldn't one
> define a new NameSpace?  

Yes, but the problem with allowing addition of values to an existing 
namespace is that current implementations wouldn't receognize the new values. 
Same problem with a new namespace, except worse.

MIke


--part1_28.23639eb1.29bfb6d3_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 3/12/02 11:58:59 AM Eastern Standard Time, K.Carlberg@cs.ucl.ac.uk writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>&gt; I thought your draft was proposing two values: "authorized-emergency" 
<BR>&gt; and "general-emergency". 
<BR>
<BR>yes, originally i did.</BLOCKQUOTE>
<BR>
<BR>Is there a new draft after version -00 from Feb 15? It proposes two values, doesn't it?
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>&gt; But the absence of a value has to default to something.
<BR>
<BR>correct. &nbsp;and it seems that it will be helpful to add text about this
<BR>in the Framework draft.
<BR>
<BR>&gt; Text is being
<BR>&gt; added to the Resource-priority draft to indicate that one of the defined 
<BR>&gt; values needs to be specified as being the default. So, in addition to
<BR>&gt; the defined levels, there needs to be a "Normal" level defined, even if
<BR>&gt; it is never signaled. 
<BR>
<BR>my assumption was that the text you speak of was in relation to the 
<BR>example DSN NameSpace. &nbsp;Otherwise, the Resource Header draft is 
<BR>implicitly requiring that any defined namespace must have at least 2 
<BR>values. &nbsp;i'd prefer not to see that constraint in there.</BLOCKQUOTE>
<BR>
<BR>In a response from Henning on Mar 7 to my following comment he wrote:
<BR>
<BR>&gt; [MAP] No, I don't think it is academic. It is worth considering. The result is that
<BR>&gt; an added item would be needed in resource-priority to designate which level
<BR>&gt; is the default for each namespace and to include the action it says above. I
<BR>&gt; don't know that this would be a "well-defined" value. It would be different
<BR>&gt; in each domain (namespace). For the dsn namespace, it would be "Routine"; for
<BR>&gt; the q735 namespace, it would be "4".
<BR>
<BR>[Henning] I've added a sentence in each about the default value.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>&gt; Right now, the Resource-header draft is not allowing expansion within a 
<BR>&gt; defined namespace. 
<BR>
<BR>this is probably a question to James and Henning, but couldn't one
<BR>define a new NameSpace? &nbsp;like MLPPv2, or extended-MLPP?</BLOCKQUOTE>
<BR>
<BR>Yes, but the problem with allowing addition of values to an existing namespace is that current implementations wouldn't receognize the new values. Same problem with a new namespace, except worse.
<BR>
<BR>MIke
<BR></FONT></HTML>

--part1_28.23639eb1.29bfb6d3_boundary--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 12 15:02:02 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04548
	for <ieprep-archive@odin.ietf.org>; Tue, 12 Mar 2002 15:02:02 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA06679;
	Tue, 12 Mar 2002 15:01:37 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA06649
	for <ieprep@optimus.ietf.org>; Tue, 12 Mar 2002 15:01:35 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04501;
	Tue, 12 Mar 2002 15:01:30 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2CK0vh15592;
	Tue, 12 Mar 2002 12:00:57 -0800 (PST)
Message-Id: <4.3.2.7.2.20020312110359.032806c8@mira-sjcm-2.cisco.com>
X-Sender: fred@mira-sjcm-2.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Tue, 12 Mar 2002 11:04:18 -0800
To: "Folts, Harold" <foltsh@ncs.gov>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] 53rd IETF - Internet Emergency Preparedness
Cc: "'agenda@ietf.org '" <agenda@ietf.org>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>,
        "'Scott_Bradner '" <sob@harvard.edu>
In-Reply-To: <AFBFF033E351D511999900204804EE54031758@rbmail104.chamb.dis
 a.mil>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 12:59 AM 3/12/2002, Folts, Harold wrote:
>Anyway, will this ID be the basis for
>the Information "requirements for ieprep"?

we need to discuss that. I don't know that it will or it won't.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 12 15:50:33 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA06628
	for <ieprep-archive@odin.ietf.org>; Tue, 12 Mar 2002 15:50:33 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA10997;
	Tue, 12 Mar 2002 15:50:12 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA10957
	for <ieprep@optimus.ietf.org>; Tue, 12 Mar 2002 15:50:09 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA06574
	for <ieprep@ietf.org>; Tue, 12 Mar 2002 15:50:06 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2CKnZh22953;
	Tue, 12 Mar 2002 12:49:35 -0800 (PST)
Message-Id: <4.3.2.7.2.20020312110543.0514e2f0@mira-sjcm-2.cisco.com>
X-Sender: fred@mira-sjcm-2.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Tue, 12 Mar 2002 12:49:21 -0800
To: <Mpierce1@aol.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
Cc: ieprep@ietf.org
In-Reply-To: <151.a4b3478.29bf81f3@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 08:08 AM 3/12/2002, Mpierce1@aol.com wrote:
>The idea is that there could be many different agencies/persons/functions 
>that need emergency access - FEMA, police, medical, Red Cross, network 
>administrators, etc. It's hard to image that all would be treated the 
>same. If there is only one possible level to apply to all, it would be 
>very difficult to decide whether or not a particular group would get the 
>authorization, e.g., is the Red Cross one that needs this? Providing a way 
>to distinguish would help. Whoever administers the authorization to use 
>emergency levels in a network would decide who gets which level. Within 
>"emergency" there needs to be 3 or 4 levels. (Remember, In MLPP, or 
>"Assured Service", one could think of the 5 levels of precedence as 
>consisting of a "Routine" level, and then four levels of "emergency".)

Perhaps, but you're mixing apples and orangutans. Providing a way to 
distinguish calls is probably a good thing. Using the standard that defines 
the label with which you identify traffic to define the policy to apply to 
those various distinguished kinds of calls dramatically reduces your 
flexibility, and brings you into a global space you are not thinking about. 
Being able to say "This is a FEMA Code 27 call" (distinguishing calls) is 
quite a bit different than saying "This is a FEMA Code 27 call, all 
telephone systems in the world including those deployed in Madagascar need 
to be able to understand what that means, and by the way that means that my 
call should be placed ahead of a call for the Red Cross but behind a call 
which might launch missiles" is quite another.

By all means, put identifying marks on calls. Leave the policy to be 
applied to those calls ("all telephone systems in the world including those 
deployed in Madagascar need to be able to understand what that means, and 
by the way that means that my call should be placed ahead of a call for the 
Red Cross but behind a call which might launch missiles") to the network.

It's not obvious to me, by the way, that '5 levels of precedence as 
consisting of a "Routine" level, and then four levels of "emergency"' is 
wise, or solves anything. Take a recent case in point: on September 11, the 
calls directing emergency traffic to the lower Manhattan or around the 
Pentagon were not "authorized" folks, they were Joe Citizen who was in no 
way special but happened to be on the scene. Cutting Joe Citizen out of the 
loop, while enabling "authorized" access to folks whose first instinct 
would have been to call home and say they were all right, would have been 
at least as disruptive to emergency services deployment as the fact that 
Joe Citizen also wanted to call home. Precedence makes sense when you have 
the power to choose; as I understand the GETS network, it doesn't choose 
whether or not to place a call as it does choose whether to let a person 
use specific network resources to accomplish that.

>>What makes sense to me in SIP is a simple statement: "I assert that this 
>>call is an emergency call". If you are not authorized to make that 
>>assertion, the correct policy is to ignore the statement. If you are 
>>authorized, the network will have a policy to apply such as "permit this 
>>call even if we are well into the overload zone" or "drop a random call 
>>in order to complete this one." The network should determine the policy, 
>>not the standard.
>
>Not sure of your point. The standard should allow a way to designate 
>sufficient levels of importance, in order for a network to apply whatever 
>its policy is.

And who decides what is important and what is not? I think you'll find that 
someone standing on the corner watching people bleed will feel their calls 
are *very* important, while someone with a different perspective might 
decide that places where lots of people are bleeding are more important 
than places where only a few are. My point is that that the policy must 
describe how to make that decision - the network policy must decide whether 
one thing is more important than another. As such, it cannot be placed in 
the hands of the actor who has only a small part of the picture. It must be 
done by the network.

Therefore, you want the person making the call to identify the issue in 
some way, and let the network apply a pre-determined policy to decide what 
resources to apply to it.

>The question is: how many distinct levels are needed?

You have yet to demonstrate that the concept of "levels" is decidable on a 
global basis. Is a FEMA call more important than its counterpart in 
Madagascar? Is that a nonsense question? If it is a nonsense question, why 
are we discussing it in the context of a standard for products that will be 
used without change in both the US and Madagascar?

>>>4.1.2 Diff-Serv
>>>
>>>[snip] In this example, there would need to be the 5 drop precedence 
>>>levels for EF, as well as for each AF class, since eventually there 
>>>should be a need to apply precedence type services to other traffic 
>>>types besides voice. Clearly, there are not enough available DSCP code points.
>>
>>I think you have a very different understanding of diffserv than I do. 
>>When you talk about a drop precedence for EF (which doesn't have a drop 
>>precedence), what are you talking about - dropping packets, or dropping calls?
>
>The current draft shows "EF for voice, with signalling of bandwidth".

There are 51 posted drafts that contain the words "voice", "signalling", 
and "bandwidth". I have just reviewed all of them and have not found those 
words. Please advise - which draft are you thinking of?

>Therefore it whould need to have multiple levels of drop precedence. This 
>refers to dropping individual packets, since that is where the marking is. 
>Actually, I think that EF only for the signaling and one of the AF classes 
>for the voice packets would make more sense, as I suggested in Annex A of 
>draft-pierce-sipping-assured-service.

As an author on AF, a material contributor to EF, and someone who has done 
a fair bit on trying to make voice work on IP, I am deeply confused by that 
statement. As near as I can tell, it results in a service that doesn't 
work. Are you available at the IETF to discuss this?

>Yes,dropping calls might make sense, but I recognize the dislike for 
>"preemption" and understand that it isn't allowed some places. The point 
>I'm making is that some resources can be stolen for the more important 
>call by causing the less important call to drop more packets, thereby 
>stealing some of its QoS. The result, when the QoS of the less important 
>call gets intolerably bad, is for the parties to give up and disconnect, 
>freeing up more resources. Meanwhile, the establishment of new calls is 
>being controlled based on precedence (importance), so their reattempt is 
>blocked.

>Diff-serv AF provides the drop precedence for what I described above, 
>without any change. Is there anything else that could do this?

That's the wrong place to start. The requirements voice has are for minimal 
jitter, latency, and loss for a data stream of defined rate. EF 
accomplishes that by measuring rate, dropping excess, and giving priority 
to the rest. The requirements for TCP are at most one drop per RTT per 
session, and preferably those drops should come from data segments (as 
opposed to Acks etc) coming from TCP speakers who are sending relatively 
large amounts of traffic. AF offers a network administration the ability to 
identify those large sessions and isolate drops to them. By the way, the 
drop precedence is within AF classes, not among them. AF1 doesn't get 
dropped before or after AF2; AFx2 gets dropped before AFx1, and AFx3 before 
AFx2.

Moving VoIP to an AF service to get the drop precedence makes no sense. You 
give up all the capabilities that make voice work, in order to get that 
capability, and the capability you are seeking makes voice not work will 
for the disadvantaged users.

If what you want to define is an EF service with multiple drop precedences, 
I would strongly suggest that you define an EF service with multiple drop 
precedences. When you do, remember that in computer science there are three 
reasonable numbers: "zero", "one", and "many". I can tell you for sure what 
will happen if you prioritize some traffic over other traffic - loss, 
delay, and variation in delay (jitter) will be transferred from the 
prioritized traffic to the other traffic. I think I can tell you what will 
happen with two priorities if the priorities map respectively to voice and 
video traffic. Folks have done a zillion simulations on priority systems 
with more levels, and have not been able to demonstrate utility on a 
per-packet level.

>I'm certainly sensitive to using a term for two things. Is the term 
>"Assured Service" used in specs, or just in everyday speech? We'll 
>consider other names.

it's mostly in speech, but it is widely used.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 12 18:12:12 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA11664
	for <ieprep-archive@odin.ietf.org>; Tue, 12 Mar 2002 18:12:11 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA22116;
	Tue, 12 Mar 2002 18:11:54 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA22087
	for <ieprep@ns.ietf.org>; Tue, 12 Mar 2002 18:11:53 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA11633
	for <ieprep@ietf.org>; Tue, 12 Mar 2002 18:11:43 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <F5YT9K92>; Tue, 12 Mar 2002 18:10:28 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410445EDC6@chntex02.is.dyncorp.com>
From: "Berg, Dennis" <Dennis.Berg@DynCorp.com>
To: "'k.carlberg@cs.ucl.ac.uk'" <k.carlberg@cs.ucl.ac.uk>
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Date: Tue, 12 Mar 2002 18:11:05 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1CA1B.2FA1DFB0"
Subject: [Ieprep] Comments on GETS in draft-carlberg-ieprep-freamework
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1CA1B.2FA1DFB0
Content-Type: text/plain;
	charset="iso-8859-1"

Ken,

Some comments and suggestions on the GETS intro to your document.

Section 1.2, paragraph "the key aspect is that GETS...." The point about
probability is important, but there are actually two contrasts here:
{probabilistic, guaranteed} and {priority, preemption}.  Suggest the
following to make both clear:
*	"The key aspect of GETS is that is supports a probabilistic approach
to call completion through priority, as opposed to guaranteed approach
through preemption. This distinction is important because emergency services
like GETS are not allowed to tear down existing calls (i.e., seize
resources) in order to establish a GETS call.  Instead, GETS increases the
probability of call completion by providing priority in the contention for
assignment of limited resources required for the call.  Thus, the GETS
features focus on increasing the probability that a particular telephone
call will be established, but cannot guarantee call completion."

Section 1.2.1. 
*	"In the presence of this code point, Local Exchange Carriers (LEC)
will attempt (if necessary and if the option is supported) to route the call
through alternate inter-exchange carriers (IXC) if it cannot complete the
call through the default IXC"

Actually the feature described, Alternate Carrier Routing (ACR), is not
based upon the HPC code point. ACR is implemented using a commercial
Advanced Intelligent Network (AIN) capability that is triggered based upon
the dialed number, in this case the GETS access number.

The two GETS features that are triggered by the HPC class mark are Trunk
Queuing (TQ) and Exemption from Restrictive Network Management Controls
(Ex-NMCs).

Suggest avoiding the complication of bringing up this other type of feature
and instead replacing the text above and to the end of the paragraph with:
*	In the presence of this code point, when a GETS call encounters a
restrictive network management control that has been activated to reduce
traffic overload to a congested route, the Local Exchange Carriers (LECs)
will provide the GETS call priority by exempting the call from this
restriction.  After receiving the exemption, if the GETS call finds all
circuits busy in the route, the LEC will provide further priority by queuing
the call for the next available circuit.  

Section 2. Final paragraph. 
*	"GETS users are authenticated using a PIN provided to the
telecommunications carrier, which signals approval back to the user's local
exchange over SS7."

Currently the IXCs authenticate GETS calls by differing methods.  There is
no communication of authentication back to the LEC where the call
originates.   However, when the call egresses the IXC to a LEC for
termination to the called party, the IXC passes the HPC class mark and the
terminating LEC provides priority treatment based upon that class mark.

Recommend deleting "...which signals approval back to the user's local
exchange over SS7" or changing to ""...which signals authentication to
subsequent networks via the HPC class mark."

Dennis Berg
GETS IC Deputy Program Manager, Engineering
DynCorp IS
703-818-4608


------_=_NextPart_001_01C1CA1B.2FA1DFB0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2653.12">
<TITLE>Comments on GETS in draft-carlberg-ieprep-freamework</TITLE>
</HEAD>
<BODY>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 =
FACE=3D"Arial">Ken,</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Some =
comments and suggestions on the GETS intro to your document.</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Section =
1.2, paragraph "the key aspect is that GETS...." The point about =
probability is important, but there are actually two contrasts here: =
{probabilistic, guaranteed} and {priority, preemption}.&nbsp; Suggest =
the following to make both clear:</FONT></P>

<P><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Symbol">&#183;<FONT =
FACE=3D"Courier New">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</FONT></FONT> =
<FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">"The key aspect of GETS =
is that is supports a probabilistic approach to call completion through =
priority, as opposed to guaranteed approach through preemption. This =
distinction is important because emergency services like GETS are not =
allowed to tear down existing calls (i.e., seize resources) in order to =
establish a GETS call.&nbsp; Instead, GETS increases the probability of =
call completion by providing priority in the contention for assignment =
of limited resources required for the call.&nbsp; Thus, the GETS =
features focus on increasing the probability that a particular =
telephone call will be established, but cannot guarantee call =
completion."</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Section =
1.2.1. </FONT></P>

<P><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Symbol">&#183;<FONT =
FACE=3D"Courier New">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</FONT></FONT> =
<FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">"In the presence of =
this code point, Local Exchange Carriers (LEC) will attempt (if =
necessary and if the option is supported) to route the call through =
alternate inter-exchange carriers (IXC) if it cannot complete the call =
through the default IXC"</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 =
FACE=3D"Arial">Actually the feature described, Alternate Carrier =
Routing (ACR), is not based upon the HPC code point. ACR is implemented =
using a commercial Advanced Intelligent Network (AIN) capability that =
is triggered based upon the dialed number, in this case the GETS access =
number.</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">The two =
GETS features that are triggered by the HPC class mark are Trunk =
Queuing (TQ) and Exemption from Restrictive Network Management Controls =
(Ex-NMCs).</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Suggest =
avoiding the complication of bringing up this other type of feature and =
instead replacing the text above and to the end of the paragraph =
with:</FONT></P>

<P><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Symbol">&#183;<FONT =
FACE=3D"Courier New">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</FONT></FONT> =
<FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">In the presence of this =
code point, when a GETS call encounters a restrictive network =
management control that has been activated to reduce traffic overload =
to a congested route, the Local Exchange Carriers (LECs) will provide =
the GETS call priority by exempting the call from this =
restriction.&nbsp; After receiving the exemption, if the GETS call =
finds all circuits busy in the route, the LEC will provide further =
priority by queuing the call for the next available circuit.&nbsp; =
</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">Section =
2. Final paragraph. </FONT></P>

<P><FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Symbol">&#183;<FONT =
FACE=3D"Courier New">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</FONT></FONT> =
<FONT COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">"GETS users are =
authenticated using a PIN provided to the telecommunications carrier, =
which signals approval back to the user's local exchange over =
SS7."</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 =
FACE=3D"Arial">Currently the IXCs authenticate GETS calls by differing =
methods.&nbsp; There is no communication of authentication back to the =
LEC where the call originates.&nbsp;&nbsp;</FONT> <FONT =
COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">However, w</FONT><FONT =
COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">hen the call egresses the IXC =
to a LEC for termination to the called party, the IXC passes the HPC =
class mark and the terminating LEC provide</FONT><FONT =
COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">s</FONT><FONT =
COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial"> priority treatment based =
upon that class mark.</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" SIZE=3D2 =
FACE=3D"Arial">Recommend deleting "...which signals approval back to =
the user's local exchange over SS7</FONT><FONT COLOR=3D"#000000" =
SIZE=3D2 FACE=3D"Arial">" or changing to ""...which signals =
authentication to subsequent networks via the HPC</FONT> <FONT =
COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">class mark</FONT><FONT =
COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">.</FONT><FONT =
COLOR=3D"#000000" SIZE=3D2 FACE=3D"Arial">"</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" FACE=3D"Times New Roman">Dennis =
Berg</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" FACE=3D"Times New Roman">GETS =
IC Deputy Program Manager, Engineering</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" FACE=3D"Times New =
Roman">DynCorp IS</FONT></P>

<P ALIGN=3DLEFT><FONT COLOR=3D"#000000" FACE=3D"Times New =
Roman">703-818-4608</FONT></P>

</BODY>
</HTML>
------_=_NextPart_001_01C1CA1B.2FA1DFB0--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 03:54:01 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA00724
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 03:54:01 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id DAA00571;
	Wed, 13 Mar 2002 03:53:12 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id DAA00539
	for <ieprep@ns.ietf.org>; Wed, 13 Mar 2002 03:53:11 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id DAA00706
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 03:53:07 -0500 (EST)
Received: from posh.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.05663-0@bells.cs.ucl.ac.uk>; Wed, 13 Mar 2002 08:52:57 +0000
X-Mailer: exmh version 2.5 07/13/2001 with version: MH 6.8.3 #25[UCI]
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-reply-to: Your message of "Tue, 12 Mar 2002 14:53:55 EST." <28.23639eb1.29bfb6d3@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Date: Wed, 13 Mar 2002 08:52:54 +0000
Message-ID: <576.1016009574@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id DAA00540
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit


> Is there a new draft after version -00 from Feb 15? It proposes two
> values, doesn't it? 

ah ha, i see the root of the confusion.  the Internet-Drafts editor
posted an old version of the Framework draft originally submitted
back in October under a different name, as opposed to the slightly
modified version i submitted in February.  ok, i'll direct my comments 
on what has been posted instead of what was submitted :-)

> > [MAP] No, I don't think it is academic. It is worth considering. The result is that 
> > an added item would be needed in resource-priority to designate which level 
> > is the default for each namespace and to include the action it says above. I 
> > don't know that this would be a "well-defined" value. It would be different 
> > in each domain (namespace). For the dsn namespace, it would be "Routine"; for 
> > the q735 namespace, it would be "4". 
> 
> [Henning] I've added a sentence in each about the default value. 

i'll try and convince him and james otherwise :-)


> > this is probably a question to James and Henning, but couldn't one 
> > define a new NameSpace?  like MLPPv2, or extended-MLPP?
>
> Yes, but the problem with allowing addition of values to an existing 
> namespace is that current implementations wouldn't receognize the new
> values. Same problem with a new namespace, except worse. 

i think there is some confusion.  in the above i said a *new* NameSpace,
I didn't say to change an existing one -- ie, one that has already been
defined and registered with IANA.  So if James registers "DSN" with 5 
values, he or someone else can later register "DSNv2" with N values if
a new version of MLPP exists with a different set of values.

which probably leads to a need for a best common practice -- choose
the namespace values carefully, and try to rely on other documents
or standards to reinforce the reasons for selecting the values
associated with the NameSpace.

cheers,

-ken



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 04:18:52 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA01086
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 04:18:52 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id EAA02049;
	Wed, 13 Mar 2002 04:18:39 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id EAA01995
	for <ieprep@ns.ietf.org>; Wed, 13 Mar 2002 04:18:35 -0500 (EST)
Received: from rbhub101.chamb.disa.mil (rbhub101.chamb.disa.mil [209.22.120.24])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA01083
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 04:18:32 -0500 (EST)
Received: by rbhub101.chamb.disa.mil with Internet Mail Service (5.5.2653.19)
	id <GWRLSXKZ>; Wed, 13 Mar 2002 04:18:34 -0500
Message-ID: <AFBFF033E351D511999900204804EE54031761@rbmail104.chamb.disa.mil>
From: "Folts, Harold" <foltsh@ncs.gov>
To: "'Mpierce1@aol.com '" <Mpierce1@aol.com>
Cc: "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
Date: Wed, 13 Mar 2002 04:24:04 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

 Mike,

Please keep in mind that this work is directed to NS/EP telecommunications
requirements, which are not necessarily the same as DoD in their dedicatated
networks.

Hal

-----Original Message-----
From: Mpierce1@aol.com
To: fred@cisco.com
Cc: ieprep@ietf.org
Sent: 3/12/02 11:08 AM
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework

In a message dated 3/11/02 7:51:13 PM Eastern Standard Time,
fred@cisco.com writes: 




At 09:33 AM 3/11/2002, Mpierce1@aol.com wrote: 
>At this point, it would be good to consider whether the possibility
should 
>be included for multiple levels of "authorized". (Note that the
parallel 
>work in SG16 on H.460.4 is leaning towards including such a
capability.) 

are these "levels of authorization" or "kinds of authorization"? My 
experience with levels is that the discussion is always about winners
and 
losers: "my stuff is more important than his stuff, so I think there
needs 
to be another level."



The idea is that there could be many different
agencies/persons/functions that need emergency access - FEMA, police,
medical, Red Cross, network administrators, etc. It's hard to image that
all would be treated the same. If there is only one possible level to
apply to all, it would be very difficult to decide whether or not a
particular group would get the authorization, e.g., is the Red Cross one
that needs this? Providing a way to distinguish would help. Whoever
administers the authorization to use emergency levels in a network would
decide who gets which level. Within "emergency" there needs to be 3 or 4
levels. (Remember, In MLPP, or "Assured Service", one could think of the
5 levels of precedence as consisting of a "Routine" level, and then four
levels of "emergency".) 




What makes sense to me in SIP is a simple statement: "I assert that this

call is an emergency call". If you are not authorized to make that 
assertion, the correct policy is to ignore the statement. If you are 
authorized, the network will have a policy to apply such as "permit this

call even if we are well into the overload zone" or "drop a random call
in 
order to complete this one." The network should determine the policy,
not 
the standard.



Not sure of your point. The standard should allow a way to designate
sufficient levels of importance, in order for a network to apply
whatever its policy is. I can certainly envision that a desired network
policy is to control access based on the relative level of the emergency
call and the authorization of the person making the attempt. The MLPP
model is that each user is authorized to use up to a certain level, and
may signal any level up to that one in a call setup. The question is:
how many distinct levels are needed? 



>4.1.2 Diff-Serv 
> 
>[snip] In this example, there would need to be the 5 drop precedence 
>levels for EF, as well as for each AF class, since eventually there
should 
>be a need to apply precedence type services to other traffic types
besides 
>voice. Clearly, there are not enough available DSCP code points. 

I think you have a very different understanding of diffserv than I do.
When 
you talk about a drop precedence for EF (which doesn't have a drop 
precedence), what are you talking about - dropping packets, or dropping
calls?



The current draft shows "EF for voice, with signalling of bandwidth".
Therefore it whould need to have multiple levels of drop precedence.
This refers to dropping individual packets, since that is where the
marking is. Actually, I think that EF only for the signaling and one of
the AF classes for the voice packets would make more sense, as I
suggested in Annex A of draft-pierce-sipping-assured-service. 




Recall that in GETS, you don't get differing qualities of circuits. You
get 
differing probabilities of getting a circuit, but once you have a
circuit, 
it is just a circuit and works exactly like any other circuit. In IP 
telephony, you might have varying call preferences, but once you decide
to 
carry voice at all, EF is the service that carries voice. If you're
trying 
to model GETS in IP Telephony, dropping packets makes no sense. Dropping

calls might make sense.



If GETS resulted in different routing, it might mean different circuit
quality (better or worse). Yes, it was "just a circuit", so there were
limited things that could be done to play with quality. IP provides
other options. 

Yes,dropping calls might make sense, but I recognize the dislike for
"preemption" and understand that it isn't allowed some places. The point
I'm making is that some resources can be stolen for the more important
call by causing the less important call to drop more packets, thereby
stealing some of its QoS. The result, when the QoS of the less important
call gets intolerably bad, is for the parties to give up and disconnect,
freeing up more resources. Meanwhile, the establishment of new calls is
being controlled based on precedence (importance), so their reattempt is
blocked. 



So it's not in the least obvious where diff-serv fits here. Please
advise?



Diff-serv AF provides the drop precedence for what I described above,
without any change. Is there anything else that could do this? 




>As described in draft-pierce-assured-service... 

is that draft-pierce-sipping-assured-service-01.txt?



Certainly. My fast typing often gets ahead of by slow brain. 




BTW, in view of RFC 2597 "Assured Forwarding PHB Group", which is often 
called "Assured Service", I'd personally appreciate it if you could find

another name. My head snaps every time I hear the name :^) 

I'm certainly sensitive to using a term for two things. Is the term
"Assured Service" used in specs, or just in everyday speech? We'll
consider other names.



Mike 


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 05:18:32 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA01782
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 05:18:32 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id FAA05467;
	Wed, 13 Mar 2002 05:18:18 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id FAA05433
	for <ieprep@optimus.ietf.org>; Wed, 13 Mar 2002 05:18:16 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id FAA01777
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 05:18:12 -0500 (EST)
Received: from posh.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.12111-0@bells.cs.ucl.ac.uk>; Wed, 13 Mar 2002 10:17:53 +0000
X-Mailer: exmh version 2.5 07/13/2001 with version: MH 6.8.3 #25[UCI]
To: "Berg, Dennis" <Dennis.Berg@DynCorp.com>
cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: Re: [Ieprep] Comments on GETS in draft-carlberg-ieprep-freamework
In-reply-to: Your message of "Tue, 12 Mar 2002 18:11:05 EST." <CBED705A7FD2D311865500508B1089410445EDC6@chntex02.is.dyncorp.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Date: Wed, 13 Mar 2002 10:17:50 +0000
Message-ID: <791.1016014670@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id FAA05434
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

Hello Dennis,

> Section 1.2, paragraph "the key aspect is that GETS...." The point about 
> probability is important, but there are actually two contrasts here: 
> {probabilistic, guaranteed} and {priority, preemption}.  Suggest the 
> following to make both clear:
> · "The key aspect of GETS is that is supports a probabilistic approach to 
> call completion through priority, as opposed to guaranteed approach through 
> preemption. This distinction is important because emergency services like 
> GETS are not allowed to tear down existing calls (i.e., seize resources) in 
> order to establish a GETS call.  Instead, GETS increases the probability of 
> call completion by providing priority in the contention for assignment of 
> limited resources required for the call.  Thus, the GETS features focus on 
> increasing the probability that a particular telephone call will be 
> established, but cannot guarantee call completion."

ok, i'll add the changes.

> · "In the presence of this code point, Local Exchange Carriers (LEC) will 
> attempt (if necessary and if the option is supported) to route the call 
> through alternate inter-exchange carriers (IXC) if it cannot complete the 
> call through the default IXC"
>
> Actually the feature described, Alternate Carrier Routing (ACR), is not
> based upon the HPC code point. ACR is implemented using a commercial
> Advanced Intelligent Network (AIN) capability that is triggered based
> upon the dialed number, in this case the GETS access number.
>
> The two GETS features that are triggered by the HPC class mark are Trunk 
> Queuing (TQ) and Exemption from Restrictive Network Management Controls 
> (Ex-NMCs).
>
> Suggest avoiding the complication of bringing up this other type of feature 
> and instead replacing the text above and to the end of the paragraph with:
> · In the presence of this code point, when a GETS call encounters a 
> restrictive network management control that has been activated to reduce 
> traffic overload to a congested route, the Local Exchange Carriers (LECs) 
> will provide the GETS call priority by exempting the call from this 
> restriction.  After receiving the exemption, if the GETS call finds all 
> circuits busy in the route, the LEC will provide further priority by
> queuing the call for the next available circuit.  

ok.  i'll include most of the above.  i'll probably alter your last 
suggested sentence because the term "priority" tends to cause some
disagreement of what that exactly means and infers.

> · "GETS users are authenticated using a PIN provided to the 
> telecommunications carrier, which signals approval back to the user's
> local exchange over SS7."
> 
> Currently the IXCs authenticate GETS calls by differing methods.  There is
> no communication of authentication back to the LEC where the call 
> originates.   However, when the call egresses the IXC to a LEC for 
> termination to the called party, the IXC passes the HPC class mark and
> the terminating LEC provides priority treatment based upon that class
>  mark.
> Recommend deleting "...which signals approval back to the user's local 
> exchange over SS7" or changing to ""...which signals authentication to 
> subsequent networks via the HPC class mark."

interesting.  the above seems to contradict earlier information i received
on the matter.  i'll cross check it just to make sure and add the suggested
changes accordingly.

thanks for the comments and suggested changes,

-ken



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 11:47:23 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10827
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 11:47:23 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA25990;
	Wed, 13 Mar 2002 11:46:44 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA25956
	for <ieprep@ns.ietf.org>; Wed, 13 Mar 2002 11:46:42 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10799
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 11:46:38 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2DGkAh18625
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 08:46:10 -0800 (PST)
Message-Id: <4.3.2.7.2.20020313084243.01bc2e10@mira-sjcm-2.cisco.com>
X-Sender: fred@mira-sjcm-2.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 13 Mar 2002 08:44:16 -0800
To: ieprep@ietf.org
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-Reply-To: <1415.1016036631@cs.ucl.ac.uk>
References: <Your message of "Wed, 13 Mar 2002 07:40:36 PST." <4.3.2.7.2.20020313073946.0358f200@mira-sjcm-2.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 08:23 AM 3/13/2002, Ken Carlberg wrote:
>http://www.cs.ucl.ac.uk/staff/k.carlberg/tmp/draft-ieprep-framework-00.txt

the version of the draft that Ken attempted to post is at the above URL. 
We'll get it posted at the IETF site as soon as the posting blackout clears.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 12:02:03 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11222
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 12:02:03 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA28202;
	Wed, 13 Mar 2002 12:01:12 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA28145
	for <ieprep@ns.ietf.org>; Wed, 13 Mar 2002 12:01:08 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id MAA11190
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 12:01:03 -0500 (EST)
Received: from posh.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.24161-0@bells.cs.ucl.ac.uk>; Wed, 13 Mar 2002 17:01:03 +0000
X-Mailer: exmh version 2.5 07/13/2001 with version: MH 6.8.3 #25[UCI]
To: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-reply-to: Your message of "Wed, 13 Mar 2002 08:44:16 PST." <4.3.2.7.2.20020313084243.01bc2e10@mira-sjcm-2.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Wed, 13 Mar 2002 17:01:02 +0000
Message-ID: <1487.1016038862@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


> >http://www.cs.ucl.ac.uk/staff/k.carlberg/tmp/draft-ieprep-framework-00.txt
> 
> the version of the draft that Ken attempted to post is at the above URL. 
> We'll get it posted at the IETF site as soon as the posting blackout clears.

just a heads up...

there isn't much difference between the currently posted version and
the one listed above.  so no need to be worried about going through it
with a fine tooth comb.  from memory, the basic changes were:

  o proposing a single value for an IEPS NameSpace defined in the 
    SIP Resource Header draft

  o an additional thought about diff-serv

  o a couple of nice references on users tolerance with respect to delay
    and loss in relation to VoIP, of which the URLs I include below.

       http://www.ietf.org/proceedings/97aug/slides/tsv/ippm-voiceip
       http://www.cs.ucl.ac.uk/staff/A.Sasse/inet95audio.ps

-ken



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 18:56:06 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA25112
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 18:56:06 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA28686;
	Wed, 13 Mar 2002 18:55:33 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA28654
	for <ieprep@ns.ietf.org>; Wed, 13 Mar 2002 18:55:32 -0500 (EST)
Received: from imo-m08.mx.aol.com (imo-m08.mx.aol.com [64.12.136.163])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA25068
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 18:55:30 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m08.mx.aol.com (mail_out_v32.5.) id i.97.244cee32 (661);
	Wed, 13 Mar 2002 18:54:55 -0500 (EST)
Message-ID: <97.244cee32.29c140cf@aol.com>
Date: Wed, 13 Mar 2002 18:54:55 EST
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
To: K.Carlberg@cs.ucl.ac.uk, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_97.244cee32.29c140cf_boundary"
X-Mailer: AOL 6.0 for Windows US sub 353
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--part1_97.244cee32.29c140cf_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 3/13/02 12:01:47 PM Eastern Standard Time, 
K.Carlberg@cs.ucl.ac.uk writes:


> 
> there isn't much difference between the currently posted version and
> the one listed above.  so no need to be worried about going through it
> with a fine tooth comb.  from memory, the basic changes were:
> 
>   o proposing a single value for an IEPS NameSpace defined in the 
>     SIP Resource Header draft
> 

I don't understand this change. I thought it had been agreed that there was a 
need to distinguish between authorized emergency calls and public emergency 
calls (e.g., 911) within a single network. Aren't both of these being 
supported, just as they are in the telephone network today?

Also, as I had commented earlier, there should be a value defined in the set 
for "normal", even if the value is never included in signaling. Otherwise, 
there is really no good way to describe the action to take when no value is 
present.

See you all in Minneapolis.

Mike Pierce
Artel


--part1_97.244cee32.29c140cf_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 3/13/02 12:01:47 PM Eastern Standard Time, K.Carlberg@cs.ucl.ac.uk writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>there isn't much difference between the currently posted version and
<BR>the one listed above. &nbsp;so no need to be worried about going through it
<BR>with a fine tooth comb. &nbsp;from memory, the basic changes were:
<BR>
<BR> &nbsp;o proposing a single value for an IEPS NameSpace defined in the 
<BR> &nbsp;&nbsp;&nbsp;SIP Resource Header draft
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>I don't understand this change. I thought it had been agreed that there was a need to distinguish between authorized emergency calls and public emergency calls (e.g., 911) within a single network. Aren't both of these being supported, just as they are in the telephone network today?
<BR>
<BR>Also, as I had commented earlier, there should be a value defined in the set for "normal", even if the value is never included in signaling. Otherwise, there is really no good way to describe the action to take when no value is present.
<BR>
<BR>See you all in Minneapolis.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_97.244cee32.29c140cf_boundary--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 18:56:24 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA25131
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 18:56:24 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA28793;
	Wed, 13 Mar 2002 18:56:06 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA28765
	for <ieprep@ns.ietf.org>; Wed, 13 Mar 2002 18:56:05 -0500 (EST)
Received: from imo-d01.mx.aol.com (imo-d01.mx.aol.com [205.188.157.33])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA25107
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 18:56:03 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d01.mx.aol.com (mail_out_v32.5.) id c.15c.9f6da0d (661);
	Wed, 13 Mar 2002 18:55:03 -0500 (EST)
Message-ID: <15c.9f6da0d.29c140d7@aol.com>
Date: Wed, 13 Mar 2002 18:55:03 EST
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
To: fred@cisco.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_15c.9f6da0d.29c140d7_boundary"
X-Mailer: AOL 6.0 for Windows US sub 353
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--part1_15c.9f6da0d.29c140d7_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 3/12/02 3:50:06 PM Eastern Standard Time, fred@cisco.com 
writes:


> At 08:08 AM 3/12/2002, Mpierce1@aol.com wrote:
> >The idea is that there could be many different agencies/persons/functions 
> >that need emergency access - FEMA, police, medical, Red Cross, network 
> >administrators, etc. It's hard to image that all would be treated the 
> >same. If there is only one possible level to apply to all, it would be 
> >very difficult to decide whether or not a particular group would get the 
> >authorization, e.g., is the Red Cross one that needs this? Providing a way 
> >to distinguish would help. Whoever administers the authorization to use 
> >emergency levels in a network would decide who gets which level. Within 
> >"emergency" there needs to be 3 or 4 levels. (Remember, In MLPP, or 
> >"Assured Service", one could think of the 5 levels of precedence as 
> >consisting of a "Routine" level, and then four levels of "emergency".)
> 
> Perhaps, but you're mixing apples and orangutans. Providing a way to 
> distinguish calls is probably a good thing. Using the standard that defines 
> the label with which you identify traffic to define the policy to apply to 
> those various distinguished kinds of calls dramatically reduces your 
> flexibility, and brings you into a global space you are not thinking about. 
> 

(See my comment below about this issue of naming labels.)

> eing able to say "This is a FEMA Code 27 call" (distinguishing calls) is 
> quite a bit different than saying "This is a FEMA Code 27 call, all 
> telephone systems in the world including those deployed in Madagascar need 
> to be able to understand what that means, and by the way that means that my 
> call should be placed ahead of a call for the Red Cross but behind a call 
> 

What I meant is that there might be 3 levels for emergency calls and FEMA is 
authorized to use the highest one and maybe the Red Cross is only authorized 
to use the lowest. When FEMA makes a call, they can designate the call as the 
highest level (which must be authenticated). The network policy knows how to 
handle it and how (and if) to give it better service than a call from the Red 
Cross which designates the lowest level. This is identical to MLPP today and 
what we need for what I called "Assured Service". Whether there are 
interworking policies with a network in Madagascar is another issue.

> 
> By all means, put identifying marks on calls. Leave the policy to be 
> applied to those calls ("all telephone systems in the world including those 
> deployed in Madagascar need to be able to understand what that means, and 
> by the way that means that my call should be placed ahead of a call for the 
> Red Cross but behind a call which might launch missiles") to the network.
> 
> It's not obvious to me, by the way, that '5 levels of precedence as 
> consisting of a "Routine" level, and then four levels of "emergency"' is 
> 


It's what the US military and many others have been using for years and it is 
the basic requirement we have for them to use Voice over IP.

  
   Take a recent case in point: on September 11, the calls directing 
emergency traffic to the lower Manhattan or around the
> Pentagon were not "authorized" folks, they were Joe Citizen who was in no 
> way special but happened to be on the scene. Cutting Joe Citizen out of the 
> loop, while enabling "authorized" access to folks whose first instinct 
> would have been to call home and say they were all right, would have been 
> at least as disruptive to emergency services deployment as the fact that 
> Joe Citizen also wanted to call home. Precedence makes sense when you have 
> the power to choose; as I understand the GETS network, it doesn't choose 
> whether or not to place a call as it does choose whether to let a person 
> 


I'm surprised. The above statement sounds like you do not support the idea of 
providing "authorized" precedence for anyone, not even emergency workers, 
just because Joe Citizen might sometimes be in a better position or because 
some "authorized" person might misuse it.

> 
> >>What makes sense to me in SIP is a simple statement: "I assert that this 
> >>call is an emergency call". If you are not authorized to make that 
> >>assertion, the correct policy is to ignore the statement. If you are 
> >>authorized, the network will have a policy to apply such as "permit this 
> >>call even if we are well into the overload zone" or "drop a random call 
> >>in order to complete this one." The network should determine the policy, 
> >>not the standard.
> >
> >Not sure of your point. The standard should allow a way to designate 
> >sufficient levels of importance, in order for a network to apply whatever 
> >its policy is.
> 
> And who decides what is important and what is not? I think you'll find that 
> someone standing on the corner watching people bleed will feel their calls 
> are *very* important, while someone with a different perspective might 
> decide that places where lots of people are bleeding are more important 
> than places where only a few are. My point is that that the policy must 
> describe how to make that decision - the network policy must decide whether 
> one thing is more important than another. As such, it cannot be placed in 
> the hands of the actor who has only a small part of the picture. It must be 
> done by the network.
> 
> Therefore, you want the person making the call to identify the issue in 
> some way, and let the network apply a pre-determined policy to decide what 
> resources to apply to it.
> 

Again, your statement sounds like to do not agree with the basic premise of 
the work - that someone (i.e., the government) decides on who has 
authorization to use the emergency level for calls. Note that the case of 
someone standing on the corner reporting someone bleeding is normal 911 call, 
which is separate from authorized emergency calls (maybe from someone in a 
position to report lots of people bleeding). I don't understand the point 
that "you want the person making the call to identify the issue". SIP 
signaling obviously can't include a description of the emergency situation so 
the network can work out relative priorities with other descriptions. The 
originator has to specify a level (maybe one of many) and the network has to 
compare levels.
 .
> >The question is: how many distinct levels are needed?
> 
> You have yet to demonstrate that the concept of "levels" is decidable on a 
> global basis. Is a FEMA call more important than its counterpart in 
> Madagascar? Is that a nonsense question? If it is a nonsense question, why 
> are we discussing it in the context of a standard for products that will be 
> 

Yes, it's a nonsense question if there is some concept of different networks 
in the two places. Each would have a definition of what the emergency levels 
are and how to handle them. A FEMA call doesn't contend with one in 
Madagascar, unless there is some SLA between the networks and they have 
agreed on the mapping of meanings. If your view is one giant, ubiquitous 
network worldwide, then I have no idea how to handle this issue based on the 
current approach. No, I dont' think the issue of "levels" is decidable on a 
global basis, and, no, we shouldn't be writing a standard that is specific to 
the US. That was the point of my comments months ago on SIP saying that we 
should define the precedence levels in the standard as 0,1,2,3,4 and let each 
networks use them as they see fit. There is no reason to believe that 
Madagascar would label their levels the same as the US.

> 
> >>>4.1.2 Diff-Serv
> >>>
> >>>[snip] In this example, there would need to be the 5 drop precedence 
> >>>levels for EF, as well as for each AF class, since eventually there 
> >>>should be a need to apply precedence type services to other traffic 
> >>>types besides voice. Clearly, there are not enough available DSCP code 
> points.
> >>
> >>I think you have a very different understanding of diffserv than I do. 
> >>When you talk about a drop precedence for EF (which doesn't have a drop 
> >>precedence), what are you talking about - dropping packets, or dropping 
> calls?
> >
> >The current draft shows "EF for voice, with signalling of bandwidth".
> 
> There are 51 posted drafts that contain the words "voice", "signalling", 
> and "bandwidth". I have just reviewed all of them and have not found those 
> 

I was referring to draft-baker-ieprep-requirements-00 in section 2.4. Sorry 
about the double "l" in signaling in my comments.

<> snip>


> If what you want to define is an EF service with multiple drop precedences, 
> I would strongly suggest that you define an EF service with multiple drop 
> precedences. When you do, remember that in computer science there are three 
> reasonable numbers: "zero", "one", and "many". I can tell you for sure what 
> will happen if you prioritize some traffic over other traffic - loss, 
> delay, and variation in delay (jitter) will be transferred from the 
> prioritized traffic to the other traffic. I think I can tell you what will 
> happen with two priorities if the priorities map respectively to voice and 
> video traffic. Folks have done a zillion simulations on priority systems 
> with more levels, and have not been able to demonstrate utility on a 
> per-packet level.
> 
By your last statement do you mean that no one has demonstrated utility for 
any per-packet priority system or one with more than 3 levels? I had no 
intention of suggesting that we apply "priority" on a per-packet basis within 
a class, only selection of what is dropped (as AF does). You noted in your 
comments that the requirement for voice is "minimal jitter, latency, and loss 
for a data stream" and "EF accomplishes that by measuring rate, dropping 
excess, and giving priority to the rest." As far as I know, there is no way 
to control which packets get dropped. We would like to add five drop 
precedences to correspond with the 5 precedence levels of calls.

See you in Minneapolis.

Mike Pierce

--part1_15c.9f6da0d.29c140d7_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 3/12/02 3:50:06 PM Eastern Standard Time, fred@cisco.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">At 08:08 AM 3/12/2002, Mpierce1@aol.com wrote:
<BR>&gt;The idea is that there could be many different agencies/persons/functions 
<BR>&gt;that need emergency access - FEMA, police, medical, Red Cross, network 
<BR>&gt;administrators, etc. It's hard to image that all would be treated the 
<BR>&gt;same. If there is only one possible level to apply to all, it would be 
<BR>&gt;very difficult to decide whether or not a particular group would get the 
<BR>&gt;authorization, e.g., is the Red Cross one that needs this? Providing a way 
<BR>&gt;to distinguish would help. Whoever administers the authorization to use 
<BR>&gt;emergency levels in a network would decide who gets which level. Within 
<BR>&gt;"emergency" there needs to be 3 or 4 levels. (Remember, In MLPP, or 
<BR>&gt;"Assured Service", one could think of the 5 levels of precedence as 
<BR>&gt;consisting of a "Routine" level, and then four levels of "emergency".)
<BR>
<BR>Perhaps, but you're mixing apples and orangutans. Providing a way to 
<BR>distinguish calls is probably a good thing. Using the standard that defines 
<BR>the label with which you identify traffic to define the policy to apply to 
<BR>those various distinguished kinds of calls dramatically reduces your 
<BR>flexibility, and brings you into a global space you are not thinking about. </FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR>
<BR>(See my comment below about this issue of naming labels.)
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>B<BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">eing able to say "This is a FEMA Code 27 call" (distinguishing calls) is 
<BR>quite a bit different than saying "This is a FEMA Code 27 call, all 
<BR>telephone systems in the world including those deployed in Madagascar need 
<BR>to be able to understand what that means, and by the way that means that my 
<BR>call should be placed ahead of a call for the Red Cross but behind a call 
<BR>which might launch missiles" is quite another.</BLOCKQUOTE>
<BR>
<BR>What I meant is that there might be 3 levels for emergency calls and FEMA is authorized to use the highest one and maybe the Red Cross is only authorized to use the lowest. When FEMA makes a call, they can designate the call as the highest level (which must be authenticated). The network policy knows how to handle it and how (and if) to give it better service than a call from the Red Cross which designates the lowest level. This is identical to MLPP today and what we need for what I called "Assured Service". Whether there are interworking policies with a network in Madagascar is another issue.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>By all means, put identifying marks on calls. Leave the policy to be 
<BR>applied to those calls ("all telephone systems in the world including those 
<BR>deployed in Madagascar need to be able to understand what that means, and 
<BR>by the way that means that my call should be placed ahead of a call for the 
<BR>Red Cross but behind a call which might launch missiles") to the network.
<BR>
<BR>It's not obvious to me, by the way, that '5 levels of precedence as 
<BR>consisting of a "Routine" level, and then four levels of "emergency"' is 
<BR>wise, or solves anything.</BLOCKQUOTE>
<BR>
<BR>
<BR>It's what the US military and many others have been using for years and it is the basic requirement we have for them to use Voice over IP.
<BR>
<BR> &nbsp;
<BR> &nbsp;&nbsp;Take a recent case in point: on September 11, the calls directing emergency traffic to the lower Manhattan or around the
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">Pentagon were not "authorized" folks, they were Joe Citizen who was in no 
<BR>way special but happened to be on the scene. Cutting Joe Citizen out of the 
<BR>loop, while enabling "authorized" access to folks whose first instinct 
<BR>would have been to call home and say they were all right, would have been 
<BR>at least as disruptive to emergency services deployment as the fact that 
<BR>Joe Citizen also wanted to call home. Precedence makes sense when you have 
<BR>the power to choose; as I understand the GETS network, it doesn't choose 
<BR>whether or not to place a call as it does choose whether to let a person 
<BR>use specific network resources to accomplish that.</BLOCKQUOTE>
<BR>
<BR>
<BR>I'm surprised. The above statement sounds like you do not support the idea of providing "authorized" precedence for anyone, not even emergency workers, just because Joe Citizen might sometimes be in a better position or because some "authorized" person might misuse it.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>&gt;&gt;What makes sense to me in SIP is a simple statement: "I assert that this 
<BR>&gt;&gt;call is an emergency call". If you are not authorized to make that 
<BR>&gt;&gt;assertion, the correct policy is to ignore the statement. If you are 
<BR>&gt;&gt;authorized, the network will have a policy to apply such as "permit this 
<BR>&gt;&gt;call even if we are well into the overload zone" or "drop a random call 
<BR>&gt;&gt;in order to complete this one." The network should determine the policy, 
<BR>&gt;&gt;not the standard.
<BR>&gt;
<BR>&gt;Not sure of your point. The standard should allow a way to designate 
<BR>&gt;sufficient levels of importance, in order for a network to apply whatever 
<BR>&gt;its policy is.
<BR>
<BR>And who decides what is important and what is not? I think you'll find that 
<BR>someone standing on the corner watching people bleed will feel their calls 
<BR>are *very* important, while someone with a different perspective might 
<BR>decide that places where lots of people are bleeding are more important 
<BR>than places where only a few are. My point is that that the policy must 
<BR>describe how to make that decision - the network policy must decide whether 
<BR>one thing is more important than another. As such, it cannot be placed in 
<BR>the hands of the actor who has only a small part of the picture. It must be 
<BR>done by the network.
<BR>
<BR>Therefore, you want the person making the call to identify the issue in 
<BR>some way, and let the network apply a pre-determined policy to decide what 
<BR>resources to apply to it.
<BR></BLOCKQUOTE>
<BR>
<BR>Again, your statement sounds like to do not agree with the basic premise of the work - that someone (i.e., the government) decides on who has authorization to use the emergency level for calls. Note that the case of someone standing on the corner reporting someone bleeding is normal 911 call, which is separate from authorized emergency calls (maybe from someone in a position to report lots of people bleeding). I don't understand the point that "you want the person making the call to identify the issue". SIP signaling obviously can't include a description of the emergency situation so the network can work out relative priorities with other descriptions. The originator has to specify a level (maybe one of many) and the network has to compare levels.
<BR>.
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">&gt;The question is: how many distinct levels are needed?
<BR>
<BR>You have yet to demonstrate that the concept of "levels" is decidable on a 
<BR>global basis. Is a FEMA call more important than its counterpart in 
<BR>Madagascar? Is that a nonsense question? If it is a nonsense question, why 
<BR>are we discussing it in the context of a standard for products that will be 
<BR>used without change in both the US and Madagascar?</BLOCKQUOTE>
<BR>
<BR>Yes, it's a nonsense question if there is some concept of different networks in the two places. Each would have a definition of what the emergency levels are and how to handle them. A FEMA call doesn't contend with one in Madagascar, unless there is some SLA between the networks and they have agreed on the mapping of meanings. If your view is one giant, ubiquitous network worldwide, then I have no idea how to handle this issue based on the current approach. No, I dont' think the issue of "levels" is decidable on a global basis, and, no, we shouldn't be writing a standard that is specific to the US. That was the point of my comments months ago on SIP saying that we should define the precedence levels in the standard as 0,1,2,3,4 and let each networks use them as they see fit. There is no reason to believe that Madagascar would label their levels the same as the US.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>&gt;&gt;&gt;4.1.2 Diff-Serv
<BR>&gt;&gt;&gt;
<BR>&gt;&gt;&gt;[snip] In this example, there would need to be the 5 drop precedence 
<BR>&gt;&gt;&gt;levels for EF, as well as for each AF class, since eventually there 
<BR>&gt;&gt;&gt;should be a need to apply precedence type services to other traffic 
<BR>&gt;&gt;&gt;types besides voice. Clearly, there are not enough available DSCP code points.
<BR>&gt;&gt;
<BR>&gt;&gt;I think you have a very different understanding of diffserv than I do. 
<BR>&gt;&gt;When you talk about a drop precedence for EF (which doesn't have a drop 
<BR>&gt;&gt;precedence), what are you talking about - dropping packets, or dropping calls?
<BR>&gt;
<BR>&gt;The current draft shows "EF for voice, with signalling of bandwidth".
<BR>
<BR>There are 51 posted drafts that contain the words "voice", "signalling", 
<BR>and "bandwidth". I have just reviewed all of them and have not found those 
<BR>words. Please advise - which draft are you thinking of?</BLOCKQUOTE>
<BR>
<BR>I was referring to draft-baker-ieprep-requirements-00 in section 2.4. Sorry about the double "l" in signaling in my comments.
<BR>
<BR>&lt;<BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">snip&gt;</BLOCKQUOTE>
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">If what you want to define is an EF service with multiple drop precedences, 
<BR>I would strongly suggest that you define an EF service with multiple drop 
<BR>precedences. When you do, remember that in computer science there are three 
<BR>reasonable numbers: "zero", "one", and "many". I can tell you for sure what 
<BR>will happen if you prioritize some traffic over other traffic - loss, 
<BR>delay, and variation in delay (jitter) will be transferred from the 
<BR>prioritized traffic to the other traffic. I think I can tell you what will 
<BR>happen with two priorities if the priorities map respectively to voice and 
<BR>video traffic. Folks have done a zillion simulations on priority systems 
<BR>with more levels, and have not been able to demonstrate utility on a 
<BR>per-packet level.
<BR></BLOCKQUOTE>
<BR>By your last statement do you mean that no one has demonstrated utility for any per-packet priority system or one with more than 3 levels? I had no intention of suggesting that we apply "priority" on a per-packet basis within a class, only selection of what is dropped (as AF does). You noted in your comments that the requirement for voice is "minimal jitter, latency, and loss for a data stream" and "EF accomplishes that by measuring rate, dropping excess, and giving priority to the rest." As far as I know, there is no way to control which packets get dropped. We would like to add five drop precedences to correspond with the 5 precedence levels of calls.
<BR>
<BR>See you in Minneapolis.
<BR>
<BR>Mike Pierce</FONT></HTML>

--part1_15c.9f6da0d.29c140d7_boundary--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 21:05:47 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA27216
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 21:05:47 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA05689;
	Wed, 13 Mar 2002 21:04:36 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA05640
	for <ieprep@ns.ietf.org>; Wed, 13 Mar 2002 21:04:33 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA27213
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 21:04:30 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2E240h18438;
	Wed, 13 Mar 2002 18:04:00 -0800 (PST)
Message-Id: <4.3.2.7.2.20020313160357.01b6ef00@mira-sjcm-2.cisco.com>
X-Sender: fred@mira-sjcm-2.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 13 Mar 2002 17:28:06 -0800
To: <Mpierce1@aol.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
Cc: ieprep@ietf.org
In-Reply-To: <15c.9f6da0d.29c140d7@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 03:55 PM 3/13/2002, Mpierce1@aol.com wrote:
>What I meant is that there might be 3 levels for emergency calls and FEMA 
>is authorized to use the highest one and maybe the Red Cross is only 
>authorized to use the lowest.

That still couples the policy (levels of emergency-ness, which translate 
into ways of servicing those calls) with the labeling scheme (four numbers, 
which you would like to raise to five),

>[one routine level plus four levels of emergency are] what the US military 
>and many others have been using for years and it is the basic requirement 
>we have for them to use Voice over IP.

That makes it right? Are you sure there is not a need for five levels of 
emergency somewhere, or that the four levels that are in use are properly used?

>The above statement sounds like you do not support the idea of providing 
>"authorized" precedence for anyone, not even emergency workers, just 
>because Joe Citizen might sometimes be in a better position or because 
>some "authorized" person might misuse it.

I am on record as not being crazy about it as stated, for reasons which I 
have articulated in the past. I do believe that policy should be applied in 
placing calls, and that this might include preferential distribution of 
service. But who says that "emergency-ness" is the only contributing 
factor, or should get a she-gets-it-he-doesn't veto capability? I rather 
strongly think that if you pull the policy out of the definition of the 
label, you can apply whatever policy you want (such as varying levels of 
emergency-ness) within your domain, and let someone else apply whatever 
policy they want in theirs.

What I have a problem with is trying to force everyone to use the same 
policy model. I can't think of any two networks in the world that have 
exactly the same policy model.

In short, if it is done, I want it to be done in a manner which not only 
works for the US Military within the DISA network, but works in other 
networks as well, and can apply whatever policy is applicable in whatever 
place. I also want it to apply on a call basis, not a packet basis, unless 
it can be shown that the packet capabilities in fact materially help.

>>You have yet to demonstrate that the concept of "levels" is decidable on 
>>a global basis. Is a FEMA call more important than its counterpart in 
>>Madagascar? Is that a nonsense question? If it is a nonsense question, 
>>why are we discussing it in the context of a standard for products that 
>>will be used without change in both the US and Madagascar?
>
>Yes, it's a nonsense question if there is some concept of different 
>networks in the two places.

>No, I dont' think the issue of "levels" is decidable on a global basis, 
>and, no, we shouldn't be writing a standard that is specific to the US. 
>That was the point of my comments months ago on SIP saying that we should 
>define the precedence levels in the standard as 0,1,2,3,4 and let each 
>networks use them as they see fit.

There is one global Internet, and the one SIP standard is used equally in 
all parts of it. That one Internet is composed of various smaller networks, 
but trust me, it is border-blind and within any large network operates in 
many countries. When you say "the government authorizes the use of various 
levels of <mumble>", in the Internet, I don't know who you're talking about 
doing the authorizing - the UN? Yes, this is a fundamental difference from 
the telephone system. It is a difference which needs to be taken into account.

Let's take two examples. Cisco is a multi-national corporation, with 
service by I don't know how many providers. It has at least four DMZs to 
the outside world, in Sydney Australia, San Jose California, RTP North 
Carolina, and Amsterdam Holland. It also has internal service in a truly 
long list of countries. Cable and Wireless (an ISP) has service in upwards 
of 70 countries run through one global ISP, and several wholly owned 
subsidiary networks. C&W and Cisco are each a single network.

>By your last statement do you mean that no one has demonstrated utility 
>for any per-packet priority system or one with more than 3 levels?

pretty much. I know that systems like that can be defined. However, it is 
in no sense obvious from simulation results they result in a significant 
change to the traffic flowing through them over what would happen with 
lower numbers of priorities.

>You noted in your comments that the requirement for voice is "minimal 
>jitter, latency, and loss for a data stream" and "EF accomplishes that by 
>measuring rate, dropping excess, and giving priority to the rest." As far 
>as I know, there is no way to control which packets get dropped. We would 
>like to add five drop precedences to correspond with the 5 precedence 
>levels of calls.

Can you give me an idea why you haven't discussed that with the authors of 
draft-ietf-diffserv-rfc2598bis-02.txt and 
draft-ietf-diffserv-ef-supplemental-01.txt, soon to be RFCs 3246/3247? If 
you could have demonstrated the case, you could have had it in the 
specification. Since we have gone to the effort to take them to RFC, at 
this point it will be an especially tough sell to re-open the documents.

Five drop precedences added to EF might make sense; as I said before, it's 
not obvious to me that it does, but I'm willing to be shown your 
simulations that prove that the service still works after it is done and 
that this materially affects something.

Understand that network bottlenecks are not generally the issue any more 
except in access paths, at least in the US and Europe. Even on Sept 11, 
according to reports, the issues in the Internet were with servers, not 
backbone traffic or even access network traffic. VoIP with *no* levels of 
priority, no bandwidth reservation, and no central switch facility worked 
quite well in lower Manhattan, providing telephone access through NYSERNET 
and other networks in the area; having lost or damaged several switches in 
the area, the standard telephone system had some serious problems to 
overcome (statement not intended to be a slam on the telephone network, but 
an accurate statement of facts - they had a tough couple of weeks and 
pulled some really long hours, for which they deserve a lot of credit).


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 21:10:17 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA27258
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 21:10:16 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA05928;
	Wed, 13 Mar 2002 21:10:02 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA05894
	for <ieprep@ns.ietf.org>; Wed, 13 Mar 2002 21:10:00 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA27246
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 21:09:56 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2E29Sh21524;
	Wed, 13 Mar 2002 18:09:28 -0800 (PST)
Message-Id: <4.3.2.7.2.20020313180708.01c24cf8@mira-sjcm-2.cisco.com>
X-Sender: fred@mira-sjcm-2.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Wed, 13 Mar 2002 18:09:20 -0800
To: <Mpierce1@aol.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
Cc: ieprep@ietf.org
In-Reply-To: <4.3.2.7.2.20020313160357.01b6ef00@mira-sjcm-2.cisco.com>
References: <15c.9f6da0d.29c140d7@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 05:28 PM 3/13/2002, Mpierce1@aol.com wrote:
>Can you give me an idea why you haven't discussed that with the authors of 
>draft-ietf-diffserv-rfc2598bis-02.txt and 
>draft-ietf-diffserv-ef-supplemental-01.txt, soon to be RFCs 3246/3247? If 
>you could have demonstrated the case, you could have had it in the 
>specification. Since we have gone to the effort to take them to RFC, at 
>this point it will be an especially tough sell to re-open the documents.

having said that, let me go on to say that there is nothing whatsoever to 
stop DISA from configuring its network with four extra local-assignment 
code points that are applied to traffic on some appropriate manner that 
accomplishes five drop precedences within EF service. Diffserv only 
recommends a code point for EF; there is nothing to say you can't use a 
different one, or five. If this only applies in your network, that doesn't 
have to be standardized.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 13 21:38:06 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA28568
	for <ieprep-archive@odin.ietf.org>; Wed, 13 Mar 2002 21:38:06 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA07118;
	Wed, 13 Mar 2002 21:37:48 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA07089
	for <ieprep@ns.ietf.org>; Wed, 13 Mar 2002 21:37:46 -0500 (EST)
Received: from newdev.harvard.edu (newdev.eecs.harvard.edu [140.247.60.212])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA28535
	for <ieprep@ietf.org>; Wed, 13 Mar 2002 21:37:43 -0500 (EST)
Received: (from sob@localhost)
	by newdev.harvard.edu (8.10.2/8.10.2) id g2E2bjk13031
	for ieprep@ietf.org; Wed, 13 Mar 2002 21:37:45 -0500 (EST)
Date: Wed, 13 Mar 2002 21:37:45 -0500 (EST)
From: Scott  Bradner <sob@harvard.edu>
Message-Id: <200203140237.g2E2bjk13031@newdev.harvard.edu>
To: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


> It's what the US military and many others have been using for years and it is 
> the basic requirement we have for them to use Voice over IP.

why

(note the answer that 'they have it now' is not legit)

i.e. - lets talk about requeirements not solutions - multiple priorities 
on packets is a solution not a requirement - what is the problem to be
solved?

Scott

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 05:18:54 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA14147
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 05:18:54 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id FAA15624;
	Thu, 14 Mar 2002 05:17:22 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id FAA15592
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 05:17:21 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id FAA14121
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 05:17:17 -0500 (EST)
Received: from posh.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.10236-0@bells.cs.ucl.ac.uk>; Thu, 14 Mar 2002 10:17:14 +0000
X-Mailer: exmh version 2.5 07/13/2001 with version: MH 6.8.3 #25[UCI]
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-reply-to: Your message of "Wed, 13 Mar 2002 18:54:55 EST." <97.244cee32.29c140cf@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 14 Mar 2002 10:17:13 +0000
Message-ID: <621.1016101033@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

hi,

> I don't understand this change. I thought it had been agreed that there 
> was a need to distinguish between authorized emergency calls and public 
> emergency calls (e.g., 911) within a single network. Aren't both of
> these being supported, just as they are in the telephone network today? 

before the SIP Resource Header draft came out, I thought it would be 
beneficial to distinguish 911-type calls from GETS-type calls.

however, in a recent exchange with Henning and James, they both expressed 
a desire not to have an additional marking of these types of calls.  one
reason was because the "sos@" address would already distinguish the SIP 
message as being a 911-type message.  their position was that the 
additional header would be redundant.

so for IEPS, this leaves us with just a GETS-type "authorized-emergency" 
call and a "normal" call.  (skipping the fact that there could be other 
types of NameSpace defined calls.)  and with protocol design/specification,
one always wants to take a minimal/non-redundant position on things.  by
having a "normal" value in the IEPS NameSpace in addition to the 
"authorized-emergency" to me seems redundant to a SIP message without
the Resource Header field.  so i took the "normal" value out.

> Also, as I had commented earlier, there should be a value defined in the
> set for "normal", even if the value is never included in signaling. 
> Otherwise, there is really no good way to describe the action to take
> when no value is present. 

i can see sympathize with having a "normal" or default value if the
Resource Header would contain two or more values that distinguish calls
as being one other than Normal.  but if one is just introducing a single
value that acts as a binary decision of normal/non-normal, then I'm
not convinced that an additional "normal" value is needed.

-ken



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 08:38:13 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA16916
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 08:38:12 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA26797;
	Thu, 14 Mar 2002 08:35:35 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA26763
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 08:35:33 -0500 (EST)
Received: from rbhub101.chamb.disa.mil (rbhub101.chamb.disa.mil [209.22.120.24])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA16869
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 08:35:29 -0500 (EST)
Received: by rbhub101.chamb.disa.mil with Internet Mail Service (5.5.2653.19)
	id <GWRLTT0A>; Thu, 14 Mar 2002 08:35:31 -0500
Message-ID: <AFBFF033E351D511999900204804EE5403176D@rbmail104.chamb.disa.mil>
From: "Folts, Harold" <foltsh@ncs.gov>
To: "'Ken Carlberg '" <K.Carlberg@cs.ucl.ac.uk>,
        "'Mpierce1@aol.com '"
	 <Mpierce1@aol.com>
Cc: "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
Date: Thu, 14 Mar 2002 08:41:04 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

I have been following this lenthy discussion during my travels. I just want
to point out that we need some flexibility with the ETS provisions in the
emerging public networks. We should not be bound by the single precedence
level that GETS has applied. That was all they were able to do or needed at
that time.

I think that it is important that we recognize the potential for multiple
levels (5 have been proposed). I do think that the 911 service should have
the option for its own precidence level. That can be a policy matter in
administering priority levels. As Ken pointed out there is an "SOS"
indication in SIP for 911 calls.

In the civil and commercial world, we should have multiple levels to select
from and then be administered through policy management when the appropriate
authorities decide who gets preference over what.

This is some high level guidance that needs to be considered in developing
technical solutions.

Hal

-----Original Message-----
From: Ken Carlberg
To: Mpierce1@aol.com
Cc: ieprep@ietf.org
Sent: 3/14/02 5:17 AM
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework

hi,

> I don't understand this change. I thought it had been agreed that
there 
> was a need to distinguish between authorized emergency calls and
public 
> emergency calls (e.g., 911) within a single network. Aren't both of
> these being supported, just as they are in the telephone network
today? 

before the SIP Resource Header draft came out, I thought it would be 
beneficial to distinguish 911-type calls from GETS-type calls.

however, in a recent exchange with Henning and James, they both
expressed 
a desire not to have an additional marking of these types of calls.  one
reason was because the "sos@" address would already distinguish the SIP 
message as being a 911-type message.  their position was that the 
additional header would be redundant.

so for IEPS, this leaves us with just a GETS-type "authorized-emergency"

call and a "normal" call.  (skipping the fact that there could be other 
types of NameSpace defined calls.)  and with protocol
design/specification,
one always wants to take a minimal/non-redundant position on things.  by
having a "normal" value in the IEPS NameSpace in addition to the 
"authorized-emergency" to me seems redundant to a SIP message without
the Resource Header field.  so i took the "normal" value out.

> Also, as I had commented earlier, there should be a value defined in
the
> set for "normal", even if the value is never included in signaling. 
> Otherwise, there is really no good way to describe the action to take
> when no value is present. 

i can see sympathize with having a "normal" or default value if the
Resource Header would contain two or more values that distinguish calls
as being one other than Normal.  but if one is just introducing a single
value that acts as a binary decision of normal/non-normal, then I'm
not convinced that an additional "normal" value is needed.

-ken



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 08:46:47 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17003
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 08:46:47 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA27406;
	Thu, 14 Mar 2002 08:45:52 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA27256
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 08:45:40 -0500 (EST)
Received: from imo-r03.mx.aol.com (imo-r03.mx.aol.com [152.163.225.99])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA16985
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 08:44:26 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r03.mx.aol.com (mail_out_v32.5.) id 5.148.b05b268 (4215);
	Thu, 14 Mar 2002 08:43:48 -0500 (EST)
Message-ID: <148.b05b268.29c20314@aol.com>
Date: Thu, 14 Mar 2002 08:43:48 EST
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
To: sob@harvard.edu, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_148.b05b268.29c20314_boundary"
X-Mailer: AOL 6.0 for Windows US sub 353
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--part1_148.b05b268.29c20314_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 3/13/02 9:39:22 PM Eastern Standard Time, sob@harvard.edu 
writes:


> > It's what the US military and many others have been using for years and it 
> is 
> > the basic requirement we have for them to use Voice over IP.
> 
> why
> 
> (note the answer that 'they have it now' is not legit)
> 
> i.e. - lets talk about requeirements not solutions - multiple priorities 
> on packets is a solution not a requirement - what is the problem to be
> solved?
> 
> Scott
> 

I agree. The reference I made about "It's what the US military does" is a 
basic requirement for voice service operation. The end user must be able to 
specify one of 5 levels for voice call setup and it must have certain effects 
on the call setup and retention. (Hopefully, the user really isn't aware of 
whether the call uses IP or circuit switched or a combination.) That's the 
requirement.

On the other hand, marking of packets with drop precedence, or whatever, is a 
possible solution. Right now we see it as proabably needed in order to meet 
the service operation requirement, but realize that this is the main part to 
discuss. Clearly, there is no claim that packet priority levels is what "they 
have now". BTW, I haven't suggested multiple priorities on packets, only 
mulltiple drop precedences. Unfortunately, some of us not familiar with the 
details of AF have gotten the two concepts mixed up at times. I certainly 
have.

The problem being solved is providing some level of preferential treatment to 
the higher priority call without resorting to preemption of lower precedence 
calls. Instead, some packets of the lower precedence call are "preempted" by 
being dropped when the offered load exceeds the capacity. I haven't seen any 
other solution. (We're talking about the case when the offered voice traffic 
exceeds the capacity of a small access link, not one in which there is still 
a whole lot of data and background traffic that can be pushed aside.)

