Minutes SIPPING WG at IETF 68 Minutes edited by Gonzalo Camarillo Based on notes by Spencer Dawkins, Bruce Lowekamp, and Tom Taylor Meeting chaired by Mary Barnes and Gonzalo Camarillo Slides presented included in the proceedings First Session MONDAY, March 19, 2007, 0900-1130 Topic: Agenda Bash Discussions led by: Chairs The chairs presented the current status of the WG. The chairs presented the last and the next planned charter updates. Topic: Configuration Framework Discussion led by: Sumanth Channabasappa Relevant documents: draft-ietf-sipping-config-framework-11.txt Sumanth presented the draft, which is the result of the efforts of a design team. Around 30 people had read this version of the draft. Some people suggested to include an executive summary and a few examples, given that the draft is quite long. The draft should also be clearer about terminology. It should be easy to understand when the draft is using already existing terms defined in other specifications (e.g., in the event notification framework) and when the draft is using newly defined terms. There is a need to involve DNS people in the discussions. The current team does not have enough expertise no the DNS area. It might be possible to remove the device ID from the local network profile. There were discussions on a mechanism to fall-back to HTTP. The draft needs to be clearer about this, because while some people thought this meant to only use HTTP without using events, others thought this was a mechanism to skip some of the initial steps in the framework. There seems to be an agreement to leave the local network profile, which is useful for session policies, in the draft. Topic: A User Agent Profile Data Set for Media Policy Discussion led by: Volker Hilt Relevant documents: draft-ietf-sipping-media-policy-dataset-03.txt The session policies framework and its associated event package are ready for WGLC. The dataset draft has still some open issues. This draft depends on the user agent dataset draft, which is not even a WG item yet. The WG needs to decide how to advance both drafts (e.g., remove the dependency or advance them as they are now). However, in order to make that decision, there is a need to have explicit discussions on the mailing list about merging. Some people thought there might be issues related to merging. Topic: Overload Design Team Status Update Discussion led by: Jonathan Rosenberg Relevant documents: draft-ietf-sipping-overload-reqs-00.txt Jonathan presented the status of this design team, which is formed by IETF people and people with background on PSTN congestion control. The design team has agreed on a simplified simulation model to be used to run simulations of different mechanisms. The design team agreed to release intermediate results when they are available. Topic: Service Identification Discussion led by: Jonathan Rosenberg Relevant documents: draft-rosenberg-sipping-service-identification-01.txt Jonathan summarized the mailing list discussions on the issues. There had been tons of emails about it. Some people had concerns about interoperability if service IDs are defined. Jonathan proposed to follow the principle "do what I say", which consists of having explicit SIP mechanisms to signal all the aspects of the session being established, as opposed to deriving them from the service ID. The chairs took a "hum" on whether or not the WG wanted to work on this and there was not a clear winner. The chairs organized an ad-hoc meeting the next day to continue discussions on this issue. The ad-hoc meeting lasted one hour. People discussed the meaning of the "do what I say" approach. Participants voted (17 against 7) to produce a draft describing the "do what I say" approach. Topic: How do we achieve WG deliverables? Discussion led by: chairs The chairs had proposed to move drafts that are not ready in time to the end of the WGLC queue. There was agreement that this is the right thing to do. Second Session FRIDAY, March 23, 2007, 0900-1030 Topic: Agenda Bash Discussions led by: Chairs The last hour of the meeting had, in principle, been reserved by the ECRIT WG. However, they did not need that time, finally. The area directors request (and got) a 5 minute slot to present the next steps for the service identification issue. The MMUSIC chairs requested (and got) a 30 minute slot to discuss issues related to IPv6 transition, ICE, and ANAT. Topic: Service Identification Discussion led by: Cullen Jennings Cullen asked for a "hum" on whether or not to add a milestone to the SIPPING charter to work on a problem statement draft. There was consensus to work on it. Cullen also talked about a draft defining a P-header and a draft to define media features for sub media types, but no "hums" were taken on these items. Topic: ICE vs ANAT in MMUSIC Discussion led by: Jonathan Rosenberg Relevant documents: draft-ietf-mmusic-ice-15.txt draft-ietf-sipping-v6-transition-04.txt RFC 4091 Jonathan presented the issue, which had already been presented in the MMUSIC session. People were concerned with interoperability and ease of deployment. There was a very rough consensus to deprecate ANAT and use ICE alone vs. use ANAT and ICE together. Topic: SIP File Directory Discussion led by: Miguel A. Garcia-Martin Relevant documents: draft-garcia-sipping-resource-sharing-framework-01.txt draft-garcia-sipping-resource-event-package-01.txt draft-garcia-sipping-resource-desc-pidf-00.txt There is interest in using SIP events for things beyond what SIP events are used for today. Some people think this is an appropriate use of SIP events while others do not think so. The area director (Cullen) pointed out that there are more groups interested in this type of work. They should coordinate among themselves. Topic: SIP End-to-End Performance Metrics Discussion led by: Daryl Malas Relevant documents: draft-malas-performance-metrics-06.txt Daryl presented the draft. People indicated that it is preferable to have a small set of metrics than defining everything imaginable. Keith Drage sent a bunch of comments to BMWG that are also applicable to this draft. The area director (Cullen) indicated that even though we have not managed to find a home for this draft yet, the work is important and should continue. Cullen will try to gather people interested in these issues to try to find the best possible way forward for this type of effort. Topic: Extensions to SIP for the support of the Call Completion Services for ETSI Discussion led by: Denis Alexeitsev Relevant documents: draft-poetzl-sipping-call-completion-02.txt Denis presented the results of the discussions in TISPAN, where they prefer the event-based solution over the BFCP-based solution. The chairs took a "hum" in the group and there were more people in favor of the event-based solution. People complained that TISPAN has not discussed their concerns about the BFCP-based solution on the SIPPING list and that they may be based on misconceptions. There is a need to have more discussions on this issue on the mailing list. Topic: Early Media in SIP; Problem Statement, Requirements, and Analysis of Solutions Discussion led by: Richard Barnes Relevant documents: draft-barnes-sip-em-ps-req-sol-00.txt Richard presented the draft and there were discussions on it. More discussions will follow on the mailing list. Topic: Early Media Indirection Discussion led by: Francois Audet Relevant documents: draft-stucker-sipping-early-media-indirection-00.txt Francois presented the draft and there were discussions on it. More discussions will follow on the mailing list. Important to keep in mind that if something new is defined, we have to be pretty sure that people are going to deploy it. Otherwise, it will be useless.