
From laura.liess.dt@googlemail.com  Thu Sep 13 05:27:30 2012
Return-Path: <laura.liess.dt@googlemail.com>
X-Original-To: salud@ietfa.amsl.com
Delivered-To: salud@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 47FF121F8499 for <salud@ietfa.amsl.com>; Thu, 13 Sep 2012 05:27:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.977
X-Spam-Level: 
X-Spam-Status: No, score=-2.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id s0dlL-AZB8Wj for <salud@ietfa.amsl.com>; Thu, 13 Sep 2012 05:27:27 -0700 (PDT)
Received: from mail-iy0-f172.google.com (mail-iy0-f172.google.com [209.85.210.172]) by ietfa.amsl.com (Postfix) with ESMTP id C00AF21F84B2 for <salud@ietf.org>; Thu, 13 Sep 2012 05:27:27 -0700 (PDT)
Received: by iabz21 with SMTP id z21so2393587iab.31 for <salud@ietf.org>; Thu, 13 Sep 2012 05:27:27 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:cc:content-type; bh=ZCM+JuIjBkO2Ey1IR95kCWlLUR9e4hxISgCtaqNv+nU=; b=tcTD0Ei9Vf7Dy/OPj6TFcE2fcqrhjU0YINIedkLyqY00l/E8BY6v9ruWXZrIh45CeN 052CMWiTX/N2+skmlTKqRvUHl7G1M2rQjE4btTtLWDPIvq5qVUllZrhhSEDMza9R7YUY OhwZWVD6/j7u69zCjYQKWIWfbyDgQyep29D+qrnS2YkAUO1cMOtINTgNPh2as7ooGc5b AGKnRVYsLP50o05iBuibZzyVThSceUTti7qyXTZ4Rfxq+5uAT+saK581Pdnze8rQj0Uh 0QfnfeMbfdM2HJQ/S+CHlO1nAATzN5rnFjcaus+8Z/lBmXw1/HGq/a/03x25yPhhudAe Lpqg==
MIME-Version: 1.0
Received: by 10.50.149.137 with SMTP id ua9mr2301134igb.65.1347539247348; Thu, 13 Sep 2012 05:27:27 -0700 (PDT)
Received: by 10.231.180.211 with HTTP; Thu, 13 Sep 2012 05:27:27 -0700 (PDT)
Date: Thu, 13 Sep 2012 14:27:27 +0200
Message-ID: <CACWXZj314Q7K3vg4yDc9gQHNd-qSjOKeFv8oq7=KG07yHbrFtw@mail.gmail.com>
From: Laura Liess <laura.liess.dt@googlemail.com>
To: Dale Worley <dworley@avaya.com>, Paul Kyzivat <pkyzivat@alum.mit.edu>,  Alan Johnston <alan.b.johnston@gmail.com>, Roland Jesske <R.Jesske@telekom.de>
Content-Type: text/plain; charset=ISO-8859-1
Cc: salud@ietf.org
Subject: [salud] Concern #2 and ABNF-syntax change proposal
X-BeenThere: salud@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Sip ALerting for User Devices working group discussion list <salud.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/salud>, <mailto:salud-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/salud>
List-Post: <mailto:salud@ietf.org>
List-Help: <mailto:salud-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/salud>, <mailto:salud-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 13 Sep 2012 12:27:30 -0000

Hi,

I have following questions/change proposals:

1) On how to deal with <private-name>s and the Concern #2:

To reach a compromise, I would propose to adopt Alfred's second
proposal to add a simple sequence number to reverse domain name, as
proposed in http://www.ietf.org/mail-archive/web/salud/current/msg00257.html
from June 29:

"IANA could delegate a branch of the salud namespace to a domain name
owner using a past year alone (that is related to activity of the
domain owner in this field), or a simple sequence number, as a
disambiguating component in the naming scheme."

An example would be
urn:alert:service:call-waiting:org.example.somecompany.1:abc.

Is this OK for you?

2) It seems to me that there are  small inconsistencies/ambiguities
concerning <private-name> in version 06 Section4, ABNF-syntax:

We have now:
           alert-URN        = "urn:alert:" alert-identifier
           alert-identifier = alert-category ":" alert-indication
           alert-category   = name
           alert-indication= name *(":" name)
           name = standard-name / private-name
           standard-name = label
           private-name = label "." label *[ "." label ]

     According to this definition:
         - org.example.somecompany  is a  <name>, so it can be an
alert-category, which I think we don't want.
         - In the example
urn:alert:service:call-waiting:org.example.somecompany.1:abc,
"call-waiting" and "abc" are both <standard-name>s. Also this is
correct from the syntax point of view,  we need to distinguish somehow
between  "call-waiting",  which is registered with IANA and  "a.b.c",
which is valid only inside somecompany.example.org and is not
registred with IANA.
Also in the same example "org.example.somecompany.1" is a
<private-name>.  I think  "org.example.somecompany.1" can not be used
alone, it is more a qualifier or prefix than a <name> ( it can not be
rendered).
In my opinion, the current syntax at least leads to some confusion, or
did I miss something here?