Mike


--part1_148.b05b268.29c20314_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 3/13/02 9:39:22 PM Eastern Standard Time, sob@harvard.edu writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">&gt; It's what the US military and many others have been using for years and it is 
<BR>&gt; the basic requirement we have for them to use Voice over IP.
<BR>
<BR>why
<BR>
<BR>(note the answer that 'they have it now' is not legit)
<BR>
<BR>i.e. - lets talk about requeirements not solutions - multiple priorities 
<BR>on packets is a solution not a requirement - what is the problem to be
<BR>solved?
<BR>
<BR>Scott
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>I agree. The reference I made about "It's what the US military does" is a basic requirement for voice service operation. The end user must be able to specify one of 5 levels for voice call setup and it must have certain effects on the call setup and retention. (Hopefully, the user really isn't aware of whether the call uses IP or circuit switched or a combination.) That's the requirement.
<BR>
<BR>On the other hand, marking of packets with drop precedence, or whatever, is a possible solution. Right now we see it as proabably needed in order to meet the service operation requirement, but realize that this is the main part to discuss. Clearly, there is no claim that packet priority levels is what "they have now". BTW, I haven't suggested multiple priorities on packets, only mulltiple drop precedences. Unfortunately, some of us not familiar with the details of AF have gotten the two concepts mixed up at times. I certainly have.
<BR>
<BR>The problem being solved is providing some level of preferential treatment to the higher priority call without resorting to preemption of lower precedence calls. Instead, some packets of the lower precedence call are "preempted" by being dropped when the offered load exceeds the capacity. I haven't seen any other solution. (We're talking about the case when the offered <U>voice</U> traffic exceeds the capacity of a small access link, not one in which there is still a whole lot of data and background traffic that can be pushed aside.)
<BR>
<BR>Mike
<BR></FONT></HTML>

--part1_148.b05b268.29c20314_boundary--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 09:20:59 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17462
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 09:20:59 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA29520;
	Thu, 14 Mar 2002 09:20:37 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA29484
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 09:20:36 -0500 (EST)
Received: from newdev.harvard.edu (newdev.eecs.harvard.edu [140.247.60.212])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17455
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 09:20:34 -0500 (EST)
Received: (from sob@localhost)
	by newdev.harvard.edu (8.10.2/8.10.2) id g2EEKaF14841
	for ieprep@ietf.org; Thu, 14 Mar 2002 09:20:36 -0500 (EST)
Date: Thu, 14 Mar 2002 09:20:36 -0500 (EST)
From: Scott  Bradner <sob@harvard.edu>
Message-Id: <200203141420.g2EEKaF14841@newdev.harvard.edu>
To: ieprep@ietf.org
Subject:  Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


Mike sez:
  The reference I made about "It's what the US military does" is a 
  basic requirement for voice service operation.

we are talking past each other

I maintain that N-levels of priority is a solution not a requirement

the question that needs to be answered is what functions are needed to
be supported - not how to do them

specifically - what problem is thought to be solved by N levels of priority?

	is it that there should be a different probibility for
	calls to be completed based on who you are or what you are doing?

	is it that the quality of the call should depends on who you
	are and what you are doing?

	or is it someting else

Scott

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 09:29:55 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17669
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 09:29:55 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA00230;
	Thu, 14 Mar 2002 09:29:38 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA00191
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 09:29:37 -0500 (EST)
Received: from rbhub101.chamb.disa.mil (rbhub101.chamb.disa.mil [209.22.120.24])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17666
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 09:29:35 -0500 (EST)
Received: by rbhub101.chamb.disa.mil with Internet Mail Service (5.5.2653.19)
	id <GWRLTV8K>; Thu, 14 Mar 2002 09:29:37 -0500
Message-ID: <AFBFF033E351D511999900204804EE54031773@rbmail104.chamb.disa.mil>
From: "Folts, Harold" <foltsh@ncs.gov>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Date: Thu, 14 Mar 2002 09:35:09 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain
Subject: [Ieprep] ieprep WG Meeting
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

I am very disappointed that I will be unable to attend the ieprep session on
Monday. I have had a sudden death in the family that I have to tend to.

I hope to have someone else there that can participate in any discussion of
my ID draft-folts-ieprep-white-paper-00.txt. I will continue my particpation
and will help develop the text for a requirements RFC.

Hal

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 09:33:06 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17757
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 09:33:06 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA00918;
	Thu, 14 Mar 2002 09:31:40 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA00878
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 09:31:38 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17718
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 09:31:36 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id HAA18560
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 07:29:50 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA35A3
          for <ieprep@ietf.org>; Thu, 14 Mar 2002 07:31:37 -0700
Received: FROM agcs.com BY mknaomi ; Thu Mar 14 07:31:35 2002 -0700
Message-ID: <3C90B448.B7AE665E@agcs.com>
Date: Thu, 14 Mar 2002 07:31:36 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "Folts, Harold" <foltsh@ncs.gov>
CC: "'Ken Carlberg '" <K.Carlberg@cs.ucl.ac.uk>,
        "'Mpierce1@aol.com '" <Mpierce1@aol.com>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
References: <AFBFF033E351D511999900204804EE5403176D@rbmail104.chamb.disa.mil>
Content-Type: multipart/mixed;
 boundary="------------2B42438263C2435FE4641C60"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------2B42438263C2435FE4641C60
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Folks,

I also have been following this discussion.

I hesitate to jump in, but perhaps I should share the outcome of the ITU-T SG 11
meeting (Feb. 18- March 1) as a reference point.

At the ITU-T SG 11 we agreed to add an optional NEW PARAMETER to the SS7 and
BICC IAM. If the new parameter is present, it identifies the call as an ETS call
(and since the parameter is not in the national values it would be passed
internationally. )

The absence of the parameter means that the call is not marked as ETS. However,
the call may still have a national CPC value such as NS/EP indicating the
national treatment required, or one of the  normal CPC value indicating
non-emergency treatment.

The advantage of a new parameter is that a call can be marked as
normal,  national emergency (based on CPC), internationally recognized emergency
(new parameter), or both national and international recognized emergency (both
the new parameter and a national CPC value).

The new optional parameter only identifies the call as ETS, but does not
currently identify levels.

"Folts, Harold" wrote:

> I have been following this lenthy discussion during my travels. I just want
> to point out that we need some flexibility with the ETS provisions in the
> emerging public networks. We should not be bound by the single precedence
> level that GETS has applied. That was all they were able to do or needed at
> that time.
>
> I think that it is important that we recognize the potential for multiple
> levels (5 have been proposed). I do think that the 911 service should have
> the option for its own precidence level. That can be a policy matter in
> administering priority levels. As Ken pointed out there is an "SOS"
> indication in SIP for 911 calls.
>
> In the civil and commercial world, we should have multiple levels to select
> from and then be administered through policy management when the appropriate
> authorities decide who gets preference over what.
>
> This is some high level guidance that needs to be considered in developing
> technical solutions.
>
> Hal
>
> -----Original Message-----
> From: Ken Carlberg
> To: Mpierce1@aol.com
> Cc: ieprep@ietf.org
> Sent: 3/14/02 5:17 AM
> Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
>
> hi,
>
> > I don't understand this change. I thought it had been agreed that
> there
> > was a need to distinguish between authorized emergency calls and
> public
> > emergency calls (e.g., 911) within a single network. Aren't both of
> > these being supported, just as they are in the telephone network
> today?
>
> before the SIP Resource Header draft came out, I thought it would be
> beneficial to distinguish 911-type calls from GETS-type calls.
>
> however, in a recent exchange with Henning and James, they both
> expressed
> a desire not to have an additional marking of these types of calls.  one
> reason was because the "sos@" address would already distinguish the SIP
> message as being a 911-type message.  their position was that the
> additional header would be redundant.
>
> so for IEPS, this leaves us with just a GETS-type "authorized-emergency"
>
> call and a "normal" call.  (skipping the fact that there could be other
> types of NameSpace defined calls.)  and with protocol
> design/specification,
> one always wants to take a minimal/non-redundant position on things.  by
> having a "normal" value in the IEPS NameSpace in addition to the
> "authorized-emergency" to me seems redundant to a SIP message without
> the Resource Header field.  so i took the "normal" value out.
>
> > Also, as I had commented earlier, there should be a value defined in
> the
> > set for "normal", even if the value is never included in signaling.
> > Otherwise, there is really no good way to describe the action to take
> > when no value is present.
>
> i can see sympathize with having a "normal" or default value if the
> Resource Header would contain two or more values that distinguish calls
> as being one other than Normal.  but if one is just introducing a single
> value that acts as a binary decision of normal/non-normal, then I'm
> not convinced that an additional "normal" value is needed.
>
> -ken
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

--------------2B42438263C2435FE4641C60
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------2B42438263C2435FE4641C60--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 10:23:34 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19061
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 10:23:34 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA03765;
	Thu, 14 Mar 2002 10:12:27 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA03729
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 10:12:24 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id KAA18862
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 10:12:21 -0500 (EST)
Received: from posh.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.05608-0@bells.cs.ucl.ac.uk>; Thu, 14 Mar 2002 15:12:19 +0000
X-Mailer: exmh version 2.5 07/13/2001 with version: MH 6.8.3 #25[UCI]
To: Scott Bradner <sob@harvard.edu>
cc: ieprep@ietf.org
In-reply-to: Your message of "Thu, 14 Mar 2002 09:20:36 EST." <200203141420.g2EEKaF14841@newdev.harvard.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 14 Mar 2002 15:12:16 +0000
Message-ID: <1260.1016118736@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Subject: [Ieprep] Levels of Priority
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


>   The reference I made about "It's what the US military does" is a 
>   basic requirement for voice service operation.
> 
> I maintain that N-levels of priority is a solution not a requirement
> 
> the question that needs to be answered is what functions are needed to
> be supported - not how to do them
>
> specifically - what problem is thought to be solved by N levels of priority?

with a slight change of subject line description, there seems to be a
couple of things to consider.

first, there is the objective of having MLPP be supported via IP 
telephony.  

given the existing 5 levels of service for MLPP defined for the 
telephony world, there would appear to be a requirement to have a 
corresponding set of 5 levels or labels of signaling for SIP.  this 
is so that IP telephony via SIP can convey labels on a one-to-one 
basis with MLPP enabled telco networks.  most of these MLPP telco nets
may be private, but folks like US DoD (and NATO, and I believe the 
Japanese MoD) which use MLPP, would like to have a standard for SIP to 
convey these labels so that they can buy SIP phones from more than one 
vendor.

The proposed DSN NameSpace of James and Henning's draft, together with 
the Resource Header field, would seem to satisfy this requirement for
SIP.

my impression of what remains an open issue that is up for a debate is 
whether there should be a corresponding set of 5 markings for the 
(non-signaling) data flows -- in particular, at the IP layer.  And I 
suspect that what Scott is asking is where is the requirement/problem 
that would prompt this solution.

I just want to make sure we don't mix apples and oranges in the debate.

-ken

ps, i was also under the impression that one or some of the public
    cellular providers were planning on adding MLPP, thus extending
    the problem space beyond just private telco nets



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 10:34:36 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19569
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 10:34:36 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA05597;
	Thu, 14 Mar 2002 10:34:15 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA05565
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 10:34:13 -0500 (EST)
Received: from newdev.harvard.edu (newdev.eecs.harvard.edu [140.247.60.212])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19548
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 10:34:11 -0500 (EST)
Received: (from sob@localhost)
	by newdev.harvard.edu (8.10.2/8.10.2) id g2EFYEu15473
	for ieprep@ietf.org; Thu, 14 Mar 2002 10:34:14 -0500 (EST)
Date: Thu, 14 Mar 2002 10:34:14 -0500 (EST)
From: Scott  Bradner <sob@harvard.edu>
Message-Id: <200203141534.g2EFYEu15473@newdev.harvard.edu>
To: ieprep@ietf.org
Subject: [Ieprep] Re: Levels of Priority
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


Ken sez:
> given the existing 5 levels of service for MLPP defined for the 
> telephony world, there would appear to be a requirement to have a 
> corresponding set of 5 levels or labels of signaling for SIP. 

1/ having an ability to transport a priority field to a SIP/PSTN
   gateway so that it can be forwarded to the PSTN makes a lot
   of sense

2/ knowing what to do with such a priority field in a SIP node
   or proxy is another thing altogether

3/ and both are VERY different than marking SIP signaling packets
   with 5 different diffserv markings

4/ and are VERY different than marking SIP voice packets with 5
   different diffserv markings


Scott

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 11:20:51 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA21190
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 11:20:50 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA09417;
	Thu, 14 Mar 2002 11:15:38 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA29800
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 09:24:31 -0500 (EST)
Received: from prserv.net (asmtp2.prserv.net [32.97.166.52])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17587
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 09:24:29 -0500 (EST)
Received: from D8J4Y211 (<unknown.domain>[212.234.161.3])
          by prserv.net (asmtp2) with SMTP
          id <2002031414242925204ais7le>
          (Authid: usinet.capthal);
          Thu, 14 Mar 2002 14:24:29 +0000
Reply-To: <catching@captainhal.com>
From: "capthal" <capthal@attglobal.net>
To: "'Scott  Bradner'" <sob@harvard.edu>, <ieprep@ietf.org>
Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
Date: Thu, 14 Mar 2002 15:23:15 +0100
Message-ID: <000201c1cb63$cc15dbf0$2d051cac@D8J4Y211>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook, Build 10.0.2627
Importance: Normal
In-Reply-To: <200203141420.g2EEKaF14841@newdev.harvard.edu>
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

That sounds on track to me.

Hal

> -----Original Message-----
> From: ieprep-admin@ietf.org [mailto:ieprep-admin@ietf.org] On Behalf
Of
> Scott Bradner
> Sent: Thursday, March 14, 2002 3:21 PM
> To: ieprep@ietf.org
> Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
> 
> 
> Mike sez:
>   The reference I made about "It's what the US military does" is a
>   basic requirement for voice service operation.
> 
> we are talking past each other
> 
> I maintain that N-levels of priority is a solution not a requirement
> 
> the question that needs to be answered is what functions are needed to
> be supported - not how to do them
> 
> specifically - what problem is thought to be solved by N levels of
> priority?
> 
> 	is it that there should be a different probibility for
> 	calls to be completed based on who you are or what you are
doing?
> 
> 	is it that the quality of the call should depends on who you
> 	are and what you are doing?
> 
> 	or is it someting else
> 
> Scott
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep




_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 11:45:11 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA22040
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 11:45:11 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA12979;
	Thu, 14 Mar 2002 11:42:36 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA12939
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 11:42:34 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA21946
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 11:42:31 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2EGfmh28059;
	Thu, 14 Mar 2002 08:41:48 -0800 (PST)
Message-Id: <4.3.2.7.2.20020314083236.023206c8@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 14 Mar 2002 08:33:09 -0800
To: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Levels of Priority
Cc: Scott Bradner <sob@harvard.edu>, ieprep@ietf.org
In-Reply-To: <1260.1016118736@cs.ucl.ac.uk>
References: <Your message of "Thu, 14 Mar 2002 09:20:36 EST." <200203141420.g2EEKaF14841@newdev.harvard.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 07:12 AM 3/14/2002, Ken Carlberg wrote:
>given the existing 5 levels of service for MLPP defined for the
>telephony world

tell me - what are these five levels of precedence? I would expect that 
they are precedence in placing calls, not precedence in dropping packets.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 11:57:01 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA22511
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 11:57:01 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA13988;
	Thu, 14 Mar 2002 11:55:58 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA13956
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 11:55:56 -0500 (EST)
Received: from unity.cisco.com (evvbu-exchange.cisco.com [64.101.140.73])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA22483
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 11:55:52 -0500 (EST)
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
Subject: RE: [Ieprep] Levels of Priority
Date: Thu, 14 Mar 2002 10:55:25 -0600
Message-ID: <E58F685255415A4EB2D4B8A02AEF32E3CC427E@evvbu-exchange.amer.unity.cisco.com>
Thread-Topic: [Ieprep] Levels of Priority
Thread-Index: AcHLd14BU6kBxR0vT0KZ5Rj3A6bhyAAAWgAA
From: "Bob Bell" <rtbell@cisco.com>
To: "Fred Baker" <fred@cisco.com>, "Ken Carlberg" <K.Carlberg@cs.ucl.ac.uk>
Cc: "Scott Bradner" <sob@harvard.edu>, <ieprep@ietf.org>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id LAA13957
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

Fred - 

Actually, they are sort of like packet dropping precidence in that if a call of a higher precidence is trying to be established, it can cause a lower precidence call to be terminated to free telephone resources (usually trunks) so that the higher precidence call can be completed.

Bob

>-----Original Message-----
>From: Fred Baker [mailto:fred@cisco.com]
>Sent: Thursday, 14 March, 2002 9:33 AM
>To: Ken Carlberg
>Cc: Scott Bradner; ieprep@ietf.org
>Subject: Re: [Ieprep] Levels of Priority
>
>
>At 07:12 AM 3/14/2002, Ken Carlberg wrote:
>>given the existing 5 levels of service for MLPP defined for the
>>telephony world
>
>tell me - what are these five levels of precedence? I would 
>expect that 
>they are precedence in placing calls, not precedence in 
>dropping packets.
>
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 12:21:49 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA23237
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 12:21:49 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA17576;
	Thu, 14 Mar 2002 12:21:09 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA17542
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 12:21:07 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA23177
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 12:21:03 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id JAA10712 for <ieprep@ietf.org>; Thu, 14 Mar 2002 09:20:35 -0800 (PST)
Message-Id: <4.1.20020314111514.02f042d8@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Thu, 14 Mar 2002 11:20:08 -0600
To: ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] Levels of Priority
In-Reply-To: <4.3.2.7.2.20020314083236.023206c8@mira-sjcm-4.cisco.com>
References: <1260.1016118736@cs.ucl.ac.uk>
 <Your message of "Thu, 14 Mar 2002 09:20:36 EST." <200203141420.g2EEKaF14841@newdev.harvard.edu>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_-1827909265==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_-1827909265==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 08:33 AM 3/14/2002 -0800, Fred Baker wrote:
>At 07:12 AM 3/14/2002, Ken Carlberg wrote:
>>given the existing 5 levels of service for MLPP defined for the
>>telephony world
>
>tell me - what are these five levels of precedence? I would expect that 
>they are precedence in placing calls, not precedence in dropping packets.

As it stands now, (I believe) these 5 levels are not the "precedence in
dropping packets", but in dropping session(s) in order to complete higher
precedence session(s), which adds a new level of fun to this mix (and a subtle
re-wording of Fred's comment) 
*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_-1827909265==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 08:33 AM 3/14/2002 -0800, Fred Baker wrote:</div>
<div>&gt;At 07:12 AM 3/14/2002, Ken Carlberg wrote:</div>
<div>&gt;&gt;given the existing 5 levels of service for MLPP defined for
the</div>
<div>&gt;&gt;telephony world</div>
<div>&gt;</div>
<div>&gt;tell me - what are these five levels of precedence? I would
expect that </div>
<div>&gt;they are precedence in placing calls, not precedence in dropping
packets.</div>
<br>
As it stands now, (I believe) these 5 levels are not the &quot;precedence
in dropping packets&quot;, but in dropping session(s) in order to
complete higher precedence session(s), which adds a new level of fun to
this mix (and a subtle re-wording of Fred's comment)
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_-1827909265==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 12:40:39 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA23627
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 12:40:39 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA19179;
	Thu, 14 Mar 2002 12:40:17 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA19143
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 12:40:15 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA23607
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 12:40:13 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2EHdih17690
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 09:39:44 -0800 (PST)
Message-Id: <4.3.2.7.2.20020314091119.01b07c10@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 14 Mar 2002 09:39:27 -0800
To: ieprep@ietf.org
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Levels of Priority
In-Reply-To: <E58F685255415A4EB2D4B8A02AEF32E3CC427E@evvbu-exchange.amer
 .unity.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 08:55 AM 3/14/2002, Bob Bell wrote:
>Actually, they are sort of like packet dropping precidence in that if a 
>call of a higher precidence is trying to be established, it can cause a 
>lower precidence call to be terminated to free telephone resources 
>(usually trunks) so that the higher precidence call can be completed.

but that's not what they are doing. They are placing the authorized call 
anyway, and then if congestion sets in (which could be in the form of 
synchronization between several data streams resulting from the use of a 
national 8 KHz clock, even in the presence of no other calls) dropping 
packets from the unauthorized call, resulting in a reduction of quality on 
that call.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 13:03:30 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24258
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 13:03:30 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA20877;
	Thu, 14 Mar 2002 13:02:12 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA20845
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 13:02:11 -0500 (EST)
Received: from unity.cisco.com (evvbu-exchange.cisco.com [64.101.140.73])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24157
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 13:02:08 -0500 (EST)
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
Subject: RE: [Ieprep] Levels of Priority
Date: Thu, 14 Mar 2002 12:01:39 -0600
Message-ID: <E58F685255415A4EB2D4B8A02AEF32E3CC4280@evvbu-exchange.amer.unity.cisco.com>
Thread-Topic: [Ieprep] Levels of Priority
Thread-Index: AcHLf16JL22mZjo3SbiQC9hcCMlRzwAAhGiA
From: "Bob Bell" <rtbell@cisco.com>
To: "Fred Baker" <fred@cisco.com>, <ieprep@ietf.org>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id NAA20846
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

Fred, et. al -

The comment I made was in relationship to MLPP in the phone system. The US public phone systems are prohibited by law from dropping a currently active call in favor of another call. MLPP was developed in the AUTOVON system (US Military) were the prohibition did not exist. I think what they are trying to accomplish is to preserve the QoS of a higher precidence call at the expense of the QoS for a lower precidence call. I guess it would make a level of sense in that you probably wouldn't want the voice call from the President telling SAC to recall the bombers and stand down the missles to be so badly garbled that the Commander of SAC thought the President told him to commit the attack. ;-)

Bob

>-----Original Message-----
>From: Fred Baker [mailto:fred@cisco.com]
>Sent: Thursday, 14 March, 2002 10:39 AM
>To: ieprep@ietf.org
>Subject: RE: [Ieprep] Levels of Priority
>
>
>At 08:55 AM 3/14/2002, Bob Bell wrote:
>>Actually, they are sort of like packet dropping precidence in 
>that if a 
>>call of a higher precidence is trying to be established, it 
>can cause a 
>>lower precidence call to be terminated to free telephone resources 
>>(usually trunks) so that the higher precidence call can be completed.
>
>but that's not what they are doing. They are placing the 
>authorized call 
>anyway, and then if congestion sets in (which could be in the form of 
>synchronization between several data streams resulting from 
>the use of a 
>national 8 KHz clock, even in the presence of no other calls) dropping 
>packets from the unauthorized call, resulting in a reduction 
>of quality on 
>that call.
>
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 13:13:33 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24504
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 13:13:33 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA21326;
	Thu, 14 Mar 2002 13:12:47 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA21294
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 13:12:46 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24484
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 13:12:43 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2EICEh17596
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 10:12:14 -0800 (PST)
Message-Id: <4.3.2.7.2.20020314100953.01b0c658@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 14 Mar 2002 10:11:34 -0800
To: ieprep@ietf.org
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Levels of Priority
In-Reply-To: <4.1.20020314111514.02f042d8@localhost>
References: <4.3.2.7.2.20020314083236.023206c8@mira-sjcm-4.cisco.com>
 <1260.1016118736@cs.ucl.ac.uk>
 <Your message of "Thu, 14 Mar 2002 09:20:36 EST." <200203141420.g2EEKaF14841@newdev.harvard.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 09:20 AM 3/14/2002, James M. Polk wrote:
>At 08:33 AM 3/14/2002 -0800, Fred Baker wrote:
> >At 07:12 AM 3/14/2002, Ken Carlberg wrote:
> >>given the existing 5 levels of service for MLPP defined for the
> >>telephony world
> >
> >tell me - what are these five levels of precedence? I would expect that
> >they are precedence in placing calls, not precedence in dropping packets.
>
>As it stands now, (I believe) these 5 levels are not the "precedence in 
>dropping packets", but in dropping session(s) in order to complete higher 
>precedence session(s),

which says to me that if the argument for some form of policy in SIP is to 
extend MLPP, the policy extension should model that of MLPP, and be in the 
form of call management, not packet management.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 13:44:07 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25366
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 13:44:07 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23557;
	Thu, 14 Mar 2002 13:43:43 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23502
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 13:43:41 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25336
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 13:43:38 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2EIh8h16033;
	Thu, 14 Mar 2002 10:43:08 -0800 (PST)
Message-Id: <4.3.2.7.2.20020314101903.04578ea0@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 14 Mar 2002 10:24:48 -0800
To: "Bob Bell" <rtbell@cisco.com>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Levels of Priority
Cc: <ieprep@ietf.org>
In-Reply-To: <E58F685255415A4EB2D4B8A02AEF32E3CC4280@evvbu-exchange.amer
 .unity.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 10:01 AM 3/14/2002, Bob Bell wrote:
>The comment I made was in relationship to MLPP in the phone system. The US 
>public phone systems are prohibited by law from dropping a currently 
>active call in favor of another call. MLPP was developed in the AUTOVON 
>system (US Military) were the prohibition did not exist.

this is an excellent example of the reason to separate the policy from the 
label.

In the US phone system, one could in some manner reserve a few circuits for 
emergency use, and dole those out only to authorized users, and then only 
if non-emergency circuits were unavailable. To do so, one needs an 
authenticated label that says "this is an emergency call", and carries with 
it the necessary credentials and so on. One could also have groups of those 
and some additional policy for the use of those groups.

In AUTOVON, the exact same label could be used with a different policy - go 
ahead and use all of the available circuits for normal telephone service, 
but when such a call is placed, find a circuit to drop and do so.

The device identifies the call and presents the credentials. The network 
decides what policy is applicable and applies it.

But you can only do so if the label is policy-free. It has to say "this is 
a <mumble> call", without any prior expectations of what that might mean to 
the network.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 13:55:02 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25850
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 13:55:02 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA24400;
	Thu, 14 Mar 2002 13:54:40 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA24369
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 13:54:39 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25839
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 13:54:35 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id LAA13109
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 11:52:49 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA579B
          for <ieprep@ietf.org>; Thu, 14 Mar 2002 11:54:36 -0700
Received: FROM agcs.com BY mknaomi ; Thu Mar 14 11:54:36 2002 -0700
Message-ID: <3C90F1EC.CE976D3F@agcs.com>
Date: Thu, 14 Mar 2002 11:54:36 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Bob Bell <rtbell@cisco.com>
CC: Fred Baker <fred@cisco.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Levels of Priority
References: <E58F685255415A4EB2D4B8A02AEF32E3CC4280@evvbu-exchange.amer.unity.cisco.com>
Content-Type: multipart/mixed;
 boundary="------------AB9910011966D20BA1CDDE41"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------AB9910011966D20BA1CDDE41
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Bob,

You make an excellent point. While the President most likely would not be using the PSTN to recall the SAC bomber, the concept that emergency calls both must be connected and must have a QoS sufficient for clear communication should not be forgotten.

Bob Bell wrote:

> Fred, et. al -
>
> The comment I made was in relationship to MLPP in the phone system. The US public phone systems are prohibited by law from dropping a currently active call in favor of another call. MLPP was developed in the AUTOVON system (US Military) were the prohibition did not exist. I think what they are trying to accomplish is to preserve the QoS of a higher precidence call at the expense of the QoS for a lower precidence call. I guess it would make a level of sense in that you probably wouldn't want the voice call from the President telling SAC to recall the bombers and stand down the missles to be so badly garbled that the Commander of SAC thought the President told him to commit the attack. ;-)
>
> Bob
>
> >-----Original Message-----
> >From: Fred Baker [mailto:fred@cisco.com]
> >Sent: Thursday, 14 March, 2002 10:39 AM
> >To: ieprep@ietf.org
> >Subject: RE: [Ieprep] Levels of Priority
> >
> >
> >At 08:55 AM 3/14/2002, Bob Bell wrote:
> >>Actually, they are sort of like packet dropping precidence in
> >that if a
> >>call of a higher precidence is trying to be established, it
> >can cause a
> >>lower precidence call to be terminated to free telephone resources
> >>(usually trunks) so that the higher precidence call can be completed.
> >
> >but that's not what they are doing. They are placing the
> >authorized call
> >anyway, and then if congestion sets in (which could be in the form of
> >synchronization between several data streams resulting from
> >the use of a
> >national 8 KHz clock, even in the presence of no other calls) dropping
> >packets from the unauthorized call, resulting in a reduction
> >of quality on
> >that call.
> >
> >
> >_______________________________________________
> >Ieprep mailing list
> >Ieprep@ietf.org
> >https://www1.ietf.org/mailman/listinfo/ieprep
> >
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

--------------AB9910011966D20BA1CDDE41
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------AB9910011966D20BA1CDDE41--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 14:11:17 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26502
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 14:11:17 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA26273;
	Thu, 14 Mar 2002 14:09:17 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA26243
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 14:09:16 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26451
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 14:09:13 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id LAA04651 for <ieprep@ietf.org>; Thu, 14 Mar 2002 11:08:32 -0800 (PST)
Message-Id: <4.1.20020314124529.02bf4730@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Thu, 14 Mar 2002 13:08:05 -0600
To: ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-Reply-To: <200203141420.g2EEKaF14841@newdev.harvard.edu>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_-1821433012==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_-1821433012==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 09:20 AM 3/14/2002 -0500, Scott  Bradner wrote:
>
>Mike sez:
>  The reference I made about "It's what the US military does" is a 
>  basic requirement for voice service operation.
>
>we are talking past each other
>
>I maintain that N-levels of priority is a solution not a requirement
>
>the question that needs to be answered is what functions are needed to
>be supported - not how to do them
>
>specifically - what problem is thought to be solved by N levels of priority?
>
>       is it that there should be a different probibility for
>       calls to be completed based on who you are or what you are doing?
>
>       is it that the quality of the call should depends on who you
>       are and what you are doing?
>
>       or is it someting else

Let's pose a strict set of circumstances to address part of this thread that's
not answered yet explicitly (to my understanding - and others I believe)

Let assume RSVP worked completely everywhere and between every phone in the
Internet. A VoIP call using RSVP would work almost identically to a circuit
switched call in taking up a defined amount of bandwidth through each transit
interface between two phones. Lets say that all VoIP calls push aside all other
(data) traffic always with no hesitation. In a case where each call is 80kbps
traversing a T1 (1.544Mbps), 19 calls are possible at any one time. Even though
there is 56kbps of bandwidth remaining, that 20th call cannot be completed if
it requires an 80kbps Reservation across this link.

This is an example where codecs aren't changeable -- I want a specific analysis
to this scenario

What Mike is suggesting is the potential for dropping < 1kbps from all 19
existing calls to make this 20th one exist (if all 19 calls have the same
precedence level and the new 20th call has a higher precedence level). That
instead of clearing one of the calls (preemption), merely degrading all 19
existing calls by about 1 packet per second (of the 50pps for a normal VoIP
call).

Is this considered unacceptable?

One existing example/analogy:  even though the Cellular industry isn't
packet-based, I believe most calls remain *up* AND loose more than 1% of the
media signal during a call (if not much more). 


>
>Scott
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep

*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_-1821433012==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 09:20 AM 3/14/2002 -0500, Scott&nbsp; Bradner 
wrote:</div>
<div>&gt;</div>
<div>&gt;Mike sez:</div>
<div>&gt;&nbsp; The reference I made about &quot;It's what the US
military does&quot; is a </div>
<div>&gt;&nbsp; basic requirement for voice service operation.</div>
<div>&gt;</div>
<div>&gt;we are talking past each other</div>
<div>&gt;</div>
<div>&gt;I maintain that N-levels of priority is a solution not a
requirement</div>
<div>&gt;</div>
<div>&gt;the question that needs to be answered is what functions are
needed to</div>
<div>&gt;be supported - not how to do them</div>
<div>&gt;</div>
<div>&gt;specifically - what problem is thought to be solved by N levels
of priority?</div>
<div>&gt;</div>
<div>&gt;<x-tab>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</x-tab>is it
that there should be a different probibility for</div>
<div>&gt;<x-tab>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</x-tab>calls
to be completed based on who you are or what you are doing?</div>
<div>&gt;</div>
<div>&gt;<x-tab>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</x-tab>is it
that the quality of the call should depends on who you</div>
<div>&gt;<x-tab>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</x-tab>are and
what you are doing?</div>
<div>&gt;</div>
<div>&gt;<x-tab>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</x-tab>or is
it someting else</div>
<br>
<div>Let's pose a strict set of circumstances to address part of this
thread that's not answered yet explicitly (to my understanding - and
others I believe)</div>
<br>
<div>Let assume RSVP worked completely everywhere and between every phone
in the Internet. A VoIP call using RSVP would work almost identically to
a circuit switched call in taking up a defined amount of bandwidth
through each transit interface between two phones. Lets say that all VoIP
calls push aside all other (data) traffic always with no hesitation. In a
case where each call is 80kbps traversing a T1 (1.544Mbps), 19 calls are
possible at any one time. Even though there is 56kbps of bandwidth
remaining, that 20th call cannot be completed if it requires an 80kbps
Reservation across this link.</div>
<br>
<div>This is an example where codecs aren't changeable -- I want a
specific analysis to this scenario</div>
<br>
<div>What Mike is suggesting is the potential for dropping &lt; 1kbps
from all 19 existing calls to make this 20th one exist (if all 19 calls
have the same precedence level and the new 20th call has a higher
precedence level). That instead of clearing one of the calls
(preemption), merely degrading all 19 existing calls by about 1 packet
per second (of the 50pps for a normal VoIP call).</div>
<br>
<div>Is this considered unacceptable?</div>
<br>
<div>One existing example/analogy:&nbsp; even though the Cellular
industry isn't packet-based, I believe most calls remain *up* AND loose
more than 1% of the media signal during a call (if not much more).
</div>
<br>
<br>
<div>&gt;</div>
<div>&gt;Scott</div>
<div>&gt;</div>
<div>&gt;_______________________________________________</div>
<div>&gt;Ieprep mailing list</div>
<div>&gt;Ieprep@ietf.org</div>
<div>&gt;<a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA=AUTOURL>https://www1.ietf.org/mailman/listinfo/ieprep</a></div>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_-1821433012==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 14:21:48 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26973
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 14:21:48 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27159;
	Thu, 14 Mar 2002 14:21:27 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27105
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 14:21:24 -0500 (EST)
Received: from unity.cisco.com (evvbu-exchange.cisco.com [64.101.140.73])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26935
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 14:21:21 -0500 (EST)
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1CB8D.5BA91BD4"
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
Date: Thu, 14 Mar 2002 13:20:53 -0600
Message-ID: <E58F685255415A4EB2D4B8A02AEF32E3CC4284@evvbu-exchange.amer.unity.cisco.com>
Thread-Topic: [Ieprep] Comments on draft-carlberg-ieprep-framework
Thread-Index: AcHLi82FH8OSuj4ATpeOP20NdMvyxgAAIb/Q
From: "Bob Bell" <rtbell@cisco.com>
To: "James Polk" <jmpolk@cisco.com>, <ieprep@ietf.org>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.

------_=_NextPart_001_01C1CB8D.5BA91BD4
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

James, et. al. -=20
=20
In the stated example, loosing 1% of the stream would probably not cause =
a great deal of distortion so as to render all conversations unusable. =
However, if you continue the process of stealing chunks of bandwidth, as =
more calls are "shoehorned" through the restriction point, you will =
quickly reach that condition. This is the reason I would believe that =
associating a precidence level with a call and using policy to terminate =
a lower precidence level session in favor of a higher precidence level =
would be more "fair". In this way, only the lowest precidence session =
would suffer, not all sessions. Even if you were to remove bandwidth =
from sessions proportinate to the their precidence (e.g. the lower the =
precidence, the more bandwidth could be stolen), unless some method of =
reducing their "billing" for this reduced clarity call is in place, then =
persons would be "paying" for services that they could not get. This is =
asking for trouble in the form of govenrment intervention and law =
suites. At least, if the lower precidence call is terminated, the =
"billing" also ends.
=20
Bob

-----Original Message-----
From: James Polk=20
Sent: Thursday, 14 March, 2002 12:08 PM
To: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework


At 09:20 AM 3/14/2002 -0500, Scott  Bradner wrote:
>
>Mike sez:
>  The reference I made about "It's what the US military does" is a=20
>  basic requirement for voice service operation.
>
>we are talking past each other
>
>I maintain that N-levels of priority is a solution not a requirement
>
>the question that needs to be answered is what functions are needed to
>be supported - not how to do them
>
>specifically - what problem is thought to be solved by N levels of =
priority?
>
>       is it that there should be a different probibility for
>       calls to be completed based on who you are or what you are =
doing?
>
>       is it that the quality of the call should depends on who you
>       are and what you are doing?
>
>       or is it someting else

Let's pose a strict set of circumstances to address part of this thread =
that's not answered yet explicitly (to my understanding - and others I =
believe)

Let assume RSVP worked completely everywhere and between every phone in =
the Internet. A VoIP call using RSVP would work almost identically to a =
circuit switched call in taking up a defined amount of bandwidth through =
each transit interface between two phones. Lets say that all VoIP calls =
push aside all other (data) traffic always with no hesitation. In a case =
where each call is 80kbps traversing a T1 (1.544Mbps), 19 calls are =
possible at any one time. Even though there is 56kbps of bandwidth =
remaining, that 20th call cannot be completed if it requires an 80kbps =
Reservation across this link.

This is an example where codecs aren't changeable -- I want a specific =
analysis to this scenario

What Mike is suggesting is the potential for dropping < 1kbps from all =
19 existing calls to make this 20th one exist (if all 19 calls have the =
same precedence level and the new 20th call has a higher precedence =
level). That instead of clearing one of the calls (preemption), merely =
degrading all 19 existing calls by about 1 packet per second (of the =
50pps for a normal VoIP call).

Is this considered unacceptable?

One existing example/analogy:  even though the Cellular industry isn't =
packet-based, I believe most calls remain *up* AND loose more than 1% of =
the media signal during a call (if not much more).=20


>
>Scott
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

*************************************
"People generally demand more respect for their own rights than they are =
willing to allow for others"


James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike=20
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com <http://www.cisco.com/> =20


------_=_NextPart_001_01C1CB8D.5BA91BD4
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">


<META content=3D"MSHTML 5.50.4913.1100" name=3DGENERATOR></HEAD>
<BODY>
<DIV><SPAN class=3D225321319-14032002><FONT face=3DArial color=3D#0000ff =
size=3D2>James,=20
et. al. - </FONT></SPAN></DIV>
<DIV><SPAN class=3D225321319-14032002><FONT face=3DArial color=3D#0000ff =

size=3D2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=3D225321319-14032002><FONT face=3DArial color=3D#0000ff =
size=3D2>In the=20
stated example, loosing 1% of the stream would probably not cause a =
great deal=20
of distortion so as to render all conversations unusable. However, if =
you=20
continue the process of stealing chunks of bandwidth, as more calls are=20
"shoehorned"&nbsp;through the restriction point, you will quickly reach =
that=20
condition. This is the reason I would believe that associating a =
precidence=20
level with a call and using policy to terminate a lower precidence level =
session=20
in favor of a higher precidence level would be more "fair". In this way, =
only=20
the lowest precidence session would suffer, not all sessions. Even if =
you were=20
to remove bandwidth from sessions proportinate to the their precidence =
(e.g. the=20
lower the precidence, the more bandwidth could be stolen), unless some =
method of=20
reducing their "billing" for this reduced clarity call is in place, then =
persons=20
would be "paying" for services that they could not get. This is asking =
for=20
trouble in the form of govenrment intervention and law suites. At least, =
if the=20
lower precidence call is terminated, the "billing" also=20
ends.</FONT></SPAN></DIV>
<DIV><SPAN class=3D225321319-14032002><FONT face=3DArial color=3D#0000ff =

size=3D2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=3D225321319-14032002><FONT face=3DArial color=3D#0000ff =

size=3D2>Bob</FONT></SPAN></DIV>
<BLOCKQUOTE=20
style=3D"PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px =
solid">
  <DIV class=3DOutlookMessageHeader dir=3Dltr align=3Dleft><FONT =
face=3DTahoma=20
  size=3D2>-----Original Message-----<BR><B>From:</B> James Polk =
<BR><B>Sent:</B>=20
  Thursday, 14 March, 2002 12:08 PM<BR><B>To:</B>=20
  ieprep@ietf.org<BR><B>Subject:</B> Re: [Ieprep] Comments on=20
  draft-carlberg-ieprep-framework<BR><BR></FONT></DIV>
  <DIV>At 09:20 AM 3/14/2002 -0500, Scott&nbsp; Bradner wrote:</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;Mike sez:</DIV>
  <DIV>&gt;&nbsp; The reference I made about "It's what the US military =
does" is=20
  a </DIV>
  <DIV>&gt;&nbsp; basic requirement for voice service operation.</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;we are talking past each other</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;I maintain that N-levels of priority is a solution not a=20
  requirement</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;the question that needs to be answered is what functions are =
needed=20
  to</DIV>
  <DIV>&gt;be supported - not how to do them</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;specifically - what problem is thought to be solved by N =
levels of=20
  priority?</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;<X-TAB>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</X-TAB>is =
it that=20
  there should be a different probibility for</DIV>
  =
<DIV>&gt;<X-TAB>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</X-TAB>calls =
to be=20
  completed based on who you are or what you are doing?</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;<X-TAB>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</X-TAB>is =
it that=20
  the quality of the call should depends on who you</DIV>
  <DIV>&gt;<X-TAB>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</X-TAB>are =
and what=20
  you are doing?</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;<X-TAB>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</X-TAB>or =
is it=20
  someting else</DIV><BR>
  <DIV>Let's pose a strict set of circumstances to address part of this =
thread=20
  that's not answered yet explicitly (to my understanding - and others I =

  believe)</DIV><BR>
  <DIV>Let assume RSVP worked completely everywhere and between every =
phone in=20
  the Internet. A VoIP call using RSVP would work almost identically to =
a=20
  circuit switched call in taking up a defined amount of bandwidth =
through each=20
  transit interface between two phones. Lets say that all VoIP calls =
push aside=20
  all other (data) traffic always with no hesitation. In a case where =
each call=20
  is 80kbps traversing a T1 (1.544Mbps), 19 calls are possible at any =
one time.=20
  Even though there is 56kbps of bandwidth remaining, that 20th call =
cannot be=20
  completed if it requires an 80kbps Reservation across this =
link.</DIV><BR>
  <DIV>This is an example where codecs aren't changeable -- I want a =
specific=20
  analysis to this scenario</DIV><BR>
  <DIV>What Mike is suggesting is the potential for dropping &lt; 1kbps =
from all=20
  19 existing calls to make this 20th one exist (if all 19 calls have =
the same=20
  precedence level and the new 20th call has a higher precedence level). =
That=20
  instead of clearing one of the calls (preemption), merely degrading =
all 19=20
  existing calls by about 1 packet per second (of the 50pps for a normal =
VoIP=20
  call).</DIV><BR>
  <DIV>Is this considered unacceptable?</DIV><BR>
  <DIV>One existing example/analogy:&nbsp; even though the Cellular =
industry=20
  isn't packet-based, I believe most calls remain *up* AND loose more =
than 1% of=20
  the media signal during a call (if not much more). </DIV><BR><BR>
  <DIV>&gt;</DIV>
  <DIV>&gt;Scott</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;_______________________________________________</DIV>
  <DIV>&gt;Ieprep mailing list</DIV>
  <DIV>&gt;Ieprep@ietf.org</DIV>
  <DIV>&gt;<A href=3D"https://www1.ietf.org/mailman/listinfo/ieprep"=20
  =
EUDORA=3D"AUTOURL">https://www1.ietf.org/mailman/listinfo/ieprep</A></DIV=
><BR>
  <DIV align=3Dcenter>*************************************<BR>"People =
generally=20
  demand more respect for their own rights than they are willing to =
allow for=20
  others"<BR><BR></DIV>James M. Polk<BR>Consulting Engineer<BR>Office of =
the=20
  CTO<BR><BR>Cisco Systems<BR>2200 East President George Bush Turnpike=20
  <BR>Richardson, TX&nbsp; 75082 USA<BR>w) 972.813.5208<BR>f)&nbsp;=20
  972.813.5280<BR><A href=3D"http://www.cisco.com/"=20
  eudora=3D"autourl">www.cisco.com</A> </BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C1CB8D.5BA91BD4--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 14:35:46 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA27582
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 14:35:45 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA29035;
	Thu, 14 Mar 2002 14:35:22 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA28995
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 14:35:20 -0500 (EST)
Received: from portal.east.saic.com (portal.east.saic.com [198.151.13.15])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id OAA27554
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 14:35:17 -0500 (EST)
Received: from mclmx.saic.com by portal.east.saic.com
          via smtpd (for odin.ietf.org [132.151.1.176]) with SMTP; 14 Mar 2002 19:35:20 UT
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Thu, 14 Mar 2002 14:35:07 -0500
Received: from MCL-ITS-EXIG01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.1.19) with SMTP id M2002031414350723648
 ; Thu, 14 Mar 2002 14:35:07 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <G7ADZCYJ>; Thu, 14 Mar 2002 14:36:02 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D6428915B6@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Bob Bell '" <rtbell@cisco.com>, "'Fred Baker '" <fred@cisco.com>,
        "'Ken Carlberg '" <K.Carlberg@cs.ucl.ac.uk>
Cc: "'Scott Bradner '" <sob@harvard.edu>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Levels of Priority
Date: Thu, 14 Mar 2002 14:35:05 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

All,

Clearly there are many misunderstandings about the MLPP/Assured Service.  A
good deal of these misunderstandings is caused because people haven't listed
their assumptions.  

Are we assuming users assert their own priority, are subject to no call
admission control, and mark their signaling and media streams with their
appropriate code point? (And because there is no call admission control
there is a need for 5 drop probabilities in EF?) 

Or are we assuming there is a call-admission control agent at the CPE
location (In H.323 parlance, this might be an enhanced gatekeeper) that
validates the priority of the call and intercepts incoming / outgoing call
setup messages?  Does any entity maintain call state about those calls
traversing the access link that could preempt a lower priority call when a
higher priority setup message was received? (And therefore all calls have
equal sound quality but the calls carried are those of highest priority.)

As was recently pointed out, before scrambling about the solution, it is
best to define the problem space and the underlying requirements.

Kimberly


-----Original Message-----
From: Bob Bell
To: Fred Baker; Ken Carlberg
Cc: Scott Bradner; ieprep@ietf.org
Sent: 3/14/2002 11:55 AM
Subject: RE: [Ieprep] Levels of Priority

Fred - 

Actually, they are sort of like packet dropping precidence in that if a
call of a higher precidence is trying to be established, it can cause a
lower precidence call to be terminated to free telephone resources
(usually trunks) so that the higher precidence call can be completed.

Bob

>-----Original Message-----
>From: Fred Baker [mailto:fred@cisco.com]
>Sent: Thursday, 14 March, 2002 9:33 AM
>To: Ken Carlberg
>Cc: Scott Bradner; ieprep@ietf.org
>Subject: Re: [Ieprep] Levels of Priority
>
>
>At 07:12 AM 3/14/2002, Ken Carlberg wrote:
>>given the existing 5 levels of service for MLPP defined for the
>>telephony world
>
>tell me - what are these five levels of precedence? I would 
>expect that 
>they are precedence in placing calls, not precedence in 
>dropping packets.
>
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 14:39:56 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA27796
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 14:39:56 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA29506;
	Thu, 14 Mar 2002 14:39:35 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA29474
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 14:39:31 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA27787
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 14:39:27 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id LAA17965 for <ieprep@ietf.org>; Thu, 14 Mar 2002 11:38:58 -0800 (PST)
Message-Id: <4.1.20020314131510.012d4b40@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Thu, 14 Mar 2002 13:38:31 -0600
To: ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-Reply-To: <15c.9f6da0d.29c140d7@aol.com>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_-1819607037==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_-1819607037==_.ALT
Content-Type: text/plain; charset="us-ascii"

At 06:55 PM 3/13/2002 -0500, Mpierce1@aol.com wrote: 
>
> In a message dated 3/12/02 3:50:06 PM Eastern Standard Time, fred@cisco.com
> writes: 
>
>>
>>
>> Therefore, you want the person making the call to identify the issue in 
>> some way, and let the network apply a pre-determined policy to decide what 
>> resources to apply to it. 
>
>
>
> Again, your statement sounds like to do not agree with the basic premise of
> the work - that someone (i.e., the government) decides on who has
> authorization to use the emergency level for calls. Note that the case of
> someone standing on the corner reporting someone bleeding is normal 911 call,
> which is separate from authorized emergency calls (maybe from someone in a
> position to report lots of people bleeding). I don't understand the point
> that "you want the person making the call to identify the issue". SIP
> signaling obviously can't include a description of the emergency situation so
> the network can work out relative priorities with other descriptions. The
> originator has to specify a level (maybe one of many) and the network has to
> compare levels. 
> . 


A possible mechanism exists within the SIP R-P Header in the case you argued
against earlier in another context, the statement allowing the response (200
OK) to set the level. Perhaps this is a case where I'm personally allowed to
make a call (but the network doesn't know it yet) and I INVITE to a specific
URI (the equvelent of today's GETS number), this could have the receiving end
change my original namespace.value up to whatever it needs to be for that call
in the cases where it's been established that "this call is should be routed to
FEMA, or to the RED CROSS, or to the LOCAL FIRE DEPARTMENT (only). SIP
redirects and Re-INVITES might be needed, but that should be unaware to the
caller making the call unless this is after some form of authentication is
required from that caller (pin # or something else).

*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_-1819607037==_.ALT
Content-Type: text/html; charset="us-ascii"

<html>
At 06:55 PM 3/13/2002 -0500, Mpierce1@aol.com wrote: <br>
<font face="arial" size=2><blockquote type=cite cite>In a message dated
3/12/02 3:50:06 PM Eastern Standard Time, fred@cisco.com writes: <br>
</font><br>
<blockquote type=cite cite><br>
<font face="Arial, Helvetica" size=2>Therefore, you want the person
making the call to identify the issue in <br>
some way, and let the network apply a pre-determined policy to decide
what <br>
resources to apply to it. </blockquote><br>
<br>
Again, your statement sounds like to do not agree with the basic premise
of the work - that someone (i.e., the government) decides on who has
authorization to use the emergency level for calls. Note that the case of
someone standing on the corner reporting someone bleeding is normal 911
call, which is separate from authorized emergency calls (maybe from
someone in a position to report lots of people bleeding). I don't
understand the point that &quot;you want the person making the call to
identify the issue&quot;. SIP signaling obviously can't include a
description of the emergency situation so the network can work out
relative priorities with other descriptions. The originator has to
specify a level (maybe one of many) and the network has to compare
levels. <br>
. <br>
</font></blockquote><br>
A possible mechanism exists within the SIP R-P Header in the case you
argued against earlier in another context, the statement allowing the
response (200 OK) to set the level. Perhaps this is a case where I'm
personally allowed to make a call (but the network doesn't know it yet)
and I INVITE to a specific URI (the equvelent of today's GETS number),
this could have the receiving end change my original namespace.value up
to whatever it needs to be for that call in the cases where it's been
established that &quot;this call is should be routed to FEMA, or to the
RED CROSS, or to the LOCAL FIRE DEPARTMENT (only). SIP redirects and
Re-INVITES might be needed, but that should be unaware to the caller
making the call unless this is after some form of authentication is
required from that caller (pin # or something else).<br>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_-1819607037==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 14:56:10 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28354
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 14:56:10 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA00695;
	Thu, 14 Mar 2002 14:54:42 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA00660
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 14:54:40 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28298
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 14:54:37 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id LAA09585 for <ieprep@ietf.org>; Thu, 14 Mar 2002 11:54:07 -0800 (PST)
Message-Id: <4.1.20020314135009.02c432b8@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Thu, 14 Mar 2002 13:53:40 -0600
To: ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-Reply-To: <28.23639eb1.29bfb6d3@aol.com>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_-1818697349==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_-1818697349==_.ALT
Content-Type: text/plain; charset="us-ascii"

At 02:53 PM 3/12/2002 -0500, Mpierce1@aol.com wrote: 
>
> In a message dated 3/12/02 11:58:59 AM Eastern Standard Time,
> K.Carlberg@cs.ucl.ac.uk writes: 
>
>
>>
>>
>> > Right now, the Resource-header draft is not allowing expansion within a 
>> > defined namespace. 
>>
>> this is probably a question to James and Henning, but couldn't one 
>> define a new NameSpace?  like MLPPv2, or extended-MLPP?
>
>
>
> Yes, but the problem with allowing addition of values to an existing
> namespace is that current implementations wouldn't receognize the new values.
> Same problem with a new namespace, except worse. 
>
> MIke 



Aaahhh but the UA's could easily understand and support multiple namespaces
(with associated values) concurrently, couldn't they? Kinda like a dual stack,
maintaining communication with existing systems while migrating to newer
one(s). 



*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_-1818697349==_.ALT
Content-Type: text/html; charset="us-ascii"

<html>
At 02:53 PM 3/12/2002 -0500, Mpierce1@aol.com wrote: <br>
<font face="arial" size=2><blockquote type=cite cite>In a message dated
3/12/02 11:58:59 AM Eastern Standard Time, K.Carlberg@cs.ucl.ac.uk
writes: <br>
<br>
<br>
<blockquote type=cite cite><br>
&gt; Right now, the Resource-header draft is not allowing expansion
within a <br>
&gt; defined namespace. <br>
<br>
this is probably a question to James and Henning, but couldn't one <br>
define a new NameSpace?&nbsp; like MLPPv2, or
extended-MLPP?</blockquote><br>
<br>
Yes, but the problem with allowing addition of values to an existing
namespace is that current implementations wouldn't receognize the new
values. Same problem with a new namespace, except worse. <br>
<br>
MIke </font></blockquote><br>
<br>
Aaahhh but the UA's could easily understand and support multiple
namespaces (with associated values) concurrently, couldn't they? Kinda
like a dual stack, maintaining communication with existing systems while
migrating to newer one(s). <br>
<br>
<br>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_-1818697349==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 15:19:11 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29247
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 15:19:11 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA03247;
	Thu, 14 Mar 2002 15:17:36 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA03215
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 15:17:34 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29171
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 15:17:30 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2EKH2h06943;
	Thu, 14 Mar 2002 12:17:02 -0800 (PST)
Message-Id: <4.3.2.7.2.20020314113905.01aff260@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 14 Mar 2002 12:16:14 -0800
To: ieprep@ietf.org
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
Cc: ieprep@ietf.org
In-Reply-To: <4.1.20020314124529.02bf4730@localhost>
References: <200203141420.g2EEKaF14841@newdev.harvard.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 11:08 AM 3/14/2002, James M. Polk wrote:
>What Mike is suggesting is the potential for dropping < 1kbps from all 19 
>existing calls to make this 20th one exist (if all 19 calls have the same 
>precedence level and the new 20th call has a higher precedence level). 
>That instead of clearing one of the calls (preemption), merely degrading 
>all 19 existing calls by about 1 packet per second (of the 50pps for a 
>normal VoIP call).

that may be what Mike is suggesting, if there are two levels of drop 
precedence. However, he is proposing five levels of drop precedence. So he 
might be suggesting that we drop .5 KBPS from 18 of the calls, and 10 KBPS 
from the 19th.

Or, there is another scenario. But for that, let's not use your proposed 
example, let's use the DOE situation that someone described to me in 
private email:

>there are only 146 trunks in one of our DOE Large PBX's servicing several 
>thousand people.

By "trunk", I assume that they mean "T-1", which translates to 24*146=3504 
DS0s. We have a national 8 KHz national reference clock, so I have to 
assume that these DS0s are all synchronized - every 125 microseconds, we 
are going to see a burst of VoIP packets, and since VoIP packets contain 20 
msec samples, there will be 3504/(8*20)=21 packets in the average burst, 
with a randomized density distribution.

Now, let's assume that the routers are configured to start dropping packets 
when their EF queue fills to 15 packets. Every time a burst is sent, you 
have to assume that the entire burst will arrive and get placed into a 
queue, and then will get fed to the line over the next few milliseconds. 
What we're saying is that the last 6 packets in each burst (which may have 
some randomness or may be very predictable depending on the specific 
structure of that burst - were all the calls in the same T-1 frame, or did 
they come from different trunks served by different DSPs?) will be subject 
to loss, and if there happens to be a beat frequency in the burst arrival 
distribution so that successive large bursts can fill a queue and keep it 
full for several milliseconds, entire bursts can be lost.

When we say that they have differential drop thresholds, we're saying that 
some packets will be lost if they arrive at a queue which is 10 packets 
deep, another set might be lost if they arrive at a queue which is 15 
packets deep, and so on; life could get really interesting and either 
unpredictable or all too predictable.

Has anyone figured out yet why I'm not crazy about this as stated? I don't 
mind applying policy to calls in some way, but with five drop thresholds, 
saying that we can shoe-horn a 20th call into the line by dropping 1 KBPS 
evenly from each of the other 19 is simplistic. It's a *lot* messier than 
that. And even with call policy, embedding a DISA-specific call policy in 
every SIP implementation in the world doesn't make sense.

I'm serious about my offer to review simulations of the behavior. Before 
five levels of diff-serv drop precedence is standardized, someone really 
ought to try it out.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 15:29:41 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29673
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 15:29:41 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA04675;
	Thu, 14 Mar 2002 15:29:08 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA04641
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 15:29:06 -0500 (EST)
Received: from unity.cisco.com (evvbu-exchange.cisco.com [64.101.140.73])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29637
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 15:29:02 -0500 (EST)
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
Date: Thu, 14 Mar 2002 14:28:34 -0600
Message-ID: <E58F685255415A4EB2D4B8A02AEF32E3CC4287@evvbu-exchange.amer.unity.cisco.com>
Thread-Topic: [Ieprep] Comments on draft-carlberg-ieprep-framework
Thread-Index: AcHLlVj4uIHi4a06TfyBEjEHk+Ri2wAAGjdA
From: "Bob Bell" <rtbell@cisco.com>
To: "Fred Baker" <fred@cisco.com>
Cc: <ieprep@ietf.org>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id PAA04642
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

Fred - 

You bring up an interesting point, but even though the samples are sync'ed by the 8KHz clock at the trunk interface, the samples are serialized at the point where they are sent on the ethernet. Particularly for large trunk interfaces (e.g. 20 T1's per interface for example), this serialization should help limit the bursts of traffic you are describing. Now if all interfaces are single T1s, and you have 146 T1 spans, and thus 146 separate sources, this might be a bigger problem. By the way, usually the term "Trunk" in a telephony sense is used to indicate a single DS0 or equivalent (e.g. an analog voice circuit) not an aggregation of DS0's as in a T1 or a T3 span. This is not universal but in the example you showed, having a 10-1 person to trunk ratio is not uncommon, so I would suspect the term "trunk" does refer to DS0's in this case.

Bob



>-----Original Message-----
>From: Fred Baker [mailto:fred@cisco.com]
>Sent: Thursday, 14 March, 2002 1:16 PM
>To: ieprep@ietf.org
>Cc: ieprep@ietf.org
>Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
>
>
>At 11:08 AM 3/14/2002, James M. Polk wrote:
>>What Mike is suggesting is the potential for dropping < 1kbps 
>from all 19 
>>existing calls to make this 20th one exist (if all 19 calls 
>have the same 
>>precedence level and the new 20th call has a higher 
>precedence level). 
>>That instead of clearing one of the calls (preemption), 
>merely degrading 
>>all 19 existing calls by about 1 packet per second (of the 
>50pps for a 
>>normal VoIP call).
>
>that may be what Mike is suggesting, if there are two levels of drop 
>precedence. However, he is proposing five levels of drop 
>precedence. So he 
>might be suggesting that we drop .5 KBPS from 18 of the calls, 
>and 10 KBPS 
>from the 19th.
>
>Or, there is another scenario. But for that, let's not use 
>your proposed 
>example, let's use the DOE situation that someone described to me in 
>private email:
>
>>there are only 146 trunks in one of our DOE Large PBX's 
>servicing several 
>>thousand people.
>
>By "trunk", I assume that they mean "T-1", which translates to 
>24*146=3504 
>DS0s. We have a national 8 KHz national reference clock, so I have to 
>assume that these DS0s are all synchronized - every 125 
>microseconds, we 
>are going to see a burst of VoIP packets, and since VoIP 
>packets contain 20 
>msec samples, there will be 3504/(8*20)=21 packets in the 
>average burst, 
>with a randomized density distribution.
>
>Now, let's assume that the routers are configured to start 
>dropping packets 
>when their EF queue fills to 15 packets. Every time a burst is 
>sent, you 
>have to assume that the entire burst will arrive and get placed into a 
>queue, and then will get fed to the line over the next few 
>milliseconds. 
>What we're saying is that the last 6 packets in each burst 
>(which may have 
>some randomness or may be very predictable depending on the specific 
>structure of that burst - were all the calls in the same T-1 
>frame, or did 
>they come from different trunks served by different DSPs?) 
>will be subject 
>to loss, and if there happens to be a beat frequency in the 
>burst arrival 
>distribution so that successive large bursts can fill a queue 
>and keep it 
>full for several milliseconds, entire bursts can be lost.
>
>When we say that they have differential drop thresholds, we're 
>saying that 
>some packets will be lost if they arrive at a queue which is 
>10 packets 
>deep, another set might be lost if they arrive at a queue which is 15 
>packets deep, and so on; life could get really interesting and either 
>unpredictable or all too predictable.
>
>Has anyone figured out yet why I'm not crazy about this as 
>stated? I don't 
>mind applying policy to calls in some way, but with five drop 
>thresholds, 
>saying that we can shoe-horn a 20th call into the line by 
>dropping 1 KBPS 
>evenly from each of the other 19 is simplistic. It's a *lot* 
>messier than 
>that. And even with call policy, embedding a DISA-specific 
>call policy in 
>every SIP implementation in the world doesn't make sense.
>
>I'm serious about my offer to review simulations of the 
>behavior. Before 
>five levels of diff-serv drop precedence is standardized, 
>someone really 
>ought to try it out.
>
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 16:10:36 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00991
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 16:10:35 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA08251;
	Thu, 14 Mar 2002 16:10:11 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA08215
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 16:10:08 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00984
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 16:10:05 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2EL9bh19405;
	Thu, 14 Mar 2002 13:09:37 -0800 (PST)
Message-Id: <4.3.2.7.2.20020314130347.024f54a8@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 14 Mar 2002 13:09:22 -0800
To: "Bob Bell" <rtbell@cisco.com>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
Cc: <ieprep@ietf.org>
In-Reply-To: <E58F685255415A4EB2D4B8A02AEF32E3CC4287@evvbu-exchange.amer
 .unity.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 12:28 PM 3/14/2002, Bob Bell wrote:
>serialization should help limit the bursts of traffic you are describing

somewhat: it spreads them out. 20 T-1s on a single Ethernet implies that it 
is a 100 MBPS or 1 GHz Ethernet; it presents ~15 MBPS of traffic in each 
direction. But my point is that it presents it very rapidly - not 
necessarily simultaneously, but in very short order. depending on the 
characteristics of the data stream and what it is competing with, it can 
have pretty much the behavior I described.

>By the way, usually the term "Trunk" in a telephony sense is used to 
>indicate a single DS0 or equivalent

Thanks for that info. That changes the ratios in amplitude; I think the 
general point stands.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 16:14:51 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01088
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 16:14:51 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA08513;
	Thu, 14 Mar 2002 16:14:26 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA08470
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 16:14:24 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01075
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 16:14:22 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([171.71.21.235])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2ELDZh22548
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 13:13:35 -0800 (PST)
Message-Id: <4.3.2.7.2.20020314131235.023acf08@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Thu, 14 Mar 2002 13:13:23 -0800
To: ieprep@ietf.org
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-Reply-To: <4.3.2.7.2.20020314130347.024f54a8@mira-sjcm-4.cisco.com>
References: <E58F685255415A4EB2D4B8A02AEF32E3CC4287@evvbu-exchange.amer .unity.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 01:09 PM 3/14/2002, Fred Baker wrote:
>20 T-1s on a single Ethernet implies that it is a 100 MBPS or 1 GHz 
>Ethernet; it presents ~15 MBPS of traffic in each direction.

duh; get the math right, fred. 20*1.536 = ~30 MBPS in each direction...


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 16:17:48 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01170
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 16:17:48 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA08870;
	Thu, 14 Mar 2002 16:17:27 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA08836
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 16:17:25 -0500 (EST)
Received: from unity.cisco.com (evvbu-exchange.cisco.com [64.101.140.73])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01162
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 16:17:23 -0500 (EST)
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
Date: Thu, 14 Mar 2002 15:16:55 -0600
Message-ID: <E58F685255415A4EB2D4B8A02AEF32E3CC4288@evvbu-exchange.amer.unity.cisco.com>
Thread-Topic: [Ieprep] Comments on draft-carlberg-ieprep-framework
Thread-Index: AcHLnUmF8K3HLNCtR12GJwxFFifbbAAADAvQ
From: "Bob Bell" <rtbell@cisco.com>
To: "Fred Baker" <fred@cisco.com>, <ieprep@ietf.org>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id QAA08837
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

Fred - 

Even Einstein had trouble balancing his checkbook. ;->

Bob

>-----Original Message-----
>From: Fred Baker [mailto:fred@cisco.com]
>Sent: Thursday, 14 March, 2002 2:13 PM
>To: ieprep@ietf.org
>Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
>
>
>At 01:09 PM 3/14/2002, Fred Baker wrote:
>>20 T-1s on a single Ethernet implies that it is a 100 MBPS or 1 GHz 
>>Ethernet; it presents ~15 MBPS of traffic in each direction.
>
>duh; get the math right, fred. 20*1.536 = ~30 MBPS in each direction...
>
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 14 16:34:05 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01697
	for <ieprep-archive@odin.ietf.org>; Thu, 14 Mar 2002 16:34:05 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA10165;
	Thu, 14 Mar 2002 16:33:35 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA10131
	for <ieprep@ns.ietf.org>; Thu, 14 Mar 2002 16:33:33 -0500 (EST)
Received: from portal.east.saic.com (portal.east.saic.com [198.151.13.15])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id QAA01684
	for <ieprep@ietf.org>; Thu, 14 Mar 2002 16:33:31 -0500 (EST)
Received: from mclmx.saic.com by portal.east.saic.com
          via smtpd (for odin.ietf.org [132.151.1.176]) with SMTP; 14 Mar 2002 21:33:34 UT
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Thu, 14 Mar 2002 16:33:24 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.1.19) with SMTP id M2002031416332302312
 ; Thu, 14 Mar 2002 16:33:23 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <GSW2L72C>; Thu, 14 Mar 2002 16:34:01 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64231DD22@mcl-its-exs01.mail.saic.com>
From: "Garrity, Jack" <JOHN.J.GARRITY@saic.com>
To: "'Bob Bell'" <rtbell@cisco.com>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Levels of Priority
Date: Thu, 14 Mar 2002 16:33:22 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

Bob,

Sorry, my upcoming comment is not material to what's being discussed -- but
it is not illegal to drop (preempt) a call in progress.  It's simply a
custom.  Like other customs, it's sometimes violated, but not with such
frequency as to raise a commoner's hackles.

