[sipcore] Revised draft-ietf-sipcore-subnot-etags to -03

Dean Willis <dean.willis@softarmor.com> Thu, 26 November 2009 01:03 UTC

Message-Id: <86B0A2FD-CB20-4E49-B69A-66162E91178D@softarmor.com>
From: Dean Willis <dean.willis@softarmor.com>
To: SIPCORE <sipcore@ietf.org>
Content-Type: text/plain; charset="US-ASCII"; format="flowed"; delsp="yes"
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v936)
Date: Wed, 25 Nov 2009 19:03:23 -0600
Subject: [sipcore] Revised draft-ietf-sipcore-subnot-etags to -03
Precedence: list

Based on discussions with Robert and Adam, I've revised the subnot- 
etags document.

There are no "real" changes, just clarifications on what an entity tag  
means in SIP, the uniqueness-scope of such tags, and very minor  
language cleanup.

Previously, we had a lot of debate on whether an eTag is truly  
"unique". It's not; it is unique only within a map of the state-vector  
space of each resource. For a given resource, each eTag is associated  
with one and only one view of  that resource's state (which we call an  
"entity".)  For a given resource, identical eTags mean identical  
entities.

Comparison of eTags between resources is meaningless, as they refer to  
disjoint state spaces. That is, if resource A as seen by Alice has an  
entity-tag of 0x77A and resource B as seen by Bob has an entity tag of  
0X77A, this doesn't mean that the two entities are the same. Doesn't  
mean they're different, either; it means absolutely nothing that they  
have entity tags with the same value.

I promise the draft now makes more sense than this synopsis.

I'm not sure I like the graphic rendering of the resource model in  
section 4. Please take a look at it and provide feedback.

--
Dean

[sipcore] Revised draft-ietf-sipcore-subnot-etags… Dean Willis