I would propose to change the ABNF-syntax so that the the meaning and
the registration process for them becomes more clear. Here is a
proposal:

           alert-identifier = alert-category ":" alert-indication
           alert-category   = standard-name/ (
reverse-domain-name-seq:private-name)
           alert-indication = (standard-name *(":" standard-name)) /
              (standard-name *(":" standard-name) 1*(":"
reverse-domain-name-seq) *(":" private-name) *(":"
reverse-subdomain-name-seq) *(":" private-name))
           standard-name =label
           private-name    = label
           reverse-domain-name-seq     = reverse-domain-name.seq-number
           reverse-domain-name= domain-name-label "."
domain-name-label *[ "." domain-name-label ]
           reverse-subdomain-name-seq     =
reverse-subdomain-name.seq-number
           reverse-subdomain-name= subdomain-name-label "."
subdomain-name-label *[ "." subdomain-name-label ]
           domain-name-label = let-dig [ *let-dig-hyp let-dig ]
           domain-name-label = let-dig [ *let-dig-hyp let-dig ]
           label            = let-dig [ *let-dig-hyp let-dig ]

The IANA registration proces is as follows:
   - <standard-name>s  must be registered with IANA.
   - <reverse-domain-name-seq>es must be registred with IANA by the
owner of the domain name obtained from <reverse-domain-name-seq> by
dropping the <seq-number> and reversing the reminder.
    - IANA delegates the registration process for <private-name>s used
behind a  <reverse-domain-name-seq> to the domain owner. When a domain
changes the owner and if the new owner intends to change the usage or
define new <private-names>, he must register a new
<reverse-domain-name-seq> with the same <reverse-domain-name> and
incremented   <seq-number>.
    - Domains owners should have an internal registration process for
<private-name>s and <reverse-subdomain-name-seq>s.
    - Domains owners may delegate the registration process for
<private-name>s used within subdomains to the subdomain owner.


The new syntax is quite long, I know, and it means a lot of changes in
the draft, but to me it is more clear this way. Maybe we can drop the
subdomain stuff and allow only one reverse-domain-name in an
alert-indication?
alert-indication = (standard-name *(":" standard-name)) /
(standard-name *(":" standard-name) 1*(":" reverse-domain-name-seq)
*(":" private-name) *)

Please let me know your opinion. I hope I am not completely wrong.


Thanks a lot
Laura

From pkyzivat@alum.mit.edu  Thu Sep 13 17:04:51 2012
Return-Path: <pkyzivat@alum.mit.edu>
X-Original-To: salud@ietfa.amsl.com
Delivered-To: salud@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0628921F8646 for <salud@ietfa.amsl.com>; Thu, 13 Sep 2012 17:04:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.428
X-Spam-Level: 
X-Spam-Status: No, score=-0.428 tagged_above=-999 required=5 tests=[AWL=0.009,  BAYES_00=-2.599, FH_RELAY_NODNS=1.451, HELO_MISMATCH_NET=0.611,  RDNS_NONE=0.1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XxEl1ZHtK0mn for <salud@ietfa.amsl.com>; Thu, 13 Sep 2012 17:04:50 -0700 (PDT)
Received: from qmta02.westchester.pa.mail.comcast.net (qmta02.westchester.pa.mail.comcast.net [IPv6:2001:558:fe14:43:76:96:62:24]) by ietfa.amsl.com (Postfix) with ESMTP id D7D9E21F8672 for <salud@ietf.org>; Thu, 13 Sep 2012 17:04:49 -0700 (PDT)
Received: from omta02.westchester.pa.mail.comcast.net ([76.96.62.19]) by qmta02.westchester.pa.mail.comcast.net with comcast id ymUG1j0050QuhwU51o4tXe; Fri, 14 Sep 2012 00:04:53 +0000
Received: from Paul-Kyzivats-MacBook-Pro.local ([50.138.229.164]) by omta02.westchester.pa.mail.comcast.net with comcast id yo4S1j00P3ZTu2S3No4Sor; Fri, 14 Sep 2012 00:04:26 +0000
Message-ID: <505274A0.4080100@alum.mit.edu>
Date: Thu, 13 Sep 2012 20:04:48 -0400
From: Paul Kyzivat <pkyzivat@alum.mit.edu>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:15.0) Gecko/20120907 Thunderbird/15.0.1
MIME-Version: 1.0
To: Laura Liess <laura.liess.dt@googlemail.com>
References: <CACWXZj314Q7K3vg4yDc9gQHNd-qSjOKeFv8oq7=KG07yHbrFtw@mail.gmail.com>
In-Reply-To: <CACWXZj314Q7K3vg4yDc9gQHNd-qSjOKeFv8oq7=KG07yHbrFtw@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: Dale Worley <dworley@avaya.com>, salud@ietf.org
Subject: Re: [salud] Concern #2 and ABNF-syntax change proposal
X-BeenThere: salud@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Sip ALerting for User Devices working group discussion list <salud.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/salud>, <mailto:salud-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/salud>
List-Post: <mailto:salud@ietf.org>
List-Help: <mailto:salud-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/salud>, <mailto:salud-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 14 Sep 2012 00:04:51 -0000

Laura,

I generally agree with where you are trying to go, but I'm not in love 
with the proposed syntax. Let me propose something else. It's more of a 
change, but I think in good ways.

alert-URN       = "urn:alert:" alert-identifier
alert-identifier= alert-category ":" alert-indication
alert-category  = label
alert-indication= standard-name / private-name
standard-name   = standard-part *(":" standard-part)
private-name    = [standard-name ":"] private-part *(":" alert-part)
alert-part      = standard-part / private-part
standard-part   = label
private-part    = label "@" dns-name ["(" version ")"]
dns-name        = label "." label *[ "." label ]
version         = DIGIT-2-9 / DIGIT-1-9 1*2DIGIT; 2...999
label           = let-dig [ *let-dig-hyp let-dig ]
let-dig-hyp     = let-dig / "-"
let-dig         = ALPHA / DIGIT
ALPHA           = %x41-5A / %x61-7A   ; A-Z / a-z
DIGIT           = %x30-39 ; 0-9
DIGIT-1-9       = %x31-39 ; 1-9
DIGIT-2-9       = %x32-39 ; 2-9

Key differences to prior attempts:

- I've switched from reverse-dns to regular dns names.
   I did this because it's more readable and I can no
   longer see any reason not to.