If anyone has a legal point, I'd be interested.  My lawyer can chat with
yours.

~~~ Jack ~~~
Work: 703/676-2358 | Fax: 703/848-0299
Home: 703/278-9250 | Cell: 703/508-0065
E-mail: garrityj@saic.com


-----Original Message-----
From: Bob Bell [mailto:rtbell@cisco.com]
Sent: Thursday, March 14, 2002 1:02 PM
To: Fred Baker; ieprep@ietf.org
Subject: RE: [Ieprep] Levels of Priority


Fred, et. al -

The comment I made was in relationship to MLPP in the phone system. The US
public phone systems are prohibited by law from dropping a currently active
call in favor of another call. MLPP was developed in the AUTOVON system (US
Military) were the prohibition did not exist. I think what they are trying
to accomplish is to preserve the QoS of a higher precidence call at the
expense of the QoS for a lower precidence call. I guess it would make a
level of sense in that you probably wouldn't want the voice call from the
President telling SAC to recall the bombers and stand down the missles to be
so badly garbled that the Commander of SAC thought the President told him to
commit the attack. ;-)

Bob

>-----Original Message-----
>From: Fred Baker [mailto:fred@cisco.com]
>Sent: Thursday, 14 March, 2002 10:39 AM
>To: ieprep@ietf.org
>Subject: RE: [Ieprep] Levels of Priority
>
>
>At 08:55 AM 3/14/2002, Bob Bell wrote:
>>Actually, they are sort of like packet dropping precidence in 
>that if a 
>>call of a higher precidence is trying to be established, it 
>can cause a 
>>lower precidence call to be terminated to free telephone resources 
>>(usually trunks) so that the higher precidence call can be completed.
>
>but that's not what they are doing. They are placing the 
>authorized call 
>anyway, and then if congestion sets in (which could be in the form of 
>synchronization between several data streams resulting from 
>the use of a 
>national 8 KHz clock, even in the presence of no other calls) dropping 
>packets from the unauthorized call, resulting in a reduction 
>of quality on 
>that call.
>
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 15 13:30:09 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06742
	for <ieprep-archive@odin.ietf.org>; Fri, 15 Mar 2002 13:30:09 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA16423;
	Fri, 15 Mar 2002 13:27:53 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA16393
	for <ieprep@optimus.ietf.org>; Fri, 15 Mar 2002 13:27:52 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06717
	for <ieprep@ietf.org>; Fri, 15 Mar 2002 13:27:49 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id KAA12039; Fri, 15 Mar 2002 10:27:18 -0800 (PST)
Message-Id: <4.1.20020315122217.02f88ba8@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Fri, 15 Mar 2002 12:26:51 -0600
To: ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: RE: [Ieprep] Levels of Priority
Cc: ieprep@ietf.org
In-Reply-To: <B8030EB94AF1D51196D70002A589D64231DD22@mcl-its-exs01.mail.
 saic.com>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_-1737506302==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_-1737506302==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 04:33 PM 3/14/2002 -0500, Garrity, Jack wrote:
>Bob,
>
>Sorry, my upcoming comment is not material to what's being discussed -- but
>it is not illegal to drop (preempt) a call in progress.  It's simply a
>custom.  Like other customs, it's sometimes violated, but not with such
>frequency as to raise a commoner's hackles.

Jack, then why must the FCC be petitioned for a waiver to the "preemption rule"
in order to have MLPP(like) services implemented in a non-private network just
as Verizon and VoiceStream had to last October and November?

I know this example is US specific, BTW. I don't know about other country's
rules regarding preemption on the public (state or country's owned) telco
networks -- but would be really interested in feedback here on this.

>
>If anyone has a legal point, I'd be interested.  My lawyer can chat with
>yours.
>
>~~~ Jack ~~~
>Work: 703/676-2358 | Fax: 703/848-0299
>Home: 703/278-9250 | Cell: 703/508-0065
>E-mail: garrityj@saic.com
>
>
>-----Original Message-----
>From: Bob Bell [mailto:rtbell@cisco.com]
>Sent: Thursday, March 14, 2002 1:02 PM
>To: Fred Baker; ieprep@ietf.org
>Subject: RE: [Ieprep] Levels of Priority
>
>
>Fred, et. al -
>
>The comment I made was in relationship to MLPP in the phone system. The US
>public phone systems are prohibited by law from dropping a currently active
>call in favor of another call. MLPP was developed in the AUTOVON system (US
>Military) were the prohibition did not exist. I think what they are trying
>to accomplish is to preserve the QoS of a higher precidence call at the
>expense of the QoS for a lower precidence call. I guess it would make a
>level of sense in that you probably wouldn't want the voice call from the
>President telling SAC to recall the bombers and stand down the missles to be
>so badly garbled that the Commander of SAC thought the President told him to
>commit the attack. ;-)
>
>Bob
>
>>-----Original Message-----
>>From: Fred Baker [mailto:fred@cisco.com]
>>Sent: Thursday, 14 March, 2002 10:39 AM
>>To: ieprep@ietf.org
>>Subject: RE: [Ieprep] Levels of Priority
>>
>>
>>At 08:55 AM 3/14/2002, Bob Bell wrote:
>>>Actually, they are sort of like packet dropping precidence in 
>>that if a 
>>>call of a higher precidence is trying to be established, it 
>>can cause a 
>>>lower precidence call to be terminated to free telephone resources 
>>>(usually trunks) so that the higher precidence call can be completed.
>>
>>but that's not what they are doing. They are placing the 
>>authorized call 
>>anyway, and then if congestion sets in (which could be in the form of 
>>synchronization between several data streams resulting from 
>>the use of a 
>>national 8 KHz clock, even in the presence of no other calls) dropping 
>>packets from the unauthorized call, resulting in a reduction 
>>of quality on 
>>that call.
>>
>>
>>_______________________________________________
>>Ieprep mailing list
>>Ieprep@ietf.org
>>https://www1.ietf.org/mailman/listinfo/ieprep
>>
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep

*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_-1737506302==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 04:33 PM 3/14/2002 -0500, Garrity, Jack wrote:</div>
<div>&gt;Bob,</div>
<div>&gt;</div>
<div>&gt;Sorry, my upcoming comment is not material to what's being
discussed -- but</div>
<div>&gt;it is not illegal to drop (preempt) a call in progress.&nbsp;
It's simply a</div>
<div>&gt;custom.&nbsp; Like other customs, it's sometimes violated, but
not with such</div>
<div>&gt;frequency as to raise a commoner's hackles.</div>
<br>
<div>Jack, then why must the FCC be petitioned for a waiver to the
&quot;preemption rule&quot; in order to have MLPP(like) services
implemented in a non-private network just as Verizon and VoiceStream had
to last October and November?</div>
<br>
<div>I know this example is US specific, BTW. I don't know about other
country's rules regarding preemption on the public (state or country's
owned) telco networks -- but would be really interested in feedback here
on this.</div>
<br>
<div>&gt;</div>
<div>&gt;If anyone has a legal point, I'd be interested.&nbsp; My lawyer
can chat with</div>
<div>&gt;yours.</div>
<div>&gt;</div>
<div>&gt;~~~ Jack ~~~</div>
<div>&gt;Work: 703/676-2358 | Fax: 703/848-0299</div>
<div>&gt;Home: 703/278-9250 | Cell: 703/508-0065</div>
<div>&gt;E-mail: garrityj@saic.com</div>
<div>&gt;</div>
<div>&gt;</div>
<div>&gt;-----Original Message-----</div>
<div>&gt;From: Bob Bell
[<a href="mailto:rtbell@cisco.com" EUDORA=AUTOURL>mailto:rtbell@cisco.com</a>]</div>
<div>&gt;Sent: Thursday, March 14, 2002 1:02 PM</div>
<div>&gt;To: Fred Baker; ieprep@ietf.org</div>
<div>&gt;Subject: RE: [Ieprep] Levels of Priority</div>
<div>&gt;</div>
<div>&gt;</div>
<div>&gt;Fred, et. al -</div>
<div>&gt;</div>
<div>&gt;The comment I made was in relationship to MLPP in the phone
system. The US</div>
<div>&gt;public phone systems are prohibited by law from dropping a
currently active</div>
<div>&gt;call in favor of another call. MLPP was developed in the AUTOVON
system (US</div>
<div>&gt;Military) were the prohibition did not exist. I think what they
are trying</div>
<div>&gt;to accomplish is to preserve the QoS of a higher precidence call
at the</div>
<div>&gt;expense of the QoS for a lower precidence call. I guess it would
make a</div>
<div>&gt;level of sense in that you probably wouldn't want the voice call
from the</div>
<div>&gt;President telling SAC to recall the bombers and stand down the
missles to be</div>
<div>&gt;so badly garbled that the Commander of SAC thought the President
told him to</div>
<div>&gt;commit the attack. ;-)</div>
<div>&gt;</div>
<div>&gt;Bob</div>
<div>&gt;</div>
<div>&gt;&gt;-----Original Message-----</div>
<div>&gt;&gt;From: Fred Baker
[<a href="mailto:fred@cisco.com" EUDORA=AUTOURL>mailto:fred@cisco.com</a>]</div>
<div>&gt;&gt;Sent: Thursday, 14 March, 2002 10:39 AM</div>
<div>&gt;&gt;To: ieprep@ietf.org</div>
<div>&gt;&gt;Subject: RE: [Ieprep] Levels of Priority</div>
<div>&gt;&gt;</div>
<div>&gt;&gt;</div>
<div>&gt;&gt;At 08:55 AM 3/14/2002, Bob Bell wrote:</div>
<div>&gt;&gt;&gt;Actually, they are sort of like packet dropping
precidence in </div>
<div>&gt;&gt;that if a </div>
<div>&gt;&gt;&gt;call of a higher precidence is trying to be established,
it </div>
<div>&gt;&gt;can cause a </div>
<div>&gt;&gt;&gt;lower precidence call to be terminated to free telephone
resources </div>
<div>&gt;&gt;&gt;(usually trunks) so that the higher precidence call can
be completed.</div>
<div>&gt;&gt;</div>
<div>&gt;&gt;but that's not what they are doing. They are placing the
</div>
<div>&gt;&gt;authorized call </div>
<div>&gt;&gt;anyway, and then if congestion sets in (which could be in
the form of </div>
<div>&gt;&gt;synchronization between several data streams resulting from
</div>
<div>&gt;&gt;the use of a </div>
<div>&gt;&gt;national 8 KHz clock, even in the presence of no other
calls) dropping </div>
<div>&gt;&gt;packets from the unauthorized call, resulting in a reduction
</div>
<div>&gt;&gt;of quality on </div>
<div>&gt;&gt;that call.</div>
<div>&gt;&gt;</div>
<div>&gt;&gt;</div>
<div>&gt;&gt;_______________________________________________</div>
<div>&gt;&gt;Ieprep mailing list</div>
<div>&gt;&gt;Ieprep@ietf.org</div>
<div>&gt;&gt;<a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA=AUTOURL>https://www1.ietf.org/mailman/listinfo/ieprep</a></div>
<div>&gt;&gt;</div>
<div>&gt;</div>
<div>&gt;_______________________________________________</div>
<div>&gt;Ieprep mailing list</div>
<div>&gt;Ieprep@ietf.org</div>
<div>&gt;<a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA=AUTOURL>https://www1.ietf.org/mailman/listinfo/ieprep</a></div>
<div>&gt;</div>
<div>&gt;_______________________________________________</div>
<div>&gt;Ieprep mailing list</div>
<div>&gt;Ieprep@ietf.org</div>
<div>&gt;<a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA=AUTOURL>https://www1.ietf.org/mailman/listinfo/ieprep</a></div>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_-1737506302==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 15 13:41:07 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06952
	for <ieprep-archive@odin.ietf.org>; Fri, 15 Mar 2002 13:41:07 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA17402;
	Fri, 15 Mar 2002 13:40:43 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA17365
	for <ieprep@optimus.ietf.org>; Fri, 15 Mar 2002 13:40:40 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06943
	for <ieprep@ietf.org>; Fri, 15 Mar 2002 13:40:37 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id KAA28164 for <ieprep@ietf.org>; Fri, 15 Mar 2002 10:40:07 -0800 (PST)
Message-Id: <4.1.20020315122811.0326fdf0@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Fri, 15 Mar 2002 12:39:40 -0600
To: ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
In-Reply-To: <4.3.2.7.2.20020314113905.01aff260@mira-sjcm-4.cisco.com>
References: <4.1.20020314124529.02bf4730@localhost>
 <200203141420.g2EEKaF14841@newdev.harvard.edu>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_-1736737537==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_-1736737537==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 12:16 PM 3/14/2002 -0800, Fred Baker wrote:

>that may be what Mike is suggesting, if there are two levels of drop 
>precedence. However, he is proposing five levels of drop precedence. So he 
>might be suggesting that we drop .5 KBPS from 18 of the calls, and 10 KBPS 
>from the 19th.

I'd call that cellphone quality -- just like I experience everyday. I believe
nearly everyone receives this .5kbps reduction on most calls as it is (as an
analogy), with some calls losing greater amounts of throughput. 

Another point that I've thought about but not proven mathematically
(simulation-wise) yet, what if adjacent priority levels lose at maximum 15% of
throughput relative to higher priority queues *that have packets queued*; that
if priority level #1's queue and #4's queue are the only queues with packets,
then this same 15% rule/algorithm applies as if they were adjacent.....  This
would reduce the relative loss from your lopsided example above by making a
curve loss fall-off, not linear, and not spiked.

  
*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_-1736737537==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 12:16 PM 3/14/2002 -0800, Fred Baker wrote:</div>
<br>
<div>&gt;that may be what Mike is suggesting, if there are two levels of
drop </div>
<div>&gt;precedence. However, he is proposing five levels of drop
precedence. So he </div>
<div>&gt;might be suggesting that we drop .5 KBPS from 18 of the calls,
and 10 KBPS </div>
<div>&gt;from the 19th.</div>
<br>
<div>I'd call that cellphone quality -- just like I experience everyday.
I believe nearly everyone receives this .5kbps reduction on most calls as
it is (as an analogy), with some calls losing greater amounts of
throughput. </div>
<br>
<div>Another point that I've thought about but not proven mathematically
(simulation-wise) yet, what if adjacent priority levels lose at maximum
15% of throughput relative to higher priority queues *that have packets
queued*; that if priority level #1's queue and #4's queue are the only
queues with packets, then this same 15% rule/algorithm applies as if they
were adjacent.....&nbsp; This would reduce the relative loss from your
lopsided example above by making a curve loss fall-off, not linear, and
not spiked.</div>
<br>
&nbsp;
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_-1736737537==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 15 14:24:23 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA07737
	for <ieprep-archive@odin.ietf.org>; Fri, 15 Mar 2002 14:24:23 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA19670;
	Fri, 15 Mar 2002 14:23:51 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA19633
	for <ieprep@optimus.ietf.org>; Fri, 15 Mar 2002 14:23:49 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA07719
	for <ieprep@ietf.org>; Fri, 15 Mar 2002 14:23:46 -0500 (EST)
Received: from FRED-W2K6.cisco.com (dhcp-128-107-163-29.cisco.com [128.107.163.29])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2FJNHh29913;
	Fri, 15 Mar 2002 11:23:17 -0800 (PST)
Message-Id: <4.3.2.7.2.20020315111338.01d230c8@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 15 Mar 2002 11:22:49 -0800
To: "James M. Polk" <jmpolk@cisco.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
Cc: ieprep@ietf.org
In-Reply-To: <4.1.20020315122811.0326fdf0@localhost>
References: <4.3.2.7.2.20020314113905.01aff260@mira-sjcm-4.cisco.com>
 <4.1.20020314124529.02bf4730@localhost>
 <200203141420.g2EEKaF14841@newdev.harvard.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 10:39 AM 3/15/2002, James M. Polk wrote:
> >might be suggesting that we drop .5 KBPS from 18 of the calls, and 10 KBPS
> >from the 19th.
>
>I'd call that cellphone quality -- just like I experience everyday. I 
>believe nearly everyone receives this .5kbps reduction on most calls as it 
>is (as an analogy), with some calls losing greater amounts of throughput.

You didn't finish reading what I said. .5 kbps loss probably is cell phone 
quality. 10 kbps loss isn't. And leaving the unlost traffic from those 
calls in the network impacts the calls that aren't losing traffic, by 
introducing jitter. If someone isn't going to be able to hold their 
conversation, just get them out of there.

What I'm telling you is that this is all a very interesting intellectual 
debate. Hard-earned experience with VoIP says that this is not something to 
be mandated because someone woke up with stars in his eyes. This stuff 
takes real analysis.

If you want to emulate MLPP, emulate MLPP. MLPP places or doesn't place 
calls based on some policy. Leave the per packet stuff out; it's a 
different service, and doesn't behave the same. Really. Really and truly.

And be smart. Just because twenty years ago we in our ignorance screwed up 
and put IP Precedence policy into IP (which we have since deprecated) 
doesn't mean we should make the same mistake again in SIP. Put a label in 
SIP that identifies the call, and present your credentials. But leave the 
policy (do we drop calls, or do we have a reserved set, or something else?) 
in the network. 


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 15 15:19:40 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA09271
	for <ieprep-archive@odin.ietf.org>; Fri, 15 Mar 2002 15:19:40 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA23124;
	Fri, 15 Mar 2002 15:18:14 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA23087
	for <ieprep@optimus.ietf.org>; Fri, 15 Mar 2002 15:18:12 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA09242
	for <ieprep@ietf.org>; Fri, 15 Mar 2002 15:18:07 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id MAA02000; Fri, 15 Mar 2002 12:17:37 -0800 (PST)
Message-Id: <4.1.20020315135346.02d3d5a8@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Fri, 15 Mar 2002 14:17:10 -0600
To: Fred Baker <fred@cisco.com>
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
Cc: ieprep@ietf.org
In-Reply-To: <4.3.2.7.2.20020315111338.01d230c8@mira-sjcm-4.cisco.com>
References: <4.1.20020315122811.0326fdf0@localhost>
 <4.3.2.7.2.20020314113905.01aff260@mira-sjcm-4.cisco.com>
 <4.1.20020314124529.02bf4730@localhost>
 <200203141420.g2EEKaF14841@newdev.harvard.edu>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_-1730887405==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_-1730887405==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 11:22 AM 3/15/2002 -0800, Fred Baker wrote:
>At 10:39 AM 3/15/2002, James M. Polk wrote:
>> >might be suggesting that we drop .5 KBPS from 18 of the calls, and 10 KBPS
>> >from the 19th.
>>
>>I'd call that cellphone quality -- just like I experience everyday. I 
>>believe nearly everyone receives this .5kbps reduction on most calls as it 
>>is (as an analogy), with some calls losing greater amounts of throughput.
>
>You didn't finish reading what I said. 

Yes I did

>.5 kbps loss probably is cell phone 
>quality. 10 kbps loss isn't. 

Have you traveled lately? For years I've been in a car with wildly varying
degrees of loss in continuous calls, yet I haven't thrown my cellphone away
because the mobility benefit outway my QoS frustration.

>And leaving the unlost traffic from those 
>calls in the network impacts the calls that aren't losing traffic, by 
>introducing jitter. If someone isn't going to be able to hold their 
>conversation, just get them out of there.

I agree with this completely and am attempting to find a way of *somehow*
creating session awareness without full nodal statefullness that will have
collateral affects few will want.

>
>What I'm telling you is that this is all a very interesting intellectual 
>debate. Hard-earned experience with VoIP says that this is not something to 
>be mandated because someone woke up with stars in his eyes. This stuff 
>takes real analysis.

I agree, numbers and charts are required

>
>If you want to emulate MLPP, emulate MLPP. 

I do, but not within this WG's area of eventual solution set. But MLPP does
give a sometimes useful model or analogy or example to reference for those
familiar with its workings.

>MLPP places or doesn't place 
>calls based on some policy. 

This is the session awareness that's required for both ETS and MLPP-like
services over IP.

>Leave the per packet stuff out; 

It's hard sometimes due to the lack of a session awareness aspect that seems to
be the obvious solution (if attainable).

>it's a 
>different service, and doesn't behave the same. Really. Really and truly.

I know, I know, I know and believe you. You have much more experience at this
than I and I can even see this to be true. It's getting there from here that's
tough sometimes.

>
>And be smart. Just because twenty years ago we in our ignorance screwed up 
>and put IP Precedence policy into IP (which we have since deprecated) 

I wish more today would realize this *is* the case instead of constantly
referring to it as if it (in IP) were still "there" and applicable.

>doesn't mean we should make the same mistake again in SIP. Put a label in 
>SIP that identifies the call, and present your credentials. 

I believe the SIP R-P ID does this to your stated expectations. I believe
policy is local in nature, but also believe policy can be grey in philosophy
for one's point of view (which yours is strong on BTW). I further believe there
should be hints for the short term as to general behaviors of labels as a
guidance -- with admittedly potentially better wording in those memos to that
affect.

>But leave the 
>policy (do we drop calls, or do we have a reserved set, or something else?) 
>in the network. 

... we need to make sure the network has the mechanisms in place to accomplish
each desired behavior in reasonable circumstances.



*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_-1730887405==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 11:22 AM 3/15/2002 -0800, Fred Baker wrote:</div>
<div>&gt;At 10:39 AM 3/15/2002, James M. Polk wrote:</div>
<div>&gt;&gt; &gt;might be suggesting that we drop .5 KBPS from 18 of the
calls, and 10 KBPS</div>
<div>&gt;&gt; &gt;from the 19th.</div>
<div>&gt;&gt;</div>
<div>&gt;&gt;I'd call that cellphone quality -- just like I experience
everyday. I </div>
<div>&gt;&gt;believe nearly everyone receives this .5kbps reduction on
most calls as it </div>
<div>&gt;&gt;is (as an analogy), with some calls losing greater amounts
of throughput.</div>
<div>&gt;</div>
<div>&gt;You didn't finish reading what I said. </div>
<br>
<div>Yes I did</div>
<br>
<div>&gt;.5 kbps loss probably is cell phone </div>
<div>&gt;quality. 10 kbps loss isn't. </div>
<br>
<div>Have you traveled lately? For years I've been in a car with wildly
varying degrees of loss in continuous calls, yet I haven't thrown my
cellphone away because the mobility benefit outway my QoS
frustration.</div>
<br>
<div>&gt;And leaving the unlost traffic from those </div>
<div>&gt;calls in the network impacts the calls that aren't losing
traffic, by </div>
<div>&gt;introducing jitter. If someone isn't going to be able to hold
their </div>
<div>&gt;conversation, just get them out of there.</div>
<br>
<div>I agree with this completely and am attempting to find a way of
*somehow* creating session awareness without full nodal statefullness
that will have collateral affects few will want.</div>
<br>
<div>&gt;</div>
<div>&gt;What I'm telling you is that this is all a very interesting
intellectual </div>
<div>&gt;debate. Hard-earned experience with VoIP says that this is not
something to </div>
<div>&gt;be mandated because someone woke up with stars in his eyes. This
stuff </div>
<div>&gt;takes real analysis.</div>
<br>
<div>I agree, numbers and charts are required</div>
<br>
<div>&gt;</div>
<div>&gt;If you want to emulate MLPP, emulate MLPP. </div>
<br>
<div>I do, but not within this WG's area of eventual solution set. But
MLPP does give a sometimes useful model or analogy or example to
reference for those familiar with its workings.</div>
<br>
<div>&gt;MLPP places or doesn't place </div>
<div>&gt;calls based on some policy. </div>
<br>
<div>This is the session awareness that's required for both ETS and
MLPP-like services over IP.</div>
<br>
<div>&gt;Leave the per packet stuff out; </div>
<br>
<div>It's hard sometimes due to the lack of a session awareness aspect
that seems to be the obvious solution (if attainable).</div>
<br>
<div>&gt;it's a </div>
<div>&gt;different service, and doesn't behave the same. Really. Really
and truly.</div>
<br>
<div>I know, I know, I know and believe you. You have much more
experience at this than I and I can even see this to be true. It's
getting there from here that's tough sometimes.</div>
<br>
<div>&gt;</div>
<div>&gt;And be smart. Just because twenty years ago we in our ignorance
screwed up </div>
<div>&gt;and put IP Precedence policy into IP (which we have since
deprecated) </div>
<br>
<div>I wish more today would realize this *is* the case instead of
constantly referring to it as if it (in IP) were still &quot;there&quot;
and applicable.</div>
<br>
<div>&gt;doesn't mean we should make the same mistake again in SIP. Put a
label in </div>
<div>&gt;SIP that identifies the call, and present your credentials.
</div>
<br>
<div>I believe the SIP R-P ID does this to your stated expectations. I
believe policy is local in nature, but also believe policy can be grey in
philosophy for one's point of view (which yours is strong on BTW). I
further believe there should be hints for the short term as to general
behaviors of labels as a guidance -- with admittedly potentially better
wording in those memos to that affect.</div>
<br>
<div>&gt;But leave the </div>
<div>&gt;policy (do we drop calls, or do we have a reserved set, or
something else?) </div>
<div>&gt;in the network. </div>
<br>
<div>... we need to make sure the network has the mechanisms in place to
accomplish each desired behavior in reasonable circumstances.</div>
<br>
<br>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_-1730887405==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 15 23:15:57 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA19258
	for <ieprep-archive@odin.ietf.org>; Fri, 15 Mar 2002 23:15:56 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id XAA15909;
	Fri, 15 Mar 2002 23:15:22 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id XAA15879
	for <ieprep@ns.ietf.org>; Fri, 15 Mar 2002 23:15:20 -0500 (EST)
Received: from imo-m03.mx.aol.com (imo-m03.mx.aol.com [64.12.136.6])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA19250
	for <ieprep@ietf.org>; Fri, 15 Mar 2002 23:15:15 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m03.mx.aol.com (mail_out_v32.5.) id c.16d.a59b7f1 (16110);
	Fri, 15 Mar 2002 23:14:39 -0500 (EST)
Received: from  web34.aolmail.aol.com (web34.aolmail.aol.com [205.188.222.10]) by air-id12.mx.aol.com (v83.45) with ESMTP id MAILINID122-0315231439; Fri, 15 Mar 2002 23:14:39 -0500
Date: Fri, 15 Mar 2002 23:14:39 EST
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
To: <fred@cisco.com>, <ieprep@ietf.org>
Cc: <ieprep@ietf.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Mailer: Unknown (No Version)
Message-ID: <16d.a59b7f1.29c420af@aol.com>
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

Fred Baker worte:

> By "trunk", I assume that they mean "T-1", which > translates to 24*146=3504 
> DS0s. We have a national 8 KHz national reference ? clock, so I have to 
? assume that these DS0s are all synchronized - every > 125 microseconds, we 
> are going to see a burst of VoIP packets, and since > VoIP packets contain 20 
> msec samples, there will be 3504/(8*20)=21 packets in > the average burst, 
> with a randomized density distribution.

I don't know where you get this view of the existing network. When someone says "trunk" you have to assume they mean a facility to carry exactly one telephone call, not a T-1. Incidentally, a T-1 carries exactly 24 DS0's, normally used for 24 "trunks". (I don't know where the mulriplier of 146 above comes from.) I am unaware of any 8khz national standard which keeps DS0 sychronized. Only the bit rate (1.544Mhz is synced). So there is nothing that would cause bursts of packets.

Mike

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 17 00:38:42 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA19152
	for <ieprep-archive@odin.ietf.org>; Sun, 17 Mar 2002 00:38:42 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id AAA24570;
	Sun, 17 Mar 2002 00:37:30 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id AAA24538
	for <ieprep@optimus.ietf.org>; Sun, 17 Mar 2002 00:37:28 -0500 (EST)
Received: from imo-r02.mx.aol.com (imo-r02.mx.aol.com [152.163.225.98])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA19135
	for <ieprep@ietf.org>; Sun, 17 Mar 2002 00:37:25 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r02.mx.aol.com (mail_out_v32.5.) id l.124.d72b016 (15874)
	 for <ieprep@ietf.org>; Sun, 17 Mar 2002 00:36:52 -0500 (EST)
Received: from  mow-m29 (mow-m29.webmail.aol.com [64.12.137.6]) by air-id07.mx.aol.com (v83.45) with ESMTP id MAILINID71-0317003652; Sun, 17 Mar 2002 00:36:52 -0500
Date: Sun, 17 Mar 2002 00:36:51 EST
To: <ieprep@ietf.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Mailer: Unknown (No Version)
Message-ID: <124.d72b016.29c58574@aol.com>
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Levels of priority and comments on ieprep framework
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

To all,

Guess it's time I said what "Mike meant", since there have been several e-mails on this subject.

First, it seems that some issues need to be put to bed for good.

1. I never suggested, nor do I think, that we should resurrect the per packet priority marking of RFC 781 or define any new per packet priority forwarding scheme.

2. The per session priority addressed in the Resource-priority draft is exactly what it says, session handling. The DoD requirement for five precedence levels is satisfied by this draft.

3. The requirement as spelled out in draft-pierce-sipping-assured-service, is to provide certain preferential treatments to a call (session) which is marked with a higher precedence. That may include preempting one or more lower precedence calls or taking resources from them. The requirement is that the attempt to set up a higher precedence call can not be blocked by the presence of lower precedence calls. So far, nothing in the requirements addresses QoS.

4. Any suggestions for per packet marking or handling is a potential solution. I've realized from Fred's comments that I should not have been suggesting the use of AF. EF is the appropriate mechanism.

Jim Polk described a case which demonstrates my suggestion for a solution using multiple drop precedences (with a minor difference).

First, we are talking of a case in which all other traffic (data, etc) has already been pushed aside, so only voice traffic is left, all using EF. In Polk's example, with 19 voice calls at the lower (session) precedence level, each using 80kbit/s on a T-1, when a new call at a higher level comes along, it can steal resources from the others without actually preempting any.

Perhaps RSVP could be used to modify the reservation of each of the existing 19 calls to reduce each by a uniform amount in order to make room to establish the 20th path.

Using Diff-serv instead, this requires multiple drop precedences (for EF), just as already defined for AF. The 20th call, at a higher precedence, is setup even though it exceeds the available bandwidth, and its packets are marked with the lower drop precedence than the other 19. Whenever congestion occurs, based on queue length or whatever (as defined in AF), individual packets of the other 19 calls are dropped. Packets for the one higher precedence call (lower drop precedence) are not dropped. Yes, the result is that the higher precedence call gets better QoS, but that is not the requirement.

(The deviation from Polk's example is that there is no requirement to drop equal numbers of packets from each of the lower precedence calls. If a disproportionate amount are dropped from one call because of the specific packet sequencing, that is acceptable. If an implementation can figure out a way to uniformly distribute the loss when using Diff-serv, that's fine.)

If there are more than two levels of drop precedence, it is always the highest which is dropped. There is no added complexity. It works exactly as defined today for AF. I have not proposed any change to this. The open question is how many drop precedence levels are needed. At least three as currently defined for each AF class makes sense, but we would prefer 5 to allow direct mapping from the five session precedence levels to the packet drop precedence levels.

Mike Pierce

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 17 17:12:30 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA11940
	for <ieprep-archive@odin.ietf.org>; Sun, 17 Mar 2002 17:12:29 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA12152;
	Sun, 17 Mar 2002 17:07:49 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA12119
	for <ieprep@optimus.ietf.org>; Sun, 17 Mar 2002 17:07:46 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA11834
	for <ieprep@ietf.org>; Sun, 17 Mar 2002 17:07:41 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id PAA07550
	for <ieprep@ietf.org>; Sun, 17 Mar 2002 15:05:52 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA15B8
          for <ieprep@ietf.org>; Sun, 17 Mar 2002 15:07:42 -0700
Received: FROM agcs.com BY mknaomi ; Sun Mar 17 15:07:40 2002 -0700
Message-ID: <3C9513AE.ED58189C@agcs.com>
Date: Sun, 17 Mar 2002 15:07:43 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "James M. Polk" <jmpolk@cisco.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] Levels of Priority
References: <4.1.20020315122217.02f88ba8@localhost>
Content-Type: multipart/mixed;
 boundary="------------589C8DBDD5E0B17E266A5F8F"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------589C8DBDD5E0B17E266A5F8F
Content-Type: multipart/alternative;
 boundary="------------9D72B771FCAD646168629A17"


--------------9D72B771FCAD646168629A17
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Folks,

Jack may very well be correct that there is no "law" against preempting
a call, but as I understand the PSTN process, each carrier has to file
tariffs with the controlling PUC(s) for that carrier that are pretty
detailed into how the service is to be provided.  I suspect that
carriers would point to such tariffs, as well as the ever present fear
of litigation, as reasons that they can not preempt calls.

So it may be less than "law", but far more than "custom".

"James M. Polk" wrote:

> At 04:33 PM 3/14/2002 -0500, Garrity, Jack wrote:>Bob,>>Sorry, my
> upcoming comment is not material to what's being discussed -- but>it
> is not illegal to drop (preempt) a call in progress.  It's simply
> a>custom.  Like other customs, it's sometimes violated, but not with
> such>frequency as to raise a commoner's hackles.Jack, then why must
> the FCC be petitioned for a waiver to the "preemption rule" in order
> to have MLPP(like) services implemented in a non-private network just
> as Verizon and VoiceStream had to last October and November?I know
> this example is US specific, BTW. I don't know about other country's
> rules regarding preemption on the public (state or country's owned)
> telco networks -- but would be really interested in feedback here on
> this.>>If anyone has a legal point, I'd be interested.  My lawyer can
> chat with>yours.>>~~~ Jack ~~~>Work: 703/676-2358 | Fax:
> 703/848-0299>Home: 703/278-9250 | Cell: 703/508-0065>E-mail:
> garrityj@saic.com>>>-----Original Message----->From: Bob Bell
> [mailto:rtbell@cisco.com]>Sent: Thursday, March 14, 2002 1:02 PM>To:
> Fred Baker; ieprep@ietf.org>Subject: RE: [Ieprep] Levels of
> Priority>>>Fred, et. al ->>The comment I made was in relationship to
> MLPP in the phone system. The US>public phone systems are prohibited
> by law from dropping a currently active>call in favor of another call.
> MLPP was developed in the AUTOVON system (US>Military) were the
> prohibition did not exist. I think what they are trying>to accomplish
> is to preserve the QoS of a higher precidence call at the>expense of
> the QoS for a lower precidence call. I guess it would make a>level of
> sense in that you probably wouldn't want the voice call from
> the>President telling SAC to recall the bombers and stand down the
> missles to be>so badly garbled that the Commander of SAC thought the
> President told him to>commit the attack. ;-)>>Bob>>>-----Original
> Message----->>From: Fred Baker [mailto:fred@cisco.com]>>Sent:
> Thursday, 14 March, 2002 10:39 AM>>To: ieprep@ietf.org>>Subject: RE:
> [Ieprep] Levels of Priority>>>>>>At 08:55 AM 3/14/2002, Bob Bell
> wrote:>>>Actually, they are sort of like packet dropping precidence
> in>>that if a>>>call of a higher precidence is trying to be
> established, it>>can cause a>>>lower precidence call to be terminated
> to free telephone resources>>>(usually trunks) so that the higher
> precidence call can be completed.>>>>but that's not what they are
> doing. They are placing the>>authorized call>>anyway, and then if
> congestion sets in (which could be in the form of>>synchronization
> between several data streams resulting from>>the use of a>>national 8
> KHz clock, even in the presence of no other calls) dropping>>packets
> from the unauthorized call, resulting in a reduction>>of quality
> on>>that
> call.>>>>>>_______________________________________________>>Ieprep
> mailing
> list>>Ieprep@ietf.org>>https://www1.ietf.org/mailman/listinfo/ieprep>>>>_______________________________________________>Ieprep
> mailing
> list>Ieprep@ietf.org>https://www1.ietf.org/mailman/listinfo/ieprep>>_______________________________________________>Ieprep
> mailing
> list>Ieprep@ietf.org>https://www1.ietf.org/mailman/listinfo/ieprep
>                 *************************************
>  "People generally demand more respect for their own rights than they
>                    are willing to allow for others"
>
>  James M. Polk
> Consulting Engineer
> Office of the CTO
>
> Cisco Systems
> 2200 East President George Bush Turnpike
> Richardson, TX  75082 USA
> w) 972.813.5208
> f)  972.813.5280
> www.cisco.com

--------------9D72B771FCAD646168629A17
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Folks,
<p>Jack may very well be correct that there is no "law" against preempting
a call, but as I understand the PSTN process, each carrier has to file
tariffs with the controlling PUC(s) for that carrier that are pretty detailed
into how the service is to be provided.&nbsp; I suspect that carriers would
point to such tariffs, as well as the ever present fear of litigation,
as reasons that they can not preempt calls.
<p>So it may be less than "law", but far more than "custom".
<p>"James M. Polk" wrote:
<blockquote TYPE=CITE>At 04:33 PM 3/14/2002 -0500, Garrity, Jack wrote:>Bob,>>Sorry,
my upcoming comment is not material to what's being discussed -- but>it
is not illegal to drop (preempt) a call in progress.&nbsp; It's simply
a>custom.&nbsp; Like other customs, it's sometimes violated, but not with
such>frequency as to raise a commoner's hackles.Jack, then why must the
FCC be petitioned for a waiver to the "preemption rule" in order to have
MLPP(like) services implemented in a non-private network just as Verizon
and VoiceStream had to last October and November?I know this example is
US specific, BTW. I don't know about other country's rules regarding preemption
on the public (state or country's owned) telco networks -- but would be
really interested in feedback here on this.>>If anyone has a legal point,
I'd be interested.&nbsp; My lawyer can chat with>yours.>>~~~ Jack ~~~>Work:
703/676-2358 | Fax: 703/848-0299>Home: 703/278-9250 | Cell: 703/508-0065>E-mail:
garrityj@saic.com>>>-----Original Message----->From: Bob Bell [<a href="mailto:rtbell@cisco.com" EUDORA="AUTOURL">mailto:rtbell@cisco.com</a>]>Sent:
Thursday, March 14, 2002 1:02 PM>To: Fred Baker; ieprep@ietf.org>Subject:
RE: [Ieprep] Levels of Priority>>>Fred, et. al ->>The comment I made was
in relationship to MLPP in the phone system. The US>public phone systems
are prohibited by law from dropping a currently active>call in favor of
another call. MLPP was developed in the AUTOVON system (US>Military) were
the prohibition did not exist. I think what they are trying>to accomplish
is to preserve the QoS of a higher precidence call at the>expense of the
QoS for a lower precidence call. I guess it would make a>level of sense
in that you probably wouldn't want the voice call from the>President telling
SAC to recall the bombers and stand down the missles to be>so badly garbled
that the Commander of SAC thought the President told him to>commit the
attack. ;-)>>Bob>>>-----Original Message----->>From: Fred Baker [<a href="mailto:fred@cisco.com" EUDORA="AUTOURL">mailto:fred@cisco.com</a>]>>Sent:
Thursday, 14 March, 2002 10:39 AM>>To: ieprep@ietf.org>>Subject: RE: [Ieprep]
Levels of Priority>>>>>>At 08:55 AM 3/14/2002, Bob Bell wrote:>>>Actually,
they are sort of like packet dropping precidence in>>that if a>>>call of
a higher precidence is trying to be established, it>>can cause a>>>lower
precidence call to be terminated to free telephone resources>>>(usually
trunks) so that the higher precidence call can be completed.>>>>but that's
not what they are doing. They are placing the>>authorized call>>anyway,
and then if congestion sets in (which could be in the form of>>synchronization
between several data streams resulting from>>the use of a>>national 8 KHz
clock, even in the presence of no other calls) dropping>>packets from the
unauthorized call, resulting in a reduction>>of quality on>>that call.>>>>>>_______________________________________________>>Ieprep
mailing list>>Ieprep@ietf.org>><a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA="AUTOURL">https://www1.ietf.org/mailman/listinfo/ieprep</a>>>>>_______________________________________________>Ieprep
mailing list>Ieprep@ietf.org><a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA="AUTOURL">https://www1.ietf.org/mailman/listinfo/ieprep</a>>>_______________________________________________>Ieprep
mailing list>Ieprep@ietf.org><a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA="AUTOURL">https://www1.ietf.org/mailman/listinfo/ieprep</a>
<center>*************************************
<br>"People generally demand more respect for their own rights than they
are willing to allow for others"</center>

<br>&nbsp;James M. Polk
<br>Consulting Engineer
<br>Office of the CTO
<p>Cisco Systems
<br>2200 East President George Bush Turnpike
<br>Richardson, TX&nbsp; 75082 USA
<br>w) 972.813.5208
<br>f)&nbsp; 972.813.5280
<br><a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></blockquote>
</html>

--------------9D72B771FCAD646168629A17--

--------------589C8DBDD5E0B17E266A5F8F
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------589C8DBDD5E0B17E266A5F8F--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 17 17:37:59 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA12340
	for <ieprep-archive@odin.ietf.org>; Sun, 17 Mar 2002 17:37:59 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA12611;
	Sun, 17 Mar 2002 17:28:26 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA12574
	for <ieprep@optimus.ietf.org>; Sun, 17 Mar 2002 17:28:24 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA12144
	for <ieprep@ietf.org>; Sun, 17 Mar 2002 17:28:21 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id PAA08106
	for <ieprep@ietf.org>; Sun, 17 Mar 2002 15:26:33 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA181F
          for <ieprep@ietf.org>; Sun, 17 Mar 2002 15:28:22 -0700
Received: FROM agcs.com BY mknaomi ; Sun Mar 17 15:28:19 2002 -0700
Message-ID: <3C951886.FB91D7CD@agcs.com>
Date: Sun, 17 Mar 2002 15:28:22 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Bob Bell <rtbell@cisco.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
References: <E58F685255415A4EB2D4B8A02AEF32E3CC4284@evvbu-exchange.amer.unity.cisco.com>
Content-Type: multipart/mixed;
 boundary="------------3A5ED5BC259D66638E200C76"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------3A5ED5BC259D66638E200C76
Content-Type: multipart/alternative;
 boundary="------------FB6094620DE807BA598587B7"


--------------FB6094620DE807BA598587B7
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Bob,

I am a little confused about what you have written.

I agree that should more and more calls just be "shoehorned" in, then
the QoS would continue to go down IF we were just looking at a SNAPSHOT
of the network where no calls ever end.

However, if we can assume that the vast bulk of the traffic is not
emergency calls, can we not also logically assume that as such non-
emergency calls individually end, the bandwidth would revert back to the
network. The existing admission policy would not allow additional non
emergency calls to be established.

So, my question is would not the QoS return to normal as soon as an
equal number of non emergency calls have completed their sessions?

It seems to me that the "overlooking" of the admission policy for
emergency calls would at the most only temporarily impact the QoS of the
network. The normal QoS would be quickly restored.  I would think that
QoS would not be more than momentarily impacted unless and until the
emergency traffic has completely displaced all of the multitude of POTS
traffic.

Am I missing something?



Bob Bell wrote:

>  James, et. al. - In the stated example, loosing 1% of the stream
> would probably not cause a great deal of distortion so as to render
> all conversations unusable. However, if you continue the process of
> stealing chunks of bandwidth, as more calls are "shoehorned" through
> the restriction point, you will quickly reach that condition. This is
> the reason I would believe that associating a precidence level with a
> call and using policy to terminate a lower precidence level session in
> favor of a higher precidence level would be more "fair". In this way,
> only the lowest precidence session would suffer, not all sessions.
> Even if you were to remove bandwidth from sessions proportinate to the
> their precidence (e.g. the lower the precidence, the more bandwidth
> could be stolen), unless some method of reducing their "billing" for
> this reduced clarity call is in place, then persons would be "paying"
> for services that they could not get. This is asking for trouble in
> the form of govenrment intervention and law suites. At least, if the
> lower precidence call is terminated, the "billing" also ends.Bob
>
>      -----Original Message-----
>      From: James Polk
>      Sent: Thursday, 14 March, 2002 12:08 PM
>      To: ieprep@ietf.org
>      Subject: Re: [Ieprep] Comments on
>      draft-carlberg-ieprep-framework
>
>      At 09:20 AM 3/14/2002 -0500, Scott  Bradner wrote:>>Mike
>      sez:>  The reference I made about "It's what the US military
>      does" is a>  basic requirement for voice service
>      operation.>>we are talking past each other>>I maintain that
>      N-levels of priority is a solution not a requirement>>the
>      question that needs to be answered is what functions are
>      needed to>be supported - not how to do them>>specifically -
>      what problem is thought to be solved by N levels of
>      priority?>>is it that there should be a different
>      probibility for>calls to be completed based on who you are
>      or what you are doing?>>is it that the quality of the call
>      should depends on who you>are and what you are doing?>>or is
>      it someting elseLet's pose a strict set of circumstances to
>      address part of this thread that's not answered yet
>      explicitly (to my understanding - and others I believe)Let
>      assume RSVP worked completely everywhere and between every
>      phone in the Internet. A VoIP call using RSVP would work
>      almost identically to a circuit switched call in taking up a
>      defined amount of bandwidth through each transit interface
>      between two phones. Lets say that all VoIP calls push aside
>      all other (data) traffic always with no hesitation. In a
>      case where each call is 80kbps traversing a T1 (1.544Mbps),
>      19 calls are possible at any one time. Even though there is
>      56kbps of bandwidth remaining, that 20th call cannot be
>      completed if it requires an 80kbps Reservation across this
>      link.This is an example where codecs aren't changeable -- I
>      want a specific analysis to this scenarioWhat Mike is
>      suggesting is the potential for dropping < 1kbps from all 19
>      existing calls to make this 20th one exist (if all 19 calls
>      have the same precedence level and the new 20th call has a
>      higher precedence level). That instead of clearing one of
>      the calls (preemption), merely degrading all 19 existing
>      calls by about 1 packet per second (of the 50pps for a
>      normal VoIP call).Is this considered unacceptable?One
>      existing example/analogy:  even though the Cellular industry
>      isn't packet-based, I believe most calls remain *up* AND
>      loose more than 1% of the media signal during a call (if not
>      much more).
>       >>Scott>>_
>      _____________________________________________>Ieprep mailing
>      list>Ieprep@ietf.org>https://www1.ietf.org/mailman/listinfo/ieprep
>                 *************************************
>       "People generally demand more respect for their own rights
>               than they are willing to allow for others"
>
>       James M. Polk
>      Consulting Engineer
>      Office of the CTO
>
>      Cisco Systems
>      2200 East President George Bush Turnpike
>      Richardson, TX  75082 USA
>      w) 972.813.5208
>      f)  972.813.5280
>      www.cisco.com
>

--------------FB6094620DE807BA598587B7
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Bob,
<p>I am a little confused about what you have written.
<p>I agree that should more and more calls just be "shoehorned" in, then
the QoS would continue to go down IF we were just looking at a SNAPSHOT
of the network where no calls ever end.
<p>However, if we can assume that the vast bulk of the traffic is not emergency
calls, can we not also logically assume that as such non- emergency calls
individually end, the bandwidth would revert back to the network. The existing
admission policy would not allow additional non emergency calls to be established.
<p>So, my question is would not the QoS return to normal as soon as an
equal number of non emergency calls have completed their sessions?
<p>It seems to me that the "overlooking" of the admission policy for emergency
calls would at the most only temporarily impact the QoS of the network.
The normal QoS would be quickly restored.&nbsp; I would think that QoS
would not be more than momentarily impacted unless and until the emergency
traffic has completely displaced all of the multitude of POTS traffic.
<p>Am I missing something?
<br>&nbsp;
<br>&nbsp;
<p>Bob Bell wrote:
<blockquote TYPE=CITE>&nbsp;<span class=225321319-14032002><font face="Arial"><font color="#0000FF"><font size=-1>James,
et. al. -&nbsp;</font></font></font></span><span class=225321319-14032002></span><span class=225321319-14032002><font face="Arial"><font color="#0000FF"><font size=-1>In
the stated example, loosing 1% of the stream would probably not cause a
great deal of distortion so as to render all conversations unusable. However,
if you continue the process of stealing chunks of bandwidth, as more calls
are "shoehorned" through the restriction point, you will quickly reach
that condition. This is the reason I would believe that associating a precidence
level with a call and using policy to terminate a lower precidence level
session in favor of a higher precidence level would be more "fair". In
this way, only the lowest precidence session would suffer, not all sessions.
Even if you were to remove bandwidth from sessions proportinate to the
their precidence (e.g. the lower the precidence, the more bandwidth could
be stolen), unless some method of reducing their "billing" for this reduced
clarity call is in place, then persons would be "paying" for services that
they could not get. This is asking for trouble in the form of govenrment
intervention and law suites. At least, if the lower precidence call is
terminated, the "billing" also ends.</font></font></font></span><span class=225321319-14032002></span><span class=225321319-14032002><font face="Arial"><font color="#0000FF"><font size=-1>Bob</font></font></font></span>
<blockquote 
style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px solid">
<div class="OutlookMessageHeader" dir="ltr"><font face="Tahoma"><font size=-1>-----Original
Message-----</font></font>
<br><font face="Tahoma"><font size=-1><b>From:</b> James Polk</font></font>
<br><font face="Tahoma"><font size=-1><b>Sent:</b> Thursday, 14 March,
2002 12:08 PM</font></font>
<br><font face="Tahoma"><font size=-1><b>To:</b> ieprep@ietf.org</font></font>
<br><font face="Tahoma"><font size=-1><b>Subject:</b> Re: [Ieprep] Comments
on draft-carlberg-ieprep-framework</font></font>
<br>&nbsp;</div>
At 09:20 AM 3/14/2002 -0500, Scott&nbsp; Bradner wrote:>>Mike sez:>&nbsp;
The reference I made about "It's what the US military does" is a>&nbsp;
basic requirement for voice service operation.>>we are talking past each
other>>I maintain that N-levels of priority is a solution not a requirement>>the
question that needs to be answered is what functions are needed to>be supported
- not how to do them>>specifically - what problem is thought to be solved
by N levels of priority?>><X-TAB></X-TAB>is it that there should be a different
probibility for><X-TAB></X-TAB>calls to be completed based on who you are
or what you are doing?>><X-TAB></X-TAB>is it that the quality of the call
should depends on who you><X-TAB></X-TAB>are and what you are doing?>><X-TAB></X-TAB>or
is it someting elseLet's pose a strict set of circumstances to address
part of this thread that's not answered yet explicitly (to my understanding
- and others I believe)Let assume RSVP worked completely everywhere and
between every phone in the Internet. A VoIP call using RSVP would work
almost identically to a circuit switched call in taking up a defined amount
of bandwidth through each transit interface between two phones. Lets say
that all VoIP calls push aside all other (data) traffic always with no
hesitation. In a case where each call is 80kbps traversing a T1 (1.544Mbps),
19 calls are possible at any one time. Even though there is 56kbps of bandwidth
remaining, that 20th call cannot be completed if it requires an 80kbps
Reservation across this link.This is an example where codecs aren't changeable
-- I want a specific analysis to this scenarioWhat Mike is suggesting is
the potential for dropping &lt; 1kbps from all 19 existing calls to make
this 20th one exist (if all 19 calls have the same precedence level and
the new 20th call has a higher precedence level). That instead of clearing
one of the calls (preemption), merely degrading all 19 existing calls by
about 1 packet per second (of the 50pps for a normal VoIP call).Is this
considered unacceptable?One existing example/analogy:&nbsp; even though
the Cellular industry isn't packet-based, I believe most calls remain *up*
AND loose more than 1% of the media signal during a call (if not much more).
<br>&nbsp;>>Scott>>_______________________________________________>Ieprep
mailing list>Ieprep@ietf.org><a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA="AUTOURL">https://www1.ietf.org/mailman/listinfo/ieprep</a>
<center>*************************************
<br>"People generally demand more respect for their own rights than they
are willing to allow for others"</center>

<br>&nbsp;James M. Polk
<br>Consulting Engineer
<br>Office of the CTO
<p>Cisco Systems
<br>2200 East President George Bush Turnpike
<br>Richardson, TX&nbsp; 75082 USA
<br>w) 972.813.5208
<br>f)&nbsp; 972.813.5280
<br><a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></blockquote>
</blockquote>
</html>

--------------FB6094620DE807BA598587B7--

--------------3A5ED5BC259D66638E200C76
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------3A5ED5BC259D66638E200C76--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 17 18:19:45 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA12970
	for <ieprep-archive@odin.ietf.org>; Sun, 17 Mar 2002 18:19:44 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA14527;
	Sun, 17 Mar 2002 18:14:00 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA14495
	for <ieprep@optimus.ietf.org>; Sun, 17 Mar 2002 18:13:58 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA12939
	for <ieprep@ietf.org>; Sun, 17 Mar 2002 18:13:54 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id PAA05841 for <ieprep@ietf.org>; Sun, 17 Mar 2002 15:13:27 -0800 (PST)
Message-Id: <4.1.20020317170728.01620e70@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Sun, 17 Mar 2002 17:12:59 -0600
To: ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] Levels of Priority
In-Reply-To: <3C9513AE.ED58189C@agcs.com>
References: <4.1.20020315122217.02f88ba8@localhost>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_7244326==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_7244326==_.ALT
Content-Type: text/plain; charset="us-ascii"

At 03:07 PM 3/17/2002 -0700, Stuart (Stu) Goldman wrote: 
>
> Folks, 
>
> Jack may very well be correct that there is no "law" against preempting a
> call, but as I understand the PSTN process, each carrier has to file tariffs
> with the controlling PUC(s) for that carrier that are pretty detailed into
> how the service is to be provided.  I suspect that carriers would point to
> such tariffs, as well as the ever present fear of litigation, as reasons that
> they can not preempt calls. 
>
> So it may be less than "law", but far more than "custom". 


Then why would the petitions go directly to the FCC from Verizon and
VoiceStream? I've got copies of the petitions and press releases... forwarded
them to Scott Bradner last December to show this to be the case based on my
comment in the mike at the BOF in SLC.

I'm not as much questioning you, as stating what I've seen and have regarding
this very topic, the current evidence is inconsistent with what the two of you
state....  :-/


*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_7244326==_.ALT
Content-Type: text/html; charset="us-ascii"

<html>
At 03:07 PM 3/17/2002 -0700, Stuart (Stu) Goldman wrote: <br>
<blockquote type=cite cite>Folks, <br>
<br>
Jack may very well be correct that there is no &quot;law&quot; against
preempting a call, but as I understand the PSTN process, each carrier has
to file tariffs with the controlling PUC(s) for that carrier that are
pretty detailed into how the service is to be provided.&nbsp; I suspect
that carriers would point to such tariffs, as well as the ever present
fear of litigation, as reasons that they can not preempt calls. <br>
<br>
So it may be less than &quot;law&quot;, but far more than
&quot;custom&quot;. <br>
</blockquote><br>
Then why would the petitions go directly to the FCC from Verizon and
VoiceStream? I've got copies of the petitions and press releases...
forwarded them to Scott Bradner last December to show this to be the case
based on my comment in the mike at the BOF in SLC.<br>
<br>
I'm not as much questioning you, as stating what I've seen and have
regarding this very topic, the current evidence is inconsistent with what
the two of you state....&nbsp; :-/<br>
<br>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_7244326==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 08:36:42 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04908
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 08:36:42 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA02943;
	Mon, 18 Mar 2002 08:34:30 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA02869
	for <ieprep@optimus.ietf.org>; Mon, 18 Mar 2002 08:34:25 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04889
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 08:34:18 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id GAA11115
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 06:32:29 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA38B8
          for <ieprep@ietf.org>; Mon, 18 Mar 2002 06:34:19 -0700
Received: FROM agcs.com BY mknaomi ; Mon Mar 18 06:34:16 2002 -0700
Message-ID: <3C95ECDE.1E6AA4AB@agcs.com>
Date: Mon, 18 Mar 2002 06:34:22 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "James M. Polk" <jmpolk@cisco.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] Levels of Priority
References: <4.1.20020315122217.02f88ba8@localhost> <4.1.20020317170728.01620e70@localhost>
Content-Type: multipart/mixed;
 boundary="------------4E96E3CA544170BEA4473E0C"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------4E96E3CA544170BEA4473E0C
Content-Type: multipart/alternative;
 boundary="------------853BDCB2F1C41F4EA726C37A"


--------------853BDCB2F1C41F4EA726C37A
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

James,

Regulatory policy is not an area of expertise for me. Could it be
possible that both mechanisms may apply? Remember we are dealing with
both LECs and IXCs.



"James M. Polk" wrote:

>  At 03:07 PM 3/17/2002 -0700, Stuart (Stu) Goldman wrote:
>
>> Folks,
>>
>> Jack may very well be correct that there is no "law" against
>> preempting a call, but as I understand the PSTN process, each
>> carrier has to file tariffs with the controlling PUC(s) for that
>> carrier that are pretty detailed into how the service is to be
>> provided.  I suspect that carriers would point to such tariffs, as
>> well as the ever present fear of litigation, as reasons that they
>> can not preempt calls.
>>
>> So it may be less than "law", but far more than "custom".
>
>
> Then why would the petitions go directly to the FCC from Verizon and
> VoiceStream? I've got copies of the petitions and press releases...
> forwarded them to Scott Bradner last December to show this to be the
> case based on my comment in the mike at the BOF in SLC.
>
> I'm not as much questioning you, as stating what I've seen and have
> regarding this very topic, the current evidence is inconsistent with
> what the two of you state....  :-/
>
>
>                 *************************************
>  "People generally demand more respect for their own rights than they
>                    are willing to allow for others"
>
>  James M. Polk
> Consulting Engineer
> Office of the CTO
>
> Cisco Systems
> 2200 East President George Bush Turnpike
> Richardson, TX  75082 USA
> w) 972.813.5208
> f)  972.813.5280
> www.cisco.com

--------------853BDCB2F1C41F4EA726C37A
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
James,
<p>Regulatory policy is not an area of expertise for me. Could it be possible
that both mechanisms may apply? Remember we are dealing with both LECs
and IXCs.
<br>&nbsp;
<br>&nbsp;
<p>"James M. Polk" wrote:
<blockquote TYPE=CITE>&nbsp;At 03:07 PM 3/17/2002 -0700, Stuart (Stu) Goldman
wrote:
<blockquote type=cite cite>Folks,
<p>Jack may very well be correct that there is no "law" against preempting
a call, but as I understand the PSTN process, each carrier has to file
tariffs with the controlling PUC(s) for that carrier that are pretty detailed
into how the service is to be provided.&nbsp; I suspect that carriers would
point to such tariffs, as well as the ever present fear of litigation,
as reasons that they can not preempt calls.
<p>So it may be less than "law", but far more than "custom".</blockquote>

<p><br>Then why would the petitions go directly to the FCC from Verizon
and VoiceStream? I've got copies of the petitions and press releases...
forwarded them to Scott Bradner last December to show this to be the case
based on my comment in the mike at the BOF in SLC.
<p>I'm not as much questioning you, as stating what I've seen and have
regarding this very topic, the current evidence is inconsistent with what
the two of you state....&nbsp; :-/
<br>&nbsp;
<br>&nbsp;
<center>*************************************
<br>"People generally demand more respect for their own rights than they
are willing to allow for others"</center>

<br>&nbsp;James M. Polk
<br>Consulting Engineer
<br>Office of the CTO
<p>Cisco Systems
<br>2200 East President George Bush Turnpike
<br>Richardson, TX&nbsp; 75082 USA
<br>w) 972.813.5208
<br>f)&nbsp; 972.813.5280
<br><a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></blockquote>
</html>

--------------853BDCB2F1C41F4EA726C37A--

--------------4E96E3CA544170BEA4473E0C
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------4E96E3CA544170BEA4473E0C--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 08:44:00 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05037
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 08:44:00 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA03331;
	Mon, 18 Mar 2002 08:43:38 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA03299
	for <ieprep@optimus.ietf.org>; Mon, 18 Mar 2002 08:43:37 -0500 (EST)
Received: from imo-r08.mx.aol.com (imo-r08.mx.aol.com [152.163.225.104])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05021
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 08:43:32 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r08.mx.aol.com (mail_out_v32.5.) id z.14f.aa26369 (4000);
	Mon, 18 Mar 2002 08:42:22 -0500 (EST)
Message-ID: <14f.aa26369.29c748bd@aol.com>
Date: Mon, 18 Mar 2002 08:42:21 EST
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
To: goldmans@agcs.com, rtbell@cisco.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_14f.aa26369.29c748bd_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 50
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--part1_14f.aa26369.29c748bd_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 3/17/2002 5:32:07 PM Eastern Standard Time, 
goldmans@agcs.com writes:


> Bob, I am a little confused about what you have written. I agree that should 
> more and more calls just be "shoehorned" in, then the QoS would continue to 
> go down IF we were just looking at a SNAPSHOT of the network where no calls 
> ever end. However, if we can assume that the vast bulk of the traffic is 
> not emergency calls, can we not also logically assume that as such non- 
> emergency calls individually end, the bandwidth would revert back to the 
> network. The existing admission policy would not allow additional non 
> emergency calls to be established. So, my question is would not the QoS 
> return to normal as soon as an equal number of non emergency calls have 
> completed their sessions? It seems to me that the "overlooking" of the 
> admission policy for emergency calls would at the most only temporarily 
> impact the QoS of the network. The normal QoS would be quickly restored.  I 
> would think that QoS would not be more than momentarily impacted unless and 
> until the emergency traffic has completely displaced all of the multitude 
> of POTS traffic. Am I missing something? 
> 
> 

Staurt,

As always, you have clearly and succinctly described what I meant in previous 
e-mails and in the Annex to draft-pierce-sipping-assured-service-01. The 
added utility to this is that the lower QoS will encourage those 
non-emergency calls to release early, while the admission policy will block 
their immediate reattempts.

Thanks.

Mike Pierce

--part1_14f.aa26369.29c748bd_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 3/17/2002 5:32:07 PM Eastern Standard Time, goldmans@agcs.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">Bob, I am a little confused about what you have written. I agree that should more and more calls just be "shoehorned" in, then the QoS would continue to go down IF we were just looking at a SNAPSHOT of the network where no calls ever end. However, if we can assume that the vast bulk of the traffic is not emergency calls, can we not also logically assume that as such non- emergency calls individually end, the bandwidth would revert back to the network. The existing admission policy would not allow additional non emergency calls to be established. So, my question is would not the QoS return to normal as soon as an equal number of non emergency calls have completed their sessions? It seems to me that the "overlooking" of the admission policy for emergency calls would at the most only temporarily impact the QoS of the network. The normal QoS would be quickly r!
 estored. &nbsp;I would think that QoS would not be more than momentarily impacted unless and until the emergency traffic has completely displaced all of the multitude of POTS traffic. Am I missing something? 
<BR>
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>Staurt,
<BR>
<BR>As always, you have clearly and succinctly described what I meant in previous e-mails and in the Annex to draft-pierce-sipping-assured-service-01. The added utility to this is that the lower QoS will encourage those non-emergency calls to release early, while the admission policy will block their immediate reattempts.
<BR>
<BR>Thanks.
<BR>
<BR>Mike Pierce</FONT></HTML>

--part1_14f.aa26369.29c748bd_boundary--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 11:34:10 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA09973
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 11:34:09 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA13644;
	Mon, 18 Mar 2002 11:25:20 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA13616
	for <ieprep@optimus.ietf.org>; Mon, 18 Mar 2002 11:25:18 -0500 (EST)
Received: from unity.cisco.com (evvbu-exchange.cisco.com [64.101.140.73])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA09540
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 11:25:14 -0500 (EST)
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1CE99.6BB2F782"
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
Subject: RE: [Ieprep] Comments on draft-carlberg-ieprep-framework
Date: Mon, 18 Mar 2002 10:24:47 -0600
Message-ID: <E58F685255415A4EB2D4B8A02AEF32E3CC4294@evvbu-exchange.amer.unity.cisco.com>
Thread-Topic: [Ieprep] Comments on draft-carlberg-ieprep-framework
Thread-Index: AcHOAxARiB26ufQJSzi7Ku0eXD1lCQAlWVgA
From: "Bob Bell" <rtbell@cisco.com>
To: <goldmans@agcs.com>
Cc: <ieprep@ietf.org>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.

------_=_NextPart_001_01C1CE99.6BB2F782
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Stu, et. al. -=20
=20
You are correct in that call duration is finite. An average internal =
call (at least from my days in circuit mode) is typically on the order =
of 1-1.5 minutes. External calls (i.e. those involving the PSTN) have an =
average duration of 3-5 minutes. Given that, 5 minutes is a very long =
time at internet speeds. Yes, call termination will free resources and =
as the general QoS degradation increases, the length of calls will go =
down. It would be interesting to see what the usage curve would look =
like. One important note in this area is that the emergency calls we are =
discussing are not isolated calls like a single instance of 911, but a =
coordinated attempt by several governmental agencies to initiate =
coordination during a disaster event when the trunk lines will be jammed =
by people calling to check on spouse, kids, parents, friends, etc. I =
don't believe that normal call dynamics apply under these conditions.
=20
Bob

-----Original Message-----
From: Stuart (Stu) Goldman [mailto:goldmans@agcs.com]
Sent: Sunday, 17 March, 2002 3:28 PM
To: Bob Bell
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework


Bob,=20

I am a little confused about what you have written.=20


I agree that should more and more calls just be "shoehorned" in, then =
the QoS would continue to go down IF we were just looking at a SNAPSHOT =
of the network where no calls ever end.=20


However, if we can assume that the vast bulk of the traffic is not =
emergency calls, can we not also logically assume that as such non- =
emergency calls individually end, the bandwidth would revert back to the =
network. The existing admission policy would not allow additional non =
emergency calls to be established.=20


So, my question is would not the QoS return to normal as soon as an =
equal number of non emergency calls have completed their sessions?=20


It seems to me that the "overlooking" of the admission policy for =
emergency calls would at the most only temporarily impact the QoS of the =
network. The normal QoS would be quickly restored.  I would think that =
QoS would not be more than momentarily impacted unless and until the =
emergency traffic has completely displaced all of the multitude of POTS =
traffic.=20


Am I missing something?=20
 =20
 =20


Bob Bell wrote:=20


 James, et. al. - In the stated example, loosing 1% of the stream would =
probably not cause a great deal of distortion so as to render all =
conversations unusable. However, if you continue the process of stealing =
chunks of bandwidth, as more calls are "shoehorned" through the =
restriction point, you will quickly reach that condition. This is the =
reason I would believe that associating a precidence level with a call =
and using policy to terminate a lower precidence level session in favor =
of a higher precidence level would be more "fair". In this way, only the =
lowest precidence session would suffer, not all sessions. Even if you =
were to remove bandwidth from sessions proportinate to the their =
precidence (e.g. the lower the precidence, the more bandwidth could be =
stolen), unless some method of reducing their "billing" for this reduced =
clarity call is in place, then persons would be "paying" for services =
that they could not get. This is asking for trouble in the form of =
govenrment intervention and law suites. At least, if the lower =
precidence call is terminated, the "billing" also ends.Bob=20

-----Original Message-----=20
From: James Polk=20
Sent: Thursday, 14 March, 2002 12:08 PM=20
To: ieprep@ietf.org=20
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework=20
=20
At 09:20 AM 3/14/2002 -0500, Scott  Bradner wrote:>>Mike sez:>  The =
reference I made about "It's what the US military does" is a>  basic =
requirement for voice service operation.>>we are talking past each =
other>>I maintain that N-levels of priority is a solution not a =
requirement>>the question that needs to be answered is what functions =
are needed to>be supported - not how to do them>>specifically - what =
problem is thought to be solved by N levels of priority?>>is it that =
there should be a different probibility for>calls to be completed based =
on who you are or what you are doing?>>is it that the quality of the =
call should depends on who you>are and what you are doing?>>or is it =
someting elseLet's pose a strict set of circumstances to address part of =
this thread that's not answered yet explicitly (to my understanding - =
and others I believe)Let assume RSVP worked completely everywhere and =
between every phone in the Internet. A VoIP call using RSVP would work =
almost identically to a circuit switched call in taking up a defined =
amount of bandwidth through each transit interface between two phones. =
Lets say that all VoIP calls push aside all other (data) traffic always =
with no hesitation. In a case where each call is 80kbps traversing a T1 =
(1.544Mbps), 19 calls are possible at any one time. Even though there is =
56kbps of bandwidth remaining, that 20th call cannot be completed if it =
requires an 80kbps Reservation across this link.This is an example where =
codecs aren't changeable -- I want a specific analysis to this =
scenarioWhat Mike is suggesting is the potential for dropping < 1kbps =
from all 19 existing calls to make this 20th one exist (if all 19 calls =
have the same precedence level and the new 20th call has a higher =
precedence level). That instead of clearing one of the calls =
(preemption), merely degrading all 19 existing calls by about 1 packet =
per second (of the 50pps for a normal VoIP call).Is this considered =
unacceptable?One existing example/analogy:  even though the Cellular =
industry isn't packet-based, I believe most calls remain *up* AND loose =
more than 1% of the media signal during a call (if not much more).=20
 >>Scott>>_______________________________________________>Ieprep mailing =
list>Ieprep@ietf.org> https://www1.ietf.org/mailman/listinfo/ieprep=20
*************************************=20
"People generally demand more respect for their own rights than they are =
willing to allow for others"

 James M. Polk=20
Consulting Engineer=20
Office of the CTO=20

Cisco Systems=20
2200 East President George Bush Turnpike=20
Richardson, TX  75082 USA=20
w) 972.813.5208=20
f)  972.813.5280=20
www.cisco.com <http://www.cisco.com/>=20


------_=_NextPart_001_01C1CE99.6BB2F782
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">


<META content=3D"MSHTML 5.50.4913.1100" name=3DGENERATOR></HEAD>
<BODY>
<DIV><SPAN class=3D770541716-18032002><FONT face=3DArial color=3D#0000ff =
size=3D2>Stu,=20
et. al. - </FONT></SPAN></DIV>
<DIV><SPAN class=3D770541716-18032002><FONT face=3DArial color=3D#0000ff =

size=3D2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=3D770541716-18032002><FONT face=3DArial color=3D#0000ff =
size=3D2>You=20
are correct in that call duration is finite. An average internal call =
(at least=20
from my days in circuit mode) is typically on the order of 1-1.5 =
minutes.=20
External calls (i.e. those involving the PSTN) have an average duration =
of 3-5=20
minutes. Given that, 5 minutes is a very long time at internet speeds. =
Yes, call=20
termination will free resources and as the general QoS degradation =
increases,=20
the length of calls will go down. It would be interesting to see what =
the usage=20
curve would look like. One important note in this area is that the =
emergency=20
calls we are discussing are not isolated calls like a single instance of =
911,=20
but a coordinated attempt by several governmental agencies to initiate=20
coordination during a disaster event when the trunk lines will be jammed =
by=20
people calling to check on spouse, kids, parents, friends, etc. I don't =
believe=20
that normal call dynamics apply under these =
conditions.</FONT></SPAN></DIV>
<DIV><SPAN class=3D770541716-18032002><FONT face=3DArial color=3D#0000ff =

size=3D2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=3D770541716-18032002><FONT face=3DArial color=3D#0000ff =

size=3D2>Bob</FONT></SPAN></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px =
solid; MARGIN-RIGHT: 0px">
  <DIV class=3DOutlookMessageHeader dir=3Dltr align=3Dleft><FONT =
face=3DTahoma=20
  size=3D2>-----Original Message-----<BR><B>From:</B> Stuart (Stu) =
Goldman=20
  [mailto:goldmans@agcs.com]<BR><B>Sent:</B> Sunday, 17 March, 2002 3:28 =

  PM<BR><B>To:</B> Bob Bell<BR><B>Cc:</B> =
ieprep@ietf.org<BR><B>Subject:</B> Re:=20
  [Ieprep] Comments on =
draft-carlberg-ieprep-framework<BR><BR></FONT></DIV>Bob,=20
  <P>I am a little confused about what you have written.=20
  <P>I agree that should more and more calls just be "shoehorned" in, =
then the=20
  QoS would continue to go down IF we were just looking at a SNAPSHOT of =
the=20
  network where no calls ever end.=20
  <P>However, if we can assume that the vast bulk of the traffic is not=20
  emergency calls, can we not also logically assume that as such non- =
emergency=20
  calls individually end, the bandwidth would revert back to the =
network. The=20
  existing admission policy would not allow additional non emergency =
calls to be=20
  established.=20
  <P>So, my question is would not the QoS return to normal as soon as an =
equal=20
  number of non emergency calls have completed their sessions?=20
  <P>It seems to me that the "overlooking" of the admission policy for =
emergency=20
  calls would at the most only temporarily impact the QoS of the =
network. The=20
  normal QoS would be quickly restored.&nbsp; I would think that QoS =
would not=20
  be more than momentarily impacted unless and until the emergency =
traffic has=20
  completely displaced all of the multitude of POTS traffic.=20
  <P>Am I missing something? <BR>&nbsp; <BR>&nbsp;=20
  <P>Bob Bell wrote:=20
  <BLOCKQUOTE TYPE=3D"CITE">&nbsp;<SPAN class=3D225321319-14032002><FONT =

    face=3DArial><FONT color=3D#0000ff><FONT size=3D-1>James, et. al.=20
    -&nbsp;</FONT></FONT></FONT></SPAN><SPAN=20
    class=3D225321319-14032002></SPAN><SPAN =
class=3D225321319-14032002><FONT=20
    face=3DArial><FONT color=3D#0000ff><FONT size=3D-1>In the stated =
example, loosing=20
    1% of the stream would probably not cause a great deal of distortion =
so as=20
    to render all conversations unusable. However, if you continue the =
process=20
    of stealing chunks of bandwidth, as more calls are "shoehorned" =
through the=20
    restriction point, you will quickly reach that condition. This is =
the reason=20
    I would believe that associating a precidence level with a call and =
using=20
    policy to terminate a lower precidence level session in favor of a =
higher=20
    precidence level would be more "fair". In this way, only the lowest=20
    precidence session would suffer, not all sessions. Even if you were =
to=20
    remove bandwidth from sessions proportinate to the their precidence =
(e.g.=20
    the lower the precidence, the more bandwidth could be stolen), =
unless some=20
    method of reducing their "billing" for this reduced clarity call is =
in=20
    place, then persons would be "paying" for services that they could =
not get.=20
    This is asking for trouble in the form of govenrment intervention =
and law=20
    suites. At least, if the lower precidence call is terminated, the =
"billing"=20
    also ends.</FONT></FONT></FONT></SPAN><SPAN=20
    class=3D225321319-14032002></SPAN><SPAN =
class=3D225321319-14032002><FONT=20
    face=3DArial><FONT color=3D#0000ff><FONT =
size=3D-1>Bob</FONT></FONT></FONT></SPAN>=20

    <BLOCKQUOTE=20
    style=3D"PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff =
2px solid">
      <DIV class=3DOutlookMessageHeader dir=3Dltr><FONT =
face=3DTahoma><FONT=20
      size=3D-1>-----Original Message-----</FONT></FONT> <BR><FONT=20
      face=3DTahoma><FONT size=3D-1><B>From:</B> James =
Polk</FONT></FONT> <BR><FONT=20
      face=3DTahoma><FONT size=3D-1><B>Sent:</B> Thursday, 14 March, =
2002 12:08=20
      PM</FONT></FONT> <BR><FONT face=3DTahoma><FONT =
size=3D-1><B>To:</B>=20
      ieprep@ietf.org</FONT></FONT> <BR><FONT face=3DTahoma><FONT=20
      size=3D-1><B>Subject:</B> Re: [Ieprep] Comments on=20
      draft-carlberg-ieprep-framework</FONT></FONT> <BR>&nbsp;</DIV>At =
09:20 AM=20
      3/14/2002 -0500, Scott&nbsp; Bradner wrote:&gt;&gt;Mike =
sez:&gt;&nbsp; The=20
      reference I made about "It's what the US military does" is =
a&gt;&nbsp;=20
      basic requirement for voice service operation.&gt;&gt;we are =
talking past=20
      each other&gt;&gt;I maintain that N-levels of priority is a =
solution not a=20
      requirement&gt;&gt;the question that needs to be answered is what=20
      functions are needed to&gt;be supported - not how to do=20
      them&gt;&gt;specifically - what problem is thought to be solved by =
N=20
      levels of priority?&gt;&gt;<X-TAB></X-TAB>is it that there should =
be a=20
      different probibility for&gt;<X-TAB></X-TAB>calls to be completed =
based on=20
      who you are or what you are doing?&gt;&gt;<X-TAB></X-TAB>is it =
that the=20
      quality of the call should depends on who =
you&gt;<X-TAB></X-TAB>are and=20
      what you are doing?&gt;&gt;<X-TAB></X-TAB>or is it someting =
elseLet's pose=20
      a strict set of circumstances to address part of this thread =
that's not=20
      answered yet explicitly (to my understanding - and others I =
believe)Let=20
      assume RSVP worked completely everywhere and between every phone =
in the=20
      Internet. A VoIP call using RSVP would work almost identically to =
a=20
      circuit switched call in taking up a defined amount of bandwidth =
through=20
      each transit interface between two phones. Lets say that all VoIP =
calls=20
      push aside all other (data) traffic always with no hesitation. In =
a case=20
      where each call is 80kbps traversing a T1 (1.544Mbps), 19 calls =
are=20
      possible at any one time. Even though there is 56kbps of bandwidth =

      remaining, that 20th call cannot be completed if it requires an =
80kbps=20
      Reservation across this link.This is an example where codecs =
aren't=20
      changeable -- I want a specific analysis to this scenarioWhat Mike =
is=20
      suggesting is the potential for dropping &lt; 1kbps from all 19 =
existing=20
      calls to make this 20th one exist (if all 19 calls have the same=20
      precedence level and the new 20th call has a higher precedence =
level).=20
      That instead of clearing one of the calls (preemption), merely =
degrading=20
      all 19 existing calls by about 1 packet per second (of the 50pps =
for a=20
      normal VoIP call).Is this considered unacceptable?One existing=20
      example/analogy:&nbsp; even though the Cellular industry isn't=20
      packet-based, I believe most calls remain *up* AND loose more than =
1% of=20
      the media signal during a call (if not much more).=20
      =
<BR>&nbsp;&gt;&gt;Scott&gt;&gt;__________________________________________=
_____&gt;Ieprep=20
      mailing list&gt;Ieprep@ietf.org&gt;<A=20
      href=3D"https://www1.ietf.org/mailman/listinfo/ieprep"=20
      =
EUDORA=3D"AUTOURL">https://www1.ietf.org/mailman/listinfo/ieprep</A>=20
      <CENTER>************************************* <BR>"People =
generally demand=20
      more respect for their own rights than they are willing to allow =
for=20
      others"</CENTER><BR>&nbsp;James M. Polk <BR>Consulting Engineer =
<BR>Office=20
      of the CTO=20
      <P>Cisco Systems <BR>2200 East President George Bush Turnpike=20
      <BR>Richardson, TX&nbsp; 75082 USA <BR>w) 972.813.5208 =
<BR>f)&nbsp;=20
      972.813.5280 <BR><A href=3D"http://www.cisco.com/"=20
      =
eudora=3D"autourl">www.cisco.com</A></P></BLOCKQUOTE></BLOCKQUOTE></BLOCK=
QUOTE></BODY></HTML>

------_=_NextPart_001_01C1CE99.6BB2F782--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 12:09:32 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11631
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 12:09:31 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA18199;
	Mon, 18 Mar 2002 12:05:36 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA18168
	for <ieprep@optimus.ietf.org>; Mon, 18 Mar 2002 12:05:35 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11410
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 12:05:30 -0500 (EST)
From: fred@cisco.com
Received: from irp-view7.cisco.com (irp-view7.cisco.com [171.69.25.24])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2IH4uh27565
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 09:04:57 -0800 (PST)
Received: (fred@localhost) by irp-view7.cisco.com (8.8.8-Cisco List Logging/CISCO.WS.1.2) id JAA28015 for ieprep@ietf.org; Mon, 18 Mar 2002 09:04:53 -0800 (PST)
Date: Mon, 18 Mar 2002 09:04:53 -0800 (PST)
Message-Id: <200203181704.JAA28015@irp-view7.cisco.com>
To: ieprep@ietf.org
Subject: [Ieprep] We have been asked to review these drafts
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

Per our illustrious Area Director,speaking in "sipping", we need to
come up with requirements that drafts of this type would address.  The
requirements should not be of the form "someone has decided that this
should be done in their private network", but rather "there is a strong
technical need for this to be done in any IP network, for specific
reasons demonstrated by user practice or simulation results."

  "SMTP Service Extension for Priority", J. Onions, G. Lunt, 05/30/2001, 
  <draft-onions-priority-smtpext-00.txt> 

  "SIP Communications Resource Priority Header", H. Schulzrinne, James 
  Polk, 01/08/2002, <draft-polk-sip-resource-01.txt> 

  "RTP Header Extension for Communications Resource Priority", James Polk, 
  11/19/2001, <draft-polk-avt-rtpext-res-pri-00.txt> 

  "SIP Communications Resource Priority Header", H. Schulzrinne, James 
  Polk, 03/06/2002, <draft-polk-sipping-resource-01.txt> 

  "Providing Emergency Call Services for SIP-based Internet Telephony", H. 
  Schulzrinne, 03/29/2001, <draft-schulzrinne-sip-911-01.txt> 

Fred

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 12:26:45 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12349
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 12:26:45 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA19553;
	Mon, 18 Mar 2002 12:22:32 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA19512
	for <ieprep@optimus.ietf.org>; Mon, 18 Mar 2002 12:22:29 -0500 (EST)
Received: from clifden.donelan.com (clifden.donelan.com [199.34.53.180])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12173
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 12:22:24 -0500 (EST)
Received: from sean (helo=localhost)
	by clifden.donelan.com with local-esmtp (Exim 3.34 #2)
	id 16n0q4-0004Du-00; Mon, 18 Mar 2002 12:22:28 -0500
Date: Mon, 18 Mar 2002 12:22:28 -0500 (EST)
From: Sean Donelan <sean@donelan.com>
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Levels of priority and comments on ieprep framework
In-Reply-To: <124.d72b016.29c58574@aol.com>
Message-ID: <Pine.GSO.4.40.0203181214070.16176-100000@clifden.donelan.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org



On Sun, 17 Mar 2002 Mpierce1@aol.com wrote:
> First, we are talking of a case in which all other traffic (data, etc)
> has already been pushed aside, so only voice traffic is left, all using
> EF.

I am assuming that this is just a small hypothetical for the purposes of
responding to Polk's specific example.  Not a statement of how such a
system should work.  Starving all other traffic for the sake of voice
is a nifty way to make the data nets perform as badly as the telephone
system does under load.  In a real emergency we may find that e-mail,
blackberry, IM or some other data traffic is more critical (not to mention
better utilitization of scarce resources) than a voice call.  During
periods of limited capacity, voice, video and other high bandwidth
consumers may have the lowest priority.

That's why folks keep saying the "policy" needs to be seperate.



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 12:31:47 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12621
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 12:31:47 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA20026;
	Mon, 18 Mar 2002 12:28:10 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA19976
	for <ieprep@optimus.ietf.org>; Mon, 18 Mar 2002 12:28:08 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12381
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 12:27:59 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id JAA23313; Mon, 18 Mar 2002 09:27:30 -0800 (PST)
Message-Id: <4.1.20020318112222.01685598@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Mon, 18 Mar 2002 11:27:03 -0600
To: ieprep@ietf.org, hgs@opus.cs.columbia.edu
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] We have been asked to review these drafts
In-Reply-To: <200203181704.JAA28015@irp-view7.cisco.com>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_72887566==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_72887566==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 09:04 AM 3/18/2002 -0800, fred@cisco.com wrote:
>Per our illustrious Area Director,speaking in "sipping", we need to
>come up with requirements that drafts of this type would address.  The
>requirements should not be of the form "someone has decided that this
>should be done in their private network", but rather "there is a strong
>technical need for this to be done in any IP network, for specific
>reasons demonstrated by user practice or simulation results."
>
>  "SMTP Service Extension for Priority", J. Onions, G. Lunt, 05/30/2001, 
>  <draft-onions-priority-smtpext-00.txt> 
>
>  "SIP Communications Resource Priority Header", H. Schulzrinne, James 
>  Polk, 01/08/2002, <draft-polk-sip-resource-01.txt> 

This ID is no longer valid - as it's been renamed to the one two below here

>
>  "RTP Header Extension for Communications Resource Priority", James Polk, 
>  11/19/2001, <draft-polk-avt-rtpext-res-pri-00.txt> 
>
>  "SIP Communications Resource Priority Header", H. Schulzrinne, James 
>  Polk, 03/06/2002, <draft-polk-sipping-resource-01.txt> 
>
>  "Providing Emergency Call Services for SIP-based Internet Telephony", H. 
>  Schulzrinne, 03/29/2001, <draft-schulzrinne-sip-911-01.txt> 

I didn't think we were grouping 911-like services and ETS/GETS-like
services..... this is a change in philosophy I'd ask for some reasoning behind

>
>Fred
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep

*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_72887566==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 09:04 AM 3/18/2002 -0800, fred@cisco.com wrote:</div>
<div>&gt;Per our illustrious Area Director,speaking in
&quot;sipping&quot;, we need to</div>
<div>&gt;come up with requirements that drafts of this type would
address.&nbsp; The</div>
<div>&gt;requirements should not be of the form &quot;someone has decided
that this</div>
<div>&gt;should be done in their private network&quot;, but rather
&quot;there is a strong</div>
<div>&gt;technical need for this to be done in any IP network, for
specific</div>
<div>&gt;reasons demonstrated by user practice or simulation
results.&quot;</div>
<div>&gt;</div>
<div>&gt;&nbsp; &quot;SMTP Service Extension for Priority&quot;, J.
Onions, G. Lunt, 05/30/2001, </div>
<div>&gt;&nbsp; &lt;draft-onions-priority-smtpext-00.txt&gt; </div>
<div>&gt;</div>
<div>&gt;&nbsp; &quot;SIP Communications Resource Priority Header&quot;,
H. Schulzrinne, James </div>
<div>&gt;&nbsp; Polk, 01/08/2002, &lt;draft-polk-sip-resource-01.txt&gt;
</div>
<br>
<div>This ID is no longer valid - as it's been renamed to the one two
below here</div>
<br>
<div>&gt;</div>
<div>&gt;&nbsp; &quot;RTP Header Extension for Communications Resource
Priority&quot;, James Polk, </div>
<div>&gt;&nbsp; 11/19/2001, &lt;draft-polk-avt-rtpext-res-pri-00.txt&gt;
</div>
<div>&gt;</div>
<div>&gt;&nbsp; &quot;SIP Communications Resource Priority Header&quot;,
H. Schulzrinne, James </div>
<div>&gt;&nbsp; Polk, 03/06/2002,
&lt;draft-polk-sipping-resource-01.txt&gt; </div>
<div>&gt;</div>
<div>&gt;&nbsp; &quot;Providing Emergency Call Services for SIP-based
Internet Telephony&quot;, H. </div>
<div>&gt;&nbsp; Schulzrinne, 03/29/2001,
&lt;draft-schulzrinne-sip-911-01.txt&gt; </div>
<br>
<div>I didn't think we were grouping 911-like services and ETS/GETS-like
services..... this is a change in philosophy I'd ask for some reasoning
behind</div>
<br>
<div>&gt;</div>
<div>&gt;Fred</div>
<div>&gt;</div>
<div>&gt;_______________________________________________</div>
<div>&gt;Ieprep mailing list</div>
<div>&gt;Ieprep@ietf.org</div>
<div>&gt;<a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA=AUTOURL>https://www1.ietf.org/mailman/listinfo/ieprep</a></div>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_72887566==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 13:00:06 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA13540
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 13:00:05 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA22194;
	Mon, 18 Mar 2002 12:56:27 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA22136
	for <ieprep@optimus.ietf.org>; Mon, 18 Mar 2002 12:56:21 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA13448
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 12:56:17 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id KAA29951
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 10:54:27 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA4329
          for <ieprep@ietf.org>; Mon, 18 Mar 2002 10:56:17 -0700
Received: FROM agcs.com BY mknaomi ; Mon Mar 18 10:56:14 2002 -0700
Message-ID: <3C962A3F.A4486731@agcs.com>
Date: Mon, 18 Mar 2002 10:56:15 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Bob Bell <rtbell@cisco.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] Comments on draft-carlberg-ieprep-framework
References: <E58F685255415A4EB2D4B8A02AEF32E3CC4294@evvbu-exchange.amer.unity.cisco.com>
Content-Type: multipart/mixed;
 boundary="------------F8EDFA1748EDF26F07010382"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------F8EDFA1748EDF26F07010382
Content-Type: multipart/alternative;
 boundary="------------95F08E5AC7B4CD07065E791E"


--------------95F08E5AC7B4CD07065E791E
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Bob,

I think we are pretty much in agreement on the facts.

5 minutes is indeed a very long time. However, since we only need the
first call release to occur to make up for the first emergency call we
let in over the limits, we do not need to wait for 5 minutes, but just
for the next one of the already established calls to release. If there
were thousands of calls in session and these calls were following the
normal statistical arrival rates and length distributions, then one
should release momentarily.

It is very true that in an emergency there very well might be a much
greater concentration of emergency calls, and following the above, on a
call by call basis, they would be displacing  non emergency calls on the
network.

I agree with you totally on the difficulty of modeling human behavior in
a future attack. Depending on its nature there could be many short calls
check if people are OK, or there could be very long calls to address
topics such as evacuation, etc.  Our best strength is in the network's
flexibility and diversity.

What I am saying about the traffic does stop working if we were to get
into a situation where the emergency traffic totally replaces the non
emergency traffic. While I think this is unlikely to occur in the
aggregate, I see that it could happen at focal points.

I just don't know what the next natural or man made disaster will look
like. We need to be prepared for the unanticipated. I know a number of
people of commented on how well the Internet stood up to the attack on
Sept. 11 th. I am sad to say that the hospitals in the New York area
also were not overwhelmed by the injured. However, I would not want to
draw the conclusion that therefore the hospitals do not need to beef up
for future attacks.


Bob Bell wrote:

>  Stu, et. al. - You are correct in that call duration is finite. An
> average internal call (at least from my days in circuit mode) is
> typically on the order of 1-1.5 minutes. External calls (i.e. those
> involving the PSTN) have an average duration of 3-5 minutes. Given
> that, 5 minutes is a very long time at internet speeds. Yes, call
> termination will free resources and as the general QoS degradation
> increases, the length of calls will go down. It would be interesting
> to see what the usage curve would look like. One important note in
> this area is that the emergency calls we are discussing are not
> isolated calls like a single instance of 911, but a coordinated
> attempt by several governmental agencies to initiate coordination
> during a disaster event when the trunk lines will be jammed by people
> calling to check on spouse, kids, parents, friends, etc. I don't
> believe that normal call dynamics apply under these conditions.Bob
>
>      -----Original Message-----
>      From: Stuart (Stu) Goldman [mailto:goldmans@agcs.com]
>      Sent: Sunday, 17 March, 2002 3:28 PM
>      To: Bob Bell
>      Cc: ieprep@ietf.org
>      Subject: Re: [Ieprep] Comments on
>      draft-carlberg-ieprep-framework
>
>      Bob,
>
>      I am a little confused about what you have written.
>
>      I agree that should more and more calls just be "shoehorned"
>      in, then the QoS would continue to go down IF we were just
>      looking at a SNAPSHOT of the network where no calls ever
>      end.
>
>      However, if we can assume that the vast bulk of the traffic
>      is not emergency calls, can we not also logically assume
>      that as such non- emergency calls individually end, the
>      bandwidth would revert back to the network. The existing
>      admission policy would not allow additional non emergency
>      calls to be established.
>
>      So, my question is would not the QoS return to normal as
>      soon as an equal number of non emergency calls have
>      completed their sessions?
>
>      It seems to me that the "overlooking" of the admission
>      policy for emergency calls would at the most only
>      temporarily impact the QoS of the network. The normal QoS
>      would be quickly restored.  I would think that QoS would not
>      be more than momentarily impacted unless and until the
>      emergency traffic has completely displaced all of the
>      multitude of POTS traffic.
>
>      Am I missing something?
>
>
>
>      Bob Bell wrote:
>
>     > James, et. al. - In the stated example, loosing 1% of the
>     > stream would probably not cause a great deal of distortion
>     > so as to render all conversations unusable. However, if
>     > you continue the process of stealing chunks of bandwidth,
>     > as more calls are "shoehorned" through the restriction
>     > point, you will quickly reach that condition. This is the
>     > reason I would believe that associating a precidence level
>     > with a call and using policy to terminate a lower
>     > precidence level session in favor of a higher precidence
>     > level would be more "fair". In this way, only the lowest
>     > precidence session would suffer, not all sessions. Even if
>     > you were to remove bandwidth from sessions proportinate to
>     > the their precidence (e.g. the lower the precidence, the
>     > more bandwidth could be stolen), unless some method of
>     > reducing their "billing" for this reduced clarity call is
>     > in place, then persons would be "paying" for services that
>     > they could not get. This is asking for trouble in the form
>     > of govenrment intervention and law suites. At least, if
>     > the lower precidence call is terminated, the "billing"
>     > also ends.Bob
>     >
>     >      -----Original Message-----
>     >      From: James Polk
>     >      Sent: Thursday, 14 March, 2002 12:08 PM
>     >      To: ieprep@ietf.org
>     >      Subject: Re: [Ieprep] Comments on
>     >      draft-carlberg-ieprep-framework
>     >      At 09:20 AM 3/14/2002 -0500, Scott  Bradner
>     >      wrote:>>Mike sez:>  The reference I made about
>     >      "It's what the US military does" is a>  basic
>     >      requirement for voice service operation.>>we are
>     >      talking past each other>>I maintain that
>     >      N-levels of priority is a solution not a
>     >      requirement>>the question that needs to be
>     >      answered is what functions are needed to>be
>     >      supported - not how to do them>>specifically -
>     >      what problem is thought to be solved by N levels
>     >      of priority?>>is it that there should be a
>     >      different probibility for>calls to be completed
>     >      based on who you are or what you are doing?>>is
>     >      it that the quality of the call should depends
>     >      on who you>are and what you are doing?>>or is it
>     >      someting elseLet's pose a strict set of
>     >      circumstances to address part of this thread
>     >      that's not answered yet explicitly (to my
>     >      understanding - and others I believe)Let assume
>     >      RSVP worked completely everywhere and between
>     >      every phone in the Internet. A VoIP call using
>     >      RSVP would work almost identically to a circuit
>     >      switched call in taking up a defined amount of
>     >      bandwidth through each transit interface between
>     >      two phones. Lets say that all VoIP calls push
>     >      aside all other (data) traffic always with no
>     >      hesitation. In a case where each call is 80kbps
>     >      traversing a T1 (1.544Mbps), 19 calls are
>     >      possible at any one time. Even though there is
>     >      56kbps of bandwidth remaining, that 20th call
>     >      cannot be completed if it requires an 80kbps
>     >      Reservation across this link.This is an example
>     >      where codecs aren't changeable -- I want a
>     >      specific analysis to this scenarioWhat Mike is
>     >      suggesting is the potential for dropping < 1kbps
>     >      from all 19 existing calls to make this 20th one
>     >      exist (if all 19 calls have the same precedence
>     >      level and the new 20th call has a higher
>     >      precedence level). That instead of clearing one
>     >      of the calls (preemption), merely degrading all
>     >      19 existing calls by about 1 packet per second
>     >      (of the 50pps for a normal VoIP call).Is this
>     >      considered unacceptable?One existing
>     >      example/analogy:  even though the Cellular
>     >      industry isn't packet-based, I believe most
>     >      calls remain *up* AND loose more than 1% of the
>     >      media signal during a call (if not much more).
>     >       >>Scott>>____________________________________
>     >      __________>Ieprep mailing
>     >      list>Ieprep@ietf.org>https://www1.ietf.org/mailman/listinfo/ieprep
>     >            *************************************
>     >       "People generally demand more respect for their
>     >        own rights than they are willing to allow for
>     >                           others"
>     >
>     >       James M. Polk
>     >      Consulting Engineer
>     >      Office of the CTO
>     >
>     >      Cisco Systems
>     >      2200 East President George Bush Turnpike
>     >      Richardson, TX  75082 USA
>     >      w) 972.813.5208
>     >      f)  972.813.5280
>     >      www.cisco.com
>     >

--------------95F08E5AC7B4CD07065E791E
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Bob,
<p>I think we are pretty much in agreement on the facts.
<p>5 minutes is indeed a very long time. However, since we only need the
first call release to occur to make up for the first emergency call we
let in over the limits, we do not need to wait for 5 minutes, but just
for the next one of the already established calls to release. If there
were thousands of calls in session and these calls were following the normal
statistical arrival rates and length distributions, then one should release
momentarily.
<p>It is very true that in an emergency there very well might be a much
greater concentration of emergency calls, and following the above, on a
call by call basis, they would be displacing&nbsp; non emergency calls
on the network.
<p>I agree with you totally on the difficulty of modeling human behavior
in a future attack. Depending on its nature there could be many short calls
check if people are OK, or there could be very long calls to address topics
such as evacuation, etc.&nbsp; Our best strength is in the network's flexibility
and diversity.
<p>What I am saying about the traffic does stop working if we were to get
into a situation where the emergency traffic totally replaces the non emergency
traffic. While I think this is unlikely to occur in the aggregate, I see
that it could happen at focal points.
<p>I just don't know what the next natural or man made disaster will look
like. We need to be prepared for the unanticipated. I know a number of
people of commented on how well the Internet stood up to the attack on
Sept. 11 th. I am sad to say that the hospitals in the New York area also
were not overwhelmed by the injured. However, I would not want to draw
the conclusion that therefore the hospitals do not need to beef up for
future attacks.
<br>&nbsp;
<p>Bob Bell wrote:
<blockquote TYPE=CITE>&nbsp;<span class=770541716-18032002><font face="Arial"><font color="#0000FF"><font size=-1>Stu,
et. al. -&nbsp;</font></font></font></span><span class=770541716-18032002></span><span class=770541716-18032002><font face="Arial"><font color="#0000FF"><font size=-1>You
are correct in that call duration is finite. An average internal call (at
least from my days in circuit mode) is typically on the order of 1-1.5
minutes. External calls (i.e. those involving the PSTN) have an average
duration of 3-5 minutes. Given that, 5 minutes is a very long time at internet
speeds. Yes, call termination will free resources and as the general QoS
degradation increases, the length of calls will go down. It would be interesting
to see what the usage curve would look like. One important note in this
area is that the emergency calls we are discussing are not isolated calls
like a single instance of 911, but a coordinated attempt by several governmental
agencies to initiate coordination during a disaster event when the trunk
lines will be jammed by people calling to check on spouse, kids, parents,
friends, etc. I don't believe that normal call dynamics apply under these
conditions.</font></font></font></span><span class=770541716-18032002></span><span class=770541716-18032002><font face="Arial"><font color="#0000FF"><font size=-1>Bob</font></font></font></span>
<blockquote dir=ltr 
style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px solid; MARGIN-RIGHT: 0px">
<div class="OutlookMessageHeader" dir="ltr"><font face="Tahoma"><font size=-1>-----Original
Message-----</font></font>
<br><font face="Tahoma"><font size=-1><b>From:</b> Stuart (Stu) Goldman
[<A HREF="mailto:goldmans@agcs.com">mailto:goldmans@agcs.com</A>]</font></font>
<br><font face="Tahoma"><font size=-1><b>Sent:</b> Sunday, 17 March, 2002
3:28 PM</font></font>
<br><font face="Tahoma"><font size=-1><b>To:</b> Bob Bell</font></font>
<br><font face="Tahoma"><font size=-1><b>Cc:</b> ieprep@ietf.org</font></font>
<br><font face="Tahoma"><font size=-1><b>Subject:</b> Re: [Ieprep] Comments
on draft-carlberg-ieprep-framework</font></font>
<br>&nbsp;</div>
Bob,
<p>I am a little confused about what you have written.
<p>I agree that should more and more calls just be "shoehorned" in, then
the QoS would continue to go down IF we were just looking at a SNAPSHOT
of the network where no calls ever end.
<p>However, if we can assume that the vast bulk of the traffic is not emergency
calls, can we not also logically assume that as such non- emergency calls
individually end, the bandwidth would revert back to the network. The existing
admission policy would not allow additional non emergency calls to be established.
<p>So, my question is would not the QoS return to normal as soon as an
equal number of non emergency calls have completed their sessions?
<p>It seems to me that the "overlooking" of the admission policy for emergency
calls would at the most only temporarily impact the QoS of the network.
The normal QoS would be quickly restored.&nbsp; I would think that QoS
would not be more than momentarily impacted unless and until the emergency
traffic has completely displaced all of the multitude of POTS traffic.
<p>Am I missing something?
<br>&nbsp;
<br>&nbsp;
<p>Bob Bell wrote:
<blockquote TYPE="CITE"><span class=225321319-14032002><font face="Arial"><font color="#0000FF"><font size=-1>James,
et. al. -&nbsp;</span><span 
    class=225321319-14032002></span><span class=225321319-14032002>In
the stated example, loosing 1% of the stream would probably not cause a
great deal of distortion so as to render all conversations unusable. However,
if you continue the process of stealing chunks of bandwidth, as more calls
are "shoehorned" through the restriction point, you will quickly reach
that condition. This is the reason I would believe that associating a precidence
level with a call and using policy to terminate a lower precidence level
session in favor of a higher precidence level would be more "fair". In
this way, only the lowest precidence session would suffer, not all sessions.
Even if you were to remove bandwidth from sessions proportinate to the
their precidence (e.g. the lower the precidence, the more bandwidth could
be stolen), unless some method of reducing their "billing" for this reduced
clarity call is in place, then persons would be "paying" for services that
they could not get. This is asking for trouble in the form of govenrment
intervention and law suites. At least, if the lower precidence call is
terminated, the "billing" also ends.</span><span 
    class=225321319-14032002></span><span class=225321319-14032002>Bob</font></font></font></span>
<blockquote 
    style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px solid">
<div class=OutlookMessageHeader dir=ltr><font face="Tahoma"><font size=-1>-----Original
Message-----</font></font>
<br><font face="Tahoma"><font size=-1><b>From:</b> James Polk</font></font>
<br><font face="Tahoma"><font size=-1><b>Sent:</b> Thursday, 14 March,
2002 12:08 PM</font></font>
<br><font face="Tahoma"><font size=-1><b>To:</b> ieprep@ietf.org</font></font>
<br><font face="Tahoma"><font size=-1><b>Subject:</b> Re: [Ieprep] Comments
on draft-carlberg-ieprep-framework</font></font></div>
At 09:20 AM 3/14/2002 -0500, Scott&nbsp; Bradner wrote:>>Mike sez:>&nbsp;
The reference I made about "It's what the US military does" is a>&nbsp;
basic requirement for voice service operation.>>we are talking past each
other>>I maintain that N-levels of priority is a solution not a requirement>>the
question that needs to be answered is what functions are needed to>be supported
- not how to do them>>specifically - what problem is thought to be solved
by N levels of priority?>><X-TAB></X-TAB>is it that there should be a different
probibility for><X-TAB></X-TAB>calls to be completed based on who you are
or what you are doing?>><X-TAB></X-TAB>is it that the quality of the call
should depends on who you><X-TAB></X-TAB>are and what you are doing?>><X-TAB></X-TAB>or
is it someting elseLet's pose a strict set of circumstances to address
part of this thread that's not answered yet explicitly (to my understanding
- and others I believe)Let assume RSVP worked completely everywhere and
between every phone in the Internet. A VoIP call using RSVP would work
almost identically to a circuit switched call in taking up a defined amount
of bandwidth through each transit interface between two phones. Lets say
that all VoIP calls push aside all other (data) traffic always with no
hesitation. In a case where each call is 80kbps traversing a T1 (1.544Mbps),
19 calls are possible at any one time. Even though there is 56kbps of bandwidth
remaining, that 20th call cannot be completed if it requires an 80kbps
Reservation across this link.This is an example where codecs aren't changeable
-- I want a specific analysis to this scenarioWhat Mike is suggesting is
the potential for dropping &lt; 1kbps from all 19 existing calls to make
this 20th one exist (if all 19 calls have the same precedence level and
the new 20th call has a higher precedence level). That instead of clearing
one of the calls (preemption), merely degrading all 19 existing calls by
about 1 packet per second (of the 50pps for a normal VoIP call).Is this
considered unacceptable?One existing example/analogy:&nbsp; even though
the Cellular industry isn't packet-based, I believe most calls remain *up*
AND loose more than 1% of the media signal during a call (if not much more).
<br>&nbsp;>>Scott>>_______________________________________________>Ieprep
mailing list>Ieprep@ietf.org><a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA="AUTOURL">https://www1.ietf.org/mailman/listinfo/ieprep</a>
<center>*************************************
<br>"People generally demand more respect for their own rights than they
are willing to allow for others"</center>

<p>&nbsp;James M. Polk
<br>Consulting Engineer
<br>Office of the CTO
<p>Cisco Systems
<br>2200 East President George Bush Turnpike
<br>Richardson, TX&nbsp; 75082 USA
<br>w) 972.813.5208
<br>f)&nbsp; 972.813.5280
<br><a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></blockquote>
</blockquote>
</blockquote>
</blockquote>
</html>

--------------95F08E5AC7B4CD07065E791E--

--------------F8EDFA1748EDF26F07010382
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------F8EDFA1748EDF26F07010382--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 15:28:37 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18855
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 15:28:37 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA03623;
	Mon, 18 Mar 2002 15:27:55 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA03583
	for <ieprep@ns.ietf.org>; Mon, 18 Mar 2002 15:27:53 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18842
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 15:27:48 -0500 (EST)
Received: from FRED-W2K6.cisco.com (sjc-vpn1-187.cisco.com [10.21.96.187])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2IKQih24330;
	Mon, 18 Mar 2002 12:26:45 -0800 (PST)
Message-Id: <4.3.2.7.2.20020318141254.02e5a998@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Mon, 18 Mar 2002 14:23:16 -0600
To: Sean Donelan <sean@donelan.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Levels of priority and comments on ieprep
  framework
Cc: ieprep@ietf.org
In-Reply-To: <Pine.GSO.4.40.0203181214070.16176-100000@clifden.donelan.c
 om>
References: <124.d72b016.29c58574@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 11:22 AM 3/18/2002, Sean Donelan wrote:
>In a real emergency we may find that e-mail,
>blackberry, IM or some other data traffic is more critical (not to mention
>better utilitization of scarce resources) than a voice call.

in point of fact, per the reports to the national academy of sciences panel 
that met two weeks ago to discuss events on 9/11, the combination of 
voice-on-IP and data (including AOL distributing blackberry-based 
email/instant messaging PDAs) running over were among the few things that 
worked really well. One outcome I (and I hope we) want to truly avoid is 
making the Internet work as poorly as the telephone system does under load.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 16:14:36 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20448
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 16:14:35 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA07187;
	Mon, 18 Mar 2002 16:14:10 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA07132
	for <ieprep@ns.ietf.org>; Mon, 18 Mar 2002 16:14:08 -0500 (EST)
Received: from kc-msxproto2.kc.umkc.edu (kc-msxproto2.kc.umkc.edu [134.193.143.159])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20407
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 16:14:02 -0500 (EST)
Received: from KC-MAIL1.kc.umkc.edu ([134.193.143.161] RDNS failed) by kc-msxproto2.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.4453);
	 Mon, 18 Mar 2002 15:14:05 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="utf-8"
Subject: RE: [Ieprep] Levels of priority and comments on ieprep framework
Date: Mon, 18 Mar 2002 15:14:05 -0600
Message-ID: <EC61AA563B48B6459EEE976942BEA11918E8DD@KC-MAIL1.kc.umkc.edu>
Thread-Topic: [Ieprep] Levels of priority and comments on ieprep framework
Thread-Index: AcHOu2QEZQ7QrzBdSBm6sC4rLROD9gABl5nS
From: "Beard, Cory" <BeardC@umkc.edu>
To: "Fred Baker" <fred@cisco.com>, "Sean Donelan" <sean@donelan.com>
Cc: <ieprep@ietf.org>
X-OriginalArrivalTime: 18 Mar 2002 21:14:05.0639 (UTC) FILETIME=[D5F04D70:01C1CEC1]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from base64 to 8bit by optimus.ietf.org id QAA07133
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

Can you point to a location where we can read these reports?
 
Thanks,
Cory Beard

	-----Original Message----- 
	From: Fred Baker [mailto:fred@cisco.com] 
	Sent: Mon 3/18/2002 2:23 PM 
	To: Sean Donelan 
	Cc: ieprep@ietf.org 
	Subject: Re: [Ieprep] Levels of priority and comments on ieprep framework
	
	

	At 11:22 AM 3/18/2002, Sean Donelan wrote:
	>In a real emergency we may find that e-mail,
	>blackberry, IM or some other data traffic is more critical (not to mention
	>better utilitization of scarce resources) than a voice call.
	
	in point of fact, per the reports to the national academy of sciences panel
	that met two weeks ago to discuss events on 9/11, the combination of
	voice-on-IP and data (including AOL distributing blackberry-based
	email/instant messaging PDAs) running over were among the few things that
	worked really well. One outcome I (and I hope we) want to truly avoid is
	making the Internet work as poorly as the telephone system does under load.
	
	
	_______________________________________________
	Ieprep mailing list
	Ieprep@ietf.org
	https://www1.ietf.org/mailman/listinfo/ieprep
	


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 16:26:35 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20952
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 16:26:35 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA07751;
	Mon, 18 Mar 2002 16:26:07 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA07717
	for <ieprep@ns.ietf.org>; Mon, 18 Mar 2002 16:26:06 -0500 (EST)
Received: from clifden.donelan.com (clifden.donelan.com [199.34.53.180])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20931
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 16:26:00 -0500 (EST)
Received: from sean (helo=localhost)
	by clifden.donelan.com with local-esmtp (Exim 3.34 #2)
	id 16n4di-0004HZ-00; Mon, 18 Mar 2002 16:25:58 -0500
Date: Mon, 18 Mar 2002 16:25:58 -0500 (EST)
From: Sean Donelan <sean@donelan.com>
To: "Beard, Cory" <BeardC@umkc.edu>
cc: Fred Baker <fred@cisco.com>, <ieprep@ietf.org>
Subject: RE: [Ieprep] Levels of priority and comments on ieprep framework
In-Reply-To: <EC61AA563B48B6459EEE976942BEA11918E8DD@KC-MAIL1.kc.umkc.edu>
Message-ID: <Pine.GSO.4.40.0203181621590.16461-100000@clifden.donelan.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


The details on the CSTB 9/11 project are at

http://www4.nationalacademies.org/cpsma/cstb.nsf/web/project_crisisconditions

CSTB hasn't published its report yet.

Individual workshop participants may or may not choose to release
their own presentations in other forums.


On Mon, 18 Mar 2002, Beard, Cory wrote:

> Can you point to a location where we can read these reports?
>
> Thanks,
> Cory Beard
>
> 	-----Original Message-----
> 	From: Fred Baker [mailto:fred@cisco.com]
> 	Sent: Mon 3/18/2002 2:23 PM
> 	To: Sean Donelan
> 	Cc: ieprep@ietf.org
> 	Subject: Re: [Ieprep] Levels of priority and comments on ieprep framework
>
>
>
> 	At 11:22 AM 3/18/2002, Sean Donelan wrote:
> 	>In a real emergency we may find that e-mail,
> 	>blackberry, IM or some other data traffic is more critical (not to mention
> 	>better utilitization of scarce resources) than a voice call.
>
> 	in point of fact, per the reports to the national academy of sciences panel
> 	that met two weeks ago to discuss events on 9/11, the combination of
> 	voice-on-IP and data (including AOL distributing blackberry-based
> 	email/instant messaging PDAs) running over were among the few things that
> 	worked really well. One outcome I (and I hope we) want to truly avoid is
> 	making the Internet work as poorly as the telephone system does under load.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 19:31:20 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA27265
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 19:31:20 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA23258;
	Mon, 18 Mar 2002 19:29:05 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA23225
	for <ieprep@ns.ietf.org>; Mon, 18 Mar 2002 19:29:04 -0500 (EST)
Received: from kc-msxproto4.kc.umkc.edu (kc-msxproto4.kc.umkc.edu [134.193.143.48])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA27147
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 19:29:01 -0500 (EST)
Received: from KC-MAIL1.kc.umkc.edu ([134.193.143.161] RDNS failed) by kc-msxproto4.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.4453);
	 Mon, 18 Mar 2002 18:29:03 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="utf-8"
Date: Mon, 18 Mar 2002 18:29:02 -0600
Message-ID: <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.edu>
Thread-Topic: Preemption
Thread-Index: AcHO3RHxuoSwUYmxRSenX6nDPUNEFg==
From: "Beard, Cory" <BeardC@umkc.edu>
To: <ieprep@ietf.org>
X-OriginalArrivalTime: 19 Mar 2002 00:29:03.0193 (UTC) FILETIME=[12396490:01C1CEDD]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from base64 to 8bit by optimus.ietf.org id TAA23226
Subject: [Ieprep] Preemption
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

In the IP Telephony Framework draft, non-preemptive action is stated as a primary objective.
 
How does this translate to packet-level dropping policies in IP?  It seems to me that RED could be seen as a form of preemption.  Should we be on record as not supporting RED, or should a distinction be made between call-level preemption and packet-level preemption?
 
IMHO we just should take no position about preemption in the framework document.  It can be discussed as a possible implementation approach (depending on jurisdictional guidelines) but should not be stated as a primary objective in the framework document.
 
Cory Beard

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 20:09:18 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA27999
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 20:09:18 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA25313;
	Mon, 18 Mar 2002 20:08:34 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA25280
	for <ieprep@ns.ietf.org>; Mon, 18 Mar 2002 20:08:32 -0500 (EST)
Received: from grok.web9.com (grok.seiri.com [204.91.156.135])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA27983
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 20:08:28 -0500 (EST)
Received: from web9.com (unknown [166.63.187.14])
	by grok.web9.com (Postfix) with ESMTP
	id 29D6E5509C; Mon, 18 Mar 2002 20:08:29 -0500 (EST)
Message-ID: <3C968F8D.E9C5B74B@web9.com>
Date: Mon, 18 Mar 2002 17:08:29 -0800
From: "Keptin Komrade Dr. Bobwrench" <bownes@web9.com>
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "Beard, Cory" <BeardC@umkc.edu>
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
References: <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit


The fact that any number of underlying protocols may be configured or
built in a preemptive manner should be addressed but not specified. We
cannot make any assumptions about the underlying protocols, any of which
may have any number of implementational effects.

Bob


"Beard, Cory" wrote:
> 
> In the IP Telephony Framework draft, non-preemptive action is stated as a primary objective.
> 
> How does this translate to packet-level dropping policies in IP?  It seems to me that RED could be seen as a form of preemption.  Should we be on record as not supporting RED, or should a distinction be made between call-level preemption and packet-level preemption?
> 
> IMHO we just should take no position about preemption in the framework document.  It can be discussed as a possible implementation approach (depending on jurisdictional guidelines) but should not be stated as a primary objective in the framework document.
> 
> Cory Beard
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 20:16:34 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA28152
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 20:16:34 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA25548;
	Mon, 18 Mar 2002 20:16:06 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA25505
	for <ieprep@ns.ietf.org>; Mon, 18 Mar 2002 20:16:04 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id UAA28145
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 20:16:00 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.24819-0@bells.cs.ucl.ac.uk>; Tue, 19 Mar 2002 01:15:54 +0000
To: "Beard, Cory" <BeardC@umkc.edu>
cc: ieprep <ieprep@ietf.org>, K.Carlberg@cs.ucl.ac.uk
Subject: Re: [Ieprep] Preemption
In-reply-to: Your message of "Mon, 18 Mar 2002 18:29:02 CST." <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.edu>
Date: Tue, 19 Mar 2002 01:15:53 +0000
Message-ID: <11843.1016500553@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

Cory,


> In the IP Telephony Framework draft, non-preemptive action is stated
> as a primary objective.  
>
> How does this translate to packet-level dropping policies in IP?  It
> seems to me that RED could be seen as a form of preemption.  Should we
> be on record as not supporting RED, or should a distinction be made
> between call-level preemption and packet-level preemption?
>
> IMHO we just should take no position about preemption in the framework
> document.  It can be discussed as a possible implementation approach
> (depending on jurisdictional guidelines) but should not be stated as a
> primary objective in the framework document.

hmmm, close, but I don't really subscribe to the notion that RED is
a form of preemption.

but instead of focusing on that debate, i think we can take a position
that at a minimum, we use non-preemption as a default goal for call
establishment.  so in a sense, it would apply to actions taken upon
call setup signaling via SIP, of H.323

-ken


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 20:50:16 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA28778
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 20:50:16 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA27157;
	Mon, 18 Mar 2002 20:47:15 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA27121
	for <ieprep@ns.ietf.org>; Mon, 18 Mar 2002 20:47:13 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA28721
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 20:47:09 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id RAA17091 for <ieprep@ietf.org>; Mon, 18 Mar 2002 17:46:40 -0800 (PST)
Message-Id: <4.1.20020318194233.016e93a8@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Mon, 18 Mar 2002 19:46:12 -0600
To: ieprep <ieprep@ietf.org>
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] Preemption
In-Reply-To: <11843.1016500553@cs.ucl.ac.uk>
References: <Your message of "Mon, 18 Mar 2002 18:29:02 CST." <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.edu>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_102837282==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_102837282==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 01:15 AM 3/19/2002 +0000, Ken Carlberg wrote:
>Cory,
>
>
>> In the IP Telephony Framework draft, non-preemptive action is stated
>> as a primary objective.  

As there is ongoing work *with* preemption (MLPP and Assured Services) that has
been frequently brought up in discussions on this list, stating that *a*
default objective is "non-preemptive" is appropriate IMO.

*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_102837282==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 01:15 AM 3/19/2002 +0000, Ken Carlberg wrote:</div>
<div>&gt;Cory,</div>
<div>&gt;</div>
<div>&gt;</div>
<div>&gt;&gt; In the IP Telephony Framework draft, non-preemptive action
is stated</div>
<div>&gt;&gt; as a primary objective.&nbsp; </div>
<br>
<div>As there is ongoing work *with* preemption (MLPP and Assured
Services) that has been frequently brought up in discussions on this
list, stating that *a* default objective is &quot;non-preemptive&quot; is
appropriate IMO.</div>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_102837282==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 18 22:52:09 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA03138
	for <ieprep-archive@odin.ietf.org>; Mon, 18 Mar 2002 22:52:09 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id WAA03576;
	Mon, 18 Mar 2002 22:51:44 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id WAA03542
	for <ieprep@ns.ietf.org>; Mon, 18 Mar 2002 22:51:42 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA03127
	for <ieprep@ietf.org>; Mon, 18 Mar 2002 22:51:37 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <F5YT9XNQ>; Mon, 18 Mar 2002 22:50:18 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C48@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'James M. Polk'" <jmpolk@cisco.com>, ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Preemption
Date: Mon, 18 Mar 2002 22:50:56 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1CEF9.462FF090"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1CEF9.462FF090
Content-Type: text/plain;
	charset="iso-8859-1"

I think the thing to remember is that pre-emption is permitted (on the PSTN)
in some nations and not permitted in others.
So it seems the first objective is not to REQUIRE anything that would look
like pre-emption when integrated with the PSTN in those countries.  In other
countries (and private networks) where (telephony) pre-emption is permitted,
it MAY be permissible to allow (but not require) something that looks like
pre-emption.
 
As an analogy, e-MLPP (in a GSM mobile network) allows pre-emption in those
juristictions which allow it, but does not require it in juristictions which
do not permit pre-emption.
 
In the telephony context, pre-emption refers to the "call".  So it is
possible that some types of packet level pre-emption (or selective packet
dropping) may be acceptable even in those contexts where call level
pre-emption is not permitted.  This would presumably be governed by SLAs for
QoS.
 
Janet Gunn

-----Original Message-----
From: James M. Polk [mailto:jmpolk@cisco.com]
Sent: Monday, March 18, 2002 8:46 PM
To: ieprep
Subject: Re: [Ieprep] Preemption


At 01:15 AM 3/19/2002 +0000, Ken Carlberg wrote:
>Cory,
>
>
>> In the IP Telephony Framework draft, non-preemptive action is stated
>> as a primary objective.  

As there is ongoing work *with* preemption (MLPP and Assured Services) that
has been frequently brought up in discussions on this list, stating that *a*
default objective is "non-preemptive" is appropriate IMO.

*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"


James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com <http://www.cisco.com/>  


------_=_NextPart_001_01C1CEF9.462FF090
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 5.50.4134.600" name=GENERATOR></HEAD>
<BODY>
<DIV><SPAN class=390083803-19032002><FONT face=Arial color=#0000ff size=2>I 
think the thing to remember is that pre-emption is permitted (on the 
PSTN)&nbsp;in some nations and not permitted in others.</FONT></SPAN></DIV>
<DIV><SPAN class=390083803-19032002><FONT face=Arial color=#0000ff size=2>So it 
seems the first objective is not to REQUIRE anything that would look like 
pre-emption when integrated with the PSTN in those countries.&nbsp; In other 
countries (and private networks) where (telephony) pre-emption is permitted, it 
MAY be permissible to allow (but not require) something that looks like 
pre-emption.</FONT></SPAN></DIV>
<DIV><SPAN class=390083803-19032002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=390083803-19032002><FONT face=Arial color=#0000ff size=2>As an 
analogy, e-MLPP (in a GSM mobile network) allows pre-emption in those 
juristictions which allow it, but does not require it in juristictions which do 
not permit pre-emption.</FONT></SPAN></DIV>
<DIV><SPAN class=390083803-19032002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=390083803-19032002><FONT face=Arial color=#0000ff size=2>In the 
telephony context, pre-emption refers to the "call".&nbsp; So it is possible 
that some&nbsp;types of packet level pre-emption (or selective packet 
dropping)&nbsp;may be acceptable even in those contexts&nbsp;where call level 
pre-emption is not permitted.&nbsp; This would presumably be governed by SLAs 
for QoS.</FONT></SPAN></DIV>
<DIV><SPAN class=390083803-19032002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=390083803-19032002><FONT face=Arial color=#0000ff size=2>Janet 
Gunn</FONT></SPAN></DIV>
<BLOCKQUOTE>
  <DIV class=OutlookMessageHeader dir=ltr align=left><FONT face=Tahoma 
  size=2>-----Original Message-----<BR><B>From:</B> James M. Polk 
  [mailto:jmpolk@cisco.com]<BR><B>Sent:</B> Monday, March 18, 2002 8:46 
  PM<BR><B>To:</B> ieprep<BR><B>Subject:</B> Re: [Ieprep] 
  Preemption<BR><BR></FONT></DIV>
  <DIV>At 01:15 AM 3/19/2002 +0000, Ken Carlberg wrote:</DIV>
  <DIV>&gt;Cory,</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;&gt; In the IP Telephony Framework draft, non-preemptive action is 
  stated</DIV>
  <DIV>&gt;&gt; as a primary objective.&nbsp; </DIV><BR>
  <DIV>As there is ongoing work *with* preemption (MLPP and Assured Services) 
  that has been frequently brought up in discussions on this list, stating that 
  *a* default objective is "non-preemptive" is appropriate IMO.</DIV><BR>
  <DIV align=center>*************************************<BR>"People generally 
  demand more respect for their own rights than they are willing to allow for 
  others"<BR><BR></DIV>James M. Polk<BR>Consulting Engineer<BR>Office of the 
  CTO<BR><BR>Cisco Systems<BR>2200 East President George Bush Turnpike 
  <BR>Richardson, TX&nbsp; 75082 USA<BR>w) 972.813.5208<BR>f)&nbsp; 
  972.813.5280<BR><A href="http://www.cisco.com/" 
  eudora="autourl">www.cisco.com</A> </BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C1CEF9.462FF090--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 19 00:20:00 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA05798
	for <ieprep-archive@odin.ietf.org>; Tue, 19 Mar 2002 00:20:00 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id AAA07251;
	Tue, 19 Mar 2002 00:15:41 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id AAA07221
	for <ieprep@ns.ietf.org>; Tue, 19 Mar 2002 00:15:39 -0500 (EST)
Received: from clifden.donelan.com (clifden.donelan.com [199.34.53.180])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA05526
	for <ieprep@ietf.org>; Tue, 19 Mar 2002 00:15:37 -0500 (EST)
Received: from sean (helo=localhost)
	by clifden.donelan.com with local-esmtp (Exim 3.34 #2)
	id 16nBxn-0004cP-00; Tue, 19 Mar 2002 00:15:11 -0500
Date: Tue, 19 Mar 2002 00:15:11 -0500 (EST)
From: Sean Donelan <sean@donelan.com>
To: "Beard, Cory" <BeardC@umkc.edu>
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
In-Reply-To: <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.edu>
Message-ID: <Pine.GSO.4.40.0203182203430.16703-100000@clifden.donelan.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


On Mon, 18 Mar 2002, Beard, Cory wrote:
> How does this translate to packet-level dropping policies in IP?
> It seems to me that RED could be seen as a form of preemption.  Should
> we be on record as not supporting RED, or should a distinction be made
> between call-level preemption and packet-level preemption?

I'm not sure if you are serious.

In an IP network, packet drops are a form of signalling.  The various
types of queue management (Tail-drop, WFQ, RED, WRED, etc) use packet
drops to signal senders how fast they can send data.  The quanitity of
different queuing disciplines demonstrates it is not a simple problem
for packet networks.

In a circuit-switched network you don't need feedback on how much
instantaneous bandwidth is available on a path once the call is
connected.  You either get a busy signal or a ringing signal.  Once
the call is connected, the capacity doesn't change until your call
ends, pre-empted or the path fails.

In a packet-switched network, the instantaneous bandwidth available may
increase or decrease during the session as the flow is re-routed or other
traffic passes through the same bottleneck.

Without packet drops, a sender would need some other method to track
of how much instaneous capacity is available along the entire path
and control its transmission rate so it doesn't overrun any buffer
anywhere along the path. Packet drops aren't the only method, but it
is the method used for many protocols built on top of IP.  IP Telephony
could develop its own bandwidth control protocol which doesn't use
packet drop signalling for congestion notification.  But it would
eventually still have to share the same network with other IP protocols
which do use packet drop.

How many phone calls can a T1 support?  23, 24, 48, 96, 200?  A circuit
switched network designer can give you a fixed answer.  A packet switched
network designer can only give you a probalistic answer.

I suspect you are stuck with packet drops assuming you want to use an
IP network.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 21 12:41:07 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA19183
	for <ieprep-archive@odin.ietf.org>; Thu, 21 Mar 2002 12:41:07 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA27321;
	Thu, 21 Mar 2002 12:39:05 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA27269
	for <ieprep@optimus.ietf.org>; Thu, 21 Mar 2002 12:39:03 -0500 (EST)
Received: from unity.cisco.com (evvbu-exchange.cisco.com [64.101.140.73])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA19095
	for <ieprep@ietf.org>; Thu, 21 Mar 2002 12:39:00 -0500 (EST)
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
Date: Thu, 21 Mar 2002 11:37:58 -0600
Message-ID: <E58F685255415A4EB2D4B8A02AEF32E3CC42A8@evvbu-exchange.amer.unity.cisco.com>
Thread-Topic: Call Preemption in Public Switched Telephone Networks
Thread-Index: AcHQR53qhbAVib5zRfGxnHFilClJ9QAtygzA
From: "Bob Bell" <rtbell@cisco.com>
To: <ieprep@ietf.org>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id MAA27270
Subject: [Ieprep] FW: Call Preemption in Public Switched Telephone Networks
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

Folks - 

I am somewhat embarrassed to state this but I was wrong relative to the legality of pre-empting a circuit mode call. As a result of the thread discussing this, I sent a query to the FCC. Here is their reply. While not absolute, it gives enough guidence to indicate that it is possible.

Bob Bell

-----Original Message-----
From: FCCTSR18 [mailto:FCCTSR18@fcc.gov]
Sent: Wednesday, 20 March, 2002 12:44 PM
To: Bob Bell
Subject: Re: Call Preemption in Public Switched Telephone Networks


Hello BOB - - hope you are well.

Bluntly stated - - I would be hard-pressed to find an FCC rule that addresses this scenario, as generally stated.  There is an appendix to Part 64 of our rules (47 CFR, Part 64, Appendix A) that addresses Telecommunications Service Priority System for National Security Emergency Preparedness; so in an emergency situation, I can understand where a scenario of this nature could take place.  

However, in normal operation of the Public Switched Telephone Network, this type of activity just doesn't seem practical.  After all, what would we do with the "busy" signal - - it wouldn't be needed any more.

It is my understanding that most carriers offer an "interrupt" service; whereby when a caller needs to make a particular phone call (usually in an emergency situation); the carrier can "interrupt" the present call.  However, this is something "requested" by a caller; not instituted by the carrier in order to serve a higher paying/priority customer; and the carrier generally levies a "hefty" charge for this interrupt service.

Further, why would a carrier want to do something of this nature?  Seems like this type of activity would eventually upset their customers (I know it would me) and then they wouldn't have enough left to worry about dropping someone's call to free up their network to process someone else's call.

If I'm "off track" here with my line of thinking; please let me know.  Also, if there are specific examples; beyond your general concerns below, that you'd care for me to address; please let me know.

Hope this helps - - thanks for contacting the FCC - - have a great day ! ! !


>>> "Bob Bell" <rtbell@cisco.com> 03/14/02 04:42PM >>>
I am sorry for the general nature of theis query, but, in the process of developing some standards for Telephony over IP, a question has been generated that I though I knew the answer to. However, several people have challenged my answer and I would appreciate it if I could get the correct answer.

The question is:

Is it legally permissable to drop a call in progress (i.e. after connection is complete) in order to free the utilized resource for use by another call?In other words, within the Public Telephone system, can an existing call be preempted by another call?

I would appreciate knowing the actual regulations if there are such.

Thank you for your time and energy in this matter.

Bob Bell
Cisco Systems, Inc.
576 S. Brentwood Ln.
Bountiful, Utah 84010
801-294-3034 (v)
801-204-3023 (f)
801-971-4200 (cell)
rtbell@cisco.com <mailto:rtbell@cisco.com> (email)



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 22 11:20:44 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25027
	for <ieprep-archive@odin.ietf.org>; Fri, 22 Mar 2002 11:20:43 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA12600;
	Fri, 22 Mar 2002 11:17:30 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA12468
	for <ieprep@optimus.ietf.org>; Fri, 22 Mar 2002 11:17:23 -0500 (EST)
Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA23387
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 10:09:28 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.9.3/8.9.3) with ESMTP id KAA11947
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 10:09:30 -0500 (EST)
Received: from cs.columbia.edu (cta.ietf53.cw.net [166.63.185.82])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.1/8.12.1) with ESMTP id g2MF9TPm022301
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT)
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 10:09:30 -0500 (EST)
Message-ID: <3C9B490F.69D80280@cs.columbia.edu>
Date: Fri, 22 Mar 2002 09:09:03 -0600
From: Henning Schulzrinne <hgs@cs.columbia.edu>
Organization: Columbia University
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: ieprep@ietf.org
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] "IEPREP" priority value for Resource-Priority
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

It has been suggested to add a namespace for GETS-like services to the
SIP Resource-Priority header. There are at least two solutions:

- Have a single namespace that covers all national GETS-like schemes,
called (for my lack of imagination) something like "emergency". The
namespace would have one value only, i.e., it's either

Resource-Priority: emergency

or no header (meaning regular call).

This does not preclude that countries could define their own namespace
if they don't want to use the generic name.

(2) Define name spaces for each national or regional emergency
telecommunication preference system. Example:

Resource-Priority: GETS

I don't see much difference, except that authorization is likely to be
specific to each system, so that it may be easier to explicitly indicate
the system to be used.

Suggestions or alternatives are appreciated.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 22 11:45:33 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25832
	for <ieprep-archive@odin.ietf.org>; Fri, 22 Mar 2002 11:45:33 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA15180;
	Fri, 22 Mar 2002 11:44:53 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA15143
	for <ieprep@optimus.ietf.org>; Fri, 22 Mar 2002 11:44:51 -0500 (EST)
Received: from pmesmtp01.wcom.com (pmesmtp01.wcom.com [199.249.20.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25825
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 11:44:48 -0500 (EST)
Received: from CONVERSION-DAEMON by firewall.wcom.com (PMDF V5.2-33 #47837)
 id <0GTD00001VTK99@firewall.wcom.com> for ieprep@ietf.org; Fri,
 22 Mar 2002 16:44:08 +0000 (GMT)
Received: from pmismtp06.wcomnet.com ([166.38.62.54])
 by firewall.wcom.com (PMDF V5.2-33 #47837)
 with ESMTP id <0GTD00N8IVTJ0Q@firewall.wcom.com>; Fri,
 22 Mar 2002 16:44:08 +0000 (GMT)
Received: from pmismtp06.wcomnet.com by pmismtp06.wcomnet.com
 (iPlanet Messaging Server 5.1 (built May  7 2001))
 with SMTP id <0GTD00401VTF9E@pmismtp06.wcomnet.com>; Fri,
 22 Mar 2002 16:44:07 +0000 (GMT)
Received: from rfreilich1 ([166.50.71.30])
 by pmismtp06.wcomnet.com (iPlanet Messaging Server 5.1 (built May  7 2001))
 with SMTP id <0GTD003BWVTJXU@pmismtp06.wcomnet.com>; Fri,
 22 Mar 2002 16:44:07 +0000 (GMT)
Date: Fri, 22 Mar 2002 10:49:09 -0600
From: Rob Freilich <robert.freilich@wcom.com>
Subject: RE: [Ieprep] "IEPREP" priority value for Resource-Priority
In-reply-to: <3C9B490F.69D80280@cs.columbia.edu>
To: "'Henning Schulzrinne'" <hgs@cs.columbia.edu>, ieprep@ietf.org
Reply-to: robert.freilich@wcom.com
Message-id: <000f01c1d1c1$7dea87a0$1e4732a6@wcomnet.com>
MIME-version: 1.0
X-MIMEOLE: Produced By Microsoft MimeOLE V5.00.2615.200
X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2911.0)
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: 7bit
Importance: Normal
X-Priority: 3 (Normal)
X-MSMail-priority: Normal
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

Henning,

It might be better for each country to define their own namespace since this
could be used as part of the key fields for authentication.  GETS uses a 10
digit PIN code, if I remember correctly.  Other similar services might use a
difference scheme altogether for authentication.  The national identifier
from the Resource-Priority header plus the code could be used in-conjunction
for authentication.  One party identifies the treatment and the other part
the actual code to authenticate.

Regards,

Rob

-----Original Message-----
From: ieprep-admin@ietf.org [mailto:ieprep-admin@ietf.org]On Behalf Of
Henning Schulzrinne
Sent: Friday, March 22, 2002 9:09 AM
To: ieprep@ietf.org
Subject: [Ieprep] "IEPREP" priority value for Resource-Priority


It has been suggested to add a namespace for GETS-like services to the
SIP Resource-Priority header. There are at least two solutions:

- Have a single namespace that covers all national GETS-like schemes,
called (for my lack of imagination) something like "emergency". The
namespace would have one value only, i.e., it's either

Resource-Priority: emergency

or no header (meaning regular call).

This does not preclude that countries could define their own namespace
if they don't want to use the generic name.

(2) Define name spaces for each national or regional emergency
telecommunication preference system. Example:

Resource-Priority: GETS

I don't see much difference, except that authorization is likely to be
specific to each system, so that it may be easier to explicitly indicate
the system to be used.

Suggestions or alternatives are appreciated.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 22 11:53:22 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25967
	for <ieprep-archive@odin.ietf.org>; Fri, 22 Mar 2002 11:53:22 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA15518;
	Fri, 22 Mar 2002 11:50:58 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA15486
	for <ieprep@optimus.ietf.org>; Fri, 22 Mar 2002 11:50:56 -0500 (EST)
Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25926
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 11:50:53 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.9.3/8.9.3) with ESMTP id LAA20392;
	Fri, 22 Mar 2002 11:50:56 -0500 (EST)
Received: from cs.columbia.edu (cta.ietf53.cw.net [166.63.185.82])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.1/8.12.1) with ESMTP id g2MGotPm026456
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Fri, 22 Mar 2002 11:50:55 -0500 (EST)
Message-ID: <3C9B60D5.951A81BB@cs.columbia.edu>
Date: Fri, 22 Mar 2002 10:50:29 -0600
From: Henning Schulzrinne <hgs@cs.columbia.edu>
Organization: Columbia University
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: robert.freilich@wcom.com
CC: ieprep@ietf.org
Subject: Re: [Ieprep] "IEPREP" priority value for Resource-Priority
References: <000f01c1d1c1$7dea87a0$1e4732a6@wcomnet.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

That was my preference as well. If any of the participants would like to
include their national scheme in the draft, please let me know.

Rob Freilich wrote:
> 
> Henning,
> 
> It might be better for each country to define their own namespace since this
> could be used as part of the key fields for authentication.  GETS uses a 10
> digit PIN code, if I remember correctly.  Other similar services might use a
> difference scheme altogether for authentication.  The national identifier
> from the Resource-Priority header plus the code could be used in-conjunction
> for authentication.  One party identifies the treatment and the other part
> the actual code to authenticate.
> 
> Regards,
> 
> Rob

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 22 15:17:38 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00683
	for <ieprep-archive@odin.ietf.org>; Fri, 22 Mar 2002 15:17:37 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA16120;
	Fri, 22 Mar 2002 15:17:09 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA16087
	for <ieprep@optimus.ietf.org>; Fri, 22 Mar 2002 15:17:07 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00677
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 15:17:04 -0500 (EST)
Received: from FRED-W2K6.cisco.com (sjc-vpn3-274.cisco.com [10.21.65.18])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2MKFwY01057;
	Fri, 22 Mar 2002 12:15:58 -0800 (PST)
Message-Id: <4.3.2.7.2.20020322134958.01f6ecd0@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 22 Mar 2002 13:50:29 -0600
To: "Beard, Cory" <BeardC@umkc.edu>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Levels of priority and comments on ieprep
  framework
Cc: "Sean Donelan" <sean@donelan.com>, <ieprep@ietf.org>
In-Reply-To: <EC61AA563B48B6459EEE976942BEA11918E8DD@KC-MAIL1.kc.umkc.ed
 u>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 03:14 PM 3/18/2002, Beard, Cory wrote:
>Can you point to a location where we can read these reports?

It's a National Academy of Science (CSTB) panel. the report will be 
published in a few months.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 22 15:19:01 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00727
	for <ieprep-archive@odin.ietf.org>; Fri, 22 Mar 2002 15:19:01 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA16067;
	Fri, 22 Mar 2002 15:16:34 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA16035
	for <ieprep@optimus.ietf.org>; Fri, 22 Mar 2002 15:16:33 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00662
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 15:16:30 -0500 (EST)
Received: from FRED-W2K6.cisco.com (sjc-vpn3-274.cisco.com [10.21.65.18])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2MKG1Y01095;
	Fri, 22 Mar 2002 12:16:01 -0800 (PST)
Message-Id: <4.3.2.7.2.20020322135058.01f745d0@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 22 Mar 2002 14:08:30 -0600
To: "Beard, Cory" <BeardC@umkc.edu>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Preemption
Cc: <ieprep@ietf.org>
In-Reply-To: <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.ed
 u>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 06:29 PM 3/18/2002, Beard, Cory wrote:
>How does this translate to packet-level dropping policies in IP?  It seems 
>to me that RED could be seen as a form of preemption.  Should we be on 
>record as not supporting RED, or should a distinction be made between 
>call-level preemption and packet-level preemption?

It's pre-emption, yes, but pre-emption of something other than what gets 
pre-empted in telco implementations. MLPP pre-emption, if implemented, is 
pre-emption of a call, which is an entire traffic stream. RED, from a VoIP 
perspective, is a random noise generator inserted in the circuit, whose 
effect is determined primarily by the competing traffic and the random 
number generator.

Personally, if I were someone trying to specify the behavior of a network 
for VoIP, I would want to know that I was implementing something that the 
best minds in the business agreed would provide good service for me. EF is 
that recommendation. I would want to have this working group specify its 
use. RFC 3246 talks about loss, and fundamentally says "it should not be 
the normal practice, apart from senders whose traffic doesn't conform to 
the service specification; a service which regularly loses conformant EF 
traffic does not conform to the EF specification." Specifying a change to 
it without a comparable level of modeling and testing to what was done to 
develop RFC 3246 seems unwise at best. You have no idea what effect your 
changes will have.

>IMHO we just should take no position about preemption in the framework 
>document.

I personally agree, if you're talking about call pre-emption. The framework 
should indicate that there will be policies applied in various networks, 
and that the exact structure of the policy is a matter of service offering 
and contract between the network provider and its customers. Pre-emption or 
lack of it is a policy which would be applied in a specific network, and on 
this list there has been discussion both of a network that does use it and 
a network that doesn't use it.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 22 15:34:13 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01146
	for <ieprep-archive@odin.ietf.org>; Fri, 22 Mar 2002 15:34:13 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA17261;
	Fri, 22 Mar 2002 15:32:18 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA17227
	for <ieprep@optimus.ietf.org>; Fri, 22 Mar 2002 15:32:16 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01104
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 15:32:13 -0500 (EST)
Received: from FRED-W2K6.cisco.com (sjc-vpn3-570.cisco.com [10.21.66.58])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2MKVCY12874;
	Fri, 22 Mar 2002 12:31:13 -0800 (PST)
Message-Id: <4.3.2.7.2.20020322142924.00b90f08@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 22 Mar 2002 14:29:58 -0600
To: Sean Donelan <sean@donelan.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Preemption
Cc: "Beard, Cory" <BeardC@umkc.edu>, ieprep@ietf.org
In-Reply-To: <Pine.GSO.4.40.0203182203430.16703-100000@clifden.donelan.c
 om>
References: <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 11:15 PM 3/18/2002, Sean Donelan wrote:
>I suspect you are stuck with packet drops assuming you want to use an
>IP network.

if you are doing call control (SIP/H.323) in an IP network, you most 
definitely can manage call count.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 22 16:20:03 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01979
	for <ieprep-archive@odin.ietf.org>; Fri, 22 Mar 2002 16:20:03 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA19653;
	Fri, 22 Mar 2002 16:19:38 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA19619
	for <ieprep@optimus.ietf.org>; Fri, 22 Mar 2002 16:19:36 -0500 (EST)
Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01970
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 16:19:32 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.9.3/8.9.3) with ESMTP id QAA10958;
	Fri, 22 Mar 2002 16:19:31 -0500 (EST)
Received: from cs.columbia.edu ([65.214.230.184])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.1/8.12.1) with ESMTP id g2MLJRPm007596
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Fri, 22 Mar 2002 16:19:29 -0500 (EST)
Message-ID: <3C9B9FC5.B57959FF@cs.columbia.edu>
Date: Fri, 22 Mar 2002 15:19:01 -0600
From: Henning Schulzrinne <hgs@cs.columbia.edu>
Organization: Columbia University
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Fred Baker <fred@cisco.com>
CC: Sean Donelan <sean@donelan.com>, "Beard, Cory" <BeardC@umkc.edu>,
        ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
References: <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.edu> <4.3.2.7.2.20020322142924.00b90f08@mira-sjcm-4.cisco.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

Fred Baker wrote:
> 
> At 11:15 PM 3/18/2002, Sean Donelan wrote:
> >I suspect you are stuck with packet drops assuming you want to use an
> >IP network.
> 
> if you are doing call control (SIP/H.323) in an IP network, you most
> definitely can manage call count.

Although translating call count to bandwidth may be an approximation
(good for voice codecs as long as users don't cheat at call setup time,
iffy for video).

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 22 23:41:21 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA09100
	for <ieprep-archive@odin.ietf.org>; Fri, 22 Mar 2002 23:41:21 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id XAA20764;
	Fri, 22 Mar 2002 23:37:37 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id XAA20732
	for <ieprep@optimus.ietf.org>; Fri, 22 Mar 2002 23:37:35 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA09052
	for <ieprep@ietf.org>; Fri, 22 Mar 2002 23:37:31 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id UAA17424; Fri, 22 Mar 2002 20:36:32 -0800 (PST)
Message-Id: <4.1.20020322160402.014f31e8@localhost>
Message-Id: <4.1.20020322160402.014f31e8@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Fri, 22 Mar 2002 16:11:37 -0600
To: Henning Schulzrinne <hgs@cs.columbia.edu>, robert.freilich@wcom.com
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] "IEPREP" priority value for Resource-Priority
Cc: ieprep@ietf.org
In-Reply-To: <3C9B60D5.951A81BB@cs.columbia.edu>
References: <000f01c1d1c1$7dea87a0$1e4732a6@wcomnet.com>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_285009==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_285009==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 10:50 AM 3/22/2002 -0600, Henning Schulzrinne wrote:
>That was my preference as well. If any of the participants would like to
>include their national scheme in the draft, please let me know.
>
>Rob Freilich wrote:
>> 
>> Henning,
>> 
>> It might be better for each country to define their own namespace since this
>> could be used as part of the key fields for authentication.  GETS uses a 10
>> digit PIN code, if I remember correctly.  Other similar services might use a
>> difference scheme altogether for authentication.  The national identifier
>> from the Resource-Priority header plus the code could be used in-conjunction
>> for authentication.  One party identifies the treatment and the other part
>> the actual code to authenticate.

Hal Folts at NCS was on a path to have IEPREP be international and call the
service "ECS" for "Emergency Communications System". I don't know where that
goal stands or if that this should be used as a namespace or was agreed upon by
each region/country, or if it needs to necessarily be. 

I have a concern *not* that each country or continent uses it's own namespace,
but in including all of them in this one ID; or is that a non-issue (and list
as many as we can know about)? It seems like listing all these different
namespace's might ought to be from within an IEPREP WG doc, which the
Resource-Priority ID isn't as wasn't planned on becoming.




>> 
>> Regards,
>> 
>> Rob
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep 
*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_285009==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 10:50 AM 3/22/2002 -0600, Henning Schulzrinne 
wrote:</div>
<div>&gt;That was my preference as well. If any of the participants would
like to</div>
<div>&gt;include their national scheme in the draft, please let me
know.</div>
<div>&gt;</div>
<div>&gt;Rob Freilich wrote:</div>
<div>&gt;&gt; </div>
<div>&gt;&gt; Henning,</div>
<div>&gt;&gt; </div>
<div>&gt;&gt; It might be better for each country to define their own
namespace since this</div>
<div>&gt;&gt; could be used as part of the key fields for
authentication.&nbsp; GETS uses a 10</div>
<div>&gt;&gt; digit PIN code, if I remember correctly.&nbsp; Other
similar services might use a</div>
<div>&gt;&gt; difference scheme altogether for authentication.&nbsp; The
national identifier</div>
<div>&gt;&gt; from the Resource-Priority header plus the code could be
used in-conjunction</div>
<div>&gt;&gt; for authentication.&nbsp; One party identifies the
treatment and the other part</div>
<div>&gt;&gt; the actual code to authenticate.</div>
<br>
<div>Hal Folts at NCS was on a path to have IEPREP be international and
call the service &quot;ECS&quot; for &quot;Emergency Communications
System&quot;. I don't know where that goal stands or if that this should
be used as a namespace or was agreed upon by each region/country, or if
it needs to necessarily be. </div>
<br>
<div>I have a concern *not* that each country or continent uses it's own
namespace, but in including all of them in this one ID; or is that a
non-issue (and list as many as we can know about)? It seems like listing
all these different namespace's might ought to be from within an IEPREP
WG doc, which the Resource-Priority ID isn't as wasn't planned on
becoming.</div>
<br>
<br>
<br>
<br>
<div>&gt;&gt; </div>
<div>&gt;&gt; Regards,</div>
<div>&gt;&gt; </div>
<div>&gt;&gt; Rob</div>
<div>&gt;</div>
<div>&gt;_______________________________________________</div>
<div>&gt;Ieprep mailing list</div>
<div>&gt;Ieprep@ietf.org</div>
&gt;<a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA=AUTOURL>https://www1.ietf.org/mailman/listinfo/ieprep</a>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_285009==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sat Mar 23 05:39:30 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA20297
	for <ieprep-archive@odin.ietf.org>; Sat, 23 Mar 2002 05:39:30 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id FAA03776;
	Sat, 23 Mar 2002 05:37:45 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id FAA03741
	for <ieprep@optimus.ietf.org>; Sat, 23 Mar 2002 05:37:44 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA20291
	for <ieprep@ietf.org>; Sat, 23 Mar 2002 05:37:39 -0500 (EST)
Received: from FRED-W2K6.cisco.com (sjc-vpn1-194.cisco.com [10.21.96.194])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2NAabY08338;
	Sat, 23 Mar 2002 02:36:38 -0800 (PST)
Message-Id: <4.3.2.7.2.20020323013032.02ad8ef0@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Sat, 23 Mar 2002 01:31:13 -0800
To: Henning Schulzrinne <hgs@cs.columbia.edu>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Preemption
Cc: ieprep@ietf.org
In-Reply-To: <3C9B9FC5.B57959FF@cs.columbia.edu>
References: <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.edu>
 <4.3.2.7.2.20020322142924.00b90f08@mira-sjcm-4.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 01:19 PM 3/22/2002, Henning Schulzrinne wrote:
>Although translating call count to bandwidth may be an approximation
>(good for voice codecs as long as users don't cheat at call setup time,
>iffy for video).

you could always signal your bandwidth, using RSVP with the EF tag on it. 
We've done a fair bit of work to make that possible...


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sat Mar 23 07:19:48 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA20957
	for <ieprep-archive@odin.ietf.org>; Sat, 23 Mar 2002 07:19:48 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id HAA14514;
	Sat, 23 Mar 2002 07:16:24 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id HAA14481
	for <ieprep@optimus.ietf.org>; Sat, 23 Mar 2002 07:16:23 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id HAA20933
	for <ieprep@ietf.org>; Sat, 23 Mar 2002 07:16:18 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.09478-0@bells.cs.ucl.ac.uk>; Sat, 23 Mar 2002 12:16:11 +0000
To: "James M. Polk" <jmpolk@cisco.com>
cc: Henning Schulzrinne <hgs@cs.columbia.edu>, robert.freilich@wcom.com,
        ieprep@ietf.org, K.Carlberg@cs.ucl.ac.uk
Subject: Re: [Ieprep] "IEPREP" priority value for Resource-Priority
In-reply-to: Your message of "Fri, 22 Mar 2002 16:11:37 CST." <4.1.20020322160402.014f31e8@localhost> <4.1.20020322160402.014f31e8@localhost>
Date: Sat, 23 Mar 2002 12:16:09 +0000
Message-ID: <15004.1016885769@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


I agree with James.  GETS is a service that uses the T1.631 High
Probability of Completion (HPC) standard that specifies  the NS/EP
code point for IAM messages.  it is this single binary code point that
we should have a corresponding value in a NameSpace for the R-P field.
other countries that use the National Security / Emergency Preparedness
(NS/EP) value is an orthogonal issue.

Internatioanl Emergency Preparedness Scheme (IEPS) is a term used in ITU
E.106, and is an attempt to extend the NS/EP to an international