- I put the version number in, but made it optional,
   with values 2...999. (You can consider the default
   to be 1.) This way you only need to insert the version
   when it's not the first. (In practice this means you
   never need to put it in.)

- I was bothered by the fact that the individual parts
   that can be stripped away in looking for something
   understood were not syntactically distinguished.
   And I saw an opportunity after switching to forward
   dns, so I changed from foo.bar:baz to baz@foo.bar.

- Everything after the <private-part> is scoped within the
   domain of the prior private name. (AFAIK this isn't
   really a change, but I want to make it clear.) The
   syntax now doesn't suggest that those parts are standard.

- I allowed switching to a second dns scope within the
   same alert-indication. I just did it because I could.
   I can remove it if desired. This allows one organization
   to "extend" the private names of another organization.

- I've used a lot of "unnecessary" symbols in the syntax.
   I did it mostly for readability. I also strived to
   leave in all those symbols that are used in the text,
   so it is hopefully largely unaffected.
   If people like this it can be cleaned up to remove the
   least necessary ones.

Some examples:

alert-info:urn:service
alert-info:urn:service:recall
alert-info:urn:alert:service:foo@example.com
alert-info:urn:alert:service:foo@example.com:bar
alert-info:urn:alert:service:foo@example.com(2):baz
alert-info:urn:alert:service:recall:foo@example.com
alert-info:urn:alert:service:foo@example.com:bar:abc@atlanta.com:baz

I see one major issue with this, or any form based on a version number:

I am concerned that we can't define the version number semantics in a 
way that will satisfy Albert. As used above, it assumes that the person 
using the dns-name has a way to know how many prior owners of the domain 
name there have been. There really isn't any way to know that.

The above syntax could be adapted to putting in something other than a 
version (such as a date). It would be nice to find something with a 
default that is sufficient most of the time. But I'm not certain that is 
possible.

	Thanks,
	Paul

On 9/13/12 8:27 AM, Laura Liess wrote:
> Hi,
>
> I have following questions/change proposals:
>
> 1) On how to deal with <private-name>s and the Concern #2:
>
> To reach a compromise, I would propose to adopt Alfred's second
> proposal to add a simple sequence number to reverse domain name, as
> proposed in http://www.ietf.org/mail-archive/web/salud/current/msg00257.html
> from June 29:
>
> "IANA could delegate a branch of the salud namespace to a domain name
> owner using a past year alone (that is related to activity of the
> domain owner in this field), or a simple sequence number, as a
> disambiguating component in the naming scheme."
>
> An example would be
> urn:alert:service:call-waiting:org.example.somecompany.1:abc.
>
> Is this OK for you?
>
> 2) It seems to me that there are  small inconsistencies/ambiguities
> concerning <private-name> in version 06 Section4, ABNF-syntax:
>
> We have now:
>             alert-URN        = "urn:alert:" alert-identifier
>             alert-identifier = alert-category ":" alert-indication
>             alert-category   = name
>             alert-indication= name *(":" name)
>             name = standard-name / private-name
>             standard-name = label
>             private-name = label "." label *[ "." label ]
>
>       According to this definition:
>           - org.example.somecompany  is a  <name>, so it can be an
> alert-category, which I think we don't want.
>           - In the example
> urn:alert:service:call-waiting:org.example.somecompany.1:abc,
> "call-waiting" and "abc" are both <standard-name>s. Also this is
> correct from the syntax point of view,  we need to distinguish somehow
> between  "call-waiting",  which is registered with IANA and  "a.b.c",
> which is valid only inside somecompany.example.org and is not
> registred with IANA.
> Also in the same example "org.example.somecompany.1" is a
> <private-name>.  I think  "org.example.somecompany.1" can not be used
> alone, it is more a qualifier or prefix than a <name> ( it can not be
> rendered).
> In my opinion, the current syntax at least leads to some confusion, or
> did I miss something here?
>
> I would propose to change the ABNF-syntax so that the the meaning and
> the registration process for them becomes more clear. Here is a
> proposal:
>
>             alert-identifier = alert-category ":" alert-indication
>             alert-category   = standard-name/ (
> reverse-domain-name-seq:private-name)
>             alert-indication = (standard-name *(":" standard-name)) /
>                (standard-name *(":" standard-name) 1*(":"
> reverse-domain-name-seq) *(":" private-name) *(":"
> reverse-subdomain-name-seq) *(":" private-name))
>             standard-name =label
>             private-name    = label
>             reverse-domain-name-seq     = reverse-domain-name.seq-number
>             reverse-domain-name= domain-name-label "."
> domain-name-label *[ "." domain-name-label ]
>             reverse-subdomain-name-seq     =
> reverse-subdomain-name.seq-number
>             reverse-subdomain-name= subdomain-name-label "."
> subdomain-name-label *[ "." subdomain-name-label ]
>             domain-name-label = let-dig [ *let-dig-hyp let-dig ]
>             domain-name-label = let-dig [ *let-dig-hyp let-dig ]
>             label            = let-dig [ *let-dig-hyp let-dig ]
>
> The IANA registration proces is as follows:
>     - <standard-name>s  must be registered with IANA.
>     - <reverse-domain-name-seq>es must be registred with IANA by the
> owner of the domain name obtained from <reverse-domain-name-seq> by
> dropping the <seq-number> and reversing the reminder.
>      - IANA delegates the registration process for <private-name>s used
> behind a  <reverse-domain-name-seq> to the domain owner. When a domain
> changes the owner and if the new owner intends to change the usage or
> define new <private-names>, he must register a new
> <reverse-domain-name-seq> with the same <reverse-domain-name> and
> incremented   <seq-number>.
>      - Domains owners should have an internal registration process for
> <private-name>s and <reverse-subdomain-name-seq>s.
>      - Domains owners may delegate the registration process for
> <private-name>s used within subdomains to the subdomain owner.
>
>
> The new syntax is quite long, I know, and it means a lot of changes in
> the draft, but to me it is more clear this way. Maybe we can drop the
> subdomain stuff and allow only one reverse-domain-name in an
> alert-indication?
> alert-indication = (standard-name *(":" standard-name)) /
> (standard-name *(":" standard-name) 1*(":" reverse-domain-name-seq)
> *(":" private-name) *)
>
> Please let me know your opinion. I hope I am not completely wrong.
>
>
> Thanks a lot
> Laura
>