perspective.  later, Hal has advocated a more general term of ECS. if
folks feel uncomfortable with the use of "IEPS" as a namespace, we could
use "HPC", or "NEPS" (national emergency preparedness scheme)....but i'd
like to give this more thought after i read the recent SG11 contribution
next week.

also, i'd strongly prefer to have the priority value associated with
the new NameSpace use the term "authorized-emergency", so that others
do not confuse the value with 911-type of emergency. (yes, i know that
there are no plans for a 911-type namespace for the R-P field, but
i want people to keep in mind the 'authorized' perspective.)

-ken


==============================================================
From:    "James M. Polk" <jmpolk@cisco.com>

>> It might be better for each country to define their own namespace since this
>> could be used as part of the key fields for authentication.  GETS uses a 10
>> digit PIN code, if I remember correctly.  Other similar services might use a
>> difference scheme altogether for authentication.  The national identifier
>> from the Resource-Priority header plus the code could be used in-conjunction
>> for authentication.  One party identifies the treatment and the other part
>> the actual code to authenticate.

Hal Folts at NCS was on a path to have IEPREP be international and call the
service "ECS" for "Emergency Communications System". I don't know where that
goal stands or if that this should be used as a namespace or was agreed upon by
each region/country, or if it needs to necessarily be. 

I have a concern *not* that each country or continent uses it's own namespace,
but in including all of them in this one ID; or is that a non-issue (and list
as many as we can know about)? It seems like listing all these different
namespace's might ought to be from within an IEPREP WG doc, which the
Resource-Priority ID isn't as wasn't planned on becoming.


James M. Polk
Consulting Engineer
Office of the CTO



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sat Mar 23 11:19:46 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA22775
	for <ieprep-archive@odin.ietf.org>; Sat, 23 Mar 2002 11:19:46 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA14020;
	Sat, 23 Mar 2002 11:18:57 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA13991
	for <ieprep@optimus.ietf.org>; Sat, 23 Mar 2002 11:18:56 -0500 (EST)
Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA22772
	for <ieprep@ietf.org>; Sat, 23 Mar 2002 11:18:51 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.9.3/8.9.3) with ESMTP id LAA27462;
	Sat, 23 Mar 2002 11:18:53 -0500 (EST)
Received: from cs.columbia.edu (cta.cs.columbia.edu [128.59.19.46])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.1/8.12.1) with ESMTP id g2NGIqPm010245
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Sat, 23 Mar 2002 11:18:52 -0500 (EST)
Message-ID: <3C9CAAD2.2577043@cs.columbia.edu>
Date: Sat, 23 Mar 2002 10:18:26 -0600
From: Henning Schulzrinne <hgs@cs.columbia.edu>
Organization: Columbia University
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Fred Baker <fred@cisco.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
References: <EC61AA563B48B6459EEE976942BEA11918E8DF@KC-MAIL1.kc.umkc.edu>
	 <4.3.2.7.2.20020322142924.00b90f08@mira-sjcm-4.cisco.com> <4.3.2.7.2.20020323013032.02ad8ef0@mira-sjcm-4.cisco.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

Yes, I (mis)understood your earlier comment as implying that
application-layer signaling could substitute for network-layer
reservations. This (mis)understanding is unfortunately a rather common
notion; H.323, to name one, proposes this and it pops up periodically on
various SIP mailing lists. I believe we're in agreement.

Fred Baker wrote:
> 
> At 01:19 PM 3/22/2002, Henning Schulzrinne wrote:
> >Although translating call count to bandwidth may be an approximation
> >(good for voice codecs as long as users don't cheat at call setup time,
> >iffy for video).
> 
> you could always signal your bandwidth, using RSVP with the EF tag on it.
> We've done a fair bit of work to make that possible...

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 24 10:02:06 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14173
	for <ieprep-archive@odin.ietf.org>; Sun, 24 Mar 2002 10:02:05 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA18655;
	Sun, 24 Mar 2002 09:45:38 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA18618
	for <ieprep@optimus.ietf.org>; Sun, 24 Mar 2002 09:45:36 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id JAA14069
	for <ieprep@ietf.org>; Sun, 24 Mar 2002 09:45:32 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.12841-0@bells.cs.ucl.ac.uk>; Sun, 24 Mar 2002 14:45:29 +0000
To: Fred Baker <fred@cisco.com>
cc: "Beard, Cory" <BeardC@umkc.edu>, ieprep <ieprep@ietf.org>
Subject: Re: [Ieprep] Preemption
In-reply-to: Your message of "Fri, 22 Mar 2002 14:08:30 CST." <4.3.2.7.2.20020322135058.01f745d0@mira-sjcm-4.cisco.com>
Date: Sun, 24 Mar 2002 14:45:27 +0000
Message-ID: <18766.1016981127@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


> >IMHO we just should take no position about preemption in the framework 
> >document.
> 
> I personally agree, if you're talking about call pre-emption. The framework 
> should indicate that there will be policies applied in various networks, 
> and that the exact structure of the policy is a matter of service offering 
> and contract between the network provider and its customers. Pre-emption or 
> lack of it is a policy which would be applied in a specific network, and on 
> this list there has been discussion both of a network that does use it and 
> a network that doesn't use it.

as it is written now, the Framework document takes the position that
non-preemptive service is the default service or mode of operation.  

there were a couple of thoughts behind this approach.  the current
deployed effort in IP telephony does not support pre-emption.  hence,
using this installed base as the default model would seem to be a nice
starting point -- much in the way that the inclusion of new service
models in IP still led the IETF to take the position that best effort
is the default service model for IP.

it also seemed that pre-emptive service in non-private networks involves
a combination of local law/regulation as well as additional mechanisms
that go beyond just labeling.  so, if we state that if for some reason a
lack of policy exists in a network for the received label, then the
default action taken should be non-preemptive.  thereby further
reinforcing the distinction between label and the action taken upon it.

i believe the framework document already articulates what Fred says
above about policies, but i'll double check to make sure.

-ken



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 24 11:13:21 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA15162
	for <ieprep-archive@odin.ietf.org>; Sun, 24 Mar 2002 11:13:21 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA04901;
	Sun, 24 Mar 2002 10:59:47 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA04859
	for <ieprep@optimus.ietf.org>; Sun, 24 Mar 2002 10:59:45 -0500 (EST)
Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14918
	for <ieprep@ietf.org>; Sun, 24 Mar 2002 10:59:42 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.9.3/8.9.3) with ESMTP id KAA21033;
	Sun, 24 Mar 2002 10:59:43 -0500 (EST)
Received: from cs.columbia.edu (cta.cs.columbia.edu [128.59.19.46])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.1/8.12.1) with ESMTP id g2OFxgPm017412
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Sun, 24 Mar 2002 10:59:43 -0500 (EST)
Message-ID: <3C9DF7D3.D5F1BF49@cs.columbia.edu>
Date: Sun, 24 Mar 2002 09:59:15 -0600
From: Henning Schulzrinne <hgs@cs.columbia.edu>
Organization: Columbia University
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
CC: "James M. Polk" <jmpolk@cisco.com>, robert.freilich@wcom.com,
        ieprep@ietf.org
Subject: Re: [Ieprep] "IEPREP" priority value for Resource-Priority
References: <15004.1016885769@cs.ucl.ac.uk>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

Given your description, I suspect that it's best to then defer this
until the issue settles down. Givent that the R-P draft has a
well-defined (I hope...) extension and registration mechanism, we can
register namespaces at any time, either national ones or a global ECS
scheme.

Ken Carlberg wrote:
> 
> I agree with James.  GETS is a service that uses the T1.631 High
> Probability of Completion (HPC) standard that specifies  the NS/EP
> code point for IAM messages.  it is this single binary code point that
> we should have a corresponding value in a NameSpace for the R-P field.
> other countries that use the National Security / Emergency Preparedness
> (NS/EP) value is an orthogonal issue.
> 
> Internatioanl Emergency Preparedness Scheme (IEPS) is a term used in ITU
> E.106, and is an attempt to extend the NS/EP to an international
> perspective.  later, Hal has advocated a more general term of ECS. if
> folks feel uncomfortable with the use of "IEPS" as a namespace, we could
> use "HPC", or "NEPS" (national emergency preparedness scheme)....but i'd
> like to give this more thought after i read the recent SG11 contribution
> next week.
> 
> also, i'd strongly prefer to have the priority value associated with
> the new NameSpace use the term "authorized-emergency", so that others
> do not confuse the value with 911-type of emergency. (yes, i know that
> there are no plans for a 911-type namespace for the R-P field, but
> i want people to keep in mind the 'authorized' perspective.)
> 
> -ken
> 
> ==============================================================
> From:    "James M. Polk" <jmpolk@cisco.com>
> 
> >> It might be better for each country to define their own namespace since this
> >> could be used as part of the key fields for authentication.  GETS uses a 10
> >> digit PIN code, if I remember correctly.  Other similar services might use a
> >> difference scheme altogether for authentication.  The national identifier
> >> from the Resource-Priority header plus the code could be used in-conjunction
> >> for authentication.  One party identifies the treatment and the other part
> >> the actual code to authenticate.
> 
> Hal Folts at NCS was on a path to have IEPREP be international and call the
> service "ECS" for "Emergency Communications System". I don't know where that
> goal stands or if that this should be used as a namespace or was agreed upon by
> each region/country, or if it needs to necessarily be.
> 
> I have a concern *not* that each country or continent uses it's own namespace,
> but in including all of them in this one ID; or is that a non-issue (and list
> as many as we can know about)? It seems like listing all these different
> namespace's might ought to be from within an IEPREP WG doc, which the
> Resource-Priority ID isn't as wasn't planned on becoming.
> 
> James M. Polk
> Consulting Engineer
> Office of the CTO
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 24 22:33:16 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA23838
	for <ieprep-archive@odin.ietf.org>; Sun, 24 Mar 2002 22:33:15 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id WAA08186;
	Sun, 24 Mar 2002 22:30:34 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id WAA08155
	for <ieprep@optimus.ietf.org>; Sun, 24 Mar 2002 22:30:33 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA23082
	for <ieprep@ietf.org>; Sun, 24 Mar 2002 22:30:28 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id UAA18467
	for <ieprep@ietf.org>; Sun, 24 Mar 2002 20:28:38 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA4E01
          for <ieprep@ietf.org>; Sun, 24 Mar 2002 20:30:30 -0700
Received: FROM agcs.com BY mknaomi ; Sun Mar 24 20:30:29 2002 -0700
Message-ID: <3C9E9256.9FA13438@agcs.com>
Date: Sun, 24 Mar 2002 19:58:30 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Bob Bell <rtbell@cisco.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] FW: Call Preemption in Public Switched Telephone Networks
References: <E58F685255415A4EB2D4B8A02AEF32E3CC42A8@evvbu-exchange.amer.unity.cisco.com>
Content-Type: multipart/mixed;
 boundary="------------7EC261FA2CBA9687D26B7156"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------7EC261FA2CBA9687D26B7156
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Bob,

I really appreciate your investigation with the FCC and your report.

Having said that, I think we are still pretty much in agreement that preempting calls within the US, is just not in the cards.

Bob Bell wrote:

> Folks -
>
> I am somewhat embarrassed to state this but I was wrong relative to the legality of pre-empting a circuit mode call. As a result of the thread discussing this, I sent a query to the FCC. Here is their reply. While not absolute, it gives enough guidence to indicate that it is possible.
>
> Bob Bell
>
> -----Original Message-----
> From: FCCTSR18 [mailto:FCCTSR18@fcc.gov]
> Sent: Wednesday, 20 March, 2002 12:44 PM
> To: Bob Bell
> Subject: Re: Call Preemption in Public Switched Telephone Networks
>
> Hello BOB - - hope you are well.
>
> Bluntly stated - - I would be hard-pressed to find an FCC rule that addresses this scenario, as generally stated.  There is an appendix to Part 64 of our rules (47 CFR, Part 64, Appendix A) that addresses Telecommunications Service Priority System for National Security Emergency Preparedness; so in an emergency situation, I can understand where a scenario of this nature could take place.
>
> However, in normal operation of the Public Switched Telephone Network, this type of activity just doesn't seem practical.  After all, what would we do with the "busy" signal - - it wouldn't be needed any more.
>
> It is my understanding that most carriers offer an "interrupt" service; whereby when a caller needs to make a particular phone call (usually in an emergency situation); the carrier can "interrupt" the present call.  However, this is something "requested" by a caller; not instituted by the carrier in order to serve a higher paying/priority customer; and the carrier generally levies a "hefty" charge for this interrupt service.
>
> Further, why would a carrier want to do something of this nature?  Seems like this type of activity would eventually upset their customers (I know it would me) and then they wouldn't have enough left to worry about dropping someone's call to free up their network to process someone else's call.
>
> If I'm "off track" here with my line of thinking; please let me know.  Also, if there are specific examples; beyond your general concerns below, that you'd care for me to address; please let me know.
>
> Hope this helps - - thanks for contacting the FCC - - have a great day ! ! !
>
> >>> "Bob Bell" <rtbell@cisco.com> 03/14/02 04:42PM >>>
> I am sorry for the general nature of theis query, but, in the process of developing some standards for Telephony over IP, a question has been generated that I though I knew the answer to. However, several people have challenged my answer and I would appreciate it if I could get the correct answer.
>
> The question is:
>
> Is it legally permissable to drop a call in progress (i.e. after connection is complete) in order to free the utilized resource for use by another call?In other words, within the Public Telephone system, can an existing call be preempted by another call?
>
> I would appreciate knowing the actual regulations if there are such.
>
> Thank you for your time and energy in this matter.
>
> Bob Bell
> Cisco Systems, Inc.
> 576 S. Brentwood Ln.
> Bountiful, Utah 84010
> 801-294-3034 (v)
> 801-204-3023 (f)
> 801-971-4200 (cell)
> rtbell@cisco.com <mailto:rtbell@cisco.com> (email)
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

--------------7EC261FA2CBA9687D26B7156
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------7EC261FA2CBA9687D26B7156--



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 24 22:41:24 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA24041
	for <ieprep-archive@odin.ietf.org>; Sun, 24 Mar 2002 22:41:24 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id WAA10647;
	Sun, 24 Mar 2002 22:40:57 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id WAA10597
	for <ieprep@optimus.ietf.org>; Sun, 24 Mar 2002 22:40:55 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA24034
	for <ieprep@ietf.org>; Sun, 24 Mar 2002 22:40:51 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id UAA18765
	for <ieprep@ietf.org>; Sun, 24 Mar 2002 20:39:01 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA4F19
          for <ieprep@ietf.org>; Sun, 24 Mar 2002 20:40:53 -0700
Received: FROM agcs.com BY mknaomi ; Sun Mar 24 20:40:53 2002 -0700
Message-ID: <3C9E9C45.ED4703DA@agcs.com>
Date: Sun, 24 Mar 2002 20:40:53 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
CC: ieprep <ieprep@ietf.org>
Subject: Re: [Ieprep] Preemption
References: <18766.1016981127@cs.ucl.ac.uk>
Content-Type: multipart/mixed;
 boundary="------------FEBB7598A50186E3E71FF8F4"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------FEBB7598A50186E3E71FF8F4
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Ken,

If I correctly understand what you have written, you are suggesting non
pre-emption is the default, but are not preventing pre-emption (but not as the
default) in those networks where it might be desired.

This sounds like an approach that everybody should be able to live with.


Ken Carlberg wrote:

> > >IMHO we just should take no position about preemption in the framework
> > >document.
> >
> > I personally agree, if you're talking about call pre-emption. The framework
> > should indicate that there will be policies applied in various networks,
> > and that the exact structure of the policy is a matter of service offering
> > and contract between the network provider and its customers. Pre-emption or
> > lack of it is a policy which would be applied in a specific network, and on
> > this list there has been discussion both of a network that does use it and
> > a network that doesn't use it.
>
> as it is written now, the Framework document takes the position that
> non-preemptive service is the default service or mode of operation.
>
> there were a couple of thoughts behind this approach.  the current
> deployed effort in IP telephony does not support pre-emption.  hence,
> using this installed base as the default model would seem to be a nice
> starting point -- much in the way that the inclusion of new service
> models in IP still led the IETF to take the position that best effort
> is the default service model for IP.
>
> it also seemed that pre-emptive service in non-private networks involves
> a combination of local law/regulation as well as additional mechanisms
> that go beyond just labeling.  so, if we state that if for some reason a
> lack of policy exists in a network for the received label, then the
> default action taken should be non-preemptive.  thereby further
> reinforcing the distinction between label and the action taken upon it.
>
> i believe the framework document already articulates what Fred says
> above about policies, but i'll double check to make sure.
>
> -ken
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

--------------FEBB7598A50186E3E71FF8F4
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------FEBB7598A50186E3E71FF8F4--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 06:20:37 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06748
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 06:20:37 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id GAA25942;
	Mon, 25 Mar 2002 06:20:22 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id GAA25901
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 06:20:21 -0500 (EST)
Received: from portal.east.saic.com (portal.east.saic.com [198.151.13.15])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id GAA06744
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 06:20:19 -0500 (EST)
Received: from mclmx.saic.com by portal.east.saic.com
          via smtpd (for odin.ietf.org [132.151.1.176]) with SMTP; 25 Mar 2002 11:20:21 UT
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 25 Mar 2002 06:20:04 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.1.19) with SMTP id M2002032506044008991
 for <ieprep@ietf.org>; Mon, 25 Mar 2002 06:04:40 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <HTRGP9A6>; Mon, 25 Mar 2002 06:04:38 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D6428915FB@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Date: Mon, 25 Mar 2002 06:04:40 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] GETS stats
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

All,

Here is some information on GETS surrounding September 11th (as requested
during the recent ieprep WG meeting).

Kimberly

___
Example call volume on September 11th
*
*Verizon's networks reached twice the normal daily volumes of 115 million
calls in New York City and 35 millions calls in the nation's capital. On a
normal business day Verizon handles 1.5 billion calls.
*
*AT&T's long distance network carried a record 431 million call attempts,
101 million more than the previous high-traffic day.

__________________

*18,000 GETS calls worldwide (2 weeks)
*10,000 GETS calls in the WTC and Pentagon areas (2 weeks)
*Completion rate exceeded 95% 
*Highest calling rate in the first 48 hours
*1,500 GETS PIN cards were used nationwide (~ 3 %)
*
*2,000 new GETS cards were issued in two weeks following 9/11
*Between 11 Sep 2001 and 31 Jan 2002, 12,000 new GETS cards distributed
nationwide
*
 
WTC/Pentagon 11 Sep 2001


GETS Calls                ~ 2,900 (88% complete)
Verizon Calls  ~ 300,000,000 (peak overload 10X)'Approx. 1 GETS call per
100,000 Verizon Public Calls'
AT&T Calls     ~ 431,000,000 (nationwide - busiest day... ever)

Wireless overload extraordinary - significant factor degrading GETS  

Nisqually Earthquake 28 Feb 2001 (Seattle)

GETS Calls                ~ 324 (88% complete)
Qwest Calls   ~ 72,000,000 (est. peak 4X )



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 06:27:24 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06786
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 06:27:24 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id GAA22930;
	Mon, 25 Mar 2002 06:06:48 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id GAA22897
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 06:06:47 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06664
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 06:06:44 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id EAA04703
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 04:04:53 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA2716
          for <ieprep@ietf.org>; Mon, 25 Mar 2002 04:06:45 -0700
Received: FROM agcs.com BY mknaomi ; Mon Mar 25 04:06:43 2002 -0700
Message-ID: <3C9F04C4.579D7DFD@agcs.com>
Date: Mon, 25 Mar 2002 04:06:44 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] "IEPREP" priority value for Resource-Priority
References: <15004.1016885769@cs.ucl.ac.uk>
Content-Type: multipart/mixed;
 boundary="------------2E924FF39B17256969994754"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------2E924FF39B17256969994754
Content-Type: multipart/alternative;
 boundary="------------DB581305CE41C28797C32FB4"


--------------DB581305CE41C28797C32FB4
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Ken,

I believe that Hal is now using the term ETS (Emergency Telecommunications Service)
rather than ECS.

At the ITU-T SG 11 meeting last month we added a new, optional parameter to the SS7
( and BICC) IAM message in the internationally defined range of values indicating
that it was an emergency service call.

Because this is a different mechanism than the Calling Party's Category in the same
SS7 IAM message, a given call can be:

   * an ordinary call
   * a nationally defined emergency call based on the national value of the CPC,
     such as NS/EP in the US, indicating a GETS HPC call
   *  an internationally recognized emergency call based on the new SS7 IAM
     parameter
   * a call that is both a nationally defined emergency call based on the national
     value of the CPC, such as NS/EP in the US, indicating a GETS HPC call, and an
     internationally recognized emergency call based on the new SS7 IAM parameter

Ken Carlberg wrote:

> I agree with James.  GETS is a service that uses the T1.631 High
> Probability of Completion (HPC) standard that specifies  the NS/EP
> code point for IAM messages.  it is this single binary code point that
> we should have a corresponding value in a NameSpace for the R-P field.
> other countries that use the National Security / Emergency Preparedness
> (NS/EP) value is an orthogonal issue.
>
> Internatioanl Emergency Preparedness Scheme (IEPS) is a term used in ITU
> E.106, and is an attempt to extend the NS/EP to an international
> perspective.  later, Hal has advocated a more general term of ECS. if
> folks feel uncomfortable with the use of "IEPS" as a namespace, we could
> use "HPC", or "NEPS" (national emergency preparedness scheme)....but i'd
> like to give this more thought after i read the recent SG11 contribution
> next week.
>
> also, i'd strongly prefer to have the priority value associated with
> the new NameSpace use the term "authorized-emergency", so that others
> do not confuse the value with 911-type of emergency. (yes, i know that
> there are no plans for a 911-type namespace for the R-P field, but
> i want people to keep in mind the 'authorized' perspective.)
>
> -ken
>
> ==============================================================
> From:    "James M. Polk" <jmpolk@cisco.com>
>
> >> It might be better for each country to define their own namespace since this
> >> could be used as part of the key fields for authentication.  GETS uses a 10
> >> digit PIN code, if I remember correctly.  Other similar services might use a
> >> difference scheme altogether for authentication.  The national identifier
> >> from the Resource-Priority header plus the code could be used in-conjunction
> >> for authentication.  One party identifies the treatment and the other part
> >> the actual code to authenticate.
>
> Hal Folts at NCS was on a path to have IEPREP be international and call the
> service "ECS" for "Emergency Communications System". I don't know where that
> goal stands or if that this should be used as a namespace or was agreed upon by
> each region/country, or if it needs to necessarily be.
>
> I have a concern *not* that each country or continent uses it's own namespace,
> but in including all of them in this one ID; or is that a non-issue (and list
> as many as we can know about)? It seems like listing all these different
> namespace's might ought to be from within an IEPREP WG doc, which the
> Resource-Priority ID isn't as wasn't planned on becoming.
>
> James M. Polk
> Consulting Engineer
> Office of the CTO
>
> _______________________________________________

--------------DB581305CE41C28797C32FB4
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Ken,
<p>I believe that Hal is now using the term ETS (Emergency Telecommunications
Service) rather than ECS.
<p>At the ITU-T SG 11 meeting last month we added a new, optional parameter
to the SS7 ( and BICC) IAM message in the internationally defined range
of values indicating that it was an emergency service call.
<p>Because this is a different mechanism than the Calling Party's Category
in the same SS7 IAM message, a given call can be:
<ul>
<li>
an ordinary call</li>

<li>
a nationally defined emergency call based on the national value of the
CPC, such as NS/EP in the US, indicating a GETS HPC call</li>

<li>
&nbsp;an internationally recognized emergency call based on the new SS7
IAM parameter</li>

<li>
a call that is both a nationally defined emergency call based on the national
value of the CPC, such as NS/EP in the US, indicating a GETS HPC call,
and an internationally recognized emergency call based on the new SS7 IAM
parameter</li>
</ul>

<p><br>Ken Carlberg wrote:
<blockquote TYPE=CITE>I agree with James.&nbsp; GETS is a service that
uses the T1.631 High
<br>Probability of Completion (HPC) standard that specifies&nbsp; the NS/EP
<br>code point for IAM messages.&nbsp; it is this single binary code point
that
<br>we should have a corresponding value in a NameSpace for the R-P field.
<br>other countries that use the National Security / Emergency Preparedness
<br>(NS/EP) value is an orthogonal issue.
<p>Internatioanl Emergency Preparedness Scheme (IEPS) is a term used in
ITU
<br>E.106, and is an attempt to extend the NS/EP to an international
<br>perspective.&nbsp; later, Hal has advocated a more general term of
ECS. if
<br>folks feel uncomfortable with the use of "IEPS" as a namespace, we
could
<br>use "HPC", or "NEPS" (national emergency preparedness scheme)....but
i'd
<br>like to give this more thought after i read the recent SG11 contribution
<br>next week.
<p>also, i'd strongly prefer to have the priority value associated with
<br>the new NameSpace use the term "authorized-emergency", so that others
<br>do not confuse the value with 911-type of emergency. (yes, i know that
<br>there are no plans for a 911-type namespace for the R-P field, but
<br>i want people to keep in mind the 'authorized' perspective.)
<p>-ken
<p>==============================================================
<br>From:&nbsp;&nbsp;&nbsp; "James M. Polk" &lt;jmpolk@cisco.com>
<p>>> It might be better for each country to define their own namespace
since this
<br>>> could be used as part of the key fields for authentication.&nbsp;
GETS uses a 10
<br>>> digit PIN code, if I remember correctly.&nbsp; Other similar services
might use a
<br>>> difference scheme altogether for authentication.&nbsp; The national
identifier
<br>>> from the Resource-Priority header plus the code could be used in-conjunction
<br>>> for authentication.&nbsp; One party identifies the treatment and
the other part
<br>>> the actual code to authenticate.
<p>Hal Folts at NCS was on a path to have IEPREP be international and call
the
<br>service "ECS" for "Emergency Communications System". I don't know where
that
<br>goal stands or if that this should be used as a namespace or was agreed
upon by
<br>each region/country, or if it needs to necessarily be.
<p>I have a concern *not* that each country or continent uses it's own
namespace,
<br>but in including all of them in this one ID; or is that a non-issue
(and list
<br>as many as we can know about)? It seems like listing all these different
<br>namespace's might ought to be from within an IEPREP WG doc, which the
<br>Resource-Priority ID isn't as wasn't planned on becoming.
<p>James M. Polk
<br>Consulting Engineer
<br>Office of the CTO
<p>_______________________________________________</blockquote>
</html>

--------------DB581305CE41C28797C32FB4--

--------------2E924FF39B17256969994754
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------2E924FF39B17256969994754--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 06:29:01 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06826
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 06:29:01 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id GAA27914;
	Mon, 25 Mar 2002 06:28:37 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id GAA27877
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 06:28:36 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA06823
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 06:28:34 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id EAA05335
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 04:26:42 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA29DF
          for <ieprep@ietf.org>; Mon, 25 Mar 2002 04:28:34 -0700
Received: FROM agcs.com BY mknaomi ; Mon Mar 25 04:28:33 2002 -0700
Message-ID: <3C9F09E2.AFDC67DE@agcs.com>
Date: Mon, 25 Mar 2002 04:28:34 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "King, Kimberly S." <KIMBERLY.S.KING@saic.com>
CC: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: Re: [Ieprep] GETS stats
References: <B8030EB94AF1D51196D70002A589D6428915FB@mcl-its-exs01.mail.saic.com>
Content-Type: multipart/mixed;
 boundary="------------930E0EDCB719FAC348996382"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------930E0EDCB719FAC348996382
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Kimberly,

I am most impressed by the 95% completion rate.



"King, Kimberly S." wrote:

> All,
>
> Here is some information on GETS surrounding September 11th (as requested
> during the recent ieprep WG meeting).
>
> Kimberly
>
> ___
> Example call volume on September 11th
> *
> *Verizon's networks reached twice the normal daily volumes of 115 million
> calls in New York City and 35 millions calls in the nation's capital. On a
> normal business day Verizon handles 1.5 billion calls.
> *
> *AT&T's long distance network carried a record 431 million call attempts,
> 101 million more than the previous high-traffic day.
>
> __________________
>
> *18,000 GETS calls worldwide (2 weeks)
> *10,000 GETS calls in the WTC and Pentagon areas (2 weeks)
> *Completion rate exceeded 95%
> *Highest calling rate in the first 48 hours
> *1,500 GETS PIN cards were used nationwide (~ 3 %)
> *
> *2,000 new GETS cards were issued in two weeks following 9/11
> *Between 11 Sep 2001 and 31 Jan 2002, 12,000 new GETS cards distributed
> nationwide
> *
>
> WTC/Pentagon 11 Sep 2001
>
> GETS Calls                ~ 2,900 (88% complete)
> Verizon Calls  ~ 300,000,000 (peak overload 10X)'Approx. 1 GETS call per
> 100,000 Verizon Public Calls'
> AT&T Calls     ~ 431,000,000 (nationwide - busiest day... ever)
>
> Wireless overload extraordinary - significant factor degrading GETS
>
> Nisqually Earthquake 28 Feb 2001 (Seattle)
>
> GETS Calls                ~ 324 (88% complete)
> Qwest Calls   ~ 72,000,000 (est. peak 4X )
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

--------------930E0EDCB719FAC348996382
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------930E0EDCB719FAC348996382--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 08:17:35 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA08927
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 08:17:35 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA23643;
	Mon, 25 Mar 2002 08:17:20 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA23595
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 08:17:18 -0500 (EST)
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA08850;
	Mon, 25 Mar 2002 08:17:16 -0500 (EST)
Message-Id: <200203251317.IAA08850@ietf.org>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; Boundary="NextPart"
To: IETF-Announce: ;
Cc: ieprep@ietf.org
From: Internet-Drafts@ietf.org
Reply-to: Internet-Drafts@ietf.org
Date: Mon, 25 Mar 2002 08:17:16 -0500
Subject: [Ieprep] I-D ACTION:draft-ietf-ieprep-framework-00.txt
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Internet Emergency Preparedness Working Group of the IETF.

	Title		: Framework for Supporting IEPS in IP Telephony
	Author(s)	: K. Carlberg, I. Brown
	Filename	: draft-ietf-ieprep-framework-00.txt
	Pages		: 26
	Date		: 21-Mar-02
	
This document presents a framework for supporting authorized
emergency related communication within the context of IP telephony.
We present a series of objectives that reflect a general view of how
authorized emergency service, in line with the International
Emergency Preparedness Scheme (IEPS), should be realized within
today's IP architecture and service models.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-ieprep-framework-00.txt

To remove yourself from the IETF Announcement list, send a message to 
ietf-announce-request with the word unsubscribe in the body of the message.

Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-ieprep-framework-00.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html 
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt


Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-ieprep-framework-00.txt".
	
NOTE:	The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.
		
		
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body;
	access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID:	<20020325080548.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-ieprep-framework-00.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-ieprep-framework-00.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<20020325080548.I-D@ietf.org>

--OtherAccess--

--NextPart--



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 12:17:37 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA20742
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 12:17:37 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA19771;
	Mon, 25 Mar 2002 11:59:50 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA19736
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 11:59:49 -0500 (EST)
Received: from web20803.mail.yahoo.com (web20803.mail.yahoo.com [216.136.226.192])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id LAA19921
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 11:59:44 -0500 (EST)
Message-ID: <20020325165947.78824.qmail@web20803.mail.yahoo.com>
Received: from [134.193.128.233] by web20803.mail.yahoo.com via HTTP; Mon, 25 Mar 2002 08:59:47 PST
Date: Mon, 25 Mar 2002 08:59:47 -0800 (PST)
From: senthil ayyasamy <mplsgeek@yahoo.com>
To: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: [Ieprep] concluded WG
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

Hi all,
The list of concluded working groups includes
*emergency information infrastructure*
(http://ietf.org/html.charters/OLD/eii-charter.html).
It was chaired by Art Botterell and concluded in
January 1997.  Did the group produce any documents? I
am curious to know about the technology employed in
1994-1997 to find a solution for emergencies.

Thanks in advance,
Senthil


=====
Ayyasamy Senthilkumar
M.S.(Computer Networking)
UMKC,MO.
-----------------------------
An expert is a man who has made all the mistakes that can be made in a very narrow field.
--Niels Bohr.

__________________________________________________
Do You Yahoo!?
Yahoo! Movies - coverage of the 74th Academy Awards®
http://movies.yahoo.com/

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 13:16:23 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA23763
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 13:16:23 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA06913;
	Mon, 25 Mar 2002 13:07:43 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA06875
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 13:07:41 -0500 (EST)
Received: from rbhub101.chamb.disa.mil (rbhub101.chamb.disa.mil [209.22.120.24])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA23258
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 13:07:38 -0500 (EST)
Received: by rbhub101.chamb.disa.mil with Internet Mail Service (5.5.2653.19)
	id <HTMJMJ91>; Mon, 25 Mar 2002 13:00:37 -0500
Message-ID: <D6EC0AD017C1D4119CAF00204804F0CC014D3D18@rbmail101.chamb.disa.mil>
From: "Choi, Don" <choid@ncr.disa.mil>
To: "'Fred Baker'" <fred@cisco.com>, "Beard, Cory" <BeardC@umkc.edu>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Preemption
Date: Mon, 25 Mar 2002 13:00:33 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

Fred and Cory,

I agree with you that how a network supports preferential treatments for
high priority calls should be up to the service provider and its customers.
We should not specify or exclude any specific mechanisms.

Don 

 

>IMHO we just should take no position about preemption in the framework 
>document.

I personally agree, if you're talking about call pre-emption. The framework 
should indicate that there will be policies applied in various networks, 
and that the exact structure of the policy is a matter of service offering 
and contract between the network provider and its customers. Pre-emption or 
lack of it is a policy which would be applied in a specific network, and on 
this list there has been discussion both of a network that does use it and 
a network that doesn't use it.




_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 13:19:46 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24019
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 13:19:46 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA09563;
	Mon, 25 Mar 2002 13:19:23 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA09522
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 13:19:21 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA23968
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 13:19:17 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id KAA21186 for <ieprep@ietf.org>; Mon, 25 Mar 2002 10:18:48 -0800 (PST)
Message-Id: <4.1.20020325121739.03225418@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Mon, 25 Mar 2002 12:18:19 -0600
To: ieprep <ieprep@ietf.org>
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] Preemption
In-Reply-To: <3C9E9C45.ED4703DA@agcs.com>
References: <18766.1016981127@cs.ucl.ac.uk>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_222425130==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_222425130==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 08:40 PM 3/24/2002 -0700, Stuart (Stu) Goldman wrote:
>Ken,
>
>If I correctly understand what you have written, you are suggesting non
>pre-emption is the default, but are not preventing pre-emption (but not as the
>default) in those networks where it might be desired.
>
>This sounds like an approach that everybody should be able to live with.

I like this approach



*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_222425130==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 08:40 PM 3/24/2002 -0700, Stuart (Stu) Goldman
wrote:</div>
<div>&gt;Ken,</div>
<div>&gt;</div>
<div>&gt;If I correctly understand what you have written, you are
suggesting non</div>
<div>&gt;pre-emption is the default, but are not preventing pre-emption
(but not as the</div>
<div>&gt;default) in those networks where it might be desired.</div>
<div>&gt;</div>
<div>&gt;This sounds like an approach that everybody should be able to
live with.</div>
<br>
<div>I like this approach</div>
<br>
<br>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_222425130==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 13:25:42 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24365
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 13:25:42 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA10962;
	Mon, 25 Mar 2002 13:25:29 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA10905
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 13:25:27 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24359
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 13:25:23 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id KAA29859 for <ieprep@ietf.org>; Mon, 25 Mar 2002 10:24:53 -0800 (PST)
Message-Id: <4.1.20020325122103.031deef0@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Mon, 25 Mar 2002 12:24:25 -0600
To: ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: Re: [Ieprep] "IEPREP" priority value for Resource-Priority
In-Reply-To: <3C9F04C4.579D7DFD@agcs.com>
References: <15004.1016885769@cs.ucl.ac.uk>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_222790806==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_222790806==_.ALT
Content-Type: text/plain; charset="us-ascii"

At 04:06 AM 3/25/2002 -0700, Stuart (Stu) Goldman wrote: 
>
> Ken, 
>
> I believe that Hal is now using the term ETS (Emergency Telecommunications
> Service) rather than ECS. 


I have an issue with this change in view of the IEPREP Charter addressing more
than Voice (ie Telecommunications) -- and of the two then prefer ECS (which
could mean anything from Voice to IM to FTP to HTTP to whatever IAA uses)


*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_222790806==_.ALT
Content-Type: text/html; charset="us-ascii"

<html>
At 04:06 AM 3/25/2002 -0700, Stuart (Stu) Goldman wrote: <br>
<blockquote type=cite cite>Ken, <br>
<br>
I believe that Hal is now using the term ETS (Emergency
Telecommunications Service) rather than ECS. </blockquote><br>
I have an issue with this change in view of the IEPREP Charter addressing
more than Voice (ie Telecommunications) -- and of the two then prefer ECS
(which could mean anything from Voice to IM to FTP to HTTP to whatever
IAA uses)<br>
<br>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_222790806==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 13:34:39 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24992
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 13:34:39 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA13125;
	Mon, 25 Mar 2002 13:34:21 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA13083
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 13:34:19 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24977
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 13:34:16 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id KAA12156; Mon, 25 Mar 2002 10:33:43 -0800 (PST)
Message-Id: <4.1.20020325122909.00a73f50@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Mon, 25 Mar 2002 12:33:15 -0600
To: "Choi, Don" <choid@ncr.disa.mil>, ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: RE: [Ieprep] Preemption
In-Reply-To: <D6EC0AD017C1D4119CAF00204804F0CC014D3D18@rbmail101.chamb.d
 isa.mil>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_223320317==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_223320317==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 01:00 PM 3/25/2002 -0500, Choi, Don wrote:
>Fred and Cory,
>
>I agree with you that how a network supports preferential treatments for
>high priority calls should be up to the service provider and its customers.
>We should not specify or exclude any specific mechanisms.

then, wouldn't it be better to acknowledge preemption and state it as being up
to local policy and mechanisms than to ignore it and not have reference(s) to
preemption within the doc (as Cory seems to indicate)? I'd strongly prefer to
inform about it, taking the position of local policy/mechanisms decide, stating
where it shouldn't or can't be used (the circumstances)...


>
>Don 
>
> 
>
>>IMHO we just should take no position about preemption in the framework 
>>document.
>
>I personally agree, if you're talking about call pre-emption. The framework 
>should indicate that there will be policies applied in various networks, 
>and that the exact structure of the policy is a matter of service offering 
>and contract between the network provider and its customers. Pre-emption or 
>lack of it is a policy which would be applied in a specific network, and on 
>this list there has been discussion both of a network that does use it and 
>a network that doesn't use it.
>
>
>
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
>https://www1.ietf.org/mailman/listinfo/ieprep

*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_223320317==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 01:00 PM 3/25/2002 -0500, Choi, Don wrote:</div>
<div>&gt;Fred and Cory,</div>
<div>&gt;</div>
<div>&gt;I agree with you that how a network supports preferential
treatments for</div>
<div>&gt;high priority calls should be up to the service provider and its
customers.</div>
<div>&gt;We should not specify or exclude any specific 
mechanisms.</div>
<br>
<div>then, wouldn't it be better to acknowledge preemption and state it
as being up to local policy and mechanisms than to ignore it and not have
reference(s) to preemption within the doc (as Cory seems to indicate)?
I'd strongly prefer to inform about it, taking the position of local
policy/mechanisms decide, stating where it shouldn't or can't be used
(the circumstances)...</div>
<br>
<br>
<div>&gt;</div>
<div>&gt;Don </div>
<div>&gt;</div>
<div>&gt; </div>
<div>&gt;</div>
<div>&gt;&gt;IMHO we just should take no position about preemption in the
framework </div>
<div>&gt;&gt;document.</div>
<div>&gt;</div>
<div>&gt;I personally agree, if you're talking about call pre-emption.
The framework </div>
<div>&gt;should indicate that there will be policies applied in various
networks, </div>
<div>&gt;and that the exact structure of the policy is a matter of
service offering </div>
<div>&gt;and contract between the network provider and its customers.
Pre-emption or </div>
<div>&gt;lack of it is a policy which would be applied in a specific
network, and on </div>
<div>&gt;this list there has been discussion both of a network that does
use it and </div>
<div>&gt;a network that doesn't use it.</div>
<div>&gt;</div>
<div>&gt;</div>
<div>&gt;</div>
<div>&gt;</div>
<div>&gt;_______________________________________________</div>
<div>&gt;Ieprep mailing list</div>
<div>&gt;Ieprep@ietf.org</div>
<div>&gt;<a href="https://www1.ietf.org/mailman/listinfo/ieprep" EUDORA=AUTOURL>https://www1.ietf.org/mailman/listinfo/ieprep</a></div>
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_223320317==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 14:47:11 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA29341
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 14:47:10 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA00845;
	Mon, 25 Mar 2002 14:44:39 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA00795
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 14:44:37 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA29152;
	Mon, 25 Mar 2002 14:44:35 -0500 (EST)
Received: from FRED-W2K6.cisco.com (sjc-vpn1-642.cisco.com [10.21.98.130])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2PJi6Y07577;
	Mon, 25 Mar 2002 11:44:06 -0800 (PST)
Message-Id: <4.3.2.7.2.20020325110547.01df7278@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Mon, 25 Mar 2002 11:10:57 -0800
To: senthil ayyasamy <mplsgeek@yahoo.com>
From: Fred Baker <fred@cisco.com>
Cc: ieprep@ietf.org, steve Coya <iesg-secretary@ietf.org>
In-Reply-To: <20020325165947.78824.qmail@web20803.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Ieprep] [iesg-secretary #1026] concluded WG
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 08:59 AM 3/25/2002, senthil ayyasamy wrote:
>The list of concluded working groups includes *emergency information 
>infrastructure* (http://ietf.org/html.charters/OLD/eii-charter.html). It 
>was chaired by Art Botterell and concluded in January 1997.  Did the group 
>produce any documents? I am curious to know about the technology employed 
>in 1994-1997 to find a solution for emergencies.

I suspect that this related to the development of Computer Emergency 
Response Teams, or CERTs. I have sent a note to Steve Coya asking for the 
history.

Unfortunately, the character string "Botterell" doesn't show up in any 
currently posted RFC. I did a search for "Emergency", and came up with 66 
RFCs, 19 of which comment on CERTs, and the rest of which use the term as a 
label in a general sense. I don't see anything directly related to the work 
of this working group.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 14:47:37 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA29398
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 14:47:37 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA01689;
	Mon, 25 Mar 2002 14:47:19 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA01644
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 14:47:18 -0500 (EST)
Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA29356;
	Mon, 25 Mar 2002 14:47:16 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.9.3/8.9.3) with ESMTP id OAA28125;
	Mon, 25 Mar 2002 14:47:06 -0500 (EST)
Received: from cs.columbia.edu (cta.cs.columbia.edu [128.59.19.46])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.1/8.12.1) with ESMTP id g2PJl6Pm012627
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Mon, 25 Mar 2002 14:47:06 -0500 (EST)
Message-ID: <3C9F7E9C.C2D14001@cs.columbia.edu>
Date: Mon, 25 Mar 2002 14:46:36 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
Organization: Columbia University
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Fred Baker <fred@cisco.com>
CC: senthil ayyasamy <mplsgeek@yahoo.com>, ieprep@ietf.org,
        steve Coya <iesg-secretary@ietf.org>
Subject: Re: [Ieprep] [iesg-secretary #1026] concluded WG
References: <4.3.2.7.2.20020325110547.01df7278@mira-sjcm-4.cisco.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

http://www.incident.com/about.html

Fred Baker wrote:
> 
> At 08:59 AM 3/25/2002, senthil ayyasamy wrote:
> >The list of concluded working groups includes *emergency information
> >infrastructure* (http://ietf.org/html.charters/OLD/eii-charter.html). It
> >was chaired by Art Botterell and concluded in January 1997.  Did the group
> >produce any documents? I am curious to know about the technology employed
> >in 1994-1997 to find a solution for emergencies.
> 
> I suspect that this related to the development of Computer Emergency
> Response Teams, or CERTs. I have sent a note to Steve Coya asking for the
> history.
> 
> Unfortunately, the character string "Botterell" doesn't show up in any
> currently posted RFC. I did a search for "Emergency", and came up with 66
> RFCs, 19 of which comment on CERTs, and the rest of which use the term as a
> label in a general sense. I don't see anything directly related to the work
> of this working group.
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 15:02:15 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00021
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 15:02:15 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA04462;
	Mon, 25 Mar 2002 14:56:53 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA04418
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 14:56:51 -0500 (EST)
Received: from web20805.mail.yahoo.com (web20805.mail.yahoo.com [216.136.226.194])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id OAA29816
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 14:56:49 -0500 (EST)
Message-ID: <20020325195651.68922.qmail@web20805.mail.yahoo.com>
Received: from [134.193.128.192] by web20805.mail.yahoo.com via HTTP; Mon, 25 Mar 2002 11:56:51 PST
Date: Mon, 25 Mar 2002 11:56:51 -0800 (PST)
From: senthil ayyasamy <mplsgeek@yahoo.com>
To: Fred Baker <fred@cisco.com>
Cc: ieprep@ietf.org, steve Coya <iesg-secretary@ietf.org>
In-Reply-To: <4.3.2.7.2.20020325110547.01df7278@mira-sjcm-4.cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: [Ieprep] Re: [iesg-secretary #1026] concluded WG
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org



> label in a general sense. I don't see anything
> directly related to the work 
> of this working group.
  Even the proceedings doesnt show any occurence.
-Senthil

=====
Ayyasamy Senthilkumar
M.S.(Computer Networking)
UMKC,MO.
-----------------------------
An expert is a man who has made all the mistakes that can be made in a very narrow field.
--Niels Bohr.

__________________________________________________
Do You Yahoo!?
Yahoo! Movies - coverage of the 74th Academy Awards®
http://movies.yahoo.com/

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 15:52:59 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA03047
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 15:52:59 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA17971;
	Mon, 25 Mar 2002 15:51:30 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA17930
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 15:51:28 -0500 (EST)
Received: from imo-d10.mx.aol.com (imo-d10.mx.aol.com [205.188.157.42])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA02975
	for <ieprep@ietf.org>; Mon, 25 Mar 2002 15:51:26 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d10.mx.aol.com (mail_out_v32.5.) id z.42.24690da8 (25512);
	Mon, 25 Mar 2002 15:50:18 -0500 (EST)
Message-ID: <42.24690da8.29d0e78a@aol.com>
Date: Mon, 25 Mar 2002 15:50:18 EST
Subject: Re: [Ieprep] Preemption
To: goldmans@agcs.com, K.Carlberg@cs.ucl.ac.uk
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_42.24690da8.29d0e78a_boundary"
X-Mailer: AOL 6.0 for Windows US sub 10524
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--part1_42.24690da8.29d0e78a_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 3/24/02 10:44:27 PM Eastern Standard Time, 
goldmans@agcs.com writes:


> Ken,
> 
> If I correctly understand what you have written, you are suggesting non
> pre-emption is the default, but are not preventing pre-emption (but not as 
> the
> default) in those networks where it might be desired.
> 
> This sounds like an approach that everybody should be able to live with.
> 

I don't understand what "default" means in this case. I think we agree that 
the "normal" or "most common" approach would be to not preempt, but I 
wouldn't describe that as a "default". It depends on the policy of the 
network.

I think it's confusing to talk of a "default" for such a thing. To me, a 
"default" operation only needs to be stated when one needs to say what to do 
if an optional parameter is missing from a received message. For example, 
default to the "Routine" precedence level if the "Resource-priority" header 
is missing.

There still is a confusion over whether you are referring to call preemption 
or packet "preemption" (dropping). Discussion of these two issues needs to be 
kept separate. Yes, call preemption in non-private networks would be subject 
to laws/regulations and, at least in the US, is unlikely to ever be allowed. 
The absence of preemption of calls is probably the presumed operation in any 
IP network. But I can't image a law addressing how we handle packets. It 
would only be a question of maintaining a required QoS level, if "regulated" 
by a tarrif, for example.

My Assured Service draft in SIPPING was explicitly presuming that packet 
level "preemption", through the use of multiple levels of selective dopping 
when congestion occurs, is required as the way to achieve the service 
requirements for the Defense Dept network. Thus, "packet preemption" would be 
the norm in that network. However, I might agree that such packet level 
"preemption" is not required to provide the authorized emergency requirements 
in the public network. But I wouldn't use the term "default".

However, I don't see why this packet level "preemption" by selective dropping 
is considered so bad. Isn't it already the described operation for the AF 
classes (with 3 drop precedences each) which might be used for the signaling 
messages? If some of these packets are dropped, might that actually prevent a 
call from being set-up, that is, look like a failure? In some networks, this 
would be worse than preemption (with its notification).

Mike Pierce



--part1_42.24690da8.29d0e78a_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 3/24/02 10:44:27 PM Eastern Standard Time, goldmans@agcs.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">Ken,
<BR>
<BR>If I correctly understand what you have written, you are suggesting non
<BR>pre-emption is the default, but are not preventing pre-emption (but not as the
<BR>default) in those networks where it might be desired.
<BR>
<BR>This sounds like an approach that everybody should be able to live with.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>I don't understand what "default" means in this case. I think we agree that the "normal" or "most common" approach would be to not preempt, but I wouldn't describe that as a "default". It depends on the policy of the network.
<BR>
<BR>I think it's confusing to talk of a "default" for such a thing. To me, a "default" operation only needs to be stated when one needs to say what to do if an optional parameter is missing from a received message. For example, default to the "Routine" precedence level if the "Resource-priority" header is missing.
<BR>
<BR>There still is a confusion over whether you are referring to call preemption or packet "preemption" (dropping). Discussion of these two issues needs to be kept separate. Yes, call preemption in non-private networks would be subject to laws/regulations and, at least in the US, is unlikely to ever be allowed. The absence of preemption of calls is probably the presumed operation in any IP network. But I can't image a law addressing how we handle packets. It would only be a question of maintaining a required QoS level, if "regulated" by a tarrif, for example.
<BR>
<BR>My Assured Service draft in SIPPING was explicitly presuming that packet level "preemption", through the use of multiple levels of selective dopping when congestion occurs, is required as the way to achieve the service requirements for the Defense Dept network. Thus, "packet preemption" would be the norm in that network. However, I might agree that such packet level "preemption" is not required to provide the authorized emergency requirements in the public network. But I wouldn't use the term "default".
<BR>
<BR>However, I don't see why this packet level "preemption" by selective dropping is considered so bad. Isn't it already the described operation for the AF classes (with 3 drop precedences each) which might be used for the signaling messages? If some of these packets are dropped, might that actually prevent a call from being set-up, that is, look like a failure? In some networks, this would be worse than preemption (with its notification).
<BR>
<BR>Mike Pierce
<BR>
<BR></FONT></HTML>

--part1_42.24690da8.29d0e78a_boundary--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Mon Mar 25 16:08:24 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA03550
	for <ieprep-archive@odin.ietf.org>; Mon, 25 Mar 2002 16:08:24 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA21845;
	Mon, 25 Mar 2002 16:07:56 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA21805
	for <ieprep@optimus.ietf.org>; Mon, 25 Mar 2002 16:07:54 -0500 (EST)
Received: from clifden.donelan.com (clifden.donelan.com [199.34.53.180])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA03521;
	Mon, 25 Mar 2002 16:07:51 -0500 (EST)
Received: from sean (helo=localhost)
	by clifden.donelan.com with local-esmtp (Exim 3.34 #2)
	id 16pbgz-0007HH-00; Mon, 25 Mar 2002 16:07:49 -0500
Date: Mon, 25 Mar 2002 16:07:49 -0500 (EST)
From: Sean Donelan <sean@donelan.com>
To: Fred Baker <fred@cisco.com>
cc: senthil ayyasamy <mplsgeek@yahoo.com>, <ieprep@ietf.org>,
        steve Coya <iesg-secretary@ietf.org>
Subject: Re: [Ieprep] [iesg-secretary #1026] concluded WG
In-Reply-To: <4.3.2.7.2.20020325110547.01df7278@mira-sjcm-4.cisco.com>
Message-ID: <Pine.GSO.4.40.0203251605350.27939-100000@clifden.donelan.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

On Mon, 25 Mar 2002, Fred Baker wrote:
> Unfortunately, the character string "Botterell" doesn't show up in any
> currently posted RFC. I did a search for "Emergency", and came up with 66
> RFCs, 19 of which comment on CERTs, and the rest of which use the term as a
> label in a general sense. I don't see anything directly related to the work
> of this working group.

Art is still active in other areas.  I believe he is currently trying
to define an XML scheme for exchanging messages between emergency
management agencies.

There are 50-60 different groups doing something containing the words
"emergency" and "internet" around the world.



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 01:04:52 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA17432
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 01:04:52 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id BAA29440;
	Tue, 26 Mar 2002 01:04:25 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id BAA29406
	for <ieprep@optimus.ietf.org>; Tue, 26 Mar 2002 01:04:23 -0500 (EST)
Received: from kc-msxproto4.kc.umkc.edu (kc-msxproto4.kc.umkc.edu [134.193.143.48])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA17416
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 01:04:22 -0500 (EST)
Received: from KC-MAIL1.kc.umkc.edu ([134.193.143.161] RDNS failed) by kc-msxproto4.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.4453);
	 Tue, 26 Mar 2002 00:04:23 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D48C.13AB688A"
Subject: RE: [Ieprep] Preemption
Date: Tue, 26 Mar 2002 00:04:23 -0600
Message-ID: <EC61AA563B48B6459EEE976942BEA11918E8E8@KC-MAIL1.kc.umkc.edu>
Thread-Topic: [Ieprep] Preemption
Thread-Index: AcHUK/nIRSBG9MYERkSTiRIf1Y/v9wAX1apg
From: "Beard, Cory" <BeardC@umkc.edu>
To: "James M. Polk" <jmpolk@cisco.com>, "Choi, Don" <choid@ncr.disa.mil>,
        <ieprep@ietf.org>
X-OriginalArrivalTime: 26 Mar 2002 06:04:23.0777 (UTC) FILETIME=[13EB1D10:01C1D48C]
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.

------_=_NextPart_001_01C1D48C.13AB688A
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

James,=20
=20
I certainly believe we should have a discussion of preemption as a =
policy option, just not state a preference.
=20
It seems there are many socio-political issues that influence setting =
policy here, ones that may likely change as people learn to operate over =
the Internet instead of the PSTN.
=20
Cory Beard
UMKC

-----Original Message-----
From: James M. Polk [mailto:jmpolk@cisco.com]
Sent: Monday, March 25, 2002 12:33 PM
To: Choi, Don; ieprep@ietf.org
Subject: RE: [Ieprep] Preemption


At 01:00 PM 3/25/2002 -0500, Choi, Don wrote:
>Fred and Cory,
>
>I agree with you that how a network supports preferential treatments =
for
>high priority calls should be up to the service provider and its =
customers.
>We should not specify or exclude any specific mechanisms.

then, wouldn't it be better to acknowledge preemption and state it as =
being up to local policy and mechanisms than to ignore it and not have =
reference(s) to preemption within the doc (as Cory seems to indicate)? =
I'd strongly prefer to inform about it, taking the position of local =
policy/mechanisms decide, stating where it shouldn't or can't be used =
(the circumstances)...


>
>Don=20
>
>=20
>
>>IMHO we just should take no position about preemption in the framework =

>>document.
>
>I personally agree, if you're talking about call pre-emption. The =
framework=20
>should indicate that there will be policies applied in various =
networks,=20
>and that the exact structure of the policy is a matter of service =
offering=20
>and contract between the network provider and its customers. =
Pre-emption or=20
>lack of it is a policy which would be applied in a specific network, =
and on=20
>this list there has been discussion both of a network that does use it =
and=20
>a network that doesn't use it.
>
>
>
>
>_______________________________________________
>Ieprep mailing list
>Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

*************************************
"People generally demand more respect for their own rights than they are =
willing to allow for others"


James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike=20
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com <http://www.cisco.com/> =20


------_=_NextPart_001_01C1D48C.13AB688A
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">


<META content=3D"MSHTML 5.50.4611.1300" name=3DGENERATOR></HEAD>
<BODY>
<DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D920372805-26032002>James, =

</SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D920372805-26032002></SPAN></FONT>&nbsp;</DIV>
<DIV><SPAN class=3D920372805-26032002><FONT face=3DArial size=3D2>I =
certainly believe=20
we&nbsp;should have&nbsp;a discussion of preemption as a =
policy&nbsp;<SPAN=20
class=3D170565805-26032002>option</SPAN>, just not state a=20
preference.</FONT></SPAN></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN=20
class=3D920372805-26032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D920372805-26032002><SPAN=20
class=3D170565805-26032002>It seems there are many socio-political =
issues that=20
influence setting policy here, ones that may likely change as people =
learn to=20
operate over the Internet instead of the =
PSTN.</SPAN></SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D920372805-26032002><SPAN=20
class=3D170565805-26032002></SPAN></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D920372805-26032002><SPAN=20
class=3D170565805-26032002>Cory Beard</SPAN></SPAN></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><SPAN class=3D920372805-26032002><SPAN=20
class=3D170565805-26032002>UMKC</SPAN></SPAN></FONT></DIV></DIV>
<BLOCKQUOTE=20
style=3D"PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px =
solid">
  <DIV class=3DOutlookMessageHeader dir=3Dltr align=3Dleft><FONT =
face=3DTahoma=20
  size=3D2>-----Original Message-----<BR><B>From:</B> James M. Polk=20
  [mailto:jmpolk@cisco.com]<BR><B>Sent:</B> Monday, March 25, 2002 12:33 =

  PM<BR><B>To:</B> Choi, Don; ieprep@ietf.org<BR><B>Subject:</B> RE: =
[Ieprep]=20
  Preemption<BR><BR></FONT></DIV>
  <DIV>At 01:00 PM 3/25/2002 -0500, Choi, Don wrote:</DIV>
  <DIV>&gt;Fred and Cory,</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;I agree with you that how a network supports preferential =
treatments=20
  for</DIV>
  <DIV>&gt;high priority calls should be up to the service provider and =
its=20
  customers.</DIV>
  <DIV>&gt;We should not specify or exclude any specific =
mechanisms.</DIV><BR>
  <DIV>then, wouldn't it be better to acknowledge preemption and state =
it as=20
  being up to local policy and mechanisms than to ignore it and not have =

  reference(s) to preemption within the doc (as Cory seems to indicate)? =
I'd=20
  strongly prefer to inform about it, taking the position of local=20
  policy/mechanisms decide, stating where it shouldn't or can't be used =
(the=20
  circumstances)...</DIV><BR><BR>
  <DIV>&gt;</DIV>
  <DIV>&gt;Don </DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt; </DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;&gt;IMHO we just should take no position about preemption in =
the=20
  framework </DIV>
  <DIV>&gt;&gt;document.</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;I personally agree, if you're talking about call pre-emption. =
The=20
  framework </DIV>
  <DIV>&gt;should indicate that there will be policies applied in =
various=20
  networks, </DIV>
  <DIV>&gt;and that the exact structure of the policy is a matter of =
service=20
  offering </DIV>
  <DIV>&gt;and contract between the network provider and its customers.=20
  Pre-emption or </DIV>
  <DIV>&gt;lack of it is a policy which would be applied in a specific =
network,=20
  and on </DIV>
  <DIV>&gt;this list there has been discussion both of a network that =
does use=20
  it and </DIV>
  <DIV>&gt;a network that doesn't use it.</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;</DIV>
  <DIV>&gt;_______________________________________________</DIV>
  <DIV>&gt;Ieprep mailing list</DIV>
  <DIV>&gt;Ieprep@ietf.org</DIV>
  <DIV>&gt;<A href=3D"https://www1.ietf.org/mailman/listinfo/ieprep"=20
  =
EUDORA=3D"AUTOURL">https://www1.ietf.org/mailman/listinfo/ieprep</A></DIV=
><BR>
  <DIV align=3Dcenter>*************************************<BR>"People =
generally=20
  demand more respect for their own rights than they are willing to =
allow for=20
  others"<BR><BR></DIV>James M. Polk<BR>Consulting Engineer<BR>Office of =
the=20
  CTO<BR><BR>Cisco Systems<BR>2200 East President George Bush Turnpike=20
  <BR>Richardson, TX&nbsp; 75082 USA<BR>w) 972.813.5208<BR>f)&nbsp;=20
  972.813.5280<BR><A href=3D"http://www.cisco.com/"=20
  eudora=3D"autourl">www.cisco.com</A> </BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C1D48C.13AB688A--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 01:09:32 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA17527
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 01:09:32 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id BAA00399;
	Tue, 26 Mar 2002 01:08:09 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id BAA00365
	for <ieprep@optimus.ietf.org>; Tue, 26 Mar 2002 01:08:07 -0500 (EST)
Received: from kc-msxproto4.kc.umkc.edu (kc-msxproto4.kc.umkc.edu [134.193.143.48])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA17479
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 01:08:06 -0500 (EST)
Received: from KC-MAIL1.kc.umkc.edu ([134.193.143.161] RDNS failed) by kc-msxproto4.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.4453);
	 Tue, 26 Mar 2002 00:07:55 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: RE: [Ieprep] Preemption
Date: Tue, 26 Mar 2002 00:07:55 -0600
Message-ID: <EC61AA563B48B6459EEE976942BEA11979807C@KC-MAIL1.kc.umkc.edu>
Thread-Topic: [Ieprep] Preemption
Thread-Index: AcHSVvcnVYGMFnkPTbSfbYxAIzj0xQCNTqJw
From: "Beard, Cory" <BeardC@umkc.edu>
To: "Fred Baker" <fred@cisco.com>, "Henning Schulzrinne" <hgs@cs.columbia.edu>
Cc: <ieprep@ietf.org>
X-OriginalArrivalTime: 26 Mar 2002 06:07:55.0852 (UTC) FILETIME=[925330C0:01C1D48C]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id BAA00366
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

> -----Original Message-----
> From: Fred Baker [mailto:fred@cisco.com]
>
> you could always signal your bandwidth, using RSVP with the 
> EF tag on it. 
> We've done a fair bit of work to make that possible...

Can we assume providers will be using RSVP and EF?  (At least in the short term)

Cory Beard

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 02:30:18 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA26589
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 02:30:18 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id CAA20414;
	Tue, 26 Mar 2002 02:29:57 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id CAA20375
	for <ieprep@optimus.ietf.org>; Tue, 26 Mar 2002 02:29:55 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA26558
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 02:29:53 -0500 (EST)
Received: from FRED-W2K6.cisco.com (sjc-vpn1-881.cisco.com [10.21.99.113])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2Q7SnY14724;
	Mon, 25 Mar 2002 23:28:49 -0800 (PST)
Message-Id: <4.3.2.7.2.20020325223111.01b47428@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Mon, 25 Mar 2002 22:33:15 -0800
To: "Beard, Cory" <BeardC@umkc.edu>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Preemption
Cc: "Henning Schulzrinne" <hgs@cs.columbia.edu>, <ieprep@ietf.org>
In-Reply-To: <EC61AA563B48B6459EEE976942BEA11979807C@KC-MAIL1.kc.umkc.ed
 u>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 10:07 PM 3/25/2002, Beard, Cory wrote:
>Can we assume providers will be using RSVP and EF?  (At least in the short 
>term)

we can mostly assume that providers will swamp the situation with bandwidth 
so that the issue is moot. You may contract the use of same on your own 
access links, though, which tends to be among the congestion points of 
concern. And frankly, what seems to be under discussion with MLPP isn't 
service provider networks, it is private networks like DISA's.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 02:42:32 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA26701
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 02:42:32 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id CAA23127;
	Tue, 26 Mar 2002 02:41:54 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id CAA23091
	for <ieprep@optimus.ietf.org>; Tue, 26 Mar 2002 02:41:52 -0500 (EST)
Received: from web20803.mail.yahoo.com (web20803.mail.yahoo.com [216.136.226.192])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id CAA26695
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 02:41:50 -0500 (EST)
Message-ID: <20020326074152.7478.qmail@web20803.mail.yahoo.com>
Received: from [134.193.6.41] by web20803.mail.yahoo.com via HTTP; Mon, 25 Mar 2002 23:41:51 PST
Date: Mon, 25 Mar 2002 23:41:51 -0800 (PST)
From: senthil ayyasamy <mplsgeek@yahoo.com>
Subject: RE: [Ieprep] Preemption
To: Fred Baker <fred@cisco.com>, "Beard, Cory" <BeardC@umkc.edu>
Cc: Henning Schulzrinne <hgs@cs.columbia.edu>, ieprep@ietf.org
In-Reply-To: <4.3.2.7.2.20020325223111.01b47428@mira-sjcm-4.cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org



> concern. And frankly, what seems to be under
> discussion with MLPP isn't 
> service provider networks, it is private networks
> like DISA's.
I strongly suggest to include mechanisms for
pre-emption into the existing framework document as an
option and leave the implementation to the discretion
of the local policy.  The James Polk draft on MLPP can
serve as a baseline.
	MLPP has Non-Preemptable Resources options which can
be explored for providing pre-emption as an
option.  This certainly prevents access pre-emption as
per the Polk draft.  But, pre-emption at the time of
congestion (network pre-emption) is as important as
access pre-emption.  So, the question is how to make
network pre-emption as an option?  This will provide
avenues for making pre-emption generic .

But still, MLPP is mandated as having influence with a
single domain based implementation only and there is
no recognized/standardized method or mechanism in the
case of which one of several lower precedence calls
gets disconnected. The working group has to focus on
these standardization issues too.Are there
possibilities for extensions to MLPP to cover more
than one domain?

- Senthil


=====
Ayyasamy Senthilkumar
M.S.(Computer Networking)
UMKC,MO.
-----------------------------
An expert is a man who has made all the mistakes that can be made in a very narrow field.
--Niels Bohr.

__________________________________________________
Do You Yahoo!?
Yahoo! Movies - coverage of the 74th Academy Awards®
http://movies.yahoo.com/

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 08:04:55 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01099
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 08:04:55 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA08076;
	Tue, 26 Mar 2002 08:04:21 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA08042
	for <ieprep@optimus.ietf.org>; Tue, 26 Mar 2002 08:04:19 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA01079
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 08:04:15 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.22349-0@bells.cs.ucl.ac.uk>; Tue, 26 Mar 2002 13:03:54 +0000
To: senthil ayyasamy <mplsgeek@yahoo.com>
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
In-reply-to: Your message of "Mon, 25 Mar 2002 23:41:51 PST." <20020326074152.7478.qmail@web20803.mail.yahoo.com>
Date: Tue, 26 Mar 2002 13:03:53 +0000
Message-ID: <19997.1017147833@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


> > concern. And frankly, what seems to be under
> > discussion with MLPP isn't=20
> > service provider networks, it is private networks
> > like DISA's.
> I strongly suggest to include mechanisms for
> pre-emption into the existing framework document as an
> option and leave the implementation to the discretion
> of the local policy.  The James Polk draft on MLPP can
> serve as a baseline.


I think this thread on viewing non-preemptive action as a default
approach is in danger of mixing apples and oranges.  lets take a step
back and review a couple of things.

in the (IEPS) Framework document, the notion of non-preemptive action 
as a default position is presented early in the draft as simply an
objective -- keeping in mind that an objective is not a solution or
mechanisim.  further, this objective doesn't discuss the mechanisms,
which is what the thread has started to dive into.  later in the draft,
when it does discuss mechanisms, the draft makes clear the use of Policy 
as the means of dictating use and applicability of actions such as 
pre-emption.  i'll refer to my earlier post on this topic as to why
i think its helpful to view non-preemptive action as a deafult position
(keeping in mind that a policy may not exist in the presence of 
signaled information)

having said that, we also shouldn't confuse the draft as one that acts
as a framework for both IEPS (GETS-like service) and MLPP.  nor, in my
opinion, should we try to evolve the draft so that it specifically
address MLPP.  if this were the case, then surely we would need to
remove the objective of being able to function in the presence of
non-ubiquitous support.  in addition, the objective of high
*probability* of call completion would not apply to MLPP.

my feeling is that a framework for MLPP should be a separate document.
one could borrow heavily from the IEPS framework draft, but it would 
seem that a framework for MLPP would have a core of different 
objectives.  a question that would remain is whether a framework for
MLPP is within the charter of IEPREP.  that decision is not my call :-)

-ken


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 08:20:26 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01457
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 08:20:26 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA11510;
	Tue, 26 Mar 2002 08:18:25 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA11445
	for <ieprep@optimus.ietf.org>; Tue, 26 Mar 2002 08:18:23 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01392
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 08:18:21 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <F5YT0HZ8>; Tue, 26 Mar 2002 08:16:44 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C52@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'senthil ayyasamy'" <mplsgeek@yahoo.com>, Fred Baker <fred@cisco.com>,
        "Beard, Cory" <BeardC@umkc.edu>
Cc: Henning Schulzrinne <hgs@cs.columbia.edu>, ieprep@ietf.org
Subject: RE: [Ieprep] Preemption
Date: Tue, 26 Mar 2002 08:17:40 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D4C8.9B1DBE50"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D4C8.9B1DBE50
Content-Type: text/plain;
	charset="iso-8859-1"

Replying to just this one point.

There is a mechanism called eMLPP (extended MLPP) which is part of the GSM
(mobile telephony) standard.  Like MLPP it includes the options for both
pre-emption and priority.  It is my understanding that it is currently in
use in some private (railway company) mobile networks in Europe.  And it is
quite possible that a service provider could/would offer it as a commercial
feature on the (non US) public mobile network.  If it is implemented in the
US (in the public network), it will almost certainly be without the
pre-emption feature.

-----Original Message-----
From: senthil ayyasamy [mailto:mplsgeek@yahoo.com]
Sent: Tuesday, March 26, 2002 2:42 AM
To: Fred Baker; Beard, Cory
Cc: Henning Schulzrinne; ieprep@ietf.org
Subject: RE: [Ieprep] Preemption

Are there
possibilities for extensions to MLPP to cover more
than one domain?

- Senthil


------_=_NextPart_001_01C1D4C8.9B1DBE50
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2653.12">
<TITLE>RE: [Ieprep] Preemption</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>Replying to just this one point.</FONT>
</P>

<P><FONT SIZE=3D2>There is a mechanism called eMLPP (extended MLPP) =
which is part of the GSM (mobile telephony) standard.&nbsp; Like MLPP =
it includes the options for both pre-emption and priority.&nbsp; It is =
my understanding that it is currently in use in some private (railway =
company) mobile networks in Europe.&nbsp; And it is quite possible that =
a service provider could/would offer it as a commercial feature on the =
(non US) public mobile network.&nbsp; If it is implemented in the US =
(in the public network), it will almost certainly be without the =
pre-emption feature.</FONT></P>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>
<BR><FONT SIZE=3D2>From: senthil ayyasamy [<A =
HREF=3D"mailto:mplsgeek@yahoo.com">mailto:mplsgeek@yahoo.com</A>]</FONT>=

<BR><FONT SIZE=3D2>Sent: Tuesday, March 26, 2002 2:42 AM</FONT>
<BR><FONT SIZE=3D2>To: Fred Baker; Beard, Cory</FONT>
<BR><FONT SIZE=3D2>Cc: Henning Schulzrinne; ieprep@ietf.org</FONT>
<BR><FONT SIZE=3D2>Subject: RE: [Ieprep] Preemption</FONT>
</P>

<P><FONT SIZE=3D2>Are there</FONT>
<BR><FONT SIZE=3D2>possibilities for extensions to MLPP to cover =
more</FONT>
<BR><FONT SIZE=3D2>than one domain?</FONT>
</P>

<P><FONT SIZE=3D2>- Senthil</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C1D4C8.9B1DBE50--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 09:03:47 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03292
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 09:03:47 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA23844;