From laura.liess.dt@googlemail.com  Tue Sep 18 07:31:52 2012
Return-Path: <laura.liess.dt@googlemail.com>
X-Original-To: salud@ietfa.amsl.com
Delivered-To: salud@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3DB3321F8750 for <salud@ietfa.amsl.com>; Tue, 18 Sep 2012 07:31:52 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.233
X-Spam-Level: 
X-Spam-Status: No, score=-2.233 tagged_above=-999 required=5 tests=[AWL=-0.744, BAYES_05=-1.11, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id y5Tf7GIoawqZ for <salud@ietfa.amsl.com>; Tue, 18 Sep 2012 07:31:51 -0700 (PDT)
Received: from mail-iy0-f172.google.com (mail-iy0-f172.google.com [209.85.210.172]) by ietfa.amsl.com (Postfix) with ESMTP id 2BF9121F874C for <salud@ietf.org>; Tue, 18 Sep 2012 07:31:51 -0700 (PDT)
Received: by iabz21 with SMTP id z21so6870091iab.31 for <salud@ietf.org>; Tue, 18 Sep 2012 07:31:50 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=7J9PVx84KVpDdlxg8aKiH7t+BnmrCmut8K7aOMoUewc=; b=xQ7yzMiNsYxTY6iJK2d/jKK4CSMqOoCEXJ3G4UamEAP2d6DTkbSlAmQ8zrV2R/h5KM uUNngyx28NnqpNzXnDiwGBUffLxz66GB2XkhgijM2QNOCQeS/6MyuKjWB+9+bZqBI6lg az7ZHmVoF4Q68DR9TmZ2R3NMQxmOtNPA6Ygb16pC1KCrFAV/u3eu24Hf7Cdzwf0gI31I gyEezUazmjPipaR2+spAr/BJ6dumCXBUsNvJpxleZx5lOBaYWyd5w7PWZR0T4DEsrxO6 Heh+nO+BdityEVWVj4VSdRiTJjuDiXAbh33zNF9e0tOZ3/j22XM+guJxOu3Xv5UTd6Nd ewCQ==
MIME-Version: 1.0
Received: by 10.50.187.233 with SMTP id fv9mr92361igc.18.1347978710414; Tue, 18 Sep 2012 07:31:50 -0700 (PDT)
Received: by 10.231.180.211 with HTTP; Tue, 18 Sep 2012 07:31:50 -0700 (PDT)
In-Reply-To: <505274A0.4080100@alum.mit.edu>
References: <CACWXZj314Q7K3vg4yDc9gQHNd-qSjOKeFv8oq7=KG07yHbrFtw@mail.gmail.com> <505274A0.4080100@alum.mit.edu>
Date: Tue, 18 Sep 2012 16:31:50 +0200
Message-ID: <CACWXZj0g2bOHEWc+53-ZnJrKDHMAJTR0pz0TjAbd7=jLbnmowg@mail.gmail.com>
From: Laura Liess <laura.liess.dt@googlemail.com>
To: Paul Kyzivat <pkyzivat@alum.mit.edu>
Content-Type: text/plain; charset=ISO-8859-1
Cc: Dale Worley <dworley@avaya.com>, salud@ietf.org
Subject: Re: [salud] Concern #2 and ABNF-syntax change proposal
X-BeenThere: salud@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Sip ALerting for User Devices working group discussion list <salud.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/salud>, <mailto:salud-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/salud>
List-Post: <mailto:salud@ietf.org>
List-Help: <mailto:salud-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/salud>, <mailto:salud-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 18 Sep 2012 14:31:52 -0000

Paul,

Thank you for the proposal and for all the work you put in it. I like
it. I like especially your idea with baz@foo.bar.
I just have some questions, to ensure I understand you correctly.

> alert-URN       = "urn:alert:" alert-identifier
> alert-identifier= alert-category ":" alert-indication
> alert-category  = label
> alert-indication= standard-name / private-name
> standard-name   = standard-part *(":" standard-part)
> private-name    = [standard-name ":"] private-part *(":" alert-part)
> alert-part      = standard-part / private-part
> standard-part   = label

[LL] What is the reason to have
     alert-part      = standard-part / private-part  instead of
     alert-part  = label / private-part  ?

My understanding here is that  <standard-part> behind a a
<private-part> is has nothing to do with the IANA registration, it is
scoped only within the domain of the prior <private-name>.  One can
use here any label he likes. Is this correct or did I miss something?

>
> private-part    = label "@" dns-name ["(" version ")"]
> dns-name        = label "." label *[ "." label ]
> version         = DIGIT-2-9 / DIGIT-1-9 1*2DIGIT; 2...999
>
> label           = let-dig [ *let-dig-hyp let-dig ]
> let-dig-hyp     = let-dig / "-"
> let-dig         = ALPHA / DIGIT
> ALPHA           = %x41-5A / %x61-7A   ; A-Z / a-z
> DIGIT           = %x30-39 ; 0-9
> DIGIT-1-9       = %x31-39 ; 1-9
> DIGIT-2-9       = %x32-39 ; 2-9
>
> Key differences to prior attempts:
>
> - I've switched from reverse-dns to regular dns names.
>   I did this because it's more readable and I can no
>   longer see any reason not to.
>
> - I put the version number in, but made it optional,
>   with values 2...999. (You can consider the default
>   to be 1.) This way you only need to insert the version
>   when it's not the first. (In practice this means you
>   never need to put it in.)
>
> - I was bothered by the fact that the individual parts
>   that can be stripped away in looking for something
>   understood were not syntactically distinguished.
>   And I saw an opportunity after switching to forward
>   dns, so I changed from foo.bar:baz to baz@foo.bar.
>
> - Everything after the <private-part> is scoped within the
>   domain of the prior private name. (AFAIK this isn't
>   really a change, but I want to make it clear.) The
>   syntax now doesn't suggest that those parts are standard.
>
> - I allowed switching to a second dns scope within the
>   same alert-indication. I just did it because I could.
>   I can remove it if desired. This allows one organization
>   to "extend" the private names of another organization.
[LL]  Great,  I think this is a realy usefull new feature :-).
>
> - I've used a lot of "unnecessary" symbols in the syntax.
>   I did it mostly for readability. I also strived to
>   leave in all those symbols that are used in the text,
>   so it is hopefully largely unaffected.
>   If people like this it can be cleaned up to remove the
>   least necessary ones.

[LL]  I would not remove anything. Readability is very important.

>
> Some examples:
>
> alert-info:urn:service
> alert-info:urn:service:recall
> alert-info:urn:alert:service:foo@example.com
> alert-info:urn:alert:service:foo@example.com:bar
> alert-info:urn:alert:service:foo@example.com(2):baz
> alert-info:urn:alert:service:recall:foo@example.com
> alert-info:urn:alert:service:foo@example.com:bar:abc@atlanta.com:baz
>
> I see one major issue with this, or any form based on a version number:
>
> I am concerned that we can't define the version number semantics in a way
> that will satisfy Albert. As used above, it assumes that the person using
> the dns-name has a way to know how many prior owners of the domain name
> there have been. There really isn't any way to know that.

[LL]. You are right. In my proposal, I assumed that domain owners
intending to use alert-info urns register the string
reverse-dns+version  with IANA  (I called it "prefix" there). Now it
is clear to me that this does not realy work.
>
> The above syntax could be adapted to putting in something other than a
> version (such as a date).
[LL] In my proposal,  I switched from "date" to "version" because
people said on the ML that "date" is too long.  Maybe we can use just
the "year"?

> It would be nice to find something with a default
> that is sufficient most of the time. But I'm not certain that is possible.
[LL] Probably there isn't anything we can do here. And you are right,
it's somehow ugly to add the date/year when they are unnecessary in
most cases ..