	Tue, 26 Mar 2002 09:02:46 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA23805
	for <ieprep@optimus.ietf.org>; Tue, 26 Mar 2002 09:02:43 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA03241
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 09:02:41 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id HAA09870
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 07:00:49 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA38EC
          for <ieprep@ietf.org>; Tue, 26 Mar 2002 07:02:42 -0700
Received: FROM agcs.com BY mknaomi ; Tue Mar 26 07:02:40 2002 -0700
Message-ID: <3CA07F81.B74F6A76@agcs.com>
Date: Tue, 26 Mar 2002 07:02:41 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
References: <19997.1017147833@cs.ucl.ac.uk>
Content-Type: multipart/mixed;
 boundary="------------AFED3D6BD533C3F5077FD7B3"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------AFED3D6BD533C3F5077FD7B3
Content-Type: multipart/alternative;
 boundary="------------A4ED1752B944A2D559AAFECA"


--------------A4ED1752B944A2D559AAFECA
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Ken,

Thanks for reminding us of the difference between policy and protocol.

It seems to me that there are four general cases:

   * If a given network never preempts, then preemption does not occur there
   * If a given network always preempts when needed, then preemption may
     occur there
   * If a given network can do either based on an mandatory indicator, then
     it does whatever the indicator directs
   * If a given network can do either based on an optional indicator, then
     the action taken in the absence of the indicator is the DEFAULT action.
     To be consistent with the policy statement in this case the default
     would seem to be to not preempt.

Is my understanding correct, or did I miss something? I  used the term
indicator since the information may be part of the protocol or could be
delivered to the network by other means. I didn't want to restrict it to
just being in the protocol.



Ken Carlberg wrote:

I think this thread on viewing non-preemptive action as a default

> approach is in danger of mixing apples and oranges.  lets take a step
> back and review a couple of things.
>
> in the (IEPS) Framework document, the notion of non-preemptive action
> as a default position is presented early in the draft as simply an
> objective -- keeping in mind that an objective is not a solution or
> mechanisim.  further, this objective doesn't discuss the mechanisms,
> which is what the thread has started to dive into.  later in the draft,
> when it does discuss mechanisms, the draft makes clear the use of Policy
> as the means of dictating use and applicability of actions such as
> pre-emption.  i'll refer to my earlier post on this topic as to why
> i think its helpful to view non-preemptive action as a deafult position
> (keeping in mind that a policy may not exist in the presence of
> signaled information)
>
> having said that, we also shouldn't confuse the draft as one that acts
> as a framework for both IEPS (GETS-like service) and MLPP.  nor, in my
> opinion, should we try to evolve the draft so that it specifically
> address MLPP.  if this were the case, then surely we would need to
> remove the objective of being able to function in the presence of
> non-ubiquitous support.  in addition, the objective of high
> *probability* of call completion would not apply to MLPP.
>
> my feeling is that a framework for MLPP should be a separate document.
> one could borrow heavily from the IEPS framework draft, but it would
> seem that a framework for MLPP would have a core of different
> objectives.  a question that would remain is whether a framework for
> MLPP is within the charter of IEPREP.  that decision is not my call :-)
>
> -ken
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

--------------A4ED1752B944A2D559AAFECA
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Ken,
<p>Thanks for reminding us of the difference between policy and protocol.
<p>It seems to me that there are four general cases:
<ul>
<li>
If a given network never preempts, then preemption does not occur there</li>

<li>
If a given network always preempts when needed, then preemption may occur
there</li>

<li>
If a given network can do either based on an mandatory indicator, then
it does whatever the indicator directs</li>

<li>
If a given network can do either based on an optional indicator, then the
action taken in the absence of the indicator is the <b>DEFAULT</b> action.
To be consistent with the policy statement in this case the default would
seem to be to not preempt.</li>
</ul>
Is my understanding correct, or did I miss something? I&nbsp; used the
term indicator since the information may be part of the protocol or could
be delivered to the network by other means. I didn't want to restrict it
to just being in the protocol.
<br>&nbsp;
<br>&nbsp;
<p>Ken Carlberg wrote:
<p>I think this thread on viewing non-preemptive action as a default
<blockquote TYPE=CITE>approach is in danger of mixing apples and oranges.&nbsp;
lets take a step
<br>back and review a couple of things.
<p>in the (IEPS) Framework document, the notion of non-preemptive action
<br>as a default position is presented early in the draft as simply an
<br>objective -- keeping in mind that an objective is not a solution or
<br>mechanisim.&nbsp; further, this objective doesn't discuss the mechanisms,
<br>which is what the thread has started to dive into.&nbsp; later in the
draft,
<br>when it does discuss mechanisms, the draft makes clear the use of Policy
<br>as the means of dictating use and applicability of actions such as
<br>pre-emption.&nbsp; i'll refer to my earlier post on this topic as to
why
<br>i think its helpful to view non-preemptive action as a deafult position
<br>(keeping in mind that a policy may not exist in the presence of
<br>signaled information)
<p>having said that, we also shouldn't confuse the draft as one that acts
<br>as a framework for both IEPS (GETS-like service) and MLPP.&nbsp; nor,
in my
<br>opinion, should we try to evolve the draft so that it specifically
<br>address MLPP.&nbsp; if this were the case, then surely we would need
to
<br>remove the objective of being able to function in the presence of
<br>non-ubiquitous support.&nbsp; in addition, the objective of high
<br>*probability* of call completion would not apply to MLPP.
<p>my feeling is that a framework for MLPP should be a separate document.
<br>one could borrow heavily from the IEPS framework draft, but it would
<br>seem that a framework for MLPP would have a core of different
<br>objectives.&nbsp; a question that would remain is whether a framework
for
<br>MLPP is within the charter of IEPREP.&nbsp; that decision is not my
call :-)
<p>-ken
<p>_______________________________________________
<br>Ieprep mailing list
<br>Ieprep@ietf.org
<br><a href="https://www1.ietf.org/mailman/listinfo/ieprep">https://www1.ietf.org/mailman/listinfo/ieprep</a></blockquote>
</html>

--------------A4ED1752B944A2D559AAFECA--

--------------AFED3D6BD533C3F5077FD7B3
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------AFED3D6BD533C3F5077FD7B3--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 12:58:36 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA16362
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 12:58:35 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA02739;
	Tue, 26 Mar 2002 12:57:51 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA02699
	for <ieprep@optimus.ietf.org>; Tue, 26 Mar 2002 12:57:50 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA16270
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 12:57:47 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id JAA22198; Tue, 26 Mar 2002 09:57:18 -0800 (PST)
Message-Id: <4.1.20020326113537.0146b630@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Tue, 26 Mar 2002 11:56:49 -0600
To: ieprep@ietf.org
From: "James M. Polk" <jmpolk@cisco.com>
Subject: RE: [Ieprep] Preemption
Cc: Henning Schulzrinne <hgs@cs.columbia.edu>
In-Reply-To: <20020326074152.7478.qmail@web20803.mail.yahoo.com>
References: <4.3.2.7.2.20020325223111.01b47428@mira-sjcm-4.cisco.com>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_307534711==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_307534711==_.ALT
Content-Type: text/plain; charset="us-ascii"


At 11:41 PM 3/25/2002 -0800, senthil ayyasamy wrote:
>
>.Are there
>possibilities for extensions to MLPP to cover more
>than one domain?

The SIP Resource Priority ID provides this domain extension to be addressed
directly through the use of a "namespace" as a domain identifier. The R-P ID
specifies a label that can/should be implemented within a single domain, or any
other domain via an SLA with other domain(s).

The case where domain (namespace) X has several priority values (immediate is
greater than routine for example). If the policy within domain X is to, during
times of congestion at any interface within that domain, determine if there are
lower relative priority existing calls in relation to a new higher priority
call, to preempt a single or more lower priority call in order to have that
higher priority call complete. 

If domain with a namespace of Y also wants to grant namespace X originated
calls the same treatment (either because they're nice folks or a formal SLA),
then all UA's, GW's and Proxies within namespace Y will also treat R-P headers
with namespace X as valid and act accordingly (similarly).

Having multiple namespaces as valid within multiple domains (UA's, GW's and
Proxies) is this extension I believe you are asking about. If not, please let
me know...

>
>- Senthil 
*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"

James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com
--=====================_307534711==_.ALT
Content-Type: text/html; charset="us-ascii"

<html><div>At 11:41 PM 3/25/2002 -0800, senthil ayyasamy wrote:</div>
<div>&gt;</div>
<div>&gt;.Are there</div>
<div>&gt;possibilities for extensions to MLPP to cover more</div>
<div>&gt;than one domain?</div>
<br>
<div>The SIP Resource Priority ID provides this domain extension to be
addressed directly through the use of a &quot;namespace&quot; as a domain
identifier. The R-P ID specifies a label that can/should be implemented
within a single domain, or any other domain via an SLA with other
domain(s).</div>
<br>
<div>The case where domain (namespace) X has several priority values
(immediate is greater than routine for example). If the policy within
domain X is to, during times of congestion at any interface within that
domain, determine if there are lower relative priority existing calls in
relation to a new higher priority call, to preempt a single or more lower
priority call in order to have that higher priority call complete.
</div>
<br>
<div>If domain with a namespace of Y also wants to grant namespace X
originated calls the same treatment (either because they're nice folks or
a formal SLA), then all UA's, GW's and Proxies within namespace Y will
also treat R-P headers with namespace X as valid and act accordingly
(similarly).</div>
<br>
<div>Having multiple namespaces as valid within multiple domains (UA's,
GW's and Proxies) is this extension I believe you are asking about. If
not, please let me know...</div>
<br>
<div>&gt;</div>
&gt;- Senthil
<br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_307534711==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 16:06:16 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA23795
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 16:06:16 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA22690;
	Tue, 26 Mar 2002 16:05:56 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA22654
	for <ieprep@ns.ietf.org>; Tue, 26 Mar 2002 16:05:55 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA23726
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 16:05:50 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVBNW>; Tue, 26 Mar 2002 16:04:23 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C55@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'goldmans@agcs.com'" <goldmans@agcs.com>,
        Ken Carlberg
	 <K.Carlberg@cs.ucl.ac.uk>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Preemption
Date: Tue, 26 Mar 2002 16:05:18 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D509.EF574360"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D509.EF574360
Content-Type: text/plain;
	charset="iso-8859-1"

It may be a bit more complicated than that.  
 
I am not sure about MLPP, but in eMLPP there are THREE indicators (in
addition to level):
Queueing allowed/Queueing NOT allowed
CAN pre-empt a lower level call/CANNOT pre-empt a call
CAN be pre-empted by a higher level call/CANNOT be preempted by another
call, no matter what the priority.
 
Typically the "Ordinary" call has the lowest priority, but may neither
pre-empt, nor be pre-empted.
 
It seems to me that there MAY be a need for all three indicators in the
context of an IP network also.
 
The indicators MIGHT be tied directly to the level, but (especially if a
single IP network is transporting calls from more than one carrier) might
also be set on a call by call basis. So, it would be desirable for the
protocol to have a way of capturing all 3 indicators.
 
Janet

-----Original Message-----
From: Stuart (Stu) Goldman [mailto:goldmans@agcs.com]
Sent: Tuesday, March 26, 2002 9:03 AM
To: Ken Carlberg
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption


Ken, 

Thanks for reminding us of the difference between policy and protocol. 


It seems to me that there are four general cases: 


*	If a given network never preempts, then preemption does not occur
there 

*	If a given network always preempts when needed, then preemption may
occur there 

*	If a given network can do either based on an mandatory indicator,
then it does whatever the indicator directs 

*	If a given network can do either based on an optional indicator,
then the action taken in the absence of the indicator is the DEFAULT action.
To be consistent with the policy statement in this case the default would
seem to be to not preempt. 

Is my understanding correct, or did I miss something? I  used the term
indicator since the information may be part of the protocol or could be
delivered to the network by other means. I didn't want to restrict it to
just being in the protocol. 
  
  


------_=_NextPart_001_01C1D509.EF574360
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 5.00.3103.1000" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=587235320-26032002>It may 
be a bit more complicated than that.&nbsp; </SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=587235320-26032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=587235320-26032002>I am 
not sure about MLPP, but in eMLPP there are THREE indicators (in addition to 
level):</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=587235320-26032002>Queueing allowed/Queueing NOT 
allowed</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=587235320-26032002>CAN 
pre-empt a lower level call/CANNOT pre-empt a call</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=587235320-26032002>CAN be 
pre-empted by a higher level call/CANNOT be preempted by another call, no matter 
what the priority.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=587235320-26032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=587235320-26032002>Typically the "Ordinary" call has the lowest priority, 
but may neither pre-empt, nor be pre-empted.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=587235320-26032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=587235320-26032002>It 
seems to me that there MAY be a need for all three indicators in the context of 
an IP network also.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=587235320-26032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=587235320-26032002>The 
indicators MIGHT be tied directly to the level, but (especially if a single IP 
network is transporting calls from more than one carrier) might also be set on a 
call by call basis. So, it would be desirable for the protocol to have a way of 
capturing all 3 indicators.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=587235320-26032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=587235320-26032002>Janet</SPAN></FONT></DIV>
<BLOCKQUOTE style="MARGIN-RIGHT: 0px">
  <DIV align=left class=OutlookMessageHeader dir=ltr><FONT face=Tahoma 
  size=2>-----Original Message-----<BR><B>From:</B> Stuart (Stu) Goldman 
  [mailto:goldmans@agcs.com]<BR><B>Sent:</B> Tuesday, March 26, 2002 9:03 
  AM<BR><B>To:</B> Ken Carlberg<BR><B>Cc:</B> ieprep@ietf.org<BR><B>Subject:</B> 
  Re: [Ieprep] Preemption<BR><BR></DIV></FONT>Ken, 
  <P>Thanks for reminding us of the difference between policy and protocol. 
  <P>It seems to me that there are four general cases: 
  <UL>
    <LI>If a given network never preempts, then preemption does not occur there 
    <LI>If a given network always preempts when needed, then preemption may 
    occur there 
    <LI>If a given network can do either based on an mandatory indicator, then 
    it does whatever the indicator directs 
    <LI>If a given network can do either based on an optional indicator, then 
    the action taken in the absence of the indicator is the <B>DEFAULT</B> 
    action. To be consistent with the policy statement in this case the default 
    would seem to be to not preempt. </LI></UL>Is my understanding correct, or did 
  I miss something? I&nbsp; used the term indicator since the information may be 
  part of the protocol or could be delivered to the network by other means. I 
  didn't want to restrict it to just being in the 
  protocol.&nbsp;<BR>&nbsp;&nbsp;<BR>&nbsp; </BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C1D509.EF574360--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 18:03:04 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA29150
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 18:03:04 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA26142;
	Tue, 26 Mar 2002 18:02:36 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA26102
	for <ieprep@ns.ietf.org>; Tue, 26 Mar 2002 18:02:34 -0500 (EST)
Received: from wells.cisco.com (wells.cisco.com [171.71.177.223])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA29130
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 18:02:31 -0500 (EST)
Received: from JMPOLK-W2K (ssh-sjc-1.cisco.com [171.68.225.134]) by wells.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with SMTP id PAA08369; Tue, 26 Mar 2002 15:01:24 -0800 (PST)
Message-Id: <4.1.20020326165702.02f61390@localhost>
X-Sender: jmpolk@localhost
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Tue, 26 Mar 2002 17:00:56 -0600
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>,
        "'goldmans@agcs.com'" <goldmans@agcs.com>,
        Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
From: "James M. Polk" <jmpolk@cisco.com>
Subject: RE: [Ieprep] Preemption
Cc: ieprep@ietf.org
In-Reply-To: <CBED705A7FD2D311865500508B10894105D48C55@chntex02.is.dynco
 rp.com>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_325781418==_.ALT"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

--=====================_325781418==_.ALT
Content-Type: text/plain; charset="us-ascii"

At 04:05 PM 3/26/2002 -0500, Gunn, Janet NE wrote: 
>
> It may be a bit more complicated than that.  
>  
> I am not sure about MLPP, but in eMLPP there are THREE indicators (in
> addition to level):
> Queueing allowed/Queueing NOT allowed
> CAN pre-empt a lower level call/CANNOT pre-empt a call
> CAN be pre-empted by a higher level call/CANNOT be preempted by another call,
> no matter what the priority.
>  
> Typically the "Ordinary" call has the lowest priority, but may neither
> pre-empt, nor be pre-empted.


I believe *all* you listed above are policy decisions that are local in nature.
If left to the end-users, they would choose "cannot preempt" every time.
Perhaps this could be dealt with via user authentication as to whether a call
from that user is or is not allowed to be preempted....

>
>  
> It seems to me that there MAY be a need for all three indicators in the
> context of an IP network also.


starting to get complicated with 3 indicators

>
>  
> The indicators MIGHT be tied directly to the level, but (especially if a
> single IP network is transporting calls from more than one carrier) might
> also be set on a call by call basis. So, it would be desirable for the
> protocol to have a way of capturing all 3 indicators.
>  
> Janet
>>
>> -----Original Message----- 
>
>
> *************************************
> "People generally demand more respect for their own rights than they are
> willing to allow for others"
>
> James M. Polk
> Consulting Engineer
> Office of the CTO
>
> Cisco Systems
> 2200 East President George Bush Turnpike 
> Richardson, TX  75082 USA
> w) 972.813.5208
> f)  972.813.5280
> www.cisco.com
--=====================_325781418==_.ALT
Content-Type: text/html; charset="us-ascii"

<html>
At 04:05 PM 3/26/2002 -0500, Gunn, Janet NE wrote: <br>
<font face="arial" size=2 color="#0000FF"><blockquote type=cite cite>It
may be a bit more complicated than that.&nbsp; </font><br>
&nbsp;<br>
<font face="arial" size=2 color="#0000FF">I am not sure about MLPP, but
in eMLPP there are THREE indicators (in addition to level):</font><br>
Queueing allowed/Queueing NOT allowed<br>
CAN pre-empt a lower level call/CANNOT pre-empt a call<br>
CAN be pre-empted by a higher level call/CANNOT be preempted by another
call, no matter what the priority.<br>
&nbsp;<br>
<font face="arial" size=2 color="#0000FF">Typically the
&quot;Ordinary&quot; call has the lowest priority, but may neither
pre-empt, nor be pre-empted.</font></blockquote><br>
I believe *all* you listed above are policy decisions that are local in
nature. If left to the end-users, they would choose &quot;cannot
preempt&quot; every time. Perhaps this could be dealt with via user
authentication as to whether a call from that user is or is not allowed
to be preempted....<br>
<br>
<blockquote type=cite cite>&nbsp;<br>
<font face="arial" size=2 color="#0000FF">It seems to me that there MAY
be a need for all three indicators in the context of an IP network
also.</font></blockquote><br>
starting to get complicated with 3 indicators<br>
<br>
<blockquote type=cite cite>&nbsp;<br>
<font face="arial" size=2 color="#0000FF">The indicators MIGHT be tied
directly to the level, but (especially if a single IP network is
transporting calls from more than one carrier) might also be set on a
call by call basis. So, it would be desirable for the protocol to have a
way of capturing all 3 indicators.</font><br>
&nbsp;<br>
<font face="arial" size=2 color="#0000FF">Janet</font><blockquote><font face="tahoma" size=2>
<dl>
<dd>-----Original Message-----</font>
</dl></blockquote><br>

<div align="center">
*************************************<br>
&quot;People generally demand more respect for their own rights than they
are willing to allow for others&quot;<br>
<br>
</div>
James M. Polk<br>
Consulting Engineer<br>
Office of the CTO<br>
<br>
Cisco Systems<br>
2200 East President George Bush Turnpike <br>
Richardson, TX&nbsp; 75082 USA<br>
w) 972.813.5208<br>
f)&nbsp; 972.813.5280<br>
<a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></html>

--=====================_325781418==_.ALT--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Tue Mar 26 18:33:31 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA29913
	for <ieprep-archive@odin.ietf.org>; Tue, 26 Mar 2002 18:33:31 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA02972;
	Tue, 26 Mar 2002 18:28:50 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA02930
	for <ieprep@ns.ietf.org>; Tue, 26 Mar 2002 18:28:49 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA29800
	for <ieprep@ietf.org>; Tue, 26 Mar 2002 18:28:46 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVBY6>; Tue, 26 Mar 2002 18:27:15 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C57@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'James M. Polk'" <jmpolk@cisco.com>,
        "'goldmans@agcs.com'"
	 <goldmans@agcs.com>,
        Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Preemption
Date: Tue, 26 Mar 2002 18:28:11 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D51D.E4C81410"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D51D.E4C81410
Content-Type: text/plain;
	charset="iso-8859-1"

To clarify- 
Yes it is a policy decision. 
The "network authority" may determine, a priori, the indicator settings
associated with each priority level- leaving no choice to the user.  Or it
may determine a range of settings that a particular user can chose from.
 
The question is, once the policy has been established, will the protocol
allow  (at a minimum) the indicators to be transported across the IP
network, and (preferably) can the IP network implement the actions
associated with the indicators?

-----Original Message-----
From: James M. Polk [mailto:jmpolk@cisco.com]
Sent: Tuesday, March 26, 2002 6:01 PM
To: Gunn, Janet NE; 'goldmans@agcs.com'; Ken Carlberg
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Preemption


At 04:05 PM 3/26/2002 -0500, Gunn, Janet NE wrote: 


It may be a bit more complicated than that.  
 
I am not sure about MLPP, but in eMLPP there are THREE indicators (in
addition to level):
Queueing allowed/Queueing NOT allowed
CAN pre-empt a lower level call/CANNOT pre-empt a call
CAN be pre-empted by a higher level call/CANNOT be preempted by another
call, no matter what the priority.
 
Typically the "Ordinary" call has the lowest priority, but may neither
pre-empt, nor be pre-empted.


I believe *all* you listed above are policy decisions that are local in
nature. If left to the end-users, they would choose "cannot preempt" every
time. Perhaps this could be dealt with via user authentication as to whether
a call from that user is or is not allowed to be preempted....




It seems to me that there MAY be a need for all three indicators in the
context of an IP network also.


starting to get complicated with 3 indicators




The indicators MIGHT be tied directly to the level, but (especially if a
single IP network is transporting calls from more than one carrier) might
also be set on a call by call basis. So, it would be desirable for the
protocol to have a way of capturing all 3 indicators.
 
Janet 



-----Original Message----- 


*************************************
"People generally demand more respect for their own rights than they are
willing to allow for others"


James M. Polk
Consulting Engineer
Office of the CTO

Cisco Systems
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA
w) 972.813.5208
f)  972.813.5280
www.cisco.com <http://www.cisco.com/>  


------_=_NextPart_001_01C1D51D.E4C81410
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 5.00.3103.1000" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=129452323-26032002>To 
clarify- </SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=129452323-26032002>Yes it 
is a policy decision. </SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=129452323-26032002>The 
"network authority" may determine, a priori, the indicator settings&nbsp; 
associated with each priority level- leaving no choice to the user.&nbsp; Or it 
may determine a range of settings that a particular user can chose 
from.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=129452323-26032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=129452323-26032002>The 
question is, once the policy has been established, will the protocol allow&nbsp; 
(at a minimum) the indicators to be transported across the IP network, and 
(preferably) can the IP network implement the actions associated with the 
indicators?</SPAN></FONT></DIV>
<BLOCKQUOTE style="MARGIN-RIGHT: 0px">
  <DIV align=left class=OutlookMessageHeader dir=ltr><FONT face=Tahoma 
  size=2>-----Original Message-----<BR><B>From:</B> James M. Polk 
  [mailto:jmpolk@cisco.com]<BR><B>Sent:</B> Tuesday, March 26, 2002 6:01 
  PM<BR><B>To:</B> Gunn, Janet NE; 'goldmans@agcs.com'; Ken 
  Carlberg<BR><B>Cc:</B> ieprep@ietf.org<BR><B>Subject:</B> RE: [Ieprep] 
  Preemption<BR><BR></DIV></FONT>At 04:05 PM 3/26/2002 -0500, Gunn, Janet NE 
  wrote: <BR><FONT color=#0000ff face=arial size=2>
  <BLOCKQUOTE cite type="cite">It may be a bit more complicated than 
    that.&nbsp; </FONT><BR>&nbsp;<BR><FONT color=#0000ff face=arial size=2>I am 
    not sure about MLPP, but in eMLPP there are THREE indicators (in addition to 
    level):</FONT><BR>Queueing allowed/Queueing NOT allowed<BR>CAN pre-empt a 
    lower level call/CANNOT pre-empt a call<BR>CAN be pre-empted by a higher 
    level call/CANNOT be preempted by another call, no matter what the 
    priority.<BR>&nbsp;<BR><FONT color=#0000ff face=arial size=2>Typically the 
    "Ordinary" call has the lowest priority, but may neither pre-empt, nor be 
    pre-empted.</FONT></BLOCKQUOTE><BR>I believe *all* you listed above are policy 
  decisions that are local in nature. If left to the end-users, they would 
  choose "cannot preempt" every time. Perhaps this could be dealt with via user 
  authentication as to whether a call from that user is or is not allowed to be 
  preempted....<BR><BR>
  <BLOCKQUOTE cite type="cite"><BR><FONT color=#0000ff face=arial size=2>It 
    seems to me that there MAY be a need for all three indicators in the context 
    of an IP network also.</FONT></BLOCKQUOTE><BR>starting to get complicated with 
  3 indicators<BR><BR>
  <BLOCKQUOTE cite type="cite"><BR><FONT color=#0000ff face=arial size=2>The 
    indicators MIGHT be tied directly to the level, but (especially if a single 
    IP network is transporting calls from more than one carrier) might also be 
    set on a call by call basis. So, it would be desirable for the protocol to 
    have a way of capturing all 3 indicators.</FONT><BR>&nbsp;<BR><FONT 
    color=#0000ff face=arial size=2>Janet</FONT>
    <BLOCKQUOTE><FONT face=tahoma size=2>
      <DL>
        <DD>-----Original Message-----</FONT> </DD></DL></BLOCKQUOTE><BR>
    <DIV align=center>*************************************<BR>"People generally 
    demand more respect for their own rights than they are willing to allow for 
    others"<BR><BR></DIV>James M. Polk<BR>Consulting Engineer<BR>Office of the 
    CTO<BR><BR>Cisco Systems<BR>2200 East President George Bush Turnpike 
    <BR>Richardson, TX&nbsp; 75082 USA<BR>w) 972.813.5208<BR>f)&nbsp; 
    972.813.5280<BR><A href="http://www.cisco.com/" 
    eudora="autourl">www.cisco.com</A> </BLOCKQUOTE></BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C1D51D.E4C81410--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 27 10:58:16 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA03903
	for <ieprep-archive@odin.ietf.org>; Wed, 27 Mar 2002 10:58:16 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA07940;
	Wed, 27 Mar 2002 10:57:41 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA07878
	for <ieprep@optimus.ietf.org>; Wed, 27 Mar 2002 10:57:38 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA03867
	for <ieprep@ietf.org>; Wed, 27 Mar 2002 10:57:35 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id IAA24849
	for <ieprep@ietf.org>; Wed, 27 Mar 2002 08:55:43 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA4B92
          for <ieprep@ietf.org>; Wed, 27 Mar 2002 08:57:35 -0700
Received: FROM agcs.com BY mknaomi ; Wed Mar 27 08:57:33 2002 -0700
Message-ID: <3CA1EBEF.9A775224@agcs.com>
Date: Wed, 27 Mar 2002 08:57:35 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
References: <CBED705A7FD2D311865500508B10894105D48C57@chntex02.is.dyncorp.com>
Content-Type: multipart/mixed;
 boundary="------------BA427E4CC9391D2E30FFF3B8"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------BA427E4CC9391D2E30FFF3B8
Content-Type: multipart/alternative;
 boundary="------------C81829FE302FBAF551E96B4E"


--------------C81829FE302FBAF551E96B4E
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Janet,

I guess I had best ask the question.

From your wording below it would seem that the model in your mind is
that of a single network, The model in my mind is that a call may cross
a series of networks to reach the termination point. Each network in the
path may have a different policy in regard to preemption. The policy
would apply to the resources used in that particular network. Thus a
given call path could be the result of the policies expressed in each
network through which the call progressed.

Could then a call preempt in network 1, but be prevented from preempting
in network 2, on its way toward termination in network 3? This would of
course mean that a call had been preempted even though the priority call
was not successfully set up. I think the answer would be yes, and that
this would be acceptable.



"Gunn, Janet NE" wrote:

>  To clarify- Yes it is a policy decision. The "network authority" may
> determine, a priori, the indicator settings  associated with each
> priority level- leaving no choice to the user.  Or it may determine a
> range of settings that a particular user can chose from.The question
> is, once the policy has been established, will the protocol allow  (at
> a minimum) the indicators to be transported across the IP network, and
> (preferably) can the IP network implement the actions associated with
> the indicators?
>
>      -----Original Message-----
>      From: James M. Polk [mailto:jmpolk@cisco.com]
>      Sent: Tuesday, March 26, 2002 6:01 PM
>      To: Gunn, Janet NE; 'goldmans@agcs.com'; Ken Carlberg
>      Cc: ieprep@ietf.org
>      Subject: RE: [Ieprep] Preemption
>
>      At 04:05 PM 3/26/2002 -0500, Gunn, Janet NE wrote:
>
>     > It may be a bit more complicated than that.
>     >
>     > I am not sure about MLPP, but in eMLPP there are THREE
>     > indicators (in addition to level):
>     > Queueing allowed/Queueing NOT allowed
>     > CAN pre-empt a lower level call/CANNOT pre-empt a call
>     > CAN be pre-empted by a higher level call/CANNOT be
>     > preempted by another call, no matter what the priority.
>     >
>     > Typically the "Ordinary" call has the lowest priority, but
>     > may neither pre-empt, nor be pre-empted.
>
>
>      I believe *all* you listed above are policy decisions that
>      are local in nature. If left to the end-users, they would
>      choose "cannot preempt" every time. Perhaps this could be
>      dealt with via user authentication as to whether a call from
>      that user is or is not allowed to be preempted....
>
>
>     >
>     > It seems to me that there MAY be a need for all three
>     > indicators in the context of an IP network also.
>
>
>      starting to get complicated with 3 indicators
>
>
>     >
>     > The indicators MIGHT be tied directly to the level, but
>     > (especially if a single IP network is transporting calls
>     > from more than one carrier) might also be set on a call by
>     > call basis. So, it would be desirable for the protocol to
>     > have a way of capturing all 3 indicators.
>     >
>     > Janet
>     >
>     >           -----Original Message-----
>     >
>     >            *************************************
>     >  "People generally demand more respect for their own rights
>     >          than they are willing to allow for others"
>     >
>     >  James M. Polk
>     > Consulting Engineer
>     > Office of the CTO
>     >
>     > Cisco Systems
>     > 2200 East President George Bush Turnpike
>     > Richardson, TX  75082 USA
>     > w) 972.813.5208
>     > f)  972.813.5280
>     > www.cisco.com
>

--------------C81829FE302FBAF551E96B4E
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Janet,
<p>I guess I had best ask the question.
<p>From your wording below it would seem that the model in your mind is
that of a single network, The model in my mind is that a call may cross
a series of networks to reach the termination point. Each network in the
path may have a different policy in regard to preemption. The policy would
apply to the resources used in that particular network. Thus a given call
path could be the result of the policies expressed in each network through
which the call progressed.
<p>Could then a call preempt in network 1, but be prevented from preempting
in network 2, on its way toward termination in network 3? This would of
course mean that a call had been preempted even though the priority call
was not successfully set up. I think the answer would be yes, and that
this would be acceptable.
<br>&nbsp;
<br>&nbsp;
<p>"Gunn, Janet NE" wrote:
<blockquote TYPE=CITE>&nbsp;<span class=129452323-26032002><font face="Arial"><font color="#0000FF"><font size=-1>To
clarify-&nbsp;</font></font></font></span><span class=129452323-26032002><font face="Arial"><font color="#0000FF"><font size=-1>Yes
it is a policy decision.&nbsp;</font></font></font></span><span class=129452323-26032002><font face="Arial"><font color="#0000FF"><font size=-1>The
"network authority" may determine, a priori, the indicator settings&nbsp;
associated with each priority level- leaving no choice to the user.&nbsp;
Or it may determine a range of settings that a particular user can chose
from.</font></font></font></span><span 
class=129452323-26032002></span><span class=129452323-26032002><font face="Arial"><font color="#0000FF"><font size=-1>The
question is, once the policy has been established, will the protocol allow&nbsp;
(at a minimum) the indicators to be transported across the IP network,
and (preferably) can the IP network implement the actions associated with
the indicators?</font></font></font></span>
<blockquote style="MARGIN-RIGHT: 0px">
<div class="OutlookMessageHeader" dir="ltr"><font face="Tahoma"><font size=-1>-----Original
Message-----</font></font>
<br><font face="Tahoma"><font size=-1><b>From:</b> James M. Polk [<A HREF="mailto:jmpolk@cisco.com">mailto:jmpolk@cisco.com</A>]</font></font>
<br><font face="Tahoma"><font size=-1><b>Sent:</b> Tuesday, March 26, 2002
6:01 PM</font></font>
<br><font face="Tahoma"><font size=-1><b>To:</b> Gunn, Janet NE; 'goldmans@agcs.com';
Ken Carlberg</font></font>
<br><font face="Tahoma"><font size=-1><b>Cc:</b> ieprep@ietf.org</font></font>
<br><font face="Tahoma"><font size=-1><b>Subject:</b> RE: [Ieprep] Preemption</font></font>
<br>&nbsp;</div>
At 04:05 PM 3/26/2002 -0500, Gunn, Janet NE wrote:
<blockquote cite type="cite"><font face="arial"><font color="#0000FF"><font size=-1>It
may be a bit more complicated than that.</font></font></font>
<p><font face="arial"><font color="#0000FF"><font size=-1>I am not sure
about MLPP, but in eMLPP there are THREE indicators (in addition to level):</font></font></font>
<br>Queueing allowed/Queueing NOT allowed
<br>CAN pre-empt a lower level call/CANNOT pre-empt a call
<br>CAN be pre-empted by a higher level call/CANNOT be preempted by another
call, no matter what the priority.
<p><font face="arial"><font color="#0000FF"><font size=-1>Typically the
"Ordinary" call has the lowest priority, but may neither pre-empt, nor
be pre-empted.</font></font></font></blockquote>

<p><br>I believe *all* you listed above are policy decisions that are local
in nature. If left to the end-users, they would choose "cannot preempt"
every time. Perhaps this could be dealt with via user authentication as
to whether a call from that user is or is not allowed to be preempted....
<br>&nbsp;
<blockquote cite type="cite">&nbsp;
<br><font face="arial"><font color="#0000FF"><font size=-1>It seems to
me that there MAY be a need for all three indicators in the context of
an IP network also.</font></font></font></blockquote>

<p><br>starting to get complicated with 3 indicators
<br>&nbsp;
<blockquote cite type="cite">&nbsp;
<br><font face="arial"><font color="#0000FF"><font size=-1>The indicators
MIGHT be tied directly to the level, but (especially if a single IP network
is transporting calls from more than one carrier) might also be set on
a call by call basis. So, it would be desirable for the protocol to have
a way of capturing all 3 indicators.</font></font></font>
<p><font face="arial"><font color="#0000FF"><font size=-1>Janet</font></font></font>
<blockquote>
<dl>
<dd>
<font face="tahoma"><font size=-1>-----Original Message-----</font></font></dd>
</dl>
</blockquote>

<center>*************************************
<br>"People generally demand more respect for their own rights than they
are willing to allow for others"</center>

<br>&nbsp;James M. Polk
<br>Consulting Engineer
<br>Office of the CTO
<p>Cisco Systems
<br>2200 East President George Bush Turnpike
<br>Richardson, TX&nbsp; 75082 USA
<br>w) 972.813.5208
<br>f)&nbsp; 972.813.5280
<br><a href="http://www.cisco.com/" eudora="autourl">www.cisco.com</a></blockquote>
</blockquote>
</blockquote>
</html>

--------------C81829FE302FBAF551E96B4E--

--------------BA427E4CC9391D2E30FFF3B8
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------BA427E4CC9391D2E30FFF3B8--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 27 11:18:15 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04853
	for <ieprep-archive@odin.ietf.org>; Wed, 27 Mar 2002 11:18:15 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA14018;
	Wed, 27 Mar 2002 11:17:45 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA13983
	for <ieprep@optimus.ietf.org>; Wed, 27 Mar 2002 11:17:44 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04825
	for <ieprep@ietf.org>; Wed, 27 Mar 2002 11:17:40 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVDBT>; Wed, 27 Mar 2002 11:16:04 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C58@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'goldmans@agcs.com'" <goldmans@agcs.com>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Preemption
Date: Wed, 27 Mar 2002 11:17:00 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D5AA.D32357F0"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D5AA.D32357F0
Content-Type: text/plain;
	charset="iso-8859-1"

It doesn't necessarily have to be a single network.  For instance, even
within the Telephony world, a call might be initiated on a GSM wireless
network which supports eMLPP, traverse the (landline) PSTN which carries the
MLPP indicators, but doesn't DO any queueing or pre-emption on the call, and
terminate on a CDMA wireless network which ignores the MLPP indicators. (Or
it could terminate on another GSM wireless network which DOES translate the
indicators from MLPP to eMLPP and acts on them.)
 
To follow your analogy, the call could pre-empt in the originaing GSM
network, and them be blocked in the  (landline) PSTN. 
 
We also can have situations where the call setup queues in one network but
not another. 
 
Yes, I would expect it to be acceptable for the same thing to happen in an
internet (littlei ), sInce this is at least as good as the telephony
network.  But it would be desirable (dependant on the even higher level
protocol of negotiating SLAs between networks) to at least have the
technical possibility of implementing a consistent policy.

-----Original Message-----
From: Stuart (Stu) Goldman [mailto:goldmans@agcs.com]
Sent: Wednesday, March 27, 2002 10:58 AM
To: Gunn, Janet NE
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption


Janet, 

I guess I had best ask the question. 


From your wording below it would seem that the model in your mind is that of
a single network, The model in my mind is that a call may cross a series of
networks to reach the termination point. Each network in the path may have a
different policy in regard to preemption. The policy would apply to the
resources used in that particular network. Thus a given call path could be
the result of the policies expressed in each network through which the call
progressed. 


Could then a call preempt in network 1, but be prevented from preempting in
network 2, on its way toward termination in network 3? This would of course
mean that a call had been preempted even though the priority call was not
successfully set up. I think the answer would be yes, and that this would be
acceptable. 
  
  


"Gunn, Janet NE" wrote: 


 To clarify- Yes it is a policy decision. The "network authority" may
determine, a priori, the indicator settings  associated with each priority
level- leaving no choice to the user.  Or it may determine a range of
settings that a particular user can chose from.The question is, once the
policy has been established, will the protocol allow  (at a minimum) the
indicators to be transported across the IP network, and (preferably) can the
IP network implement the actions associated with the indicators? 

-----Original Message----- 
From: James M. Polk [ mailto:jmpolk@cisco.com <mailto:jmpolk@cisco.com> ] 
Sent: Tuesday, March 26, 2002 6:01 PM 
To: Gunn, Janet NE; 'goldmans@agcs.com'; Ken Carlberg 
Cc: ieprep@ietf.org 
Subject: RE: [Ieprep] Preemption 
 
At 04:05 PM 3/26/2002 -0500, Gunn, Janet NE wrote: 

It may be a bit more complicated than that. 

I am not sure about MLPP, but in eMLPP there are THREE indicators (in
addition to level): 
Queueing allowed/Queueing NOT allowed 
CAN pre-empt a lower level call/CANNOT pre-empt a call 
CAN be pre-empted by a higher level call/CANNOT be preempted by another
call, no matter what the priority. 


Typically the "Ordinary" call has the lowest priority, but may neither
pre-empt, nor be pre-empted.


I believe *all* you listed above are policy decisions that are local in
nature. If left to the end-users, they would choose "cannot preempt" every
time. Perhaps this could be dealt with via user authentication as to whether
a call from that user is or is not allowed to be preempted.... 
  



It seems to me that there MAY be a need for all three indicators in the
context of an IP network also.


starting to get complicated with 3 indicators 
  



The indicators MIGHT be tied directly to the level, but (especially if a
single IP network is transporting calls from more than one carrier) might
also be set on a call by call basis. So, it would be desirable for the
protocol to have a way of capturing all 3 indicators. 

Janet 


-----Original Message----- 

************************************* 
"People generally demand more respect for their own rights than they are
willing to allow for others"

 James M. Polk 
Consulting Engineer 
Office of the CTO 

Cisco Systems 
2200 East President George Bush Turnpike 
Richardson, TX  75082 USA 
w) 972.813.5208 
f)  972.813.5280 
www.cisco.com <http://www.cisco.com/> 


------_=_NextPart_001_01C1D5AA.D32357F0
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 5.00.3103.1000" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=530350216-27032002>It 
doesn't necessarily have to be a single network.&nbsp; For instance, even within 
the Telephony world, a call might be initiated on a GSM wireless network which 
supports eMLPP, traverse the (landline) PSTN which carries the MLPP indicators, 
but doesn't DO any queueing or pre-emption on the call, and terminate on a CDMA 
wireless network which ignores the MLPP indicators. (Or it could terminate on 
another GSM wireless network which DOES translate the indicators from MLPP to 
eMLPP and acts on them.)</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=530350216-27032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=530350216-27032002>To 
follow your analogy, the call could pre-empt in the originaing GSM network, and 
them be blocked in the&nbsp; (landline) PSTN.&nbsp;</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=530350216-27032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=530350216-27032002>We 
also&nbsp;can have situations where&nbsp;the call setup queues in one network 
but not another.&nbsp;</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=530350216-27032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=530350216-27032002>Yes, I 
would expect it to be acceptable for the same thing to happen in an internet 
(littlei ), sInce this is at least as good as the telephony network.&nbsp; But 
it would be desirable (dependant on the even higher level protocol of 
negotiating SLAs between networks) to at least have the technical possibility of 
implementing a consistent policy.</SPAN></FONT></DIV>
<BLOCKQUOTE style="MARGIN-RIGHT: 0px">
  <DIV align=left class=OutlookMessageHeader dir=ltr><FONT face=Tahoma 
  size=2>-----Original Message-----<BR><B>From:</B> Stuart (Stu) Goldman 
  [mailto:goldmans@agcs.com]<BR><B>Sent:</B> Wednesday, March 27, 2002 10:58 
  AM<BR><B>To:</B> Gunn, Janet NE<BR><B>Cc:</B> 
  ieprep@ietf.org<BR><B>Subject:</B> Re: [Ieprep] 
  Preemption<BR><BR></DIV></FONT>Janet, 
  <P>I guess I had best ask the question. 
  <P>From your wording below it would seem that the model in your mind is that 
  of a single network, The model in my mind is that a call may cross a series of 
  networks to reach the termination point. Each network in the path may have a 
  different policy in regard to preemption. The policy would apply to the 
  resources used in that particular network. Thus a given call path could be the 
  result of the policies expressed in each network through which the call 
  progressed. 
  <P>Could then a call preempt in network 1, but be prevented from preempting in 
  network 2, on its way toward termination in network 3? This would of course 
  mean that a call had been preempted even though the priority call was not 
  successfully set up. I think the answer would be yes, and that this would be 
  acceptable. <BR>&nbsp; <BR>&nbsp; 
  <P>"Gunn, Janet NE" wrote: 
  <BLOCKQUOTE TYPE="CITE">&nbsp;<SPAN class=129452323-26032002><FONT 
    face=Arial><FONT color=#0000ff><FONT size=-1>To 
    clarify-&nbsp;</FONT></FONT></FONT></SPAN><SPAN 
    class=129452323-26032002><FONT face=Arial><FONT color=#0000ff><FONT 
    size=-1>Yes it is a policy decision.&nbsp;</FONT></FONT></FONT></SPAN><SPAN 
    class=129452323-26032002><FONT face=Arial><FONT color=#0000ff><FONT 
    size=-1>The "network authority" may determine, a priori, the indicator 
    settings&nbsp; associated with each priority level- leaving no choice to the 
    user.&nbsp; Or it may determine a range of settings that a particular user 
    can chose from.</FONT></FONT></FONT></SPAN><SPAN 
    class=129452323-26032002></SPAN><SPAN class=129452323-26032002><FONT 
    face=Arial><FONT color=#0000ff><FONT size=-1>The question is, once the 
    policy has been established, will the protocol allow&nbsp; (at a minimum) 
    the indicators to be transported across the IP network, and (preferably) can 
    the IP network implement the actions associated with the 
    indicators?</FONT></FONT></FONT></SPAN> 
    <BLOCKQUOTE style="MARGIN-RIGHT: 0px">
      <DIV class=OutlookMessageHeader dir=ltr><FONT face=Tahoma><FONT 
      size=-1>-----Original Message-----</FONT></FONT> <BR><FONT 
      face=Tahoma><FONT size=-1><B>From:</B> James M. Polk [<A 
      href="mailto:jmpolk@cisco.com">mailto:jmpolk@cisco.com</A>]</FONT></FONT> 
      <BR><FONT face=Tahoma><FONT size=-1><B>Sent:</B> Tuesday, March 26, 2002 
      6:01 PM</FONT></FONT> <BR><FONT face=Tahoma><FONT size=-1><B>To:</B> Gunn, 
      Janet NE; 'goldmans@agcs.com'; Ken Carlberg</FONT></FONT> <BR><FONT 
      face=Tahoma><FONT size=-1><B>Cc:</B> ieprep@ietf.org</FONT></FONT> 
      <BR><FONT face=Tahoma><FONT size=-1><B>Subject:</B> RE: [Ieprep] 
      Preemption</FONT></FONT> <BR>&nbsp;</DIV>At 04:05 PM 3/26/2002 -0500, 
      Gunn, Janet NE wrote: 
      <BLOCKQUOTE type="cite" cite><FONT face=arial><FONT color=#0000ff><FONT 
        size=-1>It may be a bit more complicated than that.</FONT></FONT></FONT> 

        <P><FONT face=arial><FONT color=#0000ff><FONT size=-1>I am not sure 
        about MLPP, but in eMLPP there are THREE indicators (in addition to 
        level):</FONT></FONT></FONT> <BR>Queueing allowed/Queueing NOT allowed 
        <BR>CAN pre-empt a lower level call/CANNOT pre-empt a call <BR>CAN be 
        pre-empted by a higher level call/CANNOT be preempted by another call, 
        no matter what the priority. 
        <P><FONT face=arial><FONT color=#0000ff><FONT size=-1>Typically the 
        "Ordinary" call has the lowest priority, but may neither pre-empt, nor 
        be pre-empted.</FONT></FONT></FONT></P></BLOCKQUOTE>
      <P><BR>I believe *all* you listed above are policy decisions that are 
      local in nature. If left to the end-users, they would choose "cannot 
      preempt" every time. Perhaps this could be dealt with via user 
      authentication as to whether a call from that user is or is not allowed to 
      be preempted.... <BR>&nbsp; 
      <BLOCKQUOTE type="cite" cite> <BR><FONT face=arial><FONT 
        color=#0000ff><FONT size=-1>It seems to me that there MAY be a need for 
        all three indicators in the context of an IP network 
        also.</FONT></FONT></FONT></BLOCKQUOTE>
      <P><BR>starting to get complicated with 3 indicators <BR>&nbsp; 
      <BLOCKQUOTE type="cite" cite> <BR><FONT face=arial><FONT 
        color=#0000ff><FONT size=-1>The indicators MIGHT be tied directly to the 
        level, but (especially if a single IP network is transporting calls from 
        more than one carrier) might also be set on a call by call basis. So, it 
        would be desirable for the protocol to have a way of capturing all 3 
        indicators.</FONT></FONT></FONT> 
        <P><FONT face=arial><FONT color=#0000ff><FONT 
        size=-1>Janet</FONT></FONT></FONT> 
        <BLOCKQUOTE>
          <DL>
            <DD><FONT face=tahoma><FONT size=-1>-----Original 
            Message-----</FONT></FONT> </DD></DL></BLOCKQUOTE>
        <CENTER>************************************* <BR>"People generally 
        demand more respect for their own rights than they are willing to allow 
        for others"</CENTER><BR>&nbsp;James M. Polk <BR>Consulting Engineer 
        <BR>Office of the CTO 
        <P>Cisco Systems <BR>2200 East President George Bush Turnpike 
        <BR>Richardson, TX&nbsp; 75082 USA <BR>w) 972.813.5208 <BR>f)&nbsp; 
        972.813.5280 <BR><A href="http://www.cisco.com/" 
        eudora="autourl">www.cisco.com</A></P></BLOCKQUOTE></BLOCKQUOTE></BLOCKQUOTE></BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C1D5AA.D32357F0--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 27 11:45:18 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA07089
	for <ieprep-archive@odin.ietf.org>; Wed, 27 Mar 2002 11:45:18 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA22152;
	Wed, 27 Mar 2002 11:44:50 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA22113
	for <ieprep@optimus.ietf.org>; Wed, 27 Mar 2002 11:44:48 -0500 (EST)
Received: from mail.kqtravel.com (uxtrav03.kqtravel.com [193.242.85.67])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id LAA07056
	for <ieprep@ietf.org>; Wed, 27 Mar 2002 11:44:42 -0500 (EST)
Received: (qmail 3294 invoked from network); 27 Mar 2002 17:36:17 -0000
Received: from jp-111.ppp.aih.net (HELO degree2.com) (202.6.255.43)
  by uxtrav03.kqtravel.com with SMTP; 27 Mar 2002 17:36:17 -0000
Message-ID: <3CA1F6F9.ABF77701@degree2.com>
Date: Wed, 27 Mar 2002 16:44:41 +0000
From: Alistair Munro <alistair.munro@degree2.com>
Organization: Degree2 Innovations Ltd.
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
CC: "'goldmans@agcs.com'" <goldmans@agcs.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
References: <CBED705A7FD2D311865500508B10894105D48C58@chntex02.is.dyncorp.com>
Content-Type: multipart/alternative;
 boundary="------------4A02699C801D35FDEB38D318"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--------------4A02699C801D35FDEB38D318
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear All,

I'm new to this list but not to GSM network design and implementation.

It's often "impossible" to transfer "standard" semantics across
inter-network interworking points because of design decisions made by
implementors of network elements and operational policies. Janet
describes one of many such situations. Isn't this one of the best
practice issues we are trying to expose?

I'm not sure if this thread is in scope - it's very useful but I can't
match it to the charter.

Regards,

Alistair

> It doesn't necessarily have to be a single network.  For instance,
> even within the Telephony world, a call might be initiated on a GSM
> wireless network which supports eMLPP, traverse the (landline) PSTN
> which carries the MLPP indicators, but doesn't DO any queueing or
> pre-emption on the call, and terminate on a CDMA wireless network
> which ignores the MLPP indicators. (Or it could terminate on another
> GSM wireless network which DOES translate the indicators from MLPP to
> eMLPP and acts on them.)To follow your analogy, the call could
> pre-empt in the originaing GSM network, and them be blocked in the
> (landline) PSTN. We also can have situations where the call setup
> queues in one network but not another. Yes, I would expect it to be
> acceptable for the same thing to happen in an internet (littlei ),
> sInce this is at least as good as the telephony network.  But it would
> be desirable (dependant on the even higher level protocol of
> negotiating SLAs between networks) to at least have the technical
> possibility of implementing a consistent policy.

--------------4A02699C801D35FDEB38D318
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Dear All,
<p>I'm new to this list but not to GSM network design and implementation.
<p>It's often "impossible" to transfer "standard" semantics across inter-network
interworking points because of design decisions made by implementors of
network elements and operational policies. Janet describes one of many
such situations. Isn't this one of the best practice issues we are trying
to expose?
<p>I'm not sure if this thread is in scope - it's very useful but I can't
match it to the charter.
<p>Regards,
<p>Alistair
<blockquote TYPE=CITE><span class=530350216-27032002><font face="Arial"><font color="#0000FF"><font size=-1>It
doesn't necessarily have to be a single network.&nbsp; For instance, even
within the Telephony world, a call might be initiated on a GSM wireless
network which supports eMLPP, traverse the (landline) PSTN which carries
the MLPP indicators, but doesn't DO any queueing or pre-emption on the
call, and terminate on a CDMA wireless network which ignores the MLPP indicators.
(Or it could terminate on another GSM wireless network which DOES translate
the indicators from MLPP to eMLPP and acts on them.)</span><span 
class=530350216-27032002></span><span class=530350216-27032002>To
follow your analogy, the call could pre-empt in the originaing GSM network,
and them be blocked in the&nbsp; (landline) PSTN.&nbsp;</span><span 
class=530350216-27032002></span><span class=530350216-27032002>We
also can have situations where the call setup queues in one network but
not another.&nbsp;</span><span 
class=530350216-27032002></span><span class=530350216-27032002>Yes,
I would expect it to be acceptable for the same thing to happen in an internet
(littlei ), sInce this is at least as good as the telephony network.&nbsp;
But it would be desirable (dependant on the even higher level protocol
of negotiating SLAs between networks) to at least have the technical possibility
of implementing a consistent policy.</font></font></font></span></blockquote>
</html>

--------------4A02699C801D35FDEB38D318--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 27 12:15:08 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA09203
	for <ieprep-archive@odin.ietf.org>; Wed, 27 Mar 2002 12:15:08 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA01001;
	Wed, 27 Mar 2002 12:14:37 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA00968
	for <ieprep@optimus.ietf.org>; Wed, 27 Mar 2002 12:14:35 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA09159
	for <ieprep@ietf.org>; Wed, 27 Mar 2002 12:14:33 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVDK2>; Wed, 27 Mar 2002 12:12:56 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C5A@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'Alistair Munro'" <alistair.munro@degree2.com>
Cc: "'goldmans@agcs.com'" <goldmans@agcs.com>, ieprep@ietf.org
Subject: RE: [Ieprep] Preemption/charter
Date: Wed, 27 Mar 2002 12:13:52 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D5B2.C4C62720"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D5B2.C4C62720
Content-Type: text/plain;
	charset="iso-8859-1"

I would have thought it fell under
"detail the specific functions and technologies needed to provide support
for Emergency Preparedness systems in the Internet. "
since Emergency Preparedness Systems (e.g., the GETS WPS, as well as some
military emergency systems) are, or will be, or may be, using MLPP and
eMLPP- like functionality.
 
But if it is out of scope, I'll be glad to drop it.

-----Original Message-----
From: Alistair Munro [mailto:alistair.munro@degree2.com]
Sent: Wednesday, March 27, 2002 11:45 AM
To: Gunn, Janet NE
Cc: 'goldmans@agcs.com'; ieprep@ietf.org
Subject: Re: [Ieprep] Preemption



I'm not sure if this thread is in scope - it's very useful but I can't match
it to the charter. 


------_=_NextPart_001_01C1D5B2.C4C62720
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 5.00.3103.1000" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=779270917-27032002>I 
would have thought it fell under</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=779270917-27032002>"detail the specific functions and technologies needed 
to provide support for Emergency Preparedness systems in the Internet. 
"</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=779270917-27032002>since 
Emergency Preparedness Systems (e.g., the&nbsp;GETS WPS, as well as some 
military emergency systems) are, or will be, or may be, using MLPP and eMLPP- 
like  functionality.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=779270917-27032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=779270917-27032002>But if 
it is out of scope, I'll be glad to drop it.</SPAN></FONT></DIV>
<BLOCKQUOTE style="MARGIN-RIGHT: 0px">
  <DIV align=left class=OutlookMessageHeader dir=ltr><FONT face=Tahoma 
  size=2>-----Original Message-----<BR><B>From:</B> Alistair Munro 
  [mailto:alistair.munro@degree2.com]<BR><B>Sent:</B> Wednesday, March 27, 2002 
  11:45 AM<BR><B>To:</B> Gunn, Janet NE<BR><B>Cc:</B> 'goldmans@agcs.com'; 
  ieprep@ietf.org<BR><B>Subject:</B> Re: [Ieprep] 
Preemption<BR><BR></DIV></FONT>
  <P>I'm not sure if this thread is in scope - it's very useful but I can't 
  match it to the charter. </P></BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C1D5B2.C4C62720--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 27 12:57:15 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11690
	for <ieprep-archive@odin.ietf.org>; Wed, 27 Mar 2002 12:57:15 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA23149;
	Wed, 27 Mar 2002 12:56:02 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA23116
	for <ieprep@optimus.ietf.org>; Wed, 27 Mar 2002 12:56:01 -0500 (EST)
Received: from mail.kqtravel.com (uxtrav03.kqtravel.com [193.242.85.67])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id MAA11568
	for <ieprep@ietf.org>; Wed, 27 Mar 2002 12:56:00 -0500 (EST)
Received: (qmail 4849 invoked from network); 27 Mar 2002 18:47:33 -0000
Received: from jp-113.ppp.aih.net (HELO degree2.com) (202.6.255.45)
  by uxtrav03.kqtravel.com with SMTP; 27 Mar 2002 18:47:33 -0000
Message-ID: <3CA207AC.7E23F835@degree2.com>
Date: Wed, 27 Mar 2002 17:55:56 +0000
From: Alistair Munro <alistair.munro@degree2.com>
Organization: Degree2 Innovations Ltd.
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
CC: "'goldmans@agcs.com'" <goldmans@agcs.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Preemption/charter
References: <CBED705A7FD2D311865500508B10894105D48C5A@chntex02.is.dyncorp.com>
Content-Type: multipart/alternative;
 boundary="------------9BA0BBD29754C4166D30C452"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--------------9BA0BBD29754C4166D30C452
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi Janet,

You describe some very relevant issues that we must take account of at
the point where the Internet "ends" and the PSTN "begins". But I don't
know what we can do about things inside the PSTN (includes GSM, ISDN and
all that) networks - it would be very optimistic to try to overhaul what
is (IMHO) worst practice - nobody's fault, just legacy systems. I hope
we can have some impact nevertheless - it would be great if I could dial
911 from my US-homed WLAN PDA in London and the emergency services in
London would respond. Try it with your GSM phone sometime... The same
MLPP rules may be accepted but parameters supplied by my PDA won't make
sense outside its home network from a PSTN viewpoint - from what I heard
in the SIPPING group, it would not be impossible in the IP domain.

Incidentally are you aware of any requirements on IP-based functions
that may be operated *inside* a PSTN-based emergency service, e.g. as a
call-centre? This does seem to be in the scope to me, but I'm happy to
be advised otherwise.

> I would have thought it fell under"detail the specific functions and
> technologies needed to provide support for Emergency Preparedness
> systems in the Internet. "since Emergency Preparedness Systems (e.g.,
> the GETS WPS, as well as some military emergency systems) are, or will
> be, or may be, using MLPP and eMLPP- like functionality.But if it is
> out of scope, I'll be glad to drop it.
>
>      -----Original Message-----
>      From: Alistair Munro [mailto:alistair.munro@degree2.com]
>      Sent: Wednesday, March 27, 2002 11:45 AM
>      To: Gunn, Janet NE
>      Cc: 'goldmans@agcs.com'; ieprep@ietf.org
>      Subject: Re: [Ieprep] Preemption
>      I'm not sure if this thread is in scope - it's very useful
>      but I can't match it to the charter.
>

Regards,

Alistair

--------------9BA0BBD29754C4166D30C452
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Hi Janet,
<p>You describe some very relevant issues that we must take account of
at the point where the Internet "ends" and the PSTN "begins". But I don't
know what we can do about things inside the PSTN (includes GSM, ISDN and
all that) networks - it would be very optimistic to try to overhaul what
is (IMHO) worst practice - nobody's fault, just legacy systems. I hope
we can have some impact nevertheless - it would be great if I could dial
911 from my US-homed WLAN PDA in London and the emergency services in London
would respond. Try it with your GSM phone sometime... The same MLPP rules
may be accepted but parameters supplied by my PDA won't make sense outside
its home network from a PSTN viewpoint - from what I heard in the SIPPING
group, it would not be impossible in the IP domain.
<p>Incidentally are you aware of any requirements on IP-based functions
that may be operated *inside* a PSTN-based emergency service, e.g. as a
call-centre? This does seem to be in the scope to me, but I'm happy to
be advised otherwise.
<blockquote TYPE=CITE><span class=779270917-27032002><font face="Arial"><font color="#0000FF"><font size=-1>I
would have thought it fell under</span><span 
class=779270917-27032002>"detail
the specific functions and technologies needed to provide support for Emergency
Preparedness systems in the Internet. "</span><span class=779270917-27032002>since
Emergency Preparedness Systems (e.g., the GETS WPS, as well as some military
emergency systems) are, or will be, or may be, using MLPP and eMLPP- like
functionality.</span><span 
class=779270917-27032002></span><span class=779270917-27032002>But
if it is out of scope, I'll be glad to drop it.</font></font></font></span>
<blockquote style="MARGIN-RIGHT: 0px">
<div class="OutlookMessageHeader" dir="ltr"><font face="Tahoma"><font size=-1>-----Original
Message-----</font></font>
<br><font face="Tahoma"><font size=-1><b>From:</b> Alistair Munro [<A HREF="mailto:alistair.munro@degree2.com">mailto:alistair.munro@degree2.com</A>]</font></font>
<br><font face="Tahoma"><font size=-1><b>Sent:</b> Wednesday, March 27,
2002 11:45 AM</font></font>
<br><font face="Tahoma"><font size=-1><b>To:</b> Gunn, Janet NE</font></font>
<br><font face="Tahoma"><font size=-1><b>Cc:</b> 'goldmans@agcs.com'; ieprep@ietf.org</font></font>
<br><font face="Tahoma"><font size=-1><b>Subject:</b> Re: [Ieprep] Preemption</font></font></div>
I'm not sure if this thread is in scope - it's very useful but I can't
match it to the charter.</blockquote>
</blockquote>

<p><br>Regards,
<p>Alistair</html>

--------------9BA0BBD29754C4166D30C452--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 27 14:46:10 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA18716
	for <ieprep-archive@odin.ietf.org>; Wed, 27 Mar 2002 14:46:10 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA04208;
	Wed, 27 Mar 2002 14:32:11 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA04181
	for <ieprep@optimus.ietf.org>; Wed, 27 Mar 2002 14:32:07 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA18118
	for <ieprep@ietf.org>; Wed, 27 Mar 2002 14:32:06 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVD82>; Wed, 27 Mar 2002 14:30:36 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C5C@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'Alistair Munro'" <alistair.munro@degree2.com>
Cc: "'goldmans@agcs.com'" <goldmans@agcs.com>, ieprep@ietf.org
Subject: RE: [Ieprep] Preemption/charter
Date: Wed, 27 Mar 2002 14:31:31 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D5C5.FF535710"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D5C5.FF535710
Content-Type: text/plain;
	charset="iso-8859-1"

Alastair,
 
I think it should be a "given" that we don't want to impose the "worst
practices" of the Telephony world on the Internet.
 
SImilarly, I don't think that this group should get involved in trying to
"fix" the compromises in the Telephony world.
 
But I do think it is useful to identify the FUNCTIONALITY that is available
in the Telephony domain, that is (or may be) used by Emergengy Preparedness
Systems, and to explore what kind of protocol support would be needed  to
provide equivalent functionality in the Internet (or an internet).
 
And I was trying to answer this question: "Are there possibilities for
extensions to MLPP to cover more than one domain? "
 
Janet

-----Original Message-----
From: Alistair Munro [mailto:alistair.munro@degree2.com]
Sent: Wednesday, March 27, 2002 12:56 PM
To: Gunn, Janet NE
Cc: 'goldmans@agcs.com'; ieprep@ietf.org
Subject: Re: [Ieprep] Preemption/charter


Hi Janet, 

You describe some very relevant issues that we must take account of at the
point where the Internet "ends" and the PSTN "begins". But I don't know what
we can do about things inside the PSTN (includes GSM, ISDN and all that)
networks - it would be very optimistic to try to overhaul what is (IMHO)
worst practice - nobody's fault, just legacy systems. I hope we can have
some impact nevertheless - it would be great if I could dial 911 from my
US-homed WLAN PDA in London and the emergency services in London would
respond. Try it with your GSM phone sometime... The same MLPP rules may be
accepted but parameters supplied by my PDA won't make sense outside its home
network from a PSTN viewpoint - from what I heard in the SIPPING group, it
would not be impossible in the IP domain. 


Incidentally are you aware of any requirements on IP-based functions that
may be operated *inside* a PSTN-based emergency service, e.g. as a
call-centre? This does seem to be in the scope to me, but I'm happy to be
advised otherwise. 


I would have thought it fell under"detail the specific functions and
technologies needed to provide support for Emergency Preparedness systems in
the Internet. "since Emergency Preparedness Systems (e.g., the GETS WPS, as
well as some military emergency systems) are, or will be, or may be, using
MLPP and eMLPP- like functionality.But if it is out of scope, I'll be glad
to drop it. 

-----Original Message----- 
From: Alistair Munro [ mailto:alistair.munro@degree2.com
<mailto:alistair.munro@degree2.com> ] 
Sent: Wednesday, March 27, 2002 11:45 AM 
To: Gunn, Janet NE 
Cc: 'goldmans@agcs.com'; ieprep@ietf.org 
Subject: Re: [Ieprep] Preemption
I'm not sure if this thread is in scope - it's very useful but I can't match
it to the charter.


Regards, 


Alistair 


------_=_NextPart_001_01C1D5C5.FF535710
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 5.00.3103.1000" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=384162319-27032002>Alastair,</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=384162319-27032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=384162319-27032002>I 
think it should be a "given" that we don't want to impose the "worst practices" 
of the Telephony world on the Internet.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=384162319-27032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=384162319-27032002>SImilarly, I don't think that this group should get 
involved in trying to "fix" the compromises in the Telephony 
world.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=384162319-27032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=384162319-27032002>But I 
do think it is useful to identify the FUNCTIONALITY that is available in the 
Telephony domain, that is (or may be) used by Emergengy Preparedness Systems, 
and to explore what kind of protocol support would be needed&nbsp; to provide 
equivalent functionality in the Internet (or an internet).</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=384162319-27032002></SPAN></FONT>&nbsp;</DIV>
<DIV><SPAN class=384162319-27032002><FONT color=#0000ff face=Arial size=2>And I 
was trying to answer this question: "</FONT><FONT size=2><FONT 
color=#0000ff><FONT face=Arial>Are there possibilities for extensions to MLPP to 
cover more than one domain?&nbsp;<SPAN 
class=384162319-27032002>"</SPAN></FONT></FONT></FONT></SPAN></DIV>
<DIV><SPAN class=384162319-27032002><FONT size=2><FONT color=#0000ff><FONT 
face=Arial><SPAN 
class=384162319-27032002></SPAN></FONT></FONT></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=384162319-27032002><FONT size=2><FONT color=#0000ff><FONT 
face=Arial><SPAN 
class=384162319-27032002>Janet</SPAN></FONT></FONT></FONT></SPAN></DIV>
<BLOCKQUOTE style="MARGIN-RIGHT: 0px">
  <DIV align=left class=OutlookMessageHeader dir=ltr><FONT face=Tahoma 
  size=2>-----Original Message-----<BR><B>From:</B> Alistair Munro 
  [mailto:alistair.munro@degree2.com]<BR><B>Sent:</B> Wednesday, March 27, 2002 
  12:56 PM<BR><B>To:</B> Gunn, Janet NE<BR><B>Cc:</B> 'goldmans@agcs.com'; 
  ieprep@ietf.org<BR><B>Subject:</B> Re: [Ieprep] 
  Preemption/charter<BR><BR></DIV></FONT>Hi Janet, 
  <P>You describe some very relevant issues that we must take account of at the 
  point where the Internet "ends" and the PSTN "begins". But I don't know what 
  we can do about things inside the PSTN (includes GSM, ISDN and all that) 
  networks - it would be very optimistic to try to overhaul what is (IMHO) worst 
  practice - nobody's fault, just legacy systems. I hope we can have some impact 
  nevertheless - it would be great if I could dial 911 from my US-homed WLAN PDA 
  in London and the emergency services in London would respond. Try it with your 
  GSM phone sometime... The same MLPP rules may be accepted but parameters 
  supplied by my PDA won't make sense outside its home network from a PSTN 
  viewpoint - from what I heard in the SIPPING group, it would not be impossible 
  in the IP domain. 
  <P>Incidentally are you aware of any requirements on IP-based functions that 
  may be operated *inside* a PSTN-based emergency service, e.g. as a 
  call-centre? This does seem to be in the scope to me, but I'm happy to be 
  advised otherwise. 
  <BLOCKQUOTE TYPE="CITE"><SPAN class=779270917-27032002><FONT 
    face=Arial><FONT color=#0000ff><FONT size=-1>I would have thought it fell 
    under</SPAN><SPAN class=779270917-27032002>"detail the specific functions 
    and technologies needed to provide support for Emergency Preparedness 
    systems in the Internet. "</SPAN><SPAN class=779270917-27032002>since 
    Emergency Preparedness Systems (e.g., the GETS WPS, as well as some military 
    emergency systems) are, or will be, or may be, using MLPP and eMLPP- like 
    functionality.</SPAN><SPAN class=779270917-27032002></SPAN><SPAN 
    class=779270917-27032002>But if it is out of scope, I'll be glad to drop 
    it.</FONT></FONT></FONT></SPAN> 
    <BLOCKQUOTE style="MARGIN-RIGHT: 0px">
      <DIV class=OutlookMessageHeader dir=ltr><FONT face=Tahoma><FONT 
      size=-1>-----Original Message-----</FONT></FONT> <BR><FONT 
      face=Tahoma><FONT size=-1><B>From:</B> Alistair Munro [<A 
      href="mailto:alistair.munro@degree2.com">mailto:alistair.munro@degree2.com</A>]</FONT></FONT> 
      <BR><FONT face=Tahoma><FONT size=-1><B>Sent:</B> Wednesday, March 27, 2002 
      11:45 AM</FONT></FONT> <BR><FONT face=Tahoma><FONT size=-1><B>To:</B> 
      Gunn, Janet NE</FONT></FONT> <BR><FONT face=Tahoma><FONT 
      size=-1><B>Cc:</B> 'goldmans@agcs.com'; ieprep@ietf.org</FONT></FONT> 
      <BR><FONT face=Tahoma><FONT size=-1><B>Subject:</B> Re: [Ieprep] 
      Preemption</FONT></FONT></DIV>I'm not sure if this thread is in scope - 
      it's very useful but I can't match it to the 
charter.</BLOCKQUOTE></BLOCKQUOTE>
  <P><BR>Regards, 
  <P>Alistair </P></BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C1D5C5.FF535710--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 27 16:51:15 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA26095
	for <ieprep-archive@odin.ietf.org>; Wed, 27 Mar 2002 16:51:15 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA16063;
	Wed, 27 Mar 2002 16:50:27 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA16029
	for <ieprep@optimus.ietf.org>; Wed, 27 Mar 2002 16:50:25 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id QAA26047
	for <ieprep@ietf.org>; Wed, 27 Mar 2002 16:50:23 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.04846-0@bells.cs.ucl.ac.uk>; Wed, 27 Mar 2002 21:50:18 +0000
To: goldmans@agcs.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
In-reply-to: Your message of "Tue, 26 Mar 2002 07:02:41 MST." <3CA07F81.B74F6A76@agcs.com>
Date: Wed, 27 Mar 2002 21:50:19 +0000
Message-ID: <18621.1017265819@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

Stu,

> It seems to me that there are four general cases:
>
>   * If a given network never preempts, then preemption does not occur there
>   * If a given network always preempts when needed, then preemption may
>     occur there
>   * If a given network can do either based on an mandatory indicator, then
>     it does whatever the indicator directs
>   * If a given network can do either based on an optional indicator, then
>     the action taken in the absence of the indicator is the DEFAULT action.
>     To be consistent with the policy statement in this case the default
>     would seem to be to not preempt.
>
> Is my understanding correct, or did I miss something? I  used the term
> indicator since the information may be part of the protocol or could be
> delivered to the network by other means. I didn't want to restrict it to
> just being in the protocol.

the above seems fine.  personally, i might express things differently,
but no need to split hairs....its seems close enough.

-ken


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Wed Mar 27 17:03:28 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA26746
	for <ieprep-archive@odin.ietf.org>; Wed, 27 Mar 2002 17:03:28 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA17840;
	Wed, 27 Mar 2002 17:01:26 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA17768
	for <ieprep@optimus.ietf.org>; Wed, 27 Mar 2002 17:01:20 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id RAA26672
	for <ieprep@ietf.org>; Wed, 27 Mar 2002 17:01:17 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.05293-0@bells.cs.ucl.ac.uk>; Wed, 27 Mar 2002 22:01:08 +0000
To: Alistair Munro <alistair.munro@degree2.com>
cc: "Gunn, Janet NE" <Janet.Gunn@dyncorp.com>,
        "'goldmans@agcs.com'" <goldmans@agcs.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
In-reply-to: Your message of "Wed, 27 Mar 2002 16:44:41 GMT." <3CA1F6F9.ABF77701@degree2.com>
Date: Wed, 27 Mar 2002 22:01:08 +0000
Message-ID: <18901.1017266468@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


> I'm not sure if this thread is in scope - it's very useful but I can't match
> it to the charter. 

at least from my perspective, don't be shy about discussing topics that
are generally related to the Charter.  its just that with specific
IEPREP related drafts, we may be a bit more careful about what goes in
(or out) of them.

-ken


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 28 08:54:40 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA25734
	for <ieprep-archive@odin.ietf.org>; Thu, 28 Mar 2002 08:54:40 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA17260;
	Thu, 28 Mar 2002 08:54:15 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA17225
	for <ieprep@optimus.ietf.org>; Thu, 28 Mar 2002 08:54:13 -0500 (EST)
Received: from mail.kqtravel.com (uxtrav03.kqtravel.com [193.242.85.67])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA25699
	for <ieprep@ietf.org>; Thu, 28 Mar 2002 08:54:05 -0500 (EST)
Received: (qmail 21336 invoked from network); 28 Mar 2002 14:45:40 -0000
Received: from jp-013.ppp.aih.net (HELO degree2.com) (202.6.255.13)
  by uxtrav03.kqtravel.com with SMTP; 28 Mar 2002 14:45:40 -0000
Message-ID: <3CA32073.D076A4D8@degree2.com>
Date: Thu, 28 Mar 2002 13:53:55 +0000
From: Alistair Munro <alistair.munro@degree2.com>
Organization: Degree2 Innovations Ltd.
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
CC: "'goldmans@agcs.com'" <goldmans@agcs.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Preemption/charter
References: <CBED705A7FD2D311865500508B10894105D48C5C@chntex02.is.dyncorp.com>
Content-Type: multipart/alternative;
 boundary="------------F5C54BA5068E931DB68B41EB"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


--------------F5C54BA5068E931DB68B41EB
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi Janet,

Sorry about the current delay in responding - I'm in Japan right now and
there is a kind of time warp in operation. I'll be back at base next
week and less disloacted.

> I think it should be a "given" that we don't want to impose the "worst
> practices" of the Telephony world on the Internet.SImilarly, I don't
> think that this group should get involved in trying to "fix" the
> compromises in the Telephony world.

I agree with this - it isn't necessary anyway. Information is always
valuable anyway.

> But I do think it is useful to identify the FUNCTIONALITY that is
> available in the Telephony domain, that is (or may be) used by
> Emergengy Preparedness Systems, and to explore what kind of protocol
> support would be needed  to provide equivalent functionality in the
> Internet (or an internet).And I was trying to answer this question:
> "Are there possibilities for extensions to MLPP to cover more than one
> domain? "

MLPP is one thing and it should be as widely applicable as possible.
I'll have to sit down and think about that - comments/questions later.

I don't know that it is necessary to provide equivalent functionality in
the Internet but there *are* some missing elements. For example, if we
can keep the statistical multiplexing gain of packet mode operation then
preemption (for example, maybe not the best nor the only telephony
feature to be examined) *may* not be so important.

Regards,

Alistair

--------------F5C54BA5068E931DB68B41EB
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Hi Janet,
<p>Sorry about the current delay in responding - I'm in Japan right now
and there is a kind of time warp in operation. I'll be back at base next
week and less disloacted.
<blockquote TYPE=CITE><span class=384162319-27032002><font face="Arial"><font color="#0000FF"><font size=-1>I
think it should be a "given" that we don't want to impose the "worst practices"
of the Telephony world on the Internet.</span><span 
class=384162319-27032002></span><span 
class=384162319-27032002>SImilarly,
I don't think that this group should get involved in trying to "fix" the