Thank you
Laura



>         Thanks,
>         Paul
>
>
> On 9/13/12 8:27 AM, Laura Liess wrote:
>>
>> Hi,
>>
>> I have following questions/change proposals:
>>
>> 1) On how to deal with <private-name>s and the Concern #2:
>>
>> To reach a compromise, I would propose to adopt Alfred's second
>> proposal to add a simple sequence number to reverse domain name, as
>> proposed in
>> http://www.ietf.org/mail-archive/web/salud/current/msg00257.html
>> from June 29:
>>
>> "IANA could delegate a branch of the salud namespace to a domain name
>> owner using a past year alone (that is related to activity of the
>> domain owner in this field), or a simple sequence number, as a
>> disambiguating component in the naming scheme."
>>
>> An example would be
>> urn:alert:service:call-waiting:org.example.somecompany.1:abc.
>>
>> Is this OK for you?
>>
>> 2) It seems to me that there are  small inconsistencies/ambiguities
>> concerning <private-name> in version 06 Section4, ABNF-syntax:
>>
>> We have now:
>>             alert-URN        = "urn:alert:" alert-identifier
>>             alert-identifier = alert-category ":" alert-indication
>>             alert-category   = name
>>             alert-indication= name *(":" name)
>>             name = standard-name / private-name
>>             standard-name = label
>>             private-name = label "." label *[ "." label ]
>>
>>       According to this definition:
>>           - org.example.somecompany  is a  <name>, so it can be an
>> alert-category, which I think we don't want.
>>           - In the example
>> urn:alert:service:call-waiting:org.example.somecompany.1:abc,
>> "call-waiting" and "abc" are both <standard-name>s. Also this is
>> correct from the syntax point of view,  we need to distinguish somehow
>> between  "call-waiting",  which is registered with IANA and  "a.b.c",
>> which is valid only inside somecompany.example.org and is not
>> registred with IANA.
>> Also in the same example "org.example.somecompany.1" is a
>> <private-name>.  I think  "org.example.somecompany.1" can not be used
>> alone, it is more a qualifier or prefix than a <name> ( it can not be
>> rendered).
>> In my opinion, the current syntax at least leads to some confusion, or
>> did I miss something here?
>>
>> I would propose to change the ABNF-syntax so that the the meaning and
>> the registration process for them becomes more clear. Here is a
>> proposal:
>>
>>             alert-identifier = alert-category ":" alert-indication
>>             alert-category   = standard-name/ (
>> reverse-domain-name-seq:private-name)
>>             alert-indication = (standard-name *(":" standard-name)) /
>>                (standard-name *(":" standard-name) 1*(":"
>> reverse-domain-name-seq) *(":" private-name) *(":"
>> reverse-subdomain-name-seq) *(":" private-name))
>>             standard-name =label
>>             private-name    = label
>>             reverse-domain-name-seq     = reverse-domain-name.seq-number
>>             reverse-domain-name= domain-name-label "."
>> domain-name-label *[ "." domain-name-label ]
>>             reverse-subdomain-name-seq     =
>> reverse-subdomain-name.seq-number
>>             reverse-subdomain-name= subdomain-name-label "."
>> subdomain-name-label *[ "." subdomain-name-label ]
>>             domain-name-label = let-dig [ *let-dig-hyp let-dig ]
>>             domain-name-label = let-dig [ *let-dig-hyp let-dig ]
>>             label            = let-dig [ *let-dig-hyp let-dig ]
>>
>> The IANA registration proces is as follows:
>>     - <standard-name>s  must be registered with IANA.
>>     - <reverse-domain-name-seq>es must be registred with IANA by the
>> owner of the domain name obtained from <reverse-domain-name-seq> by
>> dropping the <seq-number> and reversing the reminder.
>>      - IANA delegates the registration process for <private-name>s used
>> behind a  <reverse-domain-name-seq> to the domain owner. When a domain
>> changes the owner and if the new owner intends to change the usage or
>> define new <private-names>, he must register a new
>> <reverse-domain-name-seq> with the same <reverse-domain-name> and
>> incremented   <seq-number>.
>>      - Domains owners should have an internal registration process for
>> <private-name>s and <reverse-subdomain-name-seq>s.
>>      - Domains owners may delegate the registration process for
>> <private-name>s used within subdomains to the subdomain owner.
>>
>>
>> The new syntax is quite long, I know, and it means a lot of changes in
>> the draft, but to me it is more clear this way. Maybe we can drop the
>> subdomain stuff and allow only one reverse-domain-name in an
>> alert-indication?
>> alert-indication = (standard-name *(":" standard-name)) /
>> (standard-name *(":" standard-name) 1*(":" reverse-domain-name-seq)
>> *(":" private-name) *)
>>
>> Please let me know your opinion. I hope I am not completely wrong.
>>
>>
>> Thanks a lot
>> Laura
>>
>

From pkyzivat@alum.mit.edu  Tue Sep 18 10:35:39 2012
Return-Path: <pkyzivat@alum.mit.edu>
X-Original-To: salud@ietfa.amsl.com
Delivered-To: salud@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A08D521E80CB for <salud@ietfa.amsl.com>; Tue, 18 Sep 2012 10:35:39 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.437
X-Spam-Level: 
X-Spam-Status: No, score=-0.437 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FH_RELAY_NODNS=1.451, HELO_MISMATCH_NET=0.611,  RDNS_NONE=0.1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Y0YxzszL11Y9 for <salud@ietfa.amsl.com>; Tue, 18 Sep 2012 10:35:38 -0700 (PDT)
Received: from qmta05.westchester.pa.mail.comcast.net (qmta05.westchester.pa.mail.comcast.net [IPv6:2001:558:fe14:43:76:96:62:48]) by ietfa.amsl.com (Postfix) with ESMTP id 7211821E8043 for <salud@ietf.org>; Tue, 18 Sep 2012 10:35:38 -0700 (PDT)
Received: from omta07.westchester.pa.mail.comcast.net ([76.96.62.59]) by qmta05.westchester.pa.mail.comcast.net with comcast id 0dDo1k0031GhbT855hbiR5; Tue, 18 Sep 2012 17:35:42 +0000
Received: from Paul-Kyzivats-MacBook-Pro.local ([64.134.225.153]) by omta07.westchester.pa.mail.comcast.net with comcast id 0hbF1k0093KCHDc3ThbL1w; Tue, 18 Sep 2012 17:35:57 +0000
Message-ID: <5058B0BA.4070001@alum.mit.edu>
Date: Tue, 18 Sep 2012 13:34:50 -0400
From: Paul Kyzivat <pkyzivat@alum.mit.edu>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:15.0) Gecko/20120907 Thunderbird/15.0.1
MIME-Version: 1.0
To: Laura Liess <laura.liess.dt@googlemail.com>
References: <CACWXZj314Q7K3vg4yDc9gQHNd-qSjOKeFv8oq7=KG07yHbrFtw@mail.gmail.com> <505274A0.4080100@alum.mit.edu> <CACWXZj0g2bOHEWc+53-ZnJrKDHMAJTR0pz0TjAbd7=jLbnmowg@mail.gmail.com>
In-Reply-To: <CACWXZj0g2bOHEWc+53-ZnJrKDHMAJTR0pz0TjAbd7=jLbnmowg@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: Dale Worley <dworley@avaya.com>, salud@ietf.org
Subject: Re: [salud] Concern #2 and ABNF-syntax change proposal
X-BeenThere: salud@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Sip ALerting for User Devices working group discussion list <salud.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/salud>, <mailto:salud-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/salud>
List-Post: <mailto:salud@ietf.org>
List-Help: <mailto:salud-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/salud>, <mailto:salud-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 18 Sep 2012 17:35:39 -0000