compromises in the Telephony world.</font></font></font></span><span 
class=384162319-27032002></span></blockquote>
I agree with this - it isn't necessary anyway. Information is always valuable
anyway.
<blockquote TYPE=CITE><span class=384162319-27032002><font face="Arial"><font color="#0000FF"><font size=-1>But
I do think it is useful to identify the FUNCTIONALITY that is available
in the Telephony domain, that is (or may be) used by Emergengy Preparedness
Systems, and to explore what kind of protocol support would be needed&nbsp;
to provide equivalent functionality in the Internet (or an internet).</span><span 
class=384162319-27032002></span><span class=384162319-27032002>And
I was trying to answer this question: "Are there possibilities for extensions
to MLPP to cover more than one domain?&nbsp;<span 
class=384162319-27032002>"</font></font></font></span></span><span class=384162319-27032002><span 
class=384162319-27032002></span></span></blockquote>
MLPP is one thing and it should be as widely applicable as possible. I'll
have to sit down and think about that - comments/questions later.
<p>I don't know that it is necessary to provide equivalent functionality
in the Internet but there *are* some missing elements. For example, if
we can keep the statistical multiplexing gain of packet mode operation
then preemption (for example, maybe not the best nor the only telephony
feature to be examined) *may* not be so important.
<p>Regards,
<p>Alistair</html>

--------------F5C54BA5068E931DB68B41EB--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 28 09:12:29 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA26566
	for <ieprep-archive@odin.ietf.org>; Thu, 28 Mar 2002 09:12:29 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA18359;
	Thu, 28 Mar 2002 09:12:03 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA18325
	for <ieprep@optimus.ietf.org>; Thu, 28 Mar 2002 09:12:01 -0500 (EST)
Received: from mail.kqtravel.com (uxtrav03.kqtravel.com [193.242.85.67])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id JAA26562
	for <ieprep@ietf.org>; Thu, 28 Mar 2002 09:11:57 -0500 (EST)
Received: (qmail 21714 invoked from network); 28 Mar 2002 15:03:28 -0000
Received: from jp-013.ppp.aih.net (HELO degree2.com) (202.6.255.13)
  by uxtrav03.kqtravel.com with SMTP; 28 Mar 2002 15:03:28 -0000
Message-ID: <3CA324A8.CBE7ED25@degree2.com>
Date: Thu, 28 Mar 2002 14:11:53 +0000
From: Alistair Munro <alistair.munro@degree2.com>
Organization: Degree2 Innovations Ltd.
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
CC: "Gunn, Janet NE" <Janet.Gunn@dyncorp.com>,
        "'goldmans@agcs.com'" <goldmans@agcs.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
References: <18901.1017266468@cs.ucl.ac.uk>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

Hi Ken,

> > I'm not sure if this thread is in scope - it's very useful but I can't match
> > it to the charter.
>
> at least from my perspective, don't be shy about discussing topics that
> are generally related to the Charter.  its just that with specific
> IEPREP related drafts, we may be a bit more careful about what goes in
> (or out) of them.

I wonder if anyone on the operational side of PSTN, ISDN, GSM has any comment on
the preemption issue? It would certainly help my understanding of practice.

Regards,

Alistair



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 28 17:32:52 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA18247
	for <ieprep-archive@odin.ietf.org>; Thu, 28 Mar 2002 17:32:52 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA19160;
	Thu, 28 Mar 2002 17:31:57 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAB19127
	for <ieprep@optimus.ietf.org>; Thu, 28 Mar 2002 17:31:47 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA18190
	for <ieprep@ietf.org>; Thu, 28 Mar 2002 17:31:42 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVHN1>; Thu, 28 Mar 2002 17:30:04 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C60@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'Alistair Munro'" <alistair.munro@degree2.com>
Cc: "'goldmans@agcs.com'" <goldmans@agcs.com>, ieprep@ietf.org
Subject:  [Ieprep] Public Safety?
Date: Thu, 28 Mar 2002 17:31:01 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D6A8.3D3BBEE0"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D6A8.3D3BBEE0
Content-Type: text/plain;
	charset="iso-8859-1"

This (emergency service call center) sounds to me more like "public safety"
rather than "emergency preparedness".  I know this is considered
out-of-scope for GETS, but it is not clear to me whether it is in scope for
IEPREP. 
 
My comments on public safety are purely as an interested layperson, it is
not an area of expertise, so I may be talking out of the top of my hat.
 
If it is in scope, the one feature that I am sure a 911 call center would
want is a location function for IP phones that is similar to the one that
currently exists in the US for land-line phones, and is mandated (again, in
the US) for mobile phones in 2002.  I don't know if this is even plausible
in the Internet. (Does Mobile IP address physical location, or only
topological/network location?  Would topological/network location be useful
to a 911 call center?)
 
If public safety is in scope, land mobile radio interoperability (again,
specifically out of scope for GETS) is an important requirement.  During the
recovery efforts after 9/11, the the lack of interoperability between the
land mobile radio networks used by the various groups (Police, Fire, Rescue,
etc) from different jurisdictions was a major problem.  It seems plausible
that, if each of these systems had an interface to the Internet, the
Internet could provide the interoperability between the systems.
 
 
 
 -----Original Message-----
From: Alistair Munro [mailto:alistair.munro@degree2.com]
Sent: Wednesday, March 27, 2002 12:56 PM
To: Gunn, Janet NE
Cc: 'goldmans@agcs.com'; ieprep@ietf.org
Subject: Re: [Ieprep] Preemption/charter



Incidentally are you aware of any requirements on IP-based functions that
may be operated *inside* a PSTN-based emergency service, e.g. as a
call-centre? This does seem to be in the scope to me, but I'm happy to be
advised otherwise. 


------_=_NextPart_001_01C1D6A8.3D3BBEE0
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 5.00.3103.1000" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=884391822-28032002>This 
(emergency service call center) sounds to me more like "public safety" rather 
than "emergency preparedness".&nbsp; I know this is considered 
out</SPAN></FONT><FONT face=Tahoma><FONT size=2><SPAN 
class=884391822-28032002><FONT color=#0000ff face=Arial>-of-scope for GETS, but 
it is not clear to me whether it is in scope for 
IEPREP.&nbsp;</FONT></SPAN></FONT></FONT></DIV>
<DIV><FONT face=Tahoma><FONT size=2><SPAN class=884391822-28032002><FONT 
color=#0000ff face=Arial></FONT></SPAN></FONT></FONT>&nbsp;</DIV>
<DIV><FONT face=Tahoma><FONT size=2><SPAN class=884391822-28032002><FONT 
color=#0000ff face=Arial>My comments on public safety are purely as an 
interested layperson, it is not an area of expertise, so I may be talking out of 
the top of my hat.</FONT></SPAN></FONT></FONT></DIV>
<DIV><FONT face=Tahoma><FONT size=2><SPAN 
class=884391822-28032002></SPAN></FONT></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=884391822-28032002>If it 
is in scope, the one feature that I am sure a 911 call center would want is a 
location function for IP phones that is similar to the one that currently exists 
in the US for land-line phones, and is mandated (again, in the US) for mobile 
phones in 2002.&nbsp; I don't know if this is even plausible in the Internet. 
(Does Mobile IP address physical location, or only topological/network 
location?&nbsp; Would topological/network location be useful to a 911 call 
center?)</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=884391822-28032002></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=884391822-28032002>If 
public safety is in scope, land mobile radio interoperability (again, 
specifically out of scope for GETS) is an important requirement.&nbsp; During 
the recovery efforts after 9/11, the the lack of interoperability between the 
land mobile radio networks used by the various groups (Police, Fire, Rescue, 
etc) from different jurisdictions was a major problem.&nbsp; It seems plausible 
that, if each of these systems had an interface to the Internet, the Internet 
could provide the interoperability between the systems.</SPAN></FONT></DIV>
<DIV><FONT face=Tahoma><FONT size=2><SPAN 
class=884391822-28032002></SPAN></FONT></FONT>&nbsp;</DIV>
<DIV><FONT face=Tahoma><FONT size=2><SPAN 
class=884391822-28032002></SPAN></FONT></FONT>&nbsp;</DIV>
<DIV><FONT face=Tahoma><FONT size=2><SPAN 
class=884391822-28032002></SPAN></FONT></FONT>&nbsp;</DIV>
<DIV><FONT face=Tahoma><FONT size=2><SPAN 
class=884391822-28032002>&nbsp;</SPAN>-----Original Message-----<BR><B>From:</B> 
Alistair Munro [mailto:alistair.munro@degree2.com]<BR><B>Sent:</B> Wednesday, 
March 27, 2002 12:56 PM<BR><B>To:</B> Gunn, Janet NE<BR><B>Cc:</B> 
'goldmans@agcs.com'; ieprep@ietf.org<BR><B>Subject:</B> Re: [Ieprep] 
Preemption/charter<BR><BR></DIV></FONT>
<BLOCKQUOTE style="MARGIN-RIGHT: 0px"></FONT>
  <P>Incidentally are you aware of any requirements on IP-based functions that 
  may be operated *inside* a PSTN-based emergency service, e.g. as a 
  call-centre? This does seem to be in the scope to me, but I'm happy to be 
  advised otherwise. </P></BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C1D6A8.3D3BBEE0--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Thu Mar 28 18:46:43 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA19356
	for <ieprep-archive@odin.ietf.org>; Thu, 28 Mar 2002 18:46:42 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA22246;
	Thu, 28 Mar 2002 18:46:27 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA22220
	for <ieprep@optimus.ietf.org>; Thu, 28 Mar 2002 18:46:26 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA19351
	for <ieprep@ietf.org>; Thu, 28 Mar 2002 18:46:20 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id QAA07034
	for <ieprep@ietf.org>; Thu, 28 Mar 2002 16:44:30 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA81B
          for <ieprep@ietf.org>; Thu, 28 Mar 2002 16:46:23 -0700
Received: FROM agcs.com BY mknaomi ; Thu Mar 28 16:46:25 2002 -0700
Message-ID: <3CA3AB4D.CDA3CCFF@agcs.com>
Date: Thu, 28 Mar 2002 16:46:22 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
CC: "'Alistair Munro'" <alistair.munro@degree2.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Public Safety?
References: <CBED705A7FD2D311865500508B10894105D48C60@chntex02.is.dyncorp.com>
Content-Type: multipart/mixed;
 boundary="------------596529027051F01671E482B1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------596529027051F01671E482B1
Content-Type: multipart/alternative;
 boundary="------------0F32675B9D0833F20B0087DA"


--------------0F32675B9D0833F20B0087DA
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Janet,

Yes, I think you are talking about the public access to place "public"
emergency calls, rather than the preference given to emergency
responders.

Location information is indeed a problem for IP based phones.  The
problem is just compounded if one can unplug the CPE from one location
and plug it in in a different location. This makes office moves a bit
less painful for the company, but is not so good for having information
that can direct the ambulance to the right location.

Laptops are obvious a problem, and with a wireless LAN card, it becomes
yet more difficult. (As a good engineer I never ever say that something
is insolvable.)

"Gunn, Janet NE" wrote:

>  This (emergency service call center) sounds to me more like "public
> safety" rather than "emergency preparedness".  I know this is
> considered out-of-scope for GETS, but it is not clear to me whether it
> is in scope for IEPREP. My comments on public safety are purely as an
> interested layperson, it is not an area of expertise, so I may be
> talking out of the top of my hat.If it is in scope, the one feature
> that I am sure a 911 call center would want is a location function for
> IP phones that is similar to the one that currently exists in the US
> for land-line phones, and is mandated (again, in the US) for mobile
> phones in 2002.  I don't know if this is even plausible in the
> Internet. (Does Mobile IP address physical location, or only
> topological/network location?  Would topological/network location be
> useful to a 911 call center?)If public safety is in scope, land mobile
> radio interoperability (again, specifically out of scope for GETS) is
> an important requirement.  During the recovery efforts after 9/11, the
> the lack of interoperability between the land mobile radio networks
> used by the various groups (Police, Fire, Rescue, etc) from different
> jurisdictions was a major problem.  It seems plausible that, if each
> of these systems had an interface to the Internet, the Internet could
> provide the interoperability between the systems.-----Original
> Message-----
> From: Alistair Munro [mailto:alistair.munro@degree2.com]
> Sent: Wednesday, March 27, 2002 12:56 PM
> To: Gunn, Janet NE
> Cc: 'goldmans@agcs.com'; ieprep@ietf.org
> Subject: Re: [Ieprep] Preemption/charter
>
>
>      Incidentally are you aware of any requirements on IP-based
>      functions that may be operated *inside* a PSTN-based
>      emergency service, e.g. as a call-centre? This does seem to
>      be in the scope to me, but I'm happy to be advised
>      otherwise.
>

--------------0F32675B9D0833F20B0087DA
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Janet,
<p>Yes, I think you are talking about the public access to place "public"
emergency calls, rather than the preference given to emergency responders.
<p>Location information is indeed a problem for IP based phones.&nbsp;
The problem is just compounded if one can unplug the CPE from one location
and plug it in in a different location. This makes office moves a bit less
painful for the company, but is not so good for having information that
can direct the ambulance to the right location.
<p>Laptops are obvious a problem, and with a wireless LAN card, it becomes
yet more difficult. (As a good engineer I never ever say that something
is insolvable.)
<p>"Gunn, Janet NE" wrote:
<blockquote TYPE=CITE>&nbsp;<span class=884391822-28032002><font face="Arial"><font color="#0000FF"><font size=-1>This
(emergency service call center) sounds to me more like "public safety"
rather than "emergency preparedness".&nbsp; I know this is considered out</span><span 
class=884391822-28032002>-of-scope
for GETS, but it is not clear to me whether it is in scope for IEPREP.&nbsp;</font></font></font></span><span class=884391822-28032002></span><span class=884391822-28032002><font face="Arial"><font color="#0000FF"><font size=-1>My
comments on public safety are purely as an interested layperson, it is
not an area of expertise, so I may be talking out of the top of my hat.</font></font></font></span><span 
class=884391822-28032002></span><span class=884391822-28032002><font face="Arial"><font color="#0000FF"><font size=-1>If
it is in scope, the one feature that I am sure a 911 call center would
want is a location function for IP phones that is similar to the one that
currently exists in the US for land-line phones, and is mandated (again,
in the US) for mobile phones in 2002.&nbsp; I don't know if this is even
plausible in the Internet. (Does Mobile IP address physical location, or
only topological/network location?&nbsp; Would topological/network location
be useful to a 911 call center?)</font></font></font></span><span 
class=884391822-28032002></span><span class=884391822-28032002><font face="Arial"><font color="#0000FF"><font size=-1>If
public safety is in scope, land mobile radio interoperability (again, specifically
out of scope for GETS) is an important requirement.&nbsp; During the recovery
efforts after 9/11, the the lack of interoperability between the land mobile
radio networks used by the various groups (Police, Fire, Rescue, etc) from
different jurisdictions was a major problem.&nbsp; It seems plausible that,
if each of these systems had an interface to the Internet, the Internet
could provide the interoperability between the systems.</font></font></font></span><span 
class=884391822-28032002></span><span 
class=884391822-28032002></span><span 
class=884391822-28032002></span><span 
class=884391822-28032002></span><font face="Tahoma"><font size=-1>-----Original
Message-----</font></font>
<br><font face="Tahoma"><font size=-1><b>From:</b> Alistair Munro [<A HREF="mailto:alistair.munro@degree2.com">mailto:alistair.munro@degree2.com</A>]</font></font>
<br><font face="Tahoma"><font size=-1><b>Sent:</b> Wednesday, March 27,
2002 12:56 PM</font></font>
<br><font face="Tahoma"><font size=-1><b>To:</b> Gunn, Janet NE</font></font>
<br><font face="Tahoma"><font size=-1><b>Cc:</b> 'goldmans@agcs.com'; ieprep@ietf.org</font></font>
<br><font face="Tahoma"><font size=-1><b>Subject:</b> Re: [Ieprep] Preemption/charter</font></font>
<br>&nbsp;
<blockquote style="MARGIN-RIGHT: 0px">Incidentally are you aware of any
requirements on IP-based functions that may be operated *inside* a PSTN-based
emergency service, e.g. as a call-centre? This does seem to be in the scope
to me, but I'm happy to be advised otherwise.</blockquote>
</blockquote>
</html>

--------------0F32675B9D0833F20B0087DA--

--------------596529027051F01671E482B1
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------596529027051F01671E482B1--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 01:28:09 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA27324
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 01:28:09 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id BAA18769;
	Fri, 29 Mar 2002 01:27:30 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id BAA18735
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 01:27:27 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA27308
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 01:27:24 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2T6QoY03911;
	Thu, 28 Mar 2002 22:26:50 -0800 (PST)
Message-Id: <5.1.0.14.2.20020328211333.020c6c18@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com (Unverified)
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Thu, 28 Mar 2002 21:30:50 -0800
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Public Safety?
Cc: ieprep@ietf.org
In-Reply-To: <CBED705A7FD2D311865500508B10894105D48C60@chntex02.is.dynco
 rp.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 02:31 PM 3/28/2002, Gunn, Janet NE wrote:
>This (emergency service call center) sounds to me more like "public 
>safety" rather than "emergency preparedness".  I know this is considered 
>out-of-scope for GETS, but it is not clear to me whether it is in scope 
>for IEPREP.

ieprep relates to anything concerning emergencies; GETs is an obvious 
example, but just an example. If the end result of the conversation is that 
we develop requirements related to interoperability with the telephone 
network in the 911 area, we're probably pretty safe from a charter 
perspective. If it starts getting silly, I will put out the word, or Scott 
will slap my hands for not doing so.

>If it is in scope, the one feature that I am sure a 911 call center would 
>want is a location function for IP phones that is similar to the one that 
>currently exists in the US for land-line phones, and is mandated (again, 
>in the US) for mobile phones in 2002.  I don't know if this is even 
>plausible in the Internet. (Does Mobile IP address physical location, or 
>only topological/network location?  Would topological/network location be 
>useful to a 911 call center?)

Topological location is trivial ("what's your care-of address"), but I 
should think would be next to useless. The fact that I am somewhere in the 
region served by a particular prefix is very possibly less precise than the 
fact that I am in a given mobile phone cell (a set of cells might use the 
same prefix). If I'm reporting an accident, "somewhere in a certain 20 
square miles" isn't what the 911 operator wants to hear.

I believe that there is a working group addressing physical location; if it 
is a requirement, the obvious thing for us to do is state and support the 
requirement, and let said working group work out the details.

There are a number of issues in physical location. One, it tends to imply 
the use of a GPS or something like it. Land lines don't really know the 
physical location of the instrument, or at least none that I have ever 
taken apart did; the PABX etc might very well know where it is, though, and 
believe that the instrument is physically collocated. GSM and US PCS phones 
have a GPS requirement, but it is subject to similar caveats - the phone 
knows where it was last time it could see three satellites, but apart from 
that one is limited to triangulation, which has been known to be 
inaccurate. Non-mobile IP hosts tend to be IP hosts located in a building, 
meaning they will rarely if ever be able to "see" a satellite, but they may 
be near to a GPS-based time source and can report that they are within a 
certain time variance of it as determined by NTP. 802.11 cards contain no 
GPS that I know of. Mobile IP telephones should be very similar to any 
other mobile phone in this respect, however - a GPS might be mandated.

Another issue is related to privacy. If I am calling 911 to report an 
accident, I most certainly want the 911 operator to know where I am. If I 
am shopping for Christmas presents, I might not be very interested in my 
wife being able to locate me precisely. If the Christmas present is for my 
mistress (no, but imagine), all sorts of interesting legal games could come 
into play. In general, the person operating the device probably wants to be 
in control of who knows what about the location of the instrument, the 
precision of that location, and the history of its location.

>If public safety is in scope, land mobile radio interoperability (again, 
>specifically out of scope for GETS) is an important requirement.  During 
>the recovery efforts after 9/11, the the lack of interoperability between 
>the land mobile radio networks used by the various groups (Police, Fire, 
>Rescue, etc) from different jurisdictions was a major problem.  It seems 
>plausible that, if each of these systems had an interface to the Internet, 
>the Internet could provide the interoperability between the systems.

I'm in the market for some words...


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 01:28:22 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA27337
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 01:28:21 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id BAA18805;
	Fri, 29 Mar 2002 01:27:42 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id BAA18776
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 01:27:33 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA27316
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 01:27:30 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2T6QaY03822;
	Thu, 28 Mar 2002 22:26:36 -0800 (PST)
Message-Id: <5.1.0.14.2.20020328205617.01ea1368@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com (Unverified)
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Thu, 28 Mar 2002 21:02:03 -0800
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>,
        Alistair Munro <alistair.munro@degree2.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Preemption
Cc: ieprep@ietf.org
In-Reply-To: <3CA1F6F9.ABF77701@degree2.com>
References: <CBED705A7FD2D311865500508B10894105D48C58@chntex02.is.dyncorp.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 08:44 AM 3/27/2002, Alistair Munro wrote:
>I'm not sure if this thread is in scope - it's very useful but I can't 
>match it to the charter.

I would deem it in scope. We are supposed to make recommendations 
concerning "best practices" (by which we mean "the best we can think of", 
not "the best currently employed"), and set up requirements for any 
protocols that we think may need to be significantly changed from current 
definition or practice. This applies both to telephony and data traffic 
used in an emergency setting.

Janet, it sounds to me like you have a pretty good handle on some of the 
issues here.

(a)Do you feel that they are adequately covered in the framework draft? if 
not, could you propose language to go into it covering the issues?
(b)Do you see specific requirements not being met in industry standard 
protocols? If so, do you feel that either Hal Folts white paper or my draft 
cover the issues, and if so specifically which sections? In not, would you 
be willing to suggest text to cover those issues?

At 11:31 AM 3/27/2002, Gunn, Janet NE wrote:
>But I do think it is useful to identify the FUNCTIONALITY that is 
>available in the Telephony domain, that is (or may be) used by Emergengy 
>Preparedness Systems, and to explore what kind of protocol support would 
>be needed  to provide equivalent functionality in the Internet (or an 
>internet).

Yes. Specifically an example of this has been requested by the Area 
Director - we have been asked to provide requirements for the 
implementation of MLPP in SIP. I would expect that this would include

(a) identifying what problem it solves (such as "it exists in the PSTN and 
we need to be able to carry the signalling across an Internet island, 
whether we actually implement it or not", or "research/experience 
documented at <place> demonstrates that this set of functionality is 
materially helpful in solving <specified> class of problems")
(b) identifying two or more implementation approaches (such as 
authorization us use of an overhead space in GETS and preemption of lower 
priority calls in AutoVon),
(c) identifying what is the minimum necessary to implement the 
functionality in an Internet domain, and
(d) identifying what is necessary to carry the signaling from one 
PSTN/GSM/etc domain to another PSTN/GSM/etc domain across an internet domain

It would also include identification of practices that are not helpful and 
should not be replicated in the Internet, if appropriate.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 10:35:08 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19357
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 10:35:08 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA11878;
	Fri, 29 Mar 2002 10:33:58 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA11844
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 10:33:48 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19271
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 10:33:44 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLV23Z>; Fri, 29 Mar 2002 10:16:57 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C64@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'Fred Baker'" <fred@cisco.com>,
        Alistair Munro
	 <alistair.munro@degree2.com>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Preemption
Date: Fri, 29 Mar 2002 10:17:53 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D734.E5906000"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D734.E5906000
Content-Type: text/plain;
	charset="iso-8859-1"

Fred,

Some of these are already in my "to do" list (addressing the existing
documents) and I will add the others to that list, to the extent I can (for
instance, I know how GETS "does it", but I don't know how AutoVon "does
it").

Janet

-----Original Message-----
From: Fred Baker [mailto:fred@cisco.com]
Sent: Friday, March 29, 2002 12:02 AM
To: Gunn, Janet NE; Alistair Munro
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption


At 08:44 AM 3/27/2002, Alistair Munro wrote:
>I'm not sure if this thread is in scope - it's very useful but I can't 
>match it to the charter.

I would deem it in scope. We are supposed to make recommendations 
concerning "best practices" (by which we mean "the best we can think of", 
not "the best currently employed"), and set up requirements for any 
protocols that we think may need to be significantly changed from current 
definition or practice. This applies both to telephony and data traffic 
used in an emergency setting.

Janet, it sounds to me like you have a pretty good handle on some of the 
issues here.

(a)Do you feel that they are adequately covered in the framework draft? if 
not, could you propose language to go into it covering the issues?
(b)Do you see specific requirements not being met in industry standard 
protocols? If so, do you feel that either Hal Folts white paper or my draft 
cover the issues, and if so specifically which sections? In not, would you 
be willing to suggest text to cover those issues?

At 11:31 AM 3/27/2002, Gunn, Janet NE wrote:
>But I do think it is useful to identify the FUNCTIONALITY that is 
>available in the Telephony domain, that is (or may be) used by Emergengy 
>Preparedness Systems, and to explore what kind of protocol support would 
>be needed  to provide equivalent functionality in the Internet (or an 
>internet).

Yes. Specifically an example of this has been requested by the Area 
Director - we have been asked to provide requirements for the 
implementation of MLPP in SIP. I would expect that this would include

(a) identifying what problem it solves (such as "it exists in the PSTN and 
we need to be able to carry the signalling across an Internet island, 
whether we actually implement it or not", or "research/experience 
documented at <place> demonstrates that this set of functionality is 
materially helpful in solving <specified> class of problems")
(b) identifying two or more implementation approaches (such as 
authorization us use of an overhead space in GETS and preemption of lower 
priority calls in AutoVon),
(c) identifying what is the minimum necessary to implement the 
functionality in an Internet domain, and
(d) identifying what is necessary to carry the signaling from one 
PSTN/GSM/etc domain to another PSTN/GSM/etc domain across an internet domain

It would also include identification of practices that are not helpful and 
should not be replicated in the Internet, if appropriate.

------_=_NextPart_001_01C1D734.E5906000
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2653.12">
<TITLE>RE: [Ieprep] Preemption</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>Fred,</FONT>
</P>

<P><FONT SIZE=3D2>Some of these are already in my &quot;to do&quot; =
list (addressing the existing documents) and I will add the others to =
that list, to the extent I can (for instance, I know how GETS =
&quot;does it&quot;, but I don't know how AutoVon &quot;does =
it&quot;).</FONT></P>

<P><FONT SIZE=3D2>Janet</FONT>
</P>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>
<BR><FONT SIZE=3D2>From: Fred Baker [<A =
HREF=3D"mailto:fred@cisco.com">mailto:fred@cisco.com</A>]</FONT>
<BR><FONT SIZE=3D2>Sent: Friday, March 29, 2002 12:02 AM</FONT>
<BR><FONT SIZE=3D2>To: Gunn, Janet NE; Alistair Munro</FONT>
<BR><FONT SIZE=3D2>Cc: ieprep@ietf.org</FONT>
<BR><FONT SIZE=3D2>Subject: Re: [Ieprep] Preemption</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>At 08:44 AM 3/27/2002, Alistair Munro wrote:</FONT>
<BR><FONT SIZE=3D2>&gt;I'm not sure if this thread is in scope - it's =
very useful but I can't </FONT>
<BR><FONT SIZE=3D2>&gt;match it to the charter.</FONT>
</P>

<P><FONT SIZE=3D2>I would deem it in scope. We are supposed to make =
recommendations </FONT>
<BR><FONT SIZE=3D2>concerning &quot;best practices&quot; (by which we =
mean &quot;the best we can think of&quot;, </FONT>
<BR><FONT SIZE=3D2>not &quot;the best currently employed&quot;), and =
set up requirements for any </FONT>
<BR><FONT SIZE=3D2>protocols that we think may need to be significantly =
changed from current </FONT>
<BR><FONT SIZE=3D2>definition or practice. This applies both to =
telephony and data traffic </FONT>
<BR><FONT SIZE=3D2>used in an emergency setting.</FONT>
</P>

<P><FONT SIZE=3D2>Janet, it sounds to me like you have a pretty good =
handle on some of the </FONT>
<BR><FONT SIZE=3D2>issues here.</FONT>
</P>

<P><FONT SIZE=3D2>(a)Do you feel that they are adequately covered in =
the framework draft? if </FONT>
<BR><FONT SIZE=3D2>not, could you propose language to go into it =
covering the issues?</FONT>
<BR><FONT SIZE=3D2>(b)Do you see specific requirements not being met in =
industry standard </FONT>
<BR><FONT SIZE=3D2>protocols? If so, do you feel that either Hal Folts =
white paper or my draft </FONT>
<BR><FONT SIZE=3D2>cover the issues, and if so specifically which =
sections? In not, would you </FONT>
<BR><FONT SIZE=3D2>be willing to suggest text to cover those =
issues?</FONT>
</P>

<P><FONT SIZE=3D2>At 11:31 AM 3/27/2002, Gunn, Janet NE wrote:</FONT>
<BR><FONT SIZE=3D2>&gt;But I do think it is useful to identify the =
FUNCTIONALITY that is </FONT>
<BR><FONT SIZE=3D2>&gt;available in the Telephony domain, that is (or =
may be) used by Emergengy </FONT>
<BR><FONT SIZE=3D2>&gt;Preparedness Systems, and to explore what kind =
of protocol support would </FONT>
<BR><FONT SIZE=3D2>&gt;be needed&nbsp; to provide equivalent =
functionality in the Internet (or an </FONT>
<BR><FONT SIZE=3D2>&gt;internet).</FONT>
</P>

<P><FONT SIZE=3D2>Yes. Specifically an example of this has been =
requested by the Area </FONT>
<BR><FONT SIZE=3D2>Director - we have been asked to provide =
requirements for the </FONT>
<BR><FONT SIZE=3D2>implementation of MLPP in SIP. I would expect that =
this would include</FONT>
</P>

<P><FONT SIZE=3D2>(a) identifying what problem it solves (such as =
&quot;it exists in the PSTN and </FONT>
<BR><FONT SIZE=3D2>we need to be able to carry the signalling across an =
Internet island, </FONT>
<BR><FONT SIZE=3D2>whether we actually implement it or not&quot;, or =
&quot;research/experience </FONT>
<BR><FONT SIZE=3D2>documented at &lt;place&gt; demonstrates that this =
set of functionality is </FONT>
<BR><FONT SIZE=3D2>materially helpful in solving &lt;specified&gt; =
class of problems&quot;)</FONT>
<BR><FONT SIZE=3D2>(b) identifying two or more implementation =
approaches (such as </FONT>
<BR><FONT SIZE=3D2>authorization us use of an overhead space in GETS =
and preemption of lower </FONT>
<BR><FONT SIZE=3D2>priority calls in AutoVon),</FONT>
<BR><FONT SIZE=3D2>(c) identifying what is the minimum necessary to =
implement the </FONT>
<BR><FONT SIZE=3D2>functionality in an Internet domain, and</FONT>
<BR><FONT SIZE=3D2>(d) identifying what is necessary to carry the =
signaling from one </FONT>
<BR><FONT SIZE=3D2>PSTN/GSM/etc domain to another PSTN/GSM/etc domain =
across an internet domain</FONT>
</P>

<P><FONT SIZE=3D2>It would also include identification of practices =
that are not helpful and </FONT>
<BR><FONT SIZE=3D2>should not be replicated in the Internet, if =
appropriate.</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C1D734.E5906000--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 10:48:19 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19866
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 10:48:19 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA12391;
	Fri, 29 Mar 2002 10:46:57 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA12332
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 10:46:54 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19814
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 10:46:51 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330031; Fri, 29 Mar 2002 10:46:53 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] Public Safety?
To: Fred Baker <fred@cisco.com>, "Gunn, Janet NE" 
 <Janet.Gunn@DynCorp.com>
Cc: ieprep@ietf.org
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Fri, 29 Mar 2002 10:46:53 -0500
Message-ID: <web-1330031@multicasttech.com>
In-Reply-To: <5.1.0.14.2.20020328211333.020c6c18@mira-sjcm-4.cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Thu, 28 Mar 2002 21:30:50 -0800
 Fred Baker <fred@cisco.com> wrote:
> At 02:31 PM 3/28/2002, Gunn, Janet NE wrote:
<snip>

> >If public safety is in scope, land mobile radio
> interoperability (again, 
> >specifically out of scope for GETS) is an important
> requirement.  During 
> >the recovery efforts after 9/11, the the lack of
> interoperability between 
> >the land mobile radio networks used by the various
> groups (Police, Fire, 
> >Rescue, etc) from different jurisdictions was a major
> problem.  It seems 
> >plausible that, if each of these systems had an
> interface to the Internet, 
> >the Internet could provide the interoperability between
> the systems.
> 
> I'm in the market for some words...
> 

Is the focus of this group excusively on voice ?
Wouldn't a instant messager application for emergency
intercommunication be much easier to set up than
interoperating between literally thousands of legacy radio
systems ?

Regards
Marshall Eubanks


> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 11:03:16 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA20405
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 11:03:16 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA13480;
	Fri, 29 Mar 2002 11:02:52 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA13444
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 11:02:51 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA20398
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 11:02:47 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLV2WZ>; Fri, 29 Mar 2002 11:01:16 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C66@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'Marshall Eubanks'" <tme@multicasttech.com>,
        Fred Baker
	 <fred@cisco.com>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Public Safety?
Date: Fri, 29 Mar 2002 11:02:04 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D73B.11D9C010"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D73B.11D9C010
Content-Type: text/plain;
	charset="iso-8859-1"

Yes, IM would be easier to set up.  And it would work for someone at a desk,
or in a stationary vehicle, or a passenger in a moving vehicle.  But I don't
think it would work well for the police officer driving down the road, or a
fireman in a darkened building. Voice (radio) works better in those cases.

I am not actively promoting this as an approach, just brainstorming the
known problems that MIGHT be addressed by Internet functionality.

-----Original Message-----
From: Marshall Eubanks [mailto:tme@multicasttech.com]
Sent: Friday, March 29, 2002 10:47 AM
To: Fred Baker; Gunn, Janet NE
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] Public Safety?


On Thu, 28 Mar 2002 21:30:50 -0800
 Fred Baker <fred@cisco.com> wrote:
> At 02:31 PM 3/28/2002, Gunn, Janet NE wrote:
<snip>

> >If public safety is in scope, land mobile radio
> interoperability (again, 
> >specifically out of scope for GETS) is an important
> requirement.  During 
> >the recovery efforts after 9/11, the the lack of
> interoperability between 
> >the land mobile radio networks used by the various
> groups (Police, Fire, 
> >Rescue, etc) from different jurisdictions was a major
> problem.  It seems 
> >plausible that, if each of these systems had an
> interface to the Internet, 
> >the Internet could provide the interoperability between
> the systems.
> 
> I'm in the market for some words...
> 

Is the focus of this group excusively on voice ?
Wouldn't a instant messager application for emergency
intercommunication be much easier to set up than
interoperating between literally thousands of legacy radio
systems ?

Regards
Marshall Eubanks


> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

------_=_NextPart_001_01C1D73B.11D9C010
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2653.12">
<TITLE>RE: [Ieprep] Public Safety?</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>Yes, IM would be easier to set up.&nbsp; And it would =
work for someone at a desk, or in a stationary vehicle, or a passenger =
in a moving vehicle.&nbsp; But I don't think it would work well for the =
police officer driving down the road, or a fireman in a darkened =
building. Voice (radio) works better in those cases.</FONT></P>

<P><FONT SIZE=3D2>I am not actively promoting this as an approach, just =
brainstorming the known problems that MIGHT be addressed by Internet =
functionality.</FONT></P>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>
<BR><FONT SIZE=3D2>From: Marshall Eubanks [<A =
HREF=3D"mailto:tme@multicasttech.com">mailto:tme@multicasttech.com</A>]<=
/FONT>
<BR><FONT SIZE=3D2>Sent: Friday, March 29, 2002 10:47 AM</FONT>
<BR><FONT SIZE=3D2>To: Fred Baker; Gunn, Janet NE</FONT>
<BR><FONT SIZE=3D2>Cc: ieprep@ietf.org</FONT>
<BR><FONT SIZE=3D2>Subject: Re: [Ieprep] Public Safety?</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>On Thu, 28 Mar 2002 21:30:50 -0800</FONT>
<BR><FONT SIZE=3D2>&nbsp;Fred Baker &lt;fred@cisco.com&gt; =
wrote:</FONT>
<BR><FONT SIZE=3D2>&gt; At 02:31 PM 3/28/2002, Gunn, Janet NE =
wrote:</FONT>
<BR><FONT SIZE=3D2>&lt;snip&gt;</FONT>
</P>

<P><FONT SIZE=3D2>&gt; &gt;If public safety is in scope, land mobile =
radio</FONT>
<BR><FONT SIZE=3D2>&gt; interoperability (again, </FONT>
<BR><FONT SIZE=3D2>&gt; &gt;specifically out of scope for GETS) is an =
important</FONT>
<BR><FONT SIZE=3D2>&gt; requirement.&nbsp; During </FONT>
<BR><FONT SIZE=3D2>&gt; &gt;the recovery efforts after 9/11, the the =
lack of</FONT>
<BR><FONT SIZE=3D2>&gt; interoperability between </FONT>
<BR><FONT SIZE=3D2>&gt; &gt;the land mobile radio networks used by the =
various</FONT>
<BR><FONT SIZE=3D2>&gt; groups (Police, Fire, </FONT>
<BR><FONT SIZE=3D2>&gt; &gt;Rescue, etc) from different jurisdictions =
was a major</FONT>
<BR><FONT SIZE=3D2>&gt; problem.&nbsp; It seems </FONT>
<BR><FONT SIZE=3D2>&gt; &gt;plausible that, if each of these systems =
had an</FONT>
<BR><FONT SIZE=3D2>&gt; interface to the Internet, </FONT>
<BR><FONT SIZE=3D2>&gt; &gt;the Internet could provide the =
interoperability between</FONT>
<BR><FONT SIZE=3D2>&gt; the systems.</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; I'm in the market for some words...</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
</P>

<P><FONT SIZE=3D2>Is the focus of this group excusively on voice =
?</FONT>
<BR><FONT SIZE=3D2>Wouldn't a instant messager application for =
emergency</FONT>
<BR><FONT SIZE=3D2>intercommunication be much easier to set up =
than</FONT>
<BR><FONT SIZE=3D2>interoperating between literally thousands of legacy =
radio</FONT>
<BR><FONT SIZE=3D2>systems ?</FONT>
</P>

<P><FONT SIZE=3D2>Regards</FONT>
<BR><FONT SIZE=3D2>Marshall Eubanks</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; =
_______________________________________________</FONT>
<BR><FONT SIZE=3D2>&gt; Ieprep mailing list</FONT>
<BR><FONT SIZE=3D2>&gt; Ieprep@ietf.org</FONT>
<BR><FONT SIZE=3D2>&gt; <A =
HREF=3D"https://www1.ietf.org/mailman/listinfo/ieprep" =
TARGET=3D"_blank">https://www1.ietf.org/mailman/listinfo/ieprep</A></FON=
T>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C1D73B.11D9C010--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 11:24:16 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA21009
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 11:24:16 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA14377;
	Fri, 29 Mar 2002 11:23:58 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA14341
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 11:23:56 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA20996
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 11:23:52 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330048; Fri, 29 Mar 2002 11:23:55 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] Public Safety?
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>,
        "'Marshall Eubanks'" 
 <tme@multicasttech.com>,
        Fred Baker	 <fred@cisco.com>
Cc: ieprep@ietf.org
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Fri, 29 Mar 2002 11:23:55 -0500
Message-ID: <web-1330048@multicasttech.com>
In-Reply-To: <CBED705A7FD2D311865500508B10894105D48C66@chntex02.is.dyncorp.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Fri, 29 Mar 2002 11:02:04 -0500
 "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> wrote:
> Yes, IM would be easier to set up.  And it would work for
> someone at a desk,
> or in a stationary vehicle, or a passenger in a moving
> vehicle.  But I don't
> think it would work well for the police officer driving
> down the road, or a
> fireman in a darkened building. Voice (radio) works
> better in those cases.


Yes, but I suspect that it is easier and more robust to
convert a text message to voice than to set up N (or,
shudder, N^2) gateways to interconnect legacy systems.

There is also the bandwidth / latency tradeoff to consider. 
My understanding is that, after the Kobe (Japan) earthquake,
the earthquake hardened emergency phone system worked but
was immediately saturated, and therefore didn't _function_,
while the (unhardened) Internet connections available took
up the slack. I also heard that the asynchronous nature of
these connections was crucial to their functioning, as
messages might take minutes to be delivered. (Although,
thinking back, I do not remember being told _why_ this was
so.)

Regards
Marshall Eubanks

> 
> I am not actively promoting this as an approach, just
> brainstorming the
> known problems that MIGHT be addressed by Internet
> functionality.
> 
> -----Original Message-----
> From: Marshall Eubanks [mailto:tme@multicasttech.com]
> Sent: Friday, March 29, 2002 10:47 AM
> To: Fred Baker; Gunn, Janet NE
> Cc: ieprep@ietf.org
> Subject: Re: [Ieprep] Public Safety?
> 
> 
> On Thu, 28 Mar 2002 21:30:50 -0800
>  Fred Baker <fred@cisco.com> wrote:
> > At 02:31 PM 3/28/2002, Gunn, Janet NE wrote:
> <snip>
> 
> > >If public safety is in scope, land mobile radio
> > interoperability (again, 
> > >specifically out of scope for GETS) is an important
> > requirement.  During 
> > >the recovery efforts after 9/11, the the lack of
> > interoperability between 
> > >the land mobile radio networks used by the various
> > groups (Police, Fire, 
> > >Rescue, etc) from different jurisdictions was a major
> > problem.  It seems 
> > >plausible that, if each of these systems had an
> > interface to the Internet, 
> > >the Internet could provide the interoperability
> between
> > the systems.
> > 
> > I'm in the market for some words...
> > 
> 
> Is the focus of this group excusively on voice ?
> Wouldn't a instant messager application for emergency
> intercommunication be much easier to set up than
> interoperating between literally thousands of legacy
> radio
> systems ?
> 
> Regards
> Marshall Eubanks
> 
> 
> > 
> > _______________________________________________
> > Ieprep mailing list
> > Ieprep@ietf.org
> > https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 12:11:21 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA23833
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 12:11:21 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA18223;
	Fri, 29 Mar 2002 12:11:05 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA18189
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 12:11:03 -0500 (EST)
Received: from rbhub100.chamb.disa.mil (rbhub100.chamb.disa.mil [209.22.120.23])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA23796
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 12:10:58 -0500 (EST)
Received: by rbhub100.chamb.disa.mil with Internet Mail Service (5.5.2653.19)
	id <HY30DWXM>; Fri, 29 Mar 2002 12:11:02 -0500
Message-ID: <AFBFF033E351D511999900204804EE54D0039F@rbmail104.chamb.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'Marshall Eubanks'" <tme@multicasttech.com>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Public Safety?
Date: Fri, 29 Mar 2002 12:11:00 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

I think the focus of this group is to get preferential treatments for
emergency traffic in a packet-based network during congestions. These
treatments is to be provided end-to-end (access, transport, and egress). NCS
wants to use GETS as a model of how it gets preferential treatments for
voice traffic in PSTN. NCS would like to be able to do that in a packet
world. So ETS in a packet-based network would include voice and data
(including video traffic). For example, an authorized emergency  personnel
needs to be able to successfully send/distribute video clips (of the sites
he/she surveys) to others in real time during a disaster. Again, the focus
here is to include voice and data.

I agreed that IM would be a useful application to use in a emergency
situation.

Regards,

An Nguyen


-----Original Message-----
From: Marshall Eubanks [mailto:tme@multicasttech.com]
Sent: Friday, March 29, 2002 10:47 AM
To: Fred Baker; Gunn, Janet NE
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] Public Safety?


On Thu, 28 Mar 2002 21:30:50 -0800
 Fred Baker <fred@cisco.com> wrote:
> At 02:31 PM 3/28/2002, Gunn, Janet NE wrote:
<snip>

> >If public safety is in scope, land mobile radio
> interoperability (again, 
> >specifically out of scope for GETS) is an important
> requirement.  During 
> >the recovery efforts after 9/11, the the lack of
> interoperability between 
> >the land mobile radio networks used by the various
> groups (Police, Fire, 
> >Rescue, etc) from different jurisdictions was a major
> problem.  It seems 
> >plausible that, if each of these systems had an
> interface to the Internet, 
> >the Internet could provide the interoperability between
> the systems.
> 
> I'm in the market for some words...
> 

Is the focus of this group excusively on voice ?
Wouldn't a instant messager application for emergency
intercommunication be much easier to set up than
interoperating between literally thousands of legacy radio
systems ?

Regards
Marshall Eubanks


> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 12:17:54 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA24149
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 12:17:54 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA18570;
	Fri, 29 Mar 2002 12:15:59 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA18540
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 12:15:53 -0500 (EST)
Received: from sven.kdmd.net (adsl-64-160-215-222.dsl.frsn01.pacbell.net [64.160.215.222])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id MAA24035
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 12:15:48 -0500 (EST)
Received: from GW-KDMD-Message_Server by sven.kdmd.net
	with Novell_GroupWise; Fri, 29 Mar 2002 09:15:51 -0600
Message-Id: <sca430c7.080@sven.kdmd.net>
X-Mailer: Novell GroupWise Internet Agent 5.5.2.1
Date: Fri, 29 Mar 2002 09:15:17 -0600
From: "Ken Diliberto" <Ken@kdmd.net>
To: <ieprep@ietf.org>
Subject: Re: [Ieprep] Public Safety?
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id MAA18541
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

I worked for a City government for many years.  We had a backup operations center that could handle all of Police and Fire dispatch.  The potential problem was what if the phone system was down?  How would calls get to dispatch?  We had wireless data connections available to keep the dispatch system available.  But nothing for phone.

There are many cities in this area of central California that would be isolated from the phone system being down.  Is there something we can put add to address this problem with either existing voice wireless or new data wireless?  I think municipal governments could apply for grants to obtain the necessary equipment to make this happen.

Ken

>>> Fred Baker <fred@cisco.com> 03/28/02 11:30PM >>>
At 02:31 PM 3/28/2002, Gunn, Janet NE wrote:
>If public safety is in scope, land mobile radio interoperability (again, 
>specifically out of scope for GETS) is an important requirement.  During 
>the recovery efforts after 9/11, the the lack of interoperability between 
>the land mobile radio networks used by the various groups (Police, Fire, 
>Rescue, etc) from different jurisdictions was a major problem.  It seems 
>plausible that, if each of these systems had an interface to the Internet, 
>the Internet could provide the interoperability between the systems.

I'm in the market for some words...



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 12:19:26 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA24177
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 12:19:24 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA18741;
	Fri, 29 Mar 2002 12:18:58 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA18705
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 12:18:52 -0500 (EST)
Received: from sven.kdmd.net (adsl-64-160-215-222.dsl.frsn01.pacbell.net [64.160.215.222])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id MAA24169
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 12:18:47 -0500 (EST)
Received: from GW-KDMD-Message_Server by sven.kdmd.net
	with Novell_GroupWise; Fri, 29 Mar 2002 09:18:51 -0600
Message-Id: <sca4317b.081@sven.kdmd.net>
X-Mailer: Novell GroupWise Internet Agent 5.5.2.1
Date: Fri, 29 Mar 2002 09:18:37 -0600
From: "Ken Diliberto" <Ken@kdmd.net>
To: <ieprep@ietf.org>
Subject: RE: [Ieprep] Public Safety?
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id MAA18706
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

The infrastructure already exists for public safety officers to communicate with their command centers.  How about between agencies?  A dispatcher could us IM for messaging between agencies.  They already do something like this between mobile units.

Ken

>>> "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> 03/29/02 10:02AM >>>
Yes, IM would be easier to set up.  And it would work for someone at a desk,
or in a stationary vehicle, or a passenger in a moving vehicle.  But I don't
think it would work well for the police officer driving down the road, or a
fireman in a darkened building. Voice (radio) works better in those cases.

I am not actively promoting this as an approach, just brainstorming the
known problems that MIGHT be addressed by Internet functionality.

[snip]


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 12:32:58 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA24706
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 12:32:53 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA19728;
	Fri, 29 Mar 2002 12:32:30 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA19696
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 12:32:28 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA24690
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 12:32:24 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVJA6>; Fri, 29 Mar 2002 12:26:43 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C67@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'Ken Diliberto'" <Ken@kdmd.net>, ieprep@ietf.org
Subject: RE: [Ieprep] Public Safety?
Date: Fri, 29 Mar 2002 12:27:33 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D747.02A59360"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D747.02A59360
Content-Type: text/plain;
	charset="iso-8859-1"

Ken,
As I understand it, the issue following 9/11 was when groups from other
jurisdictions arrived to help.  For instance, at the Pentagon, the local MD,
DC and VA groups had interoperable land mobile radio.  But when groups from
southern Virginia and "out of state" arrived to help, interoperability
became an issue.  

My information is at second or third hand, so I may be mistaken.

-----Original Message-----
From: Ken Diliberto [mailto:Ken@kdmd.net]
Sent: Friday, March 29, 2002 10:19 AM
To: ieprep@ietf.org
Subject: RE: [Ieprep] Public Safety?


The infrastructure already exists for public safety officers to communicate
with their command centers.  How about between agencies?  A dispatcher could
us IM for messaging between agencies.  They already do something like this
between mobile units.

Ken

>>> "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> 03/29/02 10:02AM >>>
Yes, IM would be easier to set up.  And it would work for someone at a desk,
or in a stationary vehicle, or a passenger in a moving vehicle.  But I don't
think it would work well for the police officer driving down the road, or a
fireman in a darkened building. Voice (radio) works better in those cases.

I am not actively promoting this as an approach, just brainstorming the
known problems that MIGHT be addressed by Internet functionality.

[snip]


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep

------_=_NextPart_001_01C1D747.02A59360
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2653.12">
<TITLE>RE: [Ieprep] Public Safety?</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>Ken,</FONT>
<BR><FONT SIZE=3D2>As I understand it, the issue following 9/11 was =
when groups from other jurisdictions arrived to help.&nbsp; For =
instance, at the Pentagon, the local MD, DC and VA groups had =
interoperable land mobile radio.&nbsp; But when groups from southern =
Virginia and &quot;out of state&quot; arrived to help, interoperability =
became an issue.&nbsp; </FONT></P>

<P><FONT SIZE=3D2>My information is at second or third hand, so I may =
be mistaken.</FONT>
</P>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>
<BR><FONT SIZE=3D2>From: Ken Diliberto [<A =
HREF=3D"mailto:Ken@kdmd.net">mailto:Ken@kdmd.net</A>]</FONT>
<BR><FONT SIZE=3D2>Sent: Friday, March 29, 2002 10:19 AM</FONT>
<BR><FONT SIZE=3D2>To: ieprep@ietf.org</FONT>
<BR><FONT SIZE=3D2>Subject: RE: [Ieprep] Public Safety?</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>The infrastructure already exists for public safety =
officers to communicate with their command centers.&nbsp; How about =
between agencies?&nbsp; A dispatcher could us IM for messaging between =
agencies.&nbsp; They already do something like this between mobile =
units.</FONT></P>

<P><FONT SIZE=3D2>Ken</FONT>
</P>

<P><FONT SIZE=3D2>&gt;&gt;&gt; &quot;Gunn, Janet NE&quot; =
&lt;Janet.Gunn@DynCorp.com&gt; 03/29/02 10:02AM &gt;&gt;&gt;</FONT>
<BR><FONT SIZE=3D2>Yes, IM would be easier to set up.&nbsp; And it =
would work for someone at a desk,</FONT>
<BR><FONT SIZE=3D2>or in a stationary vehicle, or a passenger in a =
moving vehicle.&nbsp; But I don't</FONT>
<BR><FONT SIZE=3D2>think it would work well for the police officer =
driving down the road, or a</FONT>
<BR><FONT SIZE=3D2>fireman in a darkened building. Voice (radio) works =
better in those cases.</FONT>
</P>

<P><FONT SIZE=3D2>I am not actively promoting this as an approach, just =
brainstorming the</FONT>
<BR><FONT SIZE=3D2>known problems that MIGHT be addressed by Internet =
functionality.</FONT>
</P>

<P><FONT SIZE=3D2>[snip]</FONT>
</P>
<BR>

<P><FONT =
SIZE=3D2>_______________________________________________</FONT>
<BR><FONT SIZE=3D2>Ieprep mailing list</FONT>
<BR><FONT SIZE=3D2>Ieprep@ietf.org</FONT>
<BR><FONT SIZE=3D2><A =
HREF=3D"https://www1.ietf.org/mailman/listinfo/ieprep" =
TARGET=3D"_blank">https://www1.ietf.org/mailman/listinfo/ieprep</A></FON=
T>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C1D747.02A59360--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 12:36:59 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA24860
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 12:36:59 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA19976;
	Fri, 29 Mar 2002 12:36:41 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA19940
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 12:36:39 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA24846
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 12:36:37 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id KAA01450
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 10:34:44 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA72CB
          for <ieprep@ietf.org>; Fri, 29 Mar 2002 10:36:37 -0700
Received: FROM agcs.com BY mknaomi ; Fri Mar 29 10:36:39 2002 -0700
Message-ID: <3CA4A625.690ABB4@agcs.com>
Date: Fri, 29 Mar 2002 10:36:37 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Alistair Munro <alistair.munro@degree2.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
References: <18901.1017266468@cs.ucl.ac.uk> <3CA324A8.CBE7ED25@degree2.com> <3CA49EC8.111E9EC4@agcs.com>
Content-Type: multipart/mixed;
 boundary="------------D690CFB83222982487550FB9"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------D690CFB83222982487550FB9
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Alistair,

I am not in operations, but I can offer is that in the USA circuit switched PSTN,
preemption is not done.

> Alistair Munro wrote:
>
> > Hi Ken,
> >
> > > > I'm not sure if this thread is in scope - it's very useful but I can't match
> > > > it to the charter.
> > >
> > > at least from my perspective, don't be shy about discussing topics that
> > > are generally related to the Charter.  its just that with specific
> > > IEPREP related drafts, we may be a bit more careful about what goes in
> > > (or out) of them.
> >
> > I wonder if anyone on the operational side of PSTN, ISDN, GSM has any comment on
> > the preemption issue? It would certainly help my understanding of practice.
> >
> > Regards,
> >
> > Alistair
> >
> > _______________________________________________
> > Ieprep mailing list
> > Ieprep@ietf.org
> > https://www1.ietf.org/mailman/listinfo/ieprep

--------------D690CFB83222982487550FB9
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------D690CFB83222982487550FB9--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 12:44:39 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA25087
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 12:44:39 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA20332;
	Fri, 29 Mar 2002 12:44:00 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA20300
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 12:43:59 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA25067
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 12:43:56 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330103; Fri, 29 Mar 2002 12:43:52 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] Public Safety?
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>,
        "'Ken Diliberto'" 
 <Ken@kdmd.net>, ieprep@ietf.org
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Fri, 29 Mar 2002 12:43:52 -0500
Message-ID: <web-1330103@multicasttech.com>
In-Reply-To: <CBED705A7FD2D311865500508B10894105D48C67@chntex02.is.dyncorp.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Fri, 29 Mar 2002 12:27:33 -0500
 "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> wrote:
> Ken,
> As I understand it, the issue following 9/11 was when
> groups from other
> jurisdictions arrived to help.  For instance, at the
> Pentagon, the local MD,
> DC and VA groups had interoperable land mobile radio.
> But when groups from
> southern Virginia and "out of state" arrived to help,
> interoperability
> became an issue.  
> 

Fairfax County (where I live) has specialized emergency
teams that go all over the world for disaster relief. 
My understanding is
that interoperability is almost always an issue, even
in the US. I could go and ask them directly if this would 
be useful.

Regards
Marshall Eubanks


> My information is at second or third hand, so I may be
> mistaken.
> 
> -----Original Message-----
> From: Ken Diliberto [mailto:Ken@kdmd.net]
> Sent: Friday, March 29, 2002 10:19 AM
> To: ieprep@ietf.org
> Subject: RE: [Ieprep] Public Safety?
> 
> 
> The infrastructure already exists for public safety
> officers to communicate
> with their command centers.  How about between agencies?
> A dispatcher could
> us IM for messaging between agencies.  They already do
> something like this
> between mobile units.
> 
> Ken
> 
> >>> "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> 03/29/02
> 10:02AM >>>
> Yes, IM would be easier to set up.  And it would work for
> someone at a desk,
> or in a stationary vehicle, or a passenger in a moving
> vehicle.  But I don't
> think it would work well for the police officer driving
> down the road, or a
> fireman in a darkened building. Voice (radio) works
> better in those cases.
> 
> I am not actively promoting this as an approach, just
> brainstorming the
> known problems that MIGHT be addressed by Internet
> functionality.
> 
> [snip]
> 
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 12:47:44 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA25195
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 12:47:43 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA20461;
	Fri, 29 Mar 2002 12:47:26 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA20429
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 12:47:25 -0500 (EST)
Received: from sven.kdmd.net (adsl-64-160-215-222.dsl.frsn01.pacbell.net [64.160.215.222])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id MAA25187
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 12:47:22 -0500 (EST)
Received: from GW-KDMD-Message_Server by sven.kdmd.net
	with Novell_GroupWise; Fri, 29 Mar 2002 09:47:23 -0600
Message-Id: <sca4382b.083@sven.kdmd.net>
X-Mailer: Novell GroupWise Internet Agent 5.5.2.1
Date: Fri, 29 Mar 2002 09:47:01 -0600
From: "Ken Diliberto" <Ken@kdmd.net>
To: <ieprep@ietf.org>
Subject: RE: [Ieprep] Public Safety?
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id MAA20430
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

I worked for the City of Fresno.  I know they have radio links to some of the other local agencies but not to those who would come to help in a major disaster.

What it boils down to in many cases is politics.  A Sheriff or Chief of Police may feel they are giving up control or political power when allowing communications with other agencies.  If this is actual or just perception, I have no idea.

I think Fresno has a few shared radio channels so the mobile units can communicate, but that's about it.  What I'm thinking is if we include something in our document stating how interagency communications should be handled, the Feds may make grant money available to the local governments to purchase and install the stuff.  That is usually good grease for the wheels.

BTW,  you probably have accurate information based on my experience here.

Ken

>>> "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> 03/29/02 11:27AM >>>
Ken,
As I understand it, the issue following 9/11 was when groups from other
jurisdictions arrived to help.  For instance, at the Pentagon, the local MD,
DC and VA groups had interoperable land mobile radio.  But when groups from
southern Virginia and "out of state" arrived to help, interoperability
became an issue.  

My information is at second or third hand, so I may be mistaken.

-----Original Message-----
From: Ken Diliberto [mailto:Ken@kdmd.net] 
Sent: Friday, March 29, 2002 10:19 AM
To: ieprep@ietf.org 
Subject: RE: [Ieprep] Public Safety?


The infrastructure already exists for public safety officers to communicate
with their command centers.  How about between agencies?  A dispatcher could
us IM for messaging between agencies.  They already do something like this
between mobile units.

Ken

>>> "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> 03/29/02 10:02AM >>>
Yes, IM would be easier to set up.  And it would work for someone at a desk,
or in a stationary vehicle, or a passenger in a moving vehicle.  But I don't
think it would work well for the police officer driving down the road, or a
fireman in a darkened building. Voice (radio) works better in those cases.

I am not actively promoting this as an approach, just brainstorming the
known problems that MIGHT be addressed by Internet functionality.

[snip]


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org 
https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 12:49:48 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA25315
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 12:49:48 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA20602;
	Fri, 29 Mar 2002 12:49:21 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA20571
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 12:49:19 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA25283
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 12:49:17 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330109; Fri, 29 Mar 2002 12:49:18 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] Public Safety?
To: "Nguyen, An" <nguyena@ncs.gov>,
        "'Marshall Eubanks'" 
 <tme@multicasttech.com>
Cc: ieprep@ietf.org
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Fri, 29 Mar 2002 12:49:18 -0500
Message-ID: <web-1330109@multicasttech.com>
In-Reply-To: <AFBFF033E351D511999900204804EE54D0039F@rbmail104.chamb.disa.mil>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Fri, 29 Mar 2002 12:11:00 -0500
 "Nguyen, An" <nguyena@ncs.gov> wrote:
> I think the focus of this group is to get preferential
> treatments for
> emergency traffic in a packet-based network during
> congestions. These
> treatments is to be provided end-to-end (access,
> transport, and egress). NCS
> wants to use GETS as a model of how it gets preferential
> treatments for
> voice traffic in PSTN. NCS would like to be able to do
> that in a packet
> world. So ETS in a packet-based network would include
> voice and data
> (including video traffic). For example, an authorized
> emergency  personnel
> needs to be able to successfully send/distribute video
> clips (of the sites
> he/she surveys) to others in real time during a disaster.
> Again, the focus
> here is to include voice and data.

I would strongly urge (and am working on an I-D for) the
use of multicast to distribute video in this situation, as
it scales well and avoids edge melt-downs.
See, e.g., my presentation to NANOG-23

http://www.nanog.org/mtg-0110/eubanks.html

Regards
Marshall Eubanks


> 
> I agreed that IM would be a useful application to use in
> a emergency
> situation.
> 
> Regards,
> 
> An Nguyen
> 
> 
> -----Original Message-----
> From: Marshall Eubanks [mailto:tme@multicasttech.com]
> Sent: Friday, March 29, 2002 10:47 AM
> To: Fred Baker; Gunn, Janet NE
> Cc: ieprep@ietf.org
> Subject: Re: [Ieprep] Public Safety?
> 
> 
> On Thu, 28 Mar 2002 21:30:50 -0800
>  Fred Baker <fred@cisco.com> wrote:
> > At 02:31 PM 3/28/2002, Gunn, Janet NE wrote:
> <snip>
> 
> > >If public safety is in scope, land mobile radio
> > interoperability (again, 
> > >specifically out of scope for GETS) is an important
> > requirement.  During 
> > >the recovery efforts after 9/11, the the lack of
> > interoperability between 
> > >the land mobile radio networks used by the various
> > groups (Police, Fire, 
> > >Rescue, etc) from different jurisdictions was a major
> > problem.  It seems 
> > >plausible that, if each of these systems had an
> > interface to the Internet, 
> > >the Internet could provide the interoperability
> between
> > the systems.
> > 
> > I'm in the market for some words...
> > 
> 
> Is the focus of this group excusively on voice ?
> Wouldn't a instant messager application for emergency
> intercommunication be much easier to set up than
> interoperating between literally thousands of legacy
> radio
> systems ?
> 
> Regards
> Marshall Eubanks
> 
> 
> > 
> > _______________________________________________
> > Ieprep mailing list
> > Ieprep@ietf.org
> > https://www1.ietf.org/mailman/listinfo/ieprep
> 
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 13:19:44 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA26304
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 13:19:39 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA22038;
	Fri, 29 Mar 2002 13:19:00 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA22004
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 13:18:58 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA26286
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 13:18:56 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVJFT>; Fri, 29 Mar 2002 13:17:05 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C68@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'Marshall Eubanks'" <tme@multicasttech.com>,
        "Nguyen, An"
	 <nguyena@ncs.gov>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] multicast
Date: Fri, 29 Mar 2002 13:18:01 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D74E.0FFF3910"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D74E.0FFF3910
Content-Type: text/plain;
	charset="iso-8859-1"

What, if any, priority mechanism would be appropriate with multicast?

-----Original Message-----
From: Marshall Eubanks [mailto:tme@multicasttech.com]
Sent: Friday, March 29, 2002 12:49 PM
To: Nguyen, An; 'Marshall Eubanks'
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] Public Safety?


I would strongly urge (and am working on an I-D for) the
use of multicast to distribute video in this situation, as
it scales well and avoids edge melt-downs.
See, e.g., my presentation to NANOG-23

http://www.nanog.org/mtg-0110/eubanks.html

Regards
Marshall Eubanks


------_=_NextPart_001_01C1D74E.0FFF3910
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 5.5.2653.12">
<TITLE>RE: [Ieprep] multicast</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=2>What, if any, priority mechanism would be appropriate with multicast?</FONT>
</P>

<P><FONT SIZE=2>-----Original Message-----</FONT>
<BR><FONT SIZE=2>From: Marshall Eubanks [<A HREF="mailto:tme@multicasttech.com">mailto:tme@multicasttech.com</A>]</FONT>
<BR><FONT SIZE=2>Sent: Friday, March 29, 2002 12:49 PM</FONT>
<BR><FONT SIZE=2>To: Nguyen, An; 'Marshall Eubanks'</FONT>
<BR><FONT SIZE=2>Cc: ieprep@ietf.org</FONT>
<BR><FONT SIZE=2>Subject: Re: [Ieprep] Public Safety?</FONT>
</P>
<BR>

<P><FONT SIZE=2>I would strongly urge (and am working on an I-D for) the</FONT>
<BR><FONT SIZE=2>use of multicast to distribute video in this situation, as</FONT>
<BR><FONT SIZE=2>it scales well and avoids edge melt-downs.</FONT>
<BR><FONT SIZE=2>See, e.g., my presentation to NANOG-23</FONT>
</P>

<P><FONT SIZE=2><A HREF="http://www.nanog.org/mtg-0110/eubanks.html" TARGET="_blank">http://www.nanog.org/mtg-0110/eubanks.html</A></FONT>
</P>

<P><FONT SIZE=2>Regards</FONT>
<BR><FONT SIZE=2>Marshall Eubanks</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C1D74E.0FFF3910--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 13:37:57 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27389
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 13:37:57 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23393;
	Fri, 29 Mar 2002 13:37:40 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23358
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 13:37:38 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27381
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 13:37:35 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2TIb2Y22291;
	Fri, 29 Mar 2002 10:37:02 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329103141.04de8a40@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 10:31:52 -0800
To: "Marshall Eubanks" <tme@multicasttech.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Public Safety?
Cc: ieprep@ietf.org
In-Reply-To: <web-1330031@multicasttech.com>
References: <5.1.0.14.2.20020328211333.020c6c18@mira-sjcm-4.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 07:46 AM 3/29/2002, Marshall Eubanks wrote:
>Is the focus of this group excusively on voice ?

emphatically no.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 13:38:13 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27436
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 13:38:13 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23314;
	Fri, 29 Mar 2002 13:36:38 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23283
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 13:36:36 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27334
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 13:36:32 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330146; Fri, 29 Mar 2002 13:36:30 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] multicast
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>,
        "'Marshall Eubanks'" 
 <tme@multicasttech.com>,
        "Nguyen, An"	 <nguyena@ncs.gov>
Cc: ieprep@ietf.org
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Fri, 29 Mar 2002 13:36:30 -0500
Message-ID: <web-1330146@multicasttech.com>
In-Reply-To: <CBED705A7FD2D311865500508B10894105D48C68@chntex02.is.dyncorp.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Fri, 29 Mar 2002 13:18:01 -0500
 "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> wrote:
> What, if any, priority mechanism would be appropriate
> with multicast?

Hello;

The method that I intend to recommend is to have IANA set up
an emergency allocation block from the RFC 3138  
Class D address pool, and then set up ASM and SSM
suballocations in that block (there are at least 1000 /24's
available for this, if needed). All other multicast
addresses could then be filtered out if necessary in an
emergency, and these filters could be set up in advance.
(There could be pools set up in this for European
broadcasts, the USA, Japan, the UN, etc., etc.)

This, of course, means that priority is a manual
intervention scheme that would have to be triggered in the
NOC's or the equivalent, but a static, preconfigured,
allocation scheme offers certain advantages in an emergency,
akin to "tune to channel 2 in case of an actual emergency".

Regards
Marshall Eubanks

> 
> -----Original Message-----
> From: Marshall Eubanks [mailto:tme@multicasttech.com]
> Sent: Friday, March 29, 2002 12:49 PM
> To: Nguyen, An; 'Marshall Eubanks'
> Cc: ieprep@ietf.org
> Subject: Re: [Ieprep] Public Safety?
> 
> 
> I would strongly urge (and am working on an I-D for) the
> use of multicast to distribute video in this situation,
> as
> it scales well and avoids edge melt-downs.
> See, e.g., my presentation to NANOG-23
> 
> http://www.nanog.org/mtg-0110/eubanks.html
> 
> Regards
> Marshall Eubanks
> 


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 13:40:32 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27569
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 13:40:32 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23637;
	Fri, 29 Mar 2002 13:40:11 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23591
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 13:40:09 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27545
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 13:40:06 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2TIb3Y22297;
	Fri, 29 Mar 2002 10:37:03 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329103249.04d497a8@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 10:36:34 -0800
To: "Marshall Eubanks" <tme@multicasttech.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Public Safety?
Cc: ieprep@ietf.org
In-Reply-To: <web-1330048@multicasttech.com>
References: <CBED705A7FD2D311865500508B10894105D48C66@chntex02.is.dyncorp.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 08:23 AM 3/29/2002, Marshall Eubanks wrote:
>My understanding is that, after the Kobe (Japan) earthquake, the 
>earthquake hardened emergency phone system worked but was immediately 
>saturated, and therefore didn't _function_, while the (unhardened) 
>Internet connections available took up the slack.

I can't comment on Kobe; I know that there was infrastructure loss to the 
Internet as well. But from all reports, this was absolutely true on 
September 11 in New York and Washington. VoIP materially contributed both 
to providing phone service where switches had been compromised and to 
offloading the switches that remained operational, and data services (IM, 
"I am Alive" style ad hoc services on the web, information distribution via 
web, email, and so on) were used extensively.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 13:40:52 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27583
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 13:40:52 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23761;
	Fri, 29 Mar 2002 13:40:32 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23695
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 13:40:29 -0500 (EST)
Received: from rbhub101.chamb.disa.mil (rbhub101.chamb.disa.mil [209.22.120.24])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27564
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 13:40:25 -0500 (EST)
Received: by rbhub101.chamb.disa.mil with Internet Mail Service (5.5.2653.19)
	id <HYP4F7V6>; Fri, 29 Mar 2002 13:40:27 -0500
Message-ID: <AFBFF033E351D511999900204804EE54D003A1@rbmail104.chamb.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'Marshall Eubanks'" <tme@multicasttech.com>,
        "Nguyen, An"
	 <nguyena@ncs.gov>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Public Safety?
Date: Fri, 29 Mar 2002 13:40:25 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

The 9/11 event did not affect the multicast traffic much because it could be
a combination of good traffic provisioning and undersubsriptions of service
( I might be wrong here!). Oversubscription could change this equation very
fast when the next disaster happens. Let's say it does impact ... Is there a
mechanism in place to differentiate authorized ETS traffic from others in a
multicast network? 

Regards,

An
-----Original Message-----
From: Marshall Eubanks [mailto:tme@multicasttech.com]
Sent: Friday, March 29, 2002 12:49 PM
To: Nguyen, An; 'Marshall Eubanks'
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] Public Safety?


On Fri, 29 Mar 2002 12:11:00 -0500
 "Nguyen, An" <nguyena@ncs.gov> wrote:
> I think the focus of this group is to get preferential
> treatments for
> emergency traffic in a packet-based network during
> congestions. These
> treatments is to be provided end-to-end (access,
> transport, and egress). NCS
> wants to use GETS as a model of how it gets preferential
> treatments for
> voice traffic in PSTN. NCS would like to be able to do
> that in a packet
> world. So ETS in a packet-based network would include
> voice and data
> (including video traffic). For example, an authorized
> emergency  personnel
> needs to be able to successfully send/distribute video
> clips (of the sites
> he/she surveys) to others in real time during a disaster.
> Again, the focus
> here is to include voice and data.

I would strongly urge (and am working on an I-D for) the
use of multicast to distribute video in this situation, as
it scales well and avoids edge melt-downs.
See, e.g., my presentation to NANOG-23

http://www.nanog.org/mtg-0110/eubanks.html

Regards
Marshall Eubanks


> 
> I agreed that IM would be a useful application to use in
> a emergency
> situation.
> 
> Regards,
> 
> An Nguyen
> 
> 
> -----Original Message-----
> From: Marshall Eubanks [mailto:tme@multicasttech.com]
> Sent: Friday, March 29, 2002 10:47 AM
> To: Fred Baker; Gunn, Janet NE
> Cc: ieprep@ietf.org
> Subject: Re: [Ieprep] Public Safety?
> 
> 
> On Thu, 28 Mar 2002 21:30:50 -0800
>  Fred Baker <fred@cisco.com> wrote:
> > At 02:31 PM 3/28/2002, Gunn, Janet NE wrote:
> <snip>
> 
> > >If public safety is in scope, land mobile radio
> > interoperability (again, 
> > >specifically out of scope for GETS) is an important
> > requirement.  During 
> > >the recovery efforts after 9/11, the the lack of
> > interoperability between 
> > >the land mobile radio networks used by the various
> > groups (Police, Fire, 
> > >Rescue, etc) from different jurisdictions was a major
> > problem.  It seems 
> > >plausible that, if each of these systems had an
> > interface to the Internet, 
> > >the Internet could provide the interoperability
> between
> > the systems.
> > 
> > I'm in the market for some words...
> > 
> 
> Is the focus of this group excusively on voice ?
> Wouldn't a instant messager application for emergency
> intercommunication be much easier to set up than
> interoperating between literally thousands of legacy
> radio
> systems ?
> 
> Regards
> Marshall Eubanks
> 
> 
> > 
> > _______________________________________________
> > Ieprep mailing list
> > Ieprep@ietf.org
> > https://www1.ietf.org/mailman/listinfo/ieprep
> 
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 13:42:15 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27647
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 13:42:15 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23668;
	Fri, 29 Mar 2002 13:40:13 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA23596
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 13:40:09 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27551
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 13:40:06 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2TIaoY22179;
	Fri, 29 Mar 2002 10:36:50 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329101914.04d6a008@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 10:31:23 -0800
To: "Ken Diliberto" <Ken@kdmd.net>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Public Safety?
Cc: <ieprep@ietf.org>
In-Reply-To: <sca4317b.081@sven.kdmd.net>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 07:18 AM 3/29/2002, Ken Diliberto wrote:
>A dispatcher could us IM for messaging between agencies.  They already do 
>something like this between mobile units.

yes; are there specific requirements here that we need to comment on, or 
does IM work as is?

On September 11, AOL distributed Blackberry-based IM devices to critical 
people working in New York on a "just use it for now and we'll ask long 
term questions after we get the people out and this mess cleaned up" basis 
- one of many companies who did such laudable things. I don't know how many 
Blackberries were distributed, I just know that it happened. From what I'm 
told, this helped in field communications.

What I note is that there were no special accommodations made in that 
service for emergency use. It was simply used as-is by emergency workers. 
There was probably some additional delay in message forwarding due to load, 
but I haven't heard anyone say that the service was materially impaired by 
that. That being the case, unless someone can tell me about a serious 
problem that happened with it, my first cut suggests that this is a service 
that works fine as is.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 13:45:52 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27751
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 13:45:52 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA24158;
	Fri, 29 Mar 2002 13:45:35 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA24124
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 13:45:33 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27746
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 13:45:29 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVJ26>; Fri, 29 Mar 2002 13:43:56 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C69@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'Fred Baker'" <fred@cisco.com>, Ken Diliberto <Ken@kdmd.net>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Public Safety?
Date: Fri, 29 Mar 2002 13:44:48 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D751.CDD7F690"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D751.CDD7F690
Content-Type: text/plain;
	charset="iso-8859-1"

Would this be true if Blackberries were as ubiquitous as mobile phones?  I
just don't know.

-----Original Message-----
From: Fred Baker [mailto:fred@cisco.com]
Sent: Friday, March 29, 2002 1:31 PM
To: Ken Diliberto
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Public Safety?


On September 11, AOL distributed Blackberry-based IM devices to critical 
people working in New York 

...

That being the case, unless someone can tell me about a serious 
problem that happened with it, my first cut suggests that this is a service 
that works fine as is.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep

------_=_NextPart_001_01C1D751.CDD7F690
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2653.12">
<TITLE>RE: [Ieprep] Public Safety?</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>Would this be true if Blackberries were as ubiquitous =
as mobile phones?&nbsp; I just don't know.</FONT>
</P>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>
<BR><FONT SIZE=3D2>From: Fred Baker [<A =
HREF=3D"mailto:fred@cisco.com">mailto:fred@cisco.com</A>]</FONT>
<BR><FONT SIZE=3D2>Sent: Friday, March 29, 2002 1:31 PM</FONT>
<BR><FONT SIZE=3D2>To: Ken Diliberto</FONT>
<BR><FONT SIZE=3D2>Cc: ieprep@ietf.org</FONT>
<BR><FONT SIZE=3D2>Subject: RE: [Ieprep] Public Safety?</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>On September 11, AOL distributed Blackberry-based IM =
devices to critical </FONT>
<BR><FONT SIZE=3D2>people working in New York </FONT>
</P>

<P><FONT SIZE=3D2>...</FONT>
</P>

<P><FONT SIZE=3D2>That being the case, unless someone can tell me about =
a serious </FONT>
<BR><FONT SIZE=3D2>problem that happened with it, my first cut suggests =
that this is a service </FONT>
<BR><FONT SIZE=3D2>that works fine as is.</FONT>
</P>
<BR>

<P><FONT =
SIZE=3D2>_______________________________________________</FONT>
<BR><FONT SIZE=3D2>Ieprep mailing list</FONT>
<BR><FONT SIZE=3D2>Ieprep@ietf.org</FONT>
<BR><FONT SIZE=3D2><A =
HREF=3D"https://www1.ietf.org/mailman/listinfo/ieprep" =
TARGET=3D"_blank">https://www1.ietf.org/mailman/listinfo/ieprep</A></FON=
T>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C1D751.CDD7F690--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 13:51:10 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA28023
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 13:51:10 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA24102;
	Fri, 29 Mar 2002 13:45:30 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA24065
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 13:45:28 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27744
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 13:45:25 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2TIitY28443;
	Fri, 29 Mar 2002 10:44:55 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329103659.01ecf178@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 10:44:49 -0800
To: "Nguyen, An" <nguyena@ncs.gov>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Public Safety?
Cc: ieprep@ietf.org
In-Reply-To: <AFBFF033E351D511999900204804EE54D0039F@rbmail104.chamb.dis
 a.mil>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 09:11 AM 3/29/2002, Nguyen, An wrote:
>I think the focus of this group is to get preferential treatments for 
>emergency traffic in a packet-based network during congestions.

That is the focus of some contributors to the discussion. My focus is to 
make sure that the Internet is useful for emergency services, which means
  - doing a gap analysis to make sure that SS7 services for emergency-
    related telephony are mirrored appropriately in SIP and etc, and
  - making sure that appropriate services are available for data as well.

Note the use of the word "appropriate". I tend to think that solutions 
should solve problems. "Preferential treatment for emergency traffic" is a 
solution; absent a problem for it to solve, I'm pretty skeptical of the 
proposal.

I noted earlier that Instant Messaging (an elastic Internet data 
application) worked pretty well without special accommodations in a recent 
large-scale crisis. Had it collapsed under the load, that would have been a 
"problem", and I might find myself calling for solutions, which might 
include preferential marking of messages or some such thing. But if there 
is no problem, it seems to me that we don't need to worry about solving it. 
We might instead note that it worked well and should be on the list of 
things emergency workers might consider using.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 13:55:41 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA28145
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 13:55:41 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA24465;
	Fri, 29 Mar 2002 13:55:11 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA24431
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 13:55:09 -0500 (EST)
Received: from rbhub100.chamb.disa.mil (rbhub100.chamb.disa.mil [209.22.120.23])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA28132
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 13:55:05 -0500 (EST)
Received: by rbhub100.chamb.disa.mil with Internet Mail Service (5.5.2653.19)
	id <HY30DZ3N>; Fri, 29 Mar 2002 13:55:07 -0500
Message-ID: <AFBFF033E351D511999900204804EE54D003A2@rbmail104.chamb.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'Marshall Eubanks'" <tme@multicasttech.com>,
        "Gunn, Janet NE"
	 <Janet.Gunn@DynCorp.com>,
        "Nguyen, An" <nguyena@ncs.gov>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] multicast
Date: Fri, 29 Mar 2002 13:55:06 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

Setting aside a block of class D addresses for emergency multicast traffic
is needed. You have my vote on it.

Regards,

An Nguyen


-----Original Message-----
From: Marshall Eubanks [mailto:tme@multicasttech.com]
Sent: Friday, March 29, 2002 1:37 PM
To: Gunn, Janet NE; 'Marshall Eubanks'; Nguyen, An
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] multicast


On Fri, 29 Mar 2002 13:18:01 -0500
 "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> wrote:
> What, if any, priority mechanism would be appropriate
> with multicast?

Hello;

The method that I intend to recommend is to have IANA set up
an emergency allocation block from the RFC 3138  
Class D address pool, and then set up ASM and SSM
suballocations in that block (there are at least 1000 /24's
available for this, if needed). All other multicast
addresses could then be filtered out if necessary in an
emergency, and these filters could be set up in advance.
(There could be pools set up in this for European
broadcasts, the USA, Japan, the UN, etc., etc.)

This, of course, means that priority is a manual
intervention scheme that would have to be triggered in the
NOC's or the equivalent, but a static, preconfigured,
allocation scheme offers certain advantages in an emergency,
akin to "tune to channel 2 in case of an actual emergency".

Regards
Marshall Eubanks

> 
> -----Original Message-----
> From: Marshall Eubanks [mailto:tme@multicasttech.com]
> Sent: Friday, March 29, 2002 12:49 PM
> To: Nguyen, An; 'Marshall Eubanks'
> Cc: ieprep@ietf.org
> Subject: Re: [Ieprep] Public Safety?
> 
> 
> I would strongly urge (and am working on an I-D for) the
> use of multicast to distribute video in this situation,
> as
> it scales well and avoids edge melt-downs.
> See, e.g., my presentation to NANOG-23
> 
> http://www.nanog.org/mtg-0110/eubanks.html
> 
> Regards
> Marshall Eubanks
> 

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 14:09:54 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28898
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 14:09:54 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA25837;
	Fri, 29 Mar 2002 14:09:36 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA25802
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 14:09:35 -0500 (EST)
Received: from sven.kdmd.net (adsl-64-160-215-222.dsl.frsn01.pacbell.net [64.160.215.222])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id OAA28860
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 14:09:31 -0500 (EST)
Received: from GW-KDMD-Message_Server by sven.kdmd.net
	with Novell_GroupWise; Fri, 29 Mar 2002 11:09:33 -0600
Message-Id: <sca44b6d.085@sven.kdmd.net>
X-Mailer: Novell GroupWise Internet Agent 5.5.2.1
Date: Fri, 29 Mar 2002 11:09:16 -0600
From: "Ken Diliberto" <Ken@kdmd.net>
To: <ieprep@ietf.org>
Subject: RE: [Ieprep] Public Safety?
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by optimus.ietf.org id OAA25803
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

I think the only requirements would be along the lines of security and guaranteed delivery.  These could be simple suggestions for future discussion or they could describe how to do it.  A Blackberry would be great if some security could be built in to protect messages from interception.  My concern is the Media, but I suppose protection from terrorists should be considered in such cases.

I'll see if I can get examples of these communications.

I can see a company like Motorola creating a new public safety radio with built-in IM capabilities.  Then adding the capability to the transmitters.

Ken

>>> Fred Baker <fred@cisco.com> 03/29/02 12:31PM >>>
At 07:18 AM 3/29/2002, Ken Diliberto wrote:
>A dispatcher could us IM for messaging between agencies.  They already do 
>something like this between mobile units.

yes; are there specific requirements here that we need to comment on, or 
does IM work as is?

On September 11, AOL distributed Blackberry-based IM devices to critical 
people working in New York on a "just use it for now and we'll ask long 
term questions after we get the people out and this mess cleaned up" basis 
- one of many companies who did such laudable things. I don't know how many 
Blackberries were distributed, I just know that it happened. From what I'm 
told, this helped in field communications.

What I note is that there were no special accommodations made in that 
service for emergency use. It was simply used as-is by emergency workers. 
There was probably some additional delay in message forwarding due to load, 
but I haven't heard anyone say that the service was materially impaired by 
that. That being the case, unless someone can tell me about a serious 
problem that happened with it, my first cut suggests that this is a service 
that works fine as is.



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 14:35:03 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA00216
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 14:35:03 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA26599;
	Fri, 29 Mar 2002 14:22:22 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA26567
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 14:22:17 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA29557
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 14:22:14 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2TJLdY28179;
	Fri, 29 Mar 2002 11:21:39 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329110155.04d877c8@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 11:12:05 -0800
To: "'Marshall Eubanks'" <tme@multicasttech.com>,
        "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] multicast
Cc: ieprep@ietf.org
In-Reply-To: <CBED705A7FD2D311865500508B10894105D48C68@chntex02.is.dynco
 rp.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 10:18 AM 3/29/2002, Gunn, Janet NE wrote:
>What, if any, priority mechanism would be appropriate with multicast?

Isn't that sort of an application question? Multicast is a way to 
distribute it - you can send one stream to each recipient unicast, or one 
stream to many recipients multicast. In the latter case, you save the 
network N-1 streams worth of overall bandwidth.

RFC 3246 describes one approach to deploying QoS for multicast; RFC 2597 
describes another. There exist proponents of each. But bottom line, to get 
video to work in a network, you have to get the video frames into the 
jitter buffer before the jitter buffer is dumped into the codec. You can do 
that unicast, and you can do the same thing multicast.

Marshall, there is a niggly operational problem here, though. The 
applicability of multicast varies by use of the application. If you are 
trying to distribute CNN-style broadcasting via the Internet, it makes a 
lot of sense to send one video stream and tell viewers to "join the video 
in progress". If you have people asking to "show me that picture of the 
building collapsing", it might make sense to display it once every few 
seconds, and multicast it to a new multicast group every few seconds - if 
there is end to end service to accomplish that. The way the media have it 
set up, though, is that when you click on the image, you download (using 
TCP) a file which you now display using RealAudio or some such. According 
to the folks I have spoken with, bandwidth has not been the issue, it has 
been server resources, and a TCP download of yet another file is a pretty 
simple thing to offload to a web cache or a download server.

So I understand the *solution*: use multicast to send the stream once 
rather than bandwidth to send it many times. But what *problem* is it 
solving for emergency workers, and does it solve it in a manner useful to them?


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 14:35:34 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA00230
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 14:35:34 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27581;
	Fri, 29 Mar 2002 14:35:02 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27547
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 14:35:00 -0500 (EST)
Received: from clifden.donelan.com (clifden.donelan.com [199.34.53.180])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA00208
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 14:34:56 -0500 (EST)
Received: from sean (helo=localhost)
	by clifden.donelan.com with local-esmtp (Exim 3.34 #2)
	id 16r290-0001Y3-00; Fri, 29 Mar 2002 14:34:38 -0500
Date: Fri, 29 Mar 2002 14:34:38 -0500 (EST)
From: Sean Donelan <sean@donelan.com>
To: "Nguyen, An" <nguyena@ncs.gov>
cc: "'Marshall Eubanks'" <tme@multicasttech.com>,
        "Gunn, Janet NE" <Janet.Gunn@dyncorp.com>, <ieprep@ietf.org>
Subject: RE: [Ieprep] multicast
In-Reply-To: <AFBFF033E351D511999900204804EE54D003A2@rbmail104.chamb.disa.mil>
Message-ID: <Pine.GSO.4.40.0203291424490.5876-100000@clifden.donelan.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org



On Fri, 29 Mar 2002, Nguyen, An wrote:

> Setting aside a block of class D addresses for emergency multicast traffic
> is needed. You have my vote on it.

Setting aside a block of Class D addresses is just a waste of space
unless someone can come up with a way to use it, secure it and
activate it during an emergency.  It may be an especially bad idea
to filter out other multicast space in an emergency.

For example, during September 11 the EBS/EAS system was never activated by
the government.  In informal conversations, the broadcast community felt
the government pre-empting the response already in progress by the
commercial news community would have been more disruptive and caused more
harm than it would have helped.  Government spokespeople didn't have very
much to say.  Please standby wouldn't have been very helpful.  When
government officials did want to make a statement, they didn't seem
to have much difficulty obtaining coverage by the news organizations.
The news media was already reporting the story across multiple outlets,
often converting non-news broadcasts to simulcast of news sources.  But
some news organizations deliberately choose not to convert some outlets,
such as the cartoon channel.  If the government had activated the EBS/EAS,
they wouldn't not have had the option to keep the cartoon channel on
the air.  It would have been automatically taken off the air by the
SAGE ENDEC boxes.

Government directed pre-emption isn't always the best way to handle an
emergency.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 14:37:49 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA00299
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 14:37:48 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27702;
	Fri, 29 Mar 2002 14:37:25 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27666
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 14:37:23 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA00293
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 14:37:19 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2TJajY10589;
	Fri, 29 Mar 2002 11:36:46 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329112508.026c44b8@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 11:25:59 -0800
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Public Safety?
Cc: ieprep@ietf.org
In-Reply-To: <CBED705A7FD2D311865500508B10894105D48C69@chntex02.is.dynco
 rp.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 10:44 AM 3/29/2002, Gunn, Janet NE wrote:
>Would this be true if Blackberries were as ubiquitous as mobile phones?  I 
>just don't know.

Any problem with leaving it until we know?


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 14:38:14 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA00326
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 14:38:14 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27657;
	Fri, 29 Mar 2002 14:37:20 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27623
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 14:37:19 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA00289
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 14:37:15 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2TJakY10593;
	Fri, 29 Mar 2002 11:36:46 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329112622.01d86768@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 11:27:21 -0800
To: "Nguyen, An" <nguyena@ncs.gov>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] multicast
Cc: ieprep@ietf.org
In-Reply-To: <AFBFF033E351D511999900204804EE54D003A2@rbmail104.chamb.dis
 a.mil>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 10:55 AM 3/29/2002, Nguyen, An wrote:
>Setting aside a block of class D addresses for emergency multicast traffic
>is needed.

you have stated a solution to a problem (a lack of class D addresses 
available to emergency services at the times that they need it). Is there 
such a problem to solve? Can you demonstrate that?


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 14:45:09 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA00558
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 14:45:08 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27901;
	Fri, 29 Mar 2002 14:43:15 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27864
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 14:43:13 -0500 (EST)
Received: from chntex01.is.dyncorp.com (chntex01.is.dyncorp.com [131.131.133.210])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA00449
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 14:43:09 -0500 (EST)
Received: by chntex01.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <HWZLVJPS>; Fri, 29 Mar 2002 14:41:37 -0500
Message-ID: <CBED705A7FD2D311865500508B10894105D48C6B@chntex02.is.dyncorp.com>
From: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
To: "'Fred Baker'" <fred@cisco.com>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Public Safety?
Date: Fri, 29 Mar 2002 14:42:33 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1D759.DEB29030"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1D759.DEB29030
Content-Type: text/plain;
	charset="iso-8859-1"

No, not really.  But we don't want to get in the habit of thinking "X worked
well for 9/11, therfore it will always work well."  

Someone pointed out that the hospital system in NYC worked well, but that
was more because of the nature of the incident: so many were beyond the need
for hospital services.  But that does not, per se, mean the the hospital
system does not need to be improved.

-----Original Message-----
From: Fred Baker [mailto:fred@cisco.com]
Sent: Friday, March 29, 2002 2:26 PM
To: Gunn, Janet NE
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Public Safety?


At 10:44 AM 3/29/2002, Gunn, Janet NE wrote:
>Would this be true if Blackberries were as ubiquitous as mobile phones?  I 
>just don't know.

Any problem with leaving it until we know?

------_=_NextPart_001_01C1D759.DEB29030
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2653.12">
<TITLE>RE: [Ieprep] Public Safety?</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>No, not really.&nbsp; But we don't want to get in the =
habit of thinking &quot;X worked well for 9/11, therfore it will always =
work well.&quot;&nbsp; </FONT></P>

<P><FONT SIZE=3D2>Someone pointed out that the hospital system in NYC =
worked well, but that was more because of the nature of the incident: =
so many were beyond the need for hospital services.&nbsp; But that does =
not, per se, mean the the hospital system does not need to be =
improved.</FONT></P>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>
<BR><FONT SIZE=3D2>From: Fred Baker [<A =
HREF=3D"mailto:fred@cisco.com">mailto:fred@cisco.com</A>]</FONT>
<BR><FONT SIZE=3D2>Sent: Friday, March 29, 2002 2:26 PM</FONT>
<BR><FONT SIZE=3D2>To: Gunn, Janet NE</FONT>
<BR><FONT SIZE=3D2>Cc: ieprep@ietf.org</FONT>
<BR><FONT SIZE=3D2>Subject: RE: [Ieprep] Public Safety?</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>At 10:44 AM 3/29/2002, Gunn, Janet NE wrote:</FONT>
<BR><FONT SIZE=3D2>&gt;Would this be true if Blackberries were as =
ubiquitous as mobile phones?&nbsp; I </FONT>
<BR><FONT SIZE=3D2>&gt;just don't know.</FONT>
</P>

<P><FONT SIZE=3D2>Any problem with leaving it until we know?</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C1D759.DEB29030--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 16:38:36 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA04567
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 16:38:36 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA03303;
	Fri, 29 Mar 2002 16:37:54 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA03269
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 16:37:40 -0500 (EST)
Received: from bells.cs.ucl.ac.uk (bells.cs.ucl.ac.uk [128.16.5.31])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id QAA04491
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 16:37:35 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.04344-0@bells.cs.ucl.ac.uk>; Fri, 29 Mar 2002 21:37:23 +0000
To: Alistair Munro <alistair.munro@degree2.com>
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Preemption
In-reply-to: Your message of "Thu, 28 Mar 2002 14:11:53 GMT." <3CA324A8.CBE7ED25@degree2.com>
Date: Fri, 29 Mar 2002 21:37:21 +0000
Message-ID: <5849.1017437841@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


> I wonder if anyone on the operational side of PSTN, ISDN, GSM has any
> comment on the preemption issue? It would certainly help my
> understanding of practice.

from my experience, the preemption that has been deployed is almost
exclusive to private networks, and I'll have to pass on comments.

my understanding is that the UK, under the guidance of Oftel, has
deployed multiple priority indicators for their cellular/mobile
telephony service, but i don't believe preemption is part of that
service.  i'll look into that.

-ken



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 17:40:15 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06507
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 17:40:15 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA06895;
	Fri, 29 Mar 2002 17:39:44 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA06861
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 17:39:42 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06479
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 17:39:36 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330326; Fri, 29 Mar 2002 17:39:37 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] Public Safety?
To: "Nguyen, An" <nguyena@ncs.gov>,
        "'Marshall Eubanks'" 
 <tme@multicasttech.com>,
        "Nguyen, An"	 <nguyena@ncs.gov>
Cc: ieprep@ietf.org
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Fri, 29 Mar 2002 17:39:37 -0500
Message-ID: <web-1330326@multicasttech.com>
In-Reply-To: <AFBFF033E351D511999900204804EE54D003A1@rbmail104.chamb.disa.mil>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Fri, 29 Mar 2002 13:40:25 -0500
 "Nguyen, An" <nguyena@ncs.gov> wrote:

Dear An;

You can roughly separate the problem into 3 pieces 

first mile 
"the core"
last mile

Multicast is great on the first mile as the load never
changes, regardless of the audience. This is a great help
for huge audience spikes; they never reach the first mile.

The core was rock solid during 9/11 for every use. The
current Internet backbones tend to be way over provisioned
and are likely to stay that way in the near future. In any
case, multicast should be no more load on the core than
unicast, and quite probably less.

In the last mile, multicast is no worse than unicast (the
exact same for one video receiver on a LAN), and may be a
lot better (20 video receivers at 500 kbps will saturate a
10-Base-T LAN in unicast, but not multicast, for example). 

(There are cases, such as many legacy DSLAMs, where
multicast reverts to unicast, but, again, it should not be
worse.)

So, I cannot see a situation where multicast is
intrinsically worse than unicast, and would interested in
learning of such. Both techniques are subject to a variety
of attacks, but I do not see any intrinsic DOS superiority
either way.

> The 9/11 event did not affect the multicast traffic much
> because it could be
> a combination of good traffic provisioning and
> undersubsriptions of service
> ( I might be wrong here!). Oversubscription could change
> this equation very
> fast when the next disaster happens. Let's say it does
> impact ... Is there a

I don't see this; can you elaborate ?


> mechanism in place to differentiate authorized ETS
> traffic from others in a
> multicast network? 
> 
> Regards,
> 
> An

Regards
Marshall 


> -----Original Message-----
> From: Marshall Eubanks [mailto:tme@multicasttech.com]
> Sent: Friday, March 29, 2002 12:49 PM
> To: Nguyen, An; 'Marshall Eubanks'
> Cc: ieprep@ietf.org
> Subject: Re: [Ieprep] Public Safety?
> 
> 
> On Fri, 29 Mar 2002 12:11:00 -0500
>  "Nguyen, An" <nguyena@ncs.gov> wrote:
> > I think the focus of this group is to get preferential
> > treatments for
> > emergency traffic in a packet-based network during
> > congestions. These
> > treatments is to be provided end-to-end (access,
> > transport, and egress). NCS
> > wants to use GETS as a model of how it gets
> preferential
> > treatments for
> > voice traffic in PSTN. NCS would like to be able to do
> > that in a packet
> > world. So ETS in a packet-based network would include
> > voice and data
> > (including video traffic). For example, an authorized
> > emergency  personnel
> > needs to be able to successfully send/distribute video
> > clips (of the sites
> > he/she surveys) to others in real time during a
> disaster.
> > Again, the focus
> > here is to include voice and data.
> 
> I would strongly urge (and am working on an I-D for) the
> use of multicast to distribute video in this situation,
> as
> it scales well and avoids edge melt-downs.
> See, e.g., my presentation to NANOG-23
> 
> http://www.nanog.org/mtg-0110/eubanks.html
> 
> Regards
> Marshall Eubanks
> 
> 
> > 
> > I agreed that IM would be a useful application to use
> in
> > a emergency
> > situation.
> > 
> > Regards,
> > 
> > An Nguyen
> > 
> > 
> > -----Original Message-----
> > From: Marshall Eubanks [mailto:tme@multicasttech.com]
> > Sent: Friday, March 29, 2002 10:47 AM
> > To: Fred Baker; Gunn, Janet NE
> > Cc: ieprep@ietf.org
> > Subject: Re: [Ieprep] Public Safety?
> > 
> > 
> > On Thu, 28 Mar 2002 21:30:50 -0800
> >  Fred Baker <fred@cisco.com> wrote:
> > > At 02:31 PM 3/28/2002, Gunn, Janet NE wrote:
> > <snip>
> > 
> > > >If public safety is in scope, land mobile radio
> > > interoperability (again, 
> > > >specifically out of scope for GETS) is an important
> > > requirement.  During 
> > > >the recovery efforts after 9/11, the the lack of
> > > interoperability between 
> > > >the land mobile radio networks used by the various
> > > groups (Police, Fire, 
> > > >Rescue, etc) from different jurisdictions was a
> major
> > > problem.  It seems 
> > > >plausible that, if each of these systems had an
> > > interface to the Internet, 
> > > >the Internet could provide the interoperability
> > between
> > > the systems.
> > > 
> > > I'm in the market for some words...
> > > 
> > 
> > Is the focus of this group excusively on voice ?
> > Wouldn't a instant messager application for emergency
> > intercommunication be much easier to set up than
> > interoperating between literally thousands of legacy
> > radio
> > systems ?
> > 
> > Regards
> > Marshall Eubanks
> > 
> > 
> > > 
> > > _______________________________________________
> > > Ieprep mailing list
> > > Ieprep@ietf.org
> > > https://www1.ietf.org/mailman/listinfo/ieprep
> > 
> > 
> > _______________________________________________
> > Ieprep mailing list
> > Ieprep@ietf.org
> > https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 17:42:30 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06596
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 17:42:30 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA06677;
	Fri, 29 Mar 2002 17:33:29 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA06652
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 17:33:27 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06262
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 17:33:22 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2TMWnY05602;
	Fri, 29 Mar 2002 14:32:49 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329142415.01e9d3e0@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 14:25:06 -0800
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
From: Fred Baker <fred@cisco.com>
Subject: RE: [Ieprep] Public Safety?
Cc: ieprep@ietf.org
In-Reply-To: <CBED705A7FD2D311865500508B10894105D48C6B@chntex02.is.dynco
 rp.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 11:42 AM 3/29/2002, Gunn, Janet NE wrote:
>we don't want to get in the habit of thinking "X worked well for 9/11, 
>therfore it will always work well."
>
>Someone pointed out that the hospital system in NYC worked well, but that 
>was more because of the nature of the incident: so many were beyond the 
>need for hospital services.  But that does not, per se, mean the the 
>hospital system does not need to be improved.

agreed. On the other hand, I'm not into solving problems that aren't there. 
I'd like to focus first on problems that are demonstrable, and then ask 
ourselves what else we need to think about.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 18:20:24 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07310
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 18:20:23 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA08893;
	Fri, 29 Mar 2002 18:16:19 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA08860
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 18:16:17 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07183
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 18:16:11 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id QAA24328
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 16:14:21 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA2DE7
          for <ieprep@ietf.org>; Fri, 29 Mar 2002 16:16:13 -0700
Received: FROM agcs.com BY mknaomi ; Fri Mar 29 16:16:14 2002 -0700
Message-ID: <3CA4F5BD.FA2A7486@agcs.com>
Date: Fri, 29 Mar 2002 16:16:13 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "Nguyen, An" <nguyena@ncs.gov>
CC: "'Marshall Eubanks'" <tme@multicasttech.com>,
        "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>, ieprep@ietf.org
Subject: Re: [Ieprep] multicast
References: <AFBFF033E351D511999900204804EE54D003A2@rbmail104.chamb.disa.mil>
Content-Type: multipart/mixed;
 boundary="------------3A1758D2CF95EB3552368DFC"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------3A1758D2CF95EB3552368DFC
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

How large should such a block be?

"Nguyen, An" wrote:

> Setting aside a block of class D addresses for emergency multicast traffic
> is needed. You have my vote on it.
>
> Regards,
>
> An Nguyen
>
> -----Original Message-----
> From: Marshall Eubanks [mailto:tme@multicasttech.com]
> Sent: Friday, March 29, 2002 1:37 PM
> To: Gunn, Janet NE; 'Marshall Eubanks'; Nguyen, An
> Cc: ieprep@ietf.org
> Subject: Re: [Ieprep] multicast
>
> On Fri, 29 Mar 2002 13:18:01 -0500
>  "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> wrote:
> > What, if any, priority mechanism would be appropriate
> > with multicast?
>
> Hello;
>
> The method that I intend to recommend is to have IANA set up
> an emergency allocation block from the RFC 3138
> Class D address pool, and then set up ASM and SSM
> suballocations in that block (there are at least 1000 /24's
> available for this, if needed). All other multicast
> addresses could then be filtered out if necessary in an
> emergency, and these filters could be set up in advance.
> (There could be pools set up in this for European
> broadcasts, the USA, Japan, the UN, etc., etc.)
>
> This, of course, means that priority is a manual
> intervention scheme that would have to be triggered in the
> NOC's or the equivalent, but a static, preconfigured,
> allocation scheme offers certain advantages in an emergency,
> akin to "tune to channel 2 in case of an actual emergency".
>
> Regards
> Marshall Eubanks
>
> >
> > -----Original Message-----
> > From: Marshall Eubanks [mailto:tme@multicasttech.com]
> > Sent: Friday, March 29, 2002 12:49 PM
> > To: Nguyen, An; 'Marshall Eubanks'
> > Cc: ieprep@ietf.org
> > Subject: Re: [Ieprep] Public Safety?
> >
> >
> > I would strongly urge (and am working on an I-D for) the
> > use of multicast to distribute video in this situation,
> > as
> > it scales well and avoids edge melt-downs.
> > See, e.g., my presentation to NANOG-23
> >
> > http://www.nanog.org/mtg-0110/eubanks.html
> >
> > Regards
> > Marshall Eubanks
> >
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

--------------3A1758D2CF95EB3552368DFC
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------3A1758D2CF95EB3552368DFC--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 18:29:23 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07514
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 18:29:22 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA09520;
	Fri, 29 Mar 2002 18:29:10 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA09462
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 18:29:08 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07511
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 18:28:59 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330352; Fri, 29 Mar 2002 18:28:58 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] Public Safety?
To: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>,
        "'Fred Baker'" 
 <fred@cisco.com>, Ken Diliberto <Ken@kdmd.net>
Cc: ieprep@ietf.org
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Fri, 29 Mar 2002 18:28:58 -0500
Message-ID: <web-1330352@multicasttech.com>
In-Reply-To: <CBED705A7FD2D311865500508B10894105D48C69@chntex02.is.dyncorp.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Fri, 29 Mar 2002 13:44:48 -0500
 "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> wrote:
> Would this be true if Blackberries were as ubiquitous as
> mobile phones?  I
> just don't know.

Dear Fred;

The problems are

1.) There are _six_ different formats with some market
share.

2.) Little wars break out to deny or gain interoperability,
particularly between two companies in Redmond, Washington,
and Ashburn, Virginia. 

There are programs that will talk to all 6 services, they
are also part of the technological arms race mentioned in 
# 2.

So, this may not be as trivial as it appears on the
surface...

Regards
Marshall Eubanks

> 
> -----Original Message-----
> From: Fred Baker [mailto:fred@cisco.com]
> Sent: Friday, March 29, 2002 1:31 PM
> To: Ken Diliberto
> Cc: ieprep@ietf.org
> Subject: RE: [Ieprep] Public Safety?
> 
> 
> On September 11, AOL distributed Blackberry-based IM
> devices to critical 
> people working in New York 
> 
> ...
> 
> That being the case, unless someone can tell me about a
> serious 
> problem that happened with it, my first cut suggests that
> this is a service 
> that works fine as is.
> 
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 18:32:10 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07611
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 18:32:10 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA10096;
	Fri, 29 Mar 2002 18:31:12 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA10066
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 18:31:10 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07588
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 18:31:04 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330349; Fri, 29 Mar 2002 18:31:07 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] multicast
To: "Nguyen, An" <nguyena@ncs.gov>,
        "'Marshall Eubanks'" 
 <tme@multicasttech.com>,
        "Gunn, Janet NE"	 
 <Janet.Gunn@DynCorp.com>,
        "Nguyen, An" <nguyena@ncs.gov>
Cc: ieprep@ietf.org
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Fri, 29 Mar 2002 18:31:07 -0500
Message-ID: <web-1330349@multicasttech.com>
In-Reply-To: <AFBFF033E351D511999900204804EE54D003A2@rbmail104.chamb.disa.mil>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Fri, 29 Mar 2002 13:55:06 -0500
 "Nguyen, An" <nguyena@ncs.gov> wrote:
> Setting aside a block of class D addresses for emergency
> multicast traffic
> is needed. You have my vote on it.
> 
> Regards,
> 
> An Nguyen

Well, we have talked to RIPE and IANA about doing this.
The mechanisms are there, and I am sure that we
could get through a proposal from this WG.

Regards
Marshall 

> 
> 
> -----Original Message-----
> From: Marshall Eubanks [mailto:tme@multicasttech.com]
> Sent: Friday, March 29, 2002 1:37 PM
> To: Gunn, Janet NE; 'Marshall Eubanks'; Nguyen, An
> Cc: ieprep@ietf.org
> Subject: Re: [Ieprep] multicast
> 
> 
> On Fri, 29 Mar 2002 13:18:01 -0500
>  "Gunn, Janet NE" <Janet.Gunn@DynCorp.com> wrote:
> > What, if any, priority mechanism would be appropriate
> > with multicast?
> 
> Hello;
> 
> The method that I intend to recommend is to have IANA set
> up
> an emergency allocation block from the RFC 3138  
> Class D address pool, and then set up ASM and SSM
> suballocations in that block (there are at least 1000
> /24's
> available for this, if needed). All other multicast
> addresses could then be filtered out if necessary in an
> emergency, and these filters could be set up in advance.
> (There could be pools set up in this for European
> broadcasts, the USA, Japan, the UN, etc., etc.)
> 
> This, of course, means that priority is a manual
> intervention scheme that would have to be triggered in
> the
> NOC's or the equivalent, but a static, preconfigured,
> allocation scheme offers certain advantages in an
> emergency,
> akin to "tune to channel 2 in case of an actual
> emergency".
> 
> Regards
> Marshall Eubanks
> 
> > 
> > -----Original Message-----
> > From: Marshall Eubanks [mailto:tme@multicasttech.com]
> > Sent: Friday, March 29, 2002 12:49 PM
> > To: Nguyen, An; 'Marshall Eubanks'
> > Cc: ieprep@ietf.org
> > Subject: Re: [Ieprep] Public Safety?
> > 
> > 
> > I would strongly urge (and am working on an I-D for)
> the
> > use of multicast to distribute video in this situation,
> > as
> > it scales well and avoids edge melt-downs.
> > See, e.g., my presentation to NANOG-23
> > 
> > http://www.nanog.org/mtg-0110/eubanks.html
> > 
> > Regards
> > Marshall Eubanks
> > 
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 19:14:20 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA08109
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 19:14:20 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA11744;
	Fri, 29 Mar 2002 19:13:57 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA11700
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 19:13:55 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA08104
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 19:13:53 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330368; Fri, 29 Mar 2002 19:13:55 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] multicast
To: Fred Baker <fred@cisco.com>,
        "'Marshall Eubanks'" 
 <tme@multicasttech.com>,
        "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>
Cc: ieprep@ietf.org
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Fri, 29 Mar 2002 19:13:55 -0500
Message-ID: <web-1330368@multicasttech.com>
In-Reply-To: <5.1.0.14.2.20020329110155.04d877c8@mira-sjcm-4.cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Fri, 29 Mar 2002 11:12:05 -0800
 Fred Baker <fred@cisco.com> wrote:
> At 10:18 AM 3/29/2002, Gunn, Janet NE wrote:
> >What, if any, priority mechanism would be appropriate
> with multicast?
> 
> Isn't that sort of an application question? Multicast is
> a way to 
> distribute it - you can send one stream to each recipient
> unicast, or one 
> stream to many recipients multicast. In the latter case,
> you save the 
> network N-1 streams worth of overall bandwidth.
> 
> RFC 3246 describes one approach to deploying QoS for
> multicast; RFC 2597 
> describes another. There exist proponents of each. But
> bottom line, to get 
> video to work in a network, you have to get the video
> frames into the 
> jitter buffer before the jitter buffer is dumped into the
> codec. You can do 
> that unicast, and you can do the same thing multicast.
> 
> Marshall, there is a niggly operational problem here,
> though. The 
> applicability of multicast varies by use of the
> application. If you are 
> trying to distribute CNN-style broadcasting via the
> Internet, it makes a 
> lot of sense to send one video stream and tell viewers to
> "join the video 
> in progress". If you have people asking to "show me that
> picture of the 
> building collapsing", it might make sense to display it
> once every few 
> seconds, and multicast it to a new multicast group every
> few seconds - if 
> there is end to end service to accomplish that. The way
> the media have it 
> set up, though, is that when you click on the image, you
> download (using 
> TCP) a file which you now display using RealAudio or some
> such. According 
> to the folks I have spoken with, bandwidth has not been
> the issue, it has 
> been server resources, and a TCP download of yet another
> file is a pretty 
> simple thing to offload to a web cache or a download
> server.
> 

Dear Fred;

The trouble with that on 9/11 (in the specific cases that I
am aware of, such as the Washington Post site, which was
both Akamai'zed and totally useless, from my direct
experience), was that the source became overloaded and the
data did not flow to the cache.

In fact, on the morning and early afternoon of 9/11, the
only news sites that I could get _that had any news_, were
in Europe. My friends in NYC had the same experience, and I
received emails from there requesting news from people whose
cable TV was down and who could not get news from the
Internet.

Regards
Marshall

> So I understand the *solution*: use multicast to send the
> stream once 
> rather than bandwidth to send it many times. But what
> *problem* is it 
> solving for emergency workers, and does it solve it in a
> manner useful to them?
> 


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 19:15:57 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA08130
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 19:15:56 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA11863;
	Fri, 29 Mar 2002 19:15:42 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA11825
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 19:15:40 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA08125
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 19:15:33 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2U0F3Y06751;
	Fri, 29 Mar 2002 16:15:04 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329155930.0247fde0@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 16:06:21 -0800
To: "Marshall Eubanks" <tme@multicasttech.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Public Safety?
Cc: ieprep@ietf.org
In-Reply-To: <web-1330352@multicasttech.com>
References: <CBED705A7FD2D311865500508B10894105D48C69@chntex02.is.dyncorp.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 03:28 PM 3/29/2002, Marshall Eubanks wrote:
>1.) There are _six_ different formats with some market
>share.
>
>2.) Little wars break out to deny or gain interoperability,

Understood. That is the subject of 
http://www.ietf.org/html.charters/impp-charter.html, where the little wars 
happen in a teakettle, which in turns whistles continuously. It's quite 
noisy, as you may guess.

The question before *this* house is whether there is something specific 
that is required for IM to be useful to emergency workers - a specific 
configuration, a specific diffserv service specified on the packets (and 
please demonstrate that there is a problem that this solves), or whatever. 
I'm right with you in voting for interoperability. That's not the issue 
before this working group, except to add one set of votes for the actors in 
the IM drama to stop their silly lets-be-anything-but-interoperable games.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 19:16:03 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA08142
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 19:16:03 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA11814;
	Fri, 29 Mar 2002 19:15:37 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id TAA11782
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 19:15:36 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA08126
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 19:15:34 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2U0F4Y06768;
	Fri, 29 Mar 2002 16:15:05 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329160649.02485068@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 16:07:51 -0800
To: "Marshall Eubanks" <tme@multicasttech.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] multicast
Cc: ieprep@ietf.org
In-Reply-To: <web-1330349@multicasttech.com>
References: <AFBFF033E351D511999900204804EE54D003A2@rbmail104.chamb.disa.mil>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 03:31 PM 3/29/2002, Marshall Eubanks wrote:
>The mechanisms are there, and I am sure that we
>could get through a proposal from this WG.

Is there a shortage of multicast addresses at times when emergency workers 
need them? Can you point me to a demonstration of the fact?


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 20:06:18 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA08878
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 20:06:18 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA14367;
	Fri, 29 Mar 2002 20:03:58 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA14335
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 20:03:57 -0500 (EST)
Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA08845
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 20:03:54 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.9.3/8.9.3) with ESMTP id UAA23067;
	Fri, 29 Mar 2002 20:03:48 -0500 (EST)
Received: from cs.columbia.edu (cta.cs.columbia.edu [128.59.19.46])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.1/8.12.1) with ESMTP id g2U13lPm009553
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Fri, 29 Mar 2002 20:03:48 -0500 (EST)
Message-ID: <3CA50ED5.2E572872@cs.columbia.edu>
Date: Fri, 29 Mar 2002 20:03:17 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
Organization: Columbia University
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Fred Baker <fred@cisco.com>
CC: "Gunn, Janet NE" <Janet.Gunn@DynCorp.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Public Safety?
References: <5.1.0.14.2.20020329142415.01e9d3e0@mira-sjcm-4.cisco.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

> 
> agreed. On the other hand, I'm not into solving problems that aren't there.
> I'd like to focus first on problems that are demonstrable, and then ask
> ourselves what else we need to think about.

Having a real problem also helps with evaluating the cost-benefit
trade-off. Any priority scheme imposes costs, in terms of implementation
and testing complexity, need for strong authentication and possibility
for denial-of-service attacks both during emergencies and during regular
use. The risk is probably higher if the capability is rarely, if ever,
exercised.  Thus, this becomes a risk assessment exercise. Also, one has
to ask oneself whether spending the same amount of money on adding
resources instead of selectively dropping messages isn't a better use of
resources. [Insert repeat of QoS-vs-overprovisioning discussion here,
iteration 2457.]

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 20:19:11 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA08996
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 20:19:11 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA14847;
	Fri, 29 Mar 2002 20:14:00 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA14810
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 20:13:58 -0500 (EST)
Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA08968
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 20:13:55 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.9.3/8.9.3) with ESMTP id UAA23994;
	Fri, 29 Mar 2002 20:13:52 -0500 (EST)
Received: from cs.columbia.edu (cta.cs.columbia.edu [128.59.19.46])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.1/8.12.1) with ESMTP id g2U1DpPm009931
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Fri, 29 Mar 2002 20:13:52 -0500 (EST)
Message-ID: <3CA51131.C0D30748@cs.columbia.edu>
Date: Fri, 29 Mar 2002 20:13:21 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
Organization: Columbia University
X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Fred Baker <fred@cisco.com>
CC: Marshall Eubanks <tme@multicasttech.com>, ieprep@ietf.org
Subject: Re: [Ieprep] Public Safety?
References: <CBED705A7FD2D311865500508B10894105D48C69@chntex02.is.dyncorp.com> <5.1.0.14.2.20020329155930.0247fde0@mira-sjcm-4.cisco.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 7bit

> Understood. That is the subject of
> http://www.ietf.org/html.charters/impp-charter.html, where the little wars
> happen in a teakettle, which in turns whistles continuously. It's quite
> noisy, as you may guess.

For what it's worth, IMPP has long moved beyond the warring-camp stage.
Whether it's output (data object definitions for interoperation) is
useful at this point, given changes in the environment, is a separate
discussion.

> 
> The question before *this* house is whether there is something specific
> that is required for IM to be useful to emergency workers - a specific
> configuration, a specific diffserv service specified on the packets (and
> please demonstrate that there is a problem that this solves), or whatever.
> I'm right with you in voting for interoperability. That's not the issue
> before this working group, except to add one set of votes for the actors in
> the IM drama to stop their silly lets-be-anything-but-interoperable games.

Unfortunately (or fortunately, depending on your perspective), most of
these actors seem to have little interest in 'playing' in the IETF in
any event.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 20:29:34 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA09121
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 20:29:34 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA15177;
	Fri, 29 Mar 2002 20:28:43 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA15145
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 20:28:41 -0500 (EST)
Received: from newdev.harvard.edu (newdev.eecs.harvard.edu [140.247.60.212])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA09112
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 20:28:39 -0500 (EST)
Received: (from sob@localhost)
	by newdev.harvard.edu (8.10.2/8.10.2) id g2U1Sgp22848
	for ieprep@ietf.org; Fri, 29 Mar 2002 20:28:42 -0500 (EST)
Date: Fri, 29 Mar 2002 20:28:42 -0500 (EST)
From: Scott  Bradner <sob@harvard.edu>
Message-Id: <200203300128.g2U1Sgp22848@newdev.harvard.edu>
To: ieprep@ietf.org
Subject: Re: [Ieprep] multicast
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


Marshall sez:
> The trouble with that on 9/11 (in the specific cases that I
> am aware of, such as the Washington Post site, which was
> both Akamai'zed and totally useless, from my direct
> experience), was that the source became overloaded and the
> data did not flow to the cache.
> 
> In fact, on the morning and early afternoon of 9/11, the
> only news sites that I could get _that had any news_, were
> in Europe. My friends in NYC had the same experience, and I
> received emails from there requesting news from people whose
> cable TV was down and who could not get news from the
> Internet.

sorry - I do not see how the above statement leads one to the conclusion
that setting aside a block of multicast addresses is a useful thing to do

it would be one thing if there were significant experience with 
successful ISP-based IP-multicast applications and services to point
to, but since there are not, it seems to be quite a leap

Scott

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 20:36:45 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA09249
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 20:36:45 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA15738;
	Fri, 29 Mar 2002 20:33:24 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA15702
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 20:33:22 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA09179
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 20:33:19 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2U1WoY12721;
	Fri, 29 Mar 2002 17:32:51 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329170055.01e806a0@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 17:20:43 -0800
To: "Marshall Eubanks" <tme@multicasttech.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] multicast
Cc: ieprep@ietf.org
In-Reply-To: <web-1330368@multicasttech.com>
References: <5.1.0.14.2.20020329110155.04d877c8@mira-sjcm-4.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

At 04:13 PM 3/29/2002, Marshall Eubanks wrote:
>The trouble with that on 9/11 (in the specific cases that I am aware of, 
>such as the Washington Post site, which was both Akamai'zed and totally 
>useless, from my direct
>experience), was that the source became overloaded and the data did not 
>flow to the cache.

Well documented and well understood. That's not the point, or at least not 
my point.

If it did get to the cache - and the various services tell me that they 
have every reason to believe that they would now be able to handle what 
happened to them on that day - would the type of service being offered be 
consistent with the use of multicast? Multicast is a great solution which 
may or may not answer the need of the application that you are proposing it 
for. Remember, bandwidth was *not*the*issue*, and multicast as a solution 
solves the problems it solves by reducing the bandwidth consumed.

Plus, it's not obvious that the emergency workers were depending on CNN 
etc. They had other things on their minds. The discussion topic of ieprep 
is the needs of emergency workers and how the Internet might meet them. 
Multicast, to be relevant in this working group, needs to address a problem 
the emergency workers had, or will have in a future emergency.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Fri Mar 29 20:51:25 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA09405
	for <ieprep-archive@odin.ietf.org>; Fri, 29 Mar 2002 20:51:25 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA16342;
	Fri, 29 Mar 2002 20:50:55 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA16310
	for <ieprep@optimus.ietf.org>; Fri, 29 Mar 2002 20:50:54 -0500 (EST)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.24.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA09399
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 20:50:51 -0500 (EST)
Received: from FRED-W2K6.cisco.com ([10.25.10.242])
	by sj-msg-core-2.cisco.com (8.11.3/8.9.1) with ESMTP id g2U1oNY19387
	for <ieprep@ietf.org>; Fri, 29 Mar 2002 17:50:23 -0800 (PST)
Message-Id: <5.1.0.14.2.20020329172047.01d66930@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Fri, 29 Mar 2002 17:39:17 -0800
To: ieprep@ietf.org
From: Fred Baker <fred@cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Ieprep] a little homily
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

I feel like I'm being a broken record here, and I'm getting as tired of 
writing it as everyone else is probably getting with reading it. But we are 
here, first and foremost, to make emergency services work well on the 
Internet. The point is not to promote solutions as much as it is to 
identify problems and solve them.

Let's imagine you have a sink backed up in your house. You call the 
plumber, who comes over to look at it. S/he walks in and starts singing the 
praises of three inch lines, plastic pipe, and pointing out the fact that 
hot air rises towards your ceiling and your house could be heated more 
efficiently if there were a fan up there blowing the air down to the floor. 
Your probable reaction? "Yes, but I called you to unclog my sink. Could I 
trouble you to pull out the snake and address the problem at hand?"

As with any engineering exercise, there are a lot of pet solutions out 
there that people want to see employed - my favorites du jour include 
explicit congestion notification for TCP/SCTP QoS, and certain extensions 
to OSPF V3 for routing in certain classes of mobile ad hoc networks. But if 
you can't tell me "this happened in a past emergency or there is a good 
rationale supported by the following data and argument that it could happen 
in the next one, and that would affect the effectiveness of emergency 
services", or "there is a service in the telephone world that needs to be 
able to signal across an Internet island to support emergency services, 
whether it is an issue in the Internet or not", you haven't got a problem 
that the ieprep working group needs to solve. In such a case, I 
fundamentally don't need to hear about your favorite solution - even if I 
agree (as I do in the case of multicast) that it is a great solution which 
I would like to see more use of.

Brass tacks. We need to get down to brass tacks. Show me a problem, a 
problem that affects emergency services, or affects telephone networks of 
different technologies when they cross an Internet island and so affects 
emergency services using those telephone networks. A problem is, by 
definition, not "my pet solution has not been adopted", but "X breaks when 
bad things happen, and as a result emergency services are less effectively 
deployed and used" or "If X were adopted, emergency services would be 
enhanced in this readily demonstrable way." *Then*, lets discuss potential 
solutions, and talk about whether or not they fix the problems before the 
house. Lets adopt the solution if and only if it actually fixes the problem. 


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 31 11:33:04 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25216
	for <ieprep-archive@odin.ietf.org>; Sun, 31 Mar 2002 11:33:04 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA29847;
	Sun, 31 Mar 2002 11:32:04 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA29813
	for <ieprep@optimus.ietf.org>; Sun, 31 Mar 2002 11:32:02 -0500 (EST)
Received: from bootstrap.agcs.com (bootstrap.agcs.com [130.131.48.11])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25206
	for <ieprep@ietf.org>; Sun, 31 Mar 2002 11:31:59 -0500 (EST)
Received: from pxmail2.agcs.com (pxmail2.agcs.com [130.131.52.2])
	by bootstrap.agcs.com (Pro-8.9.3/8.9.3) with ESMTP id JAA08414
	for <ieprep@ietf.org>; Sun, 31 Mar 2002 09:30:07 -0700 (MST)
Received: from mknaomi ([130.131.166.58]) by pxmail2.agcs.com
          (Netscape Messaging Server 3.61)  with SMTP id AAA25E6
          for <ieprep@ietf.org>; Sun, 31 Mar 2002 09:32:00 -0700
Received: FROM agcs.com BY mknaomi ; Sun Mar 31 09:31:57 2002 -0700
Message-ID: <3CA73A01.CBB63A40@agcs.com>
Date: Sun, 31 Mar 2002 09:32:02 -0700
From: "Stuart (Stu) Goldman" <goldmans@agcs.com>
Reply-To: goldmans@agcs.com
Organization: AG Communication Systems
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Fred Baker <fred@cisco.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] a little homily
References: <5.1.0.14.2.20020329172047.01d66930@mira-sjcm-4.cisco.com>
Content-Type: multipart/mixed;
 boundary="------------33EEF381927F82BBDDC05024"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

This is a multi-part message in MIME format.
--------------33EEF381927F82BBDDC05024
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Fred,

Is it possible that the trouble is that we all are just asking the wrong
question? It seems to me that we are asking. "if we should work on known
problems, or if we should work on adding more tools to the toolbox available for
use in emergency situations?"

Perhaps we instead could ask how we can accomplish both.

I have heard no argument at all against fixing known problems that would
adversely affect emergency calls during a disaster. So, I think the speakers are
in full agreement on this point.

The remaining portion of the question revolves around adding additional tools to
the tool box, just in case.

I have to admit that the first picture that came to my mind when the argument
was made that "proof of a problem" is needed, was that of the debate on global
warming. By the time the proof becomes indisputable, it will simply be too late
to apply the remedy to save the planet.  (Since global warming is not within the
charter, I hope not to have started a new thread here.)

Since we can not predict what parameters the next disaster will have, the more
tools we have in the box for our disposal, the better. Not only do we have the
traditional natural disasters, but we also now have significant man-made acts of
terror that hold "unexpectedness" as a valuable component.

Having said all that, I realize that just adding tools costs money and there is
no unlimited budget here.

Perhaps a reasonable path forward would be to include investigation of new tools
that can at least show some logical basis that they might possibly be of some
help in a disaster, even if no complete, definitive proof can be offered that
guaranty they will be needed.

I would much rather error on the side of having an unused tool in the box, than
finding I needed a tool I didn't have in the event of a disaster. The stakes in
this exercise are just too big to allow such a failure.



Fred Baker wrote:

> I feel like I'm being a broken record here, and I'm getting as tired of
> writing it as everyone else is probably getting with reading it. But we are
> here, first and foremost, to make emergency services work well on the
> Internet. The point is not to promote solutions as much as it is to
> identify problems and solve them.
>
> Let's imagine you have a sink backed up in your house. You call the
> plumber, who comes over to look at it. S/he walks in and starts singing the
> praises of three inch lines, plastic pipe, and pointing out the fact that
> hot air rises towards your ceiling and your house could be heated more
> efficiently if there were a fan up there blowing the air down to the floor.
> Your probable reaction? "Yes, but I called you to unclog my sink. Could I
> trouble you to pull out the snake and address the problem at hand?"
>
> As with any engineering exercise, there are a lot of pet solutions out
> there that people want to see employed - my favorites du jour include
> explicit congestion notification for TCP/SCTP QoS, and certain extensions
> to OSPF V3 for routing in certain classes of mobile ad hoc networks. But if
> you can't tell me "this happened in a past emergency or there is a good
> rationale supported by the following data and argument that it could happen
> in the next one, and that would affect the effectiveness of emergency
> services", or "there is a service in the telephone world that needs to be
> able to signal across an Internet island to support emergency services,
> whether it is an issue in the Internet or not", you haven't got a problem
> that the ieprep working group needs to solve. In such a case, I
> fundamentally don't need to hear about your favorite solution - even if I
> agree (as I do in the case of multicast) that it is a great solution which
> I would like to see more use of.
>
> Brass tacks. We need to get down to brass tacks. Show me a problem, a
> problem that affects emergency services, or affects telephone networks of
> different technologies when they cross an Internet island and so affects
> emergency services using those telephone networks. A problem is, by
> definition, not "my pet solution has not been adopted", but "X breaks when
> bad things happen, and as a result emergency services are less effectively
> deployed and used" or "If X were adopted, emergency services would be
> enhanced in this readily demonstrable way." *Then*, lets discuss potential
> solutions, and talk about whether or not they fix the problems before the
> house. Lets adopt the solution if and only if it actually fixes the problem.
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

--------------33EEF381927F82BBDDC05024
Content-Type: text/x-vcard; charset=us-ascii;
 name="goldmans.vcf"
Content-Description: Card for Stuart (Stu) Goldman
Content-Disposition: attachment;
 filename="goldmans.vcf"
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Goldman;-Stuart
tel;cell:623 582 7136
tel;fax:623 582 7111
tel;home:602 493 8438  
tel;work:623 582 7136
x-mozilla-html:TRUE
org:<A HREF="http://www.agcs.com"><IMG SRC="http://www.agcs.com/graphics/agcslogo.gif" BORDER="0"></A> 
adr:;;P.O. Box 52179;Phoenix;AZ;85072-2179;USA
version:2.1
email;internet:goldmans@agcs.com
x-mozilla-cpt:;23888
fn:Stuart Goldman
end:vcard

--------------33EEF381927F82BBDDC05024--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 31 12:11:35 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA26557
	for <ieprep-archive@odin.ietf.org>; Sun, 31 Mar 2002 12:11:35 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA01759;
	Sun, 31 Mar 2002 12:10:39 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA01723
	for <ieprep@optimus.ietf.org>; Sun, 31 Mar 2002 12:10:37 -0500 (EST)
Received: from newdev.harvard.edu (newdev.eecs.harvard.edu [140.247.60.212])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA26530
	for <ieprep@ietf.org>; Sun, 31 Mar 2002 12:10:35 -0500 (EST)
Received: (from sob@localhost)
	by newdev.harvard.edu (8.10.2/8.10.2) id g2VHAPT26181;
	Sun, 31 Mar 2002 12:10:25 -0500 (EST)
Date: Sun, 31 Mar 2002 12:10:25 -0500 (EST)
From: Scott  Bradner <sob@harvard.edu>
Message-Id: <200203311710.g2VHAPT26181@newdev.harvard.edu>
To: goldmans@agcs.com
Subject: Re: [Ieprep] a little homily
Cc: ieprep@ietf.org
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org


> I would much rather error on the side of having an unused tool in the box, 
>  than finding I needed a tool I didn't have in the event of a disaster. 
> The stakes inthis exercise are just too big to allow such a failure.

without some understanding of problems what tools are worth spending time
on?  blowing our resources on a tool that winds up not being useful is
just the type of failure you worry about if it means that we do not
have the resources to develop the tools that would be useful.

Scott

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 31 14:25:06 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28047
	for <ieprep-archive@odin.ietf.org>; Sun, 31 Mar 2002 14:25:06 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA06754;
	Sun, 31 Mar 2002 14:24:40 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA06723
	for <ieprep@optimus.ietf.org>; Sun, 31 Mar 2002 14:24:38 -0500 (EST)
Received: from multicasttech.com (IDENT:root@lennon.multicasttech.com [63.105.122.7])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28036
	for <ieprep@ietf.org>; Sun, 31 Mar 2002 14:24:36 -0500 (EST)
Received: from [24.28.197.67] (account <marshall_eubanks@multicasttech.com>)
  by multicasttech.com (CommuniGate Pro WebUser 3.4.8)
  with HTTP id 1330710; Sun, 31 Mar 2002 14:24:31 -0500
From: "Marshall Eubanks" <tme@multicasttech.com>
Subject: Re: [Ieprep] multicast
To: Sean Donelan <sean@donelan.com>, "Nguyen, An" <nguyena@ncs.gov>
Cc: "'Marshall Eubanks'" <tme@multicasttech.com>,
        "Gunn, Janet NE" 
 <Janet.Gunn@dyncorp.com>, <ieprep@ietf.org>
X-Mailer: CommuniGate Pro Web Mailer v.3.4.8
Date: Sun, 31 Mar 2002 14:24:31 -0500
Message-ID: <web-1330710@multicasttech.com>
In-Reply-To: <Pine.GSO.4.40.0203291424490.5876-100000@clifden.donelan.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org
Content-Transfer-Encoding: 8bit

On Fri, 29 Mar 2002 14:34:38 -0500 (EST)
 Sean Donelan <sean@donelan.com> wrote:
> 
> 

Dear Sean;

I agree with you. I believe that government preemption of
regular communications channels in a crises is almost
certainly a bad idea.

I thought, however, that we were discussing protecting
emergency communications between members of emergency
services. If their communications require priority, then (in
the Internet context) filtering is one way of doing this. I
am not sure about the desirability of filtering even in this
case, however.

The general experience with crises is that any activity that
is not planned and rehearsed in advanced will be messed up
when the moment comes. By this light, I doubt that
government news channels, activated only during a crises,
would be superior to the commercial channels (which do deal
with regular crisis situations). Filtering should also be
viewed dubiously unless it is routinely practiced. 

So, IMHO, what's needed are routine means of communications
between emergency services, or between them and the public,
that are in more or less continual use, and can be hardened
for use in a crisis. 

As for what these should be, engineering experience says
that extreme events, by their nature, are very hard to
predict and simulate properly, and engineering practice says
that what worked and what didn't work in an extreme
situation should be studied seriously for a guide to what is
likely to work in future such situations. This is routinely
done in civil engineering, and should be applied here.

On 9/11 I was a direct observer of the reaction of the
Internet to a massive demand for information, as I am sure
you all were. I was also a participant, like most or all of
you, in that I was part of that demand for information.
Somewhat like a cardiologist who starts taking notes during
his own heart attack, I found myself professionally
surprised by what was going on.

I noticed, and subsequent surveys and discussions
confirmed, that what worked was 

- e-mail
- multicast

and what weren't very useful at peak were

- news related web sites
- unicast streaming sites

(At the time I was not part of instant messengering, so I
made no direct observations of that communications channel.
I would be curious to hear of reports on its use and
usability in the crisis.)

Avi Freedman of Akamai told me that the Akamai
infrastructure functioned well on 9/11, so the problems that
I and many others noticed were presumablely between sources
and caches.

I am well aware that there have been "spins" of the 9/11
Internet status in other directions; all I can say is that
some of these disagree with my direct observations.

The technology exists to set up video channels on the
Internet between any emergency services that wish it. It
seems clear to me, based on what happened on 9/11, that
multicasting should be given serious consideration for the
transport for these channels. If these are set up and put
into use now, they can be available during the next crisis.
It seems to me that this fits squarely within the charter of
the working group.

Regards
Marshall Eubanks

> On Fri, 29 Mar 2002, Nguyen, An wrote:
> 
> > Setting aside a block of class D addresses for
> emergency multicast traffic
> > is needed. You have my vote on it.
> 
> Setting aside a block of Class D addresses is just a
> waste of space
> unless someone can come up with a way to use it, secure
> it and
> activate it during an emergency.  It may be an especially
> bad idea
> to filter out other multicast space in an emergency.
> 
> For example, during September 11 the EBS/EAS system was
> never activated by
> the government.  In informal conversations, the broadcast
> community felt
> the government pre-empting the response already in
> progress by the
> commercial news community would have been more disruptive
> and caused more
> harm than it would have helped.  Government spokespeople
> didn't have very
> much to say.  Please standby wouldn't have been very
> helpful.  When
> government officials did want to make a statement, they
> didn't seem
> to have much difficulty obtaining coverage by the news
> organizations.
> The news media was already reporting the story across
> multiple outlets,
> often converting non-news broadcasts to simulcast of news
> sources.  But
> some news organizations deliberately choose not to
> convert some outlets,
> such as the cartoon channel.  If the government had
> activated the EBS/EAS,
> they wouldn't not have had the option to keep the cartoon
> channel on
> the air.  It would have been automatically taken off the
> air by the
> SAGE ENDEC boxes.
> 
> Government directed pre-emption isn't always the best way
> to handle an
> emergency.
> 


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 31 20:55:16 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA01595
	for <ieprep-archive@odin.ietf.org>; Sun, 31 Mar 2002 20:55:16 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA21523;
	Sun, 31 Mar 2002 20:52:54 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id UAA21494
	for <ieprep@optimus.ietf.org>; Sun, 31 Mar 2002 20:52:53 -0500 (EST)
Received: from mail.mazunetworks.com (mazu.mazunetworks.com [4.19.249.110])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA01563
	for <ieprep@ietf.org>; Sun, 31 Mar 2002 20:52:51 -0500 (EST)
Received: from volta.mazunetworks.com (146-115-41-110.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com [146.115.41.110])
	by mail.mazunetworks.com (8.11.2/8.11.2) with ESMTP id g311qMa27080;
	Sun, 31 Mar 2002 20:52:22 -0500
Received: from volta (dean@localhost)
	by volta.mazunetworks.com (8.11.0/8.11.0) with ESMTP id g311q5T10350;
	Sun, 31 Mar 2002 20:52:05 -0500
Message-Id: <200204010152.g311q5T10350@volta.mazunetworks.com>
To: Scott Bradner <sob@harvard.edu>
cc: goldmans@agcs.com, ieprep@ietf.org
Subject: Re: [Ieprep] a little homily 
In-Reply-To: Message from Scott Bradner <sob@harvard.edu> 
   of "Sun, 31 Mar 2002 12:10:25 EST." <200203311710.g2VHAPT26181@newdev.harvard.edu> 
Date: Sun, 31 Mar 2002 20:52:05 -0500
From: Dean Bogdanovic <dean@mazunetworks.com>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

Reading this discussion in the last week on several issues on this
list, I have to agree with Scott. Usually in situations of emergency,
communication is critical and availability is reduced. If we do not
develop optimal solution (it does not have to be the techincaly best,
compromises will be needed), tools can fail in the situations when
people are expecting them to work. In order to find that solution, lot
of time and energy will be spent, so lets try to gather information
from the past events and use that experience in best way and not to
waste resources.

--br

Dean
> 
> > I would much rather error on the side of having an unused tool in the box, 
> >  than finding I needed a tool I didn't have in the event of a disaster. 
> > The stakes inthis exercise are just too big to allow such a failure.
> 
> without some understanding of problems what tools are worth spending time
> on?  blowing our resources on a tool that winds up not being useful is
> just the type of failure you worry about if it means that we do not
> have the resources to develop the tools that would be useful.
> 
> Scott
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


From ieprep-admin@ietf.org  Sun Mar 31 21:18:02 2002
Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA01865
	for <ieprep-archive@odin.ietf.org>; Sun, 31 Mar 2002 21:18:02 -0500 (EST)
Received: from optimus.ietf.org (localhost [127.0.0.1])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA22708;
	Sun, 31 Mar 2002 21:17:40 -0500 (EST)
Received: from ietf.org (odin [132.151.1.176])
	by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA22672
	for <ieprep@optimus.ietf.org>; Sun, 31 Mar 2002 21:17:38 -0500 (EST)
Received: from mail.mazunetworks.com (mazu.mazunetworks.com [4.19.249.110])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA01858
	for <ieprep@ietf.org>; Sun, 31 Mar 2002 21:17:37 -0500 (EST)
Received: from volta.mazunetworks.com (146-115-41-110.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com [146.115.41.110])
	by mail.mazunetworks.com (8.11.2/8.11.2) with ESMTP id g312H8a27440;
	Sun, 31 Mar 2002 21:17:08 -0500
Received: from volta (dean@localhost)
	by volta.mazunetworks.com (8.11.0/8.11.0) with ESMTP id g312Gpq10410;
	Sun, 31 Mar 2002 21:16:51 -0500
Message-Id: <200204010216.g312Gpq10410@volta.mazunetworks.com>
To: "Marshall Eubanks" <tme@multicasttech.com>
cc: Sean Donelan <sean@donelan.com>, "Nguyen, An" <nguyena@ncs.gov>,
        "Gunn, Janet NE" <Janet.Gunn@dyncorp.com>, ieprep@ietf.org
Subject: Re: [Ieprep] multicast 
In-Reply-To: Message from "Marshall Eubanks" <tme@multicasttech.com> 
   of "Sun, 31 Mar 2002 14:24:31 EST." <web-1330710@multicasttech.com> 
Date: Sun, 31 Mar 2002 21:16:51 -0500
From: Dean Bogdanovic <dean@mazunetworks.com>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
X-BeenThere: ieprep@ietf.org

> (At the time I was not part of instant messengering, so I
> made no direct observations of that communications channel.
> I would be curious to hear of reports on its use and
> usability in the crisis.)

I can tell from own experience and from other people that I have heard
that IM was working well. I can not say that was the general case, but
in cases that I know, it was working well. I believe that IM could be
a good solution for emergency communications (either data or voice),
expecially if endusers can establish sessions without a middlebox and
that, from a technical perspective is possible today. It requires end
user device to be more detailed configured, but it makes it pretty
independent as long the basic infrastucture (IP connectivity) is
available.
Using SIMPLE today (I am not saying that is the best solution, but
something to start with), endusers can configure without a middlebox
and establish end to end sessions as long the IP communications is
present. The enduser device needs to have all necessary configuration
available and that device is pretty independent. 
This makes the emergency communications pretty hard to stop by DoS
attacks, because it is distributed without a central point of failure
(no addtional services are needed, but usefull if available).

> 
> Avi Freedman of Akamai told me that the Akamai
> infrastructure functioned well on 9/11, so the problems that
> I and many others noticed were presumablely between sources
> and caches.

Akamai infrastructure works well for static content, but there are
problems for dymanic content delivery or when there are lot of changes
at the source, as stated above.

> The technology exists to set up video channels on the
> Internet between any emergency services that wish it. It
> seems clear to me, based on what happened on 9/11, that
> multicasting should be given serious consideration for the
> transport for these channels. If these are set up and put
> into use now, they can be available during the next crisis.
> It seems to me that this fits squarely within the charter of
> the working group.
> 

Multicast is a nice technology and I agree that it should be
considered, but it is also stopable with good designed DoS attacks
(delivery comes from few sources, so those sources need good
protection. I do not believe that akamaizing would help, because of
dynamic nature of content delivery).

--br

Dean

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