On 9/18/12 10:31 AM, Laura Liess wrote:
> Paul,
>
> Thank you for the proposal and for all the work you put in it. I like
> it. I like especially your idea with baz@foo.bar.
> I just have some questions, to ensure I understand you correctly.
>
>> alert-URN       = "urn:alert:" alert-identifier
>> alert-identifier= alert-category ":" alert-indication
>> alert-category  = label
>> alert-indication= standard-name / private-name
>> standard-name   = standard-part *(":" standard-part)
>> private-name    = [standard-name ":"] private-part *(":" alert-part)
>> alert-part      = standard-part / private-part
>> standard-part   = label
>
> [LL] What is the reason to have
>       alert-part      = standard-part / private-part  instead of
>       alert-part  = label / private-part  ?

Those are equivalent representations. I just did it for clarity.
("standard-part" has some implied semantics, while "label" not so much.)

If/when we come to an agreement that the ABNF describes what we want it 
to describe, then the ABNF can be tweaked in whats that don't change the 
set of valid strings, to suit everyone's biases. :-)

> My understanding here is that  <standard-part> behind a a
> <private-part> is has nothing to do with the IANA registration, it is
> scoped only within the domain of the prior <private-name>.  One can
> use here any label he likes. Is this correct or did I miss something?

OK. Now I see why you asked the question above. Yes, I agree and 
intended that a standard-part behind a private-part is not "standard".
So my attempt to get the implied semantics right failed here. :-(

I can work on that later. As long as we agree on the intended semantics 
we can continue to discuss if the proposed *syntax* meets our needs.

>> private-part    = label "@" dns-name ["(" version ")"]
>> dns-name        = label "." label *[ "." label ]
>> version         = DIGIT-2-9 / DIGIT-1-9 1*2DIGIT; 2...999
>>
>> label           = let-dig [ *let-dig-hyp let-dig ]
>> let-dig-hyp     = let-dig / "-"
>> let-dig         = ALPHA / DIGIT
>> ALPHA           = %x41-5A / %x61-7A   ; A-Z / a-z
>> DIGIT           = %x30-39 ; 0-9
>> DIGIT-1-9       = %x31-39 ; 1-9
>> DIGIT-2-9       = %x32-39 ; 2-9
>>
>> Key differences to prior attempts:
>>
>> - I've switched from reverse-dns to regular dns names.
>>    I did this because it's more readable and I can no
>>    longer see any reason not to.
>>
>> - I put the version number in, but made it optional,
>>    with values 2...999. (You can consider the default
>>    to be 1.) This way you only need to insert the version
>>    when it's not the first. (In practice this means you
>>    never need to put it in.)
>>
>> - I was bothered by the fact that the individual parts
>>    that can be stripped away in looking for something
>>    understood were not syntactically distinguished.
>>    And I saw an opportunity after switching to forward
>>    dns, so I changed from foo.bar:baz to baz@foo.bar.
>>
>> - Everything after the <private-part> is scoped within the
>>    domain of the prior private name. (AFAIK this isn't
>>    really a change, but I want to make it clear.) The
>>    syntax now doesn't suggest that those parts are standard.
>>
>> - I allowed switching to a second dns scope within the
>>    same alert-indication. I just did it because I could.
>>    I can remove it if desired. This allows one organization
>>    to "extend" the private names of another organization.
> [LL]  Great,  I think this is a realy usefull new feature :-).
>>
>> - I've used a lot of "unnecessary" symbols in the syntax.
>>    I did it mostly for readability. I also strived to
>>    leave in all those symbols that are used in the text,
>>    so it is hopefully largely unaffected.
>>    If people like this it can be cleaned up to remove the
>>    least necessary ones.
>
> [LL]  I would not remove anything. Readability is very important.
>
>>
>> Some examples:
>>
>> alert-info:urn:service
>> alert-info:urn:service:recall
>> alert-info:urn:alert:service:foo@example.com
>> alert-info:urn:alert:service:foo@example.com:bar
>> alert-info:urn:alert:service:foo@example.com(2):baz
>> alert-info:urn:alert:service:recall:foo@example.com
>> alert-info:urn:alert:service:foo@example.com:bar:abc@atlanta.com:baz
>>
>> I see one major issue with this, or any form based on a version number:
>>
>> I am concerned that we can't define the version number semantics in a way
>> that will satisfy Albert. As used above, it assumes that the person using
>> the dns-name has a way to know how many prior owners of the domain name
>> there have been. There really isn't any way to know that.
>
> [LL]. You are right. In my proposal, I assumed that domain owners
> intending to use alert-info urns register the string
> reverse-dns+version  with IANA  (I called it "prefix" there). Now it
> is clear to me that this does not realy work.

It could work if they register it. But I think we have a goal that they 
not need to register it. (Because they probably won't!) If they are 
willing to register, then there are easier schemes we can use.

Perhaps we should have the discussion of whether some system involving 
registration would be acceptable. Such a scheme would no longer need to 
involve dns names at all.

Maybe we could introduce a one-time registry for an organization, for an 
id that could be used to introduce a private name anywhere in here. Or 
perhaps we can find some existing identifier that many likely users of 
this would already have that we could use for this purpose. (Aren't 
there some global "organization" codes?)

	Thanks,
	Paul

>> The above syntax could be adapted to putting in something other than a
>> version (such as a date).
> [LL] In my proposal,  I switched from "date" to "version" because
> people said on the ML that "date" is too long.  Maybe we can use just
> the "year"?
>
>> It would be nice to find something with a default
>> that is sufficient most of the time. But I'm not certain that is possible.
> [LL] Probably there isn't anything we can do here. And you are right,
> it's somehow ugly to add the date/year when they are unnecessary in
> most cases ..
>
> Thank you
> Laura
>
>
>
>>          Thanks,
>>          Paul
>>
>>
>> On 9/13/12 8:27 AM, Laura Liess wrote:
>>>
>>> Hi,
>>>
>>> I have following questions/change proposals:
>>>
>>> 1) On how to deal with <private-name>s and the Concern #2:
>>>
>>> To reach a compromise, I would propose to adopt Alfred's second
>>> proposal to add a simple sequence number to reverse domain name, as
>>> proposed in
>>> http://www.ietf.org/mail-archive/web/salud/current/msg00257.html
>>> from June 29:
>>>
>>> "IANA could delegate a branch of the salud namespace to a domain name
>>> owner using a past year alone (that is related to activity of the
>>> domain owner in this field), or a simple sequence number, as a
>>> disambiguating component in the naming scheme."
>>>
>>> An example would be
>>> urn:alert:service:call-waiting:org.example.somecompany.1:abc.
>>>
>>> Is this OK for you?
>>>
>>> 2) It seems to me that there are  small inconsistencies/ambiguities
>>> concerning <private-name> in version 06 Section4, ABNF-syntax:
>>>
>>> We have now:
>>>              alert-URN        = "urn:alert:" alert-identifier
>>>              alert-identifier = alert-category ":" alert-indication
>>>              alert-category   = name
>>>              alert-indication= name *(":" name)
>>>              name = standard-name / private-name
>>>              standard-name = label
>>>              private-name = label "." label *[ "." label ]
>>>
>>>        According to this definition:
>>>            - org.example.somecompany  is a  <name>, so it can be an
>>> alert-category, which I think we don't want.
>>>            - In the example
>>> urn:alert:service:call-waiting:org.example.somecompany.1:abc,
>>> "call-waiting" and "abc" are both <standard-name>s. Also this is
>>> correct from the syntax point of view,  we need to distinguish somehow
>>> between  "call-waiting",  which is registered with IANA and  "a.b.c",
>>> which is valid only inside somecompany.example.org and is not
>>> registred with IANA.
>>> Also in the same example "org.example.somecompany.1" is a
>>> <private-name>.  I think  "org.example.somecompany.1" can not be used
>>> alone, it is more a qualifier or prefix than a <name> ( it can not be
>>> rendered).
>>> In my opinion, the current syntax at least leads to some confusion, or
>>> did I miss something here?
>>>
>>> I would propose to change the ABNF-syntax so that the the meaning and
>>> the registration process for them becomes more clear. Here is a
>>> proposal:
>>>
>>>              alert-identifier = alert-category ":" alert-indication
>>>              alert-category   = standard-name/ (
>>> reverse-domain-name-seq:private-name)
>>>              alert-indication = (standard-name *(":" standard-name)) /
>>>                 (standard-name *(":" standard-name) 1*(":"
>>> reverse-domain-name-seq) *(":" private-name) *(":"
>>> reverse-subdomain-name-seq) *(":" private-name))
>>>              standard-name =label
>>>              private-name    = label
>>>              reverse-domain-name-seq     = reverse-domain-name.seq-number
>>>              reverse-domain-name= domain-name-label "."
>>> domain-name-label *[ "." domain-name-label ]
>>>              reverse-subdomain-name-seq     =
>>> reverse-subdomain-name.seq-number
>>>              reverse-subdomain-name= subdomain-name-label "."
>>> subdomain-name-label *[ "." subdomain-name-label ]
>>>              domain-name-label = let-dig [ *let-dig-hyp let-dig ]
>>>              domain-name-label = let-dig [ *let-dig-hyp let-dig ]
>>>              label            = let-dig [ *let-dig-hyp let-dig ]
>>>
>>> The IANA registration proces is as follows:
>>>      - <standard-name>s  must be registered with IANA.
>>>      - <reverse-domain-name-seq>es must be registred with IANA by the
>>> owner of the domain name obtained from <reverse-domain-name-seq> by
>>> dropping the <seq-number> and reversing the reminder.
>>>       - IANA delegates the registration process for <private-name>s used
>>> behind a  <reverse-domain-name-seq> to the domain owner. When a domain
>>> changes the owner and if the new owner intends to change the usage or
>>> define new <private-names>, he must register a new
>>> <reverse-domain-name-seq> with the same <reverse-domain-name> and
>>> incremented   <seq-number>.
>>>       - Domains owners should have an internal registration process for
>>> <private-name>s and <reverse-subdomain-name-seq>s.
>>>       - Domains owners may delegate the registration process for
>>> <private-name>s used within subdomains to the subdomain owner.
>>>
>>>
>>> The new syntax is quite long, I know, and it means a lot of changes in
>>> the draft, but to me it is more clear this way. Maybe we can drop the
>>> subdomain stuff and allow only one reverse-domain-name in an
>>> alert-indication?
>>> alert-indication = (standard-name *(":" standard-name)) /
>>> (standard-name *(":" standard-name) 1*(":" reverse-domain-name-seq)
>>> *(":" private-name) *)
>>>
>>> Please let me know your opinion. I hope I am not completely wrong.
>>>
>>>
>>> Thanks a lot
>>> Laura
>>>
>>
>

