Minutes SIPPING WG at IETF 64 Minutes edited by Gonzalo Camarillo Based on notes by Hisham Khartabil, Dean Willis, Jim McEachern, Dorothy Gellert, and Steve S. Meeting chaired by Dean Willis, Rohan Mahy, and Gonzalo Camarillo Slides presented included in the proceedings MONDAY, November 7, 2005, 1510-1710 Topic: Agenda Bash Discussions led by: Chairs Two ad-hoc meetings were announced: one on TISPAN-related issues and another on peer-to-peer SIP. Oscar Novo was presented as the new SIPPING secretary. The chairs presented the status of all the SIPPING WG items. The URI-list services drafts have passed WGLC and are ready, but they are waiting for the consent framework to be ready. New revisions of the RTCP summary and trait-based authorization drafts will be submitted shortly. The chairs will request their publication as soon as the new revisions appear in the archives. The chairs presented the updated charter and milestones. It was noted that the dialog usage draft has disappeared. The chairs will investigate what happened with that draft. Topic: Configuration Framework and Data Sets Discussion led by: Dan Petrie Relevant documents: draft-ietf-sipping-config-framework-07.txt draft-petrie-sipping-profile-datasets-03.txt draft-petrie-sipping-identity-dataset-00.txt draft-petrie-sipping-voip-features-dataset-00.txt draft-petrie-sipping-sip-dataset-01.txt There were discussions on whether the text on merging should be removed from the drafts. It seems easier to construct a data model such that merging is easy than provide all nodes with a general merging engine. There were discussions on how much XML validation should be required in the clients. The conclusion was to using the XML schema as used in SIMPLE. At this point, there are no requirements for strong XML validation at the client side. There were discussions without a clear conclusion on whether, for certain configuration data, user agents can be provided with the URI of a single hop or they need a routing vector. Data sets dealing with configurable elements defined by SIP or SIPPING WG items will be WG items as well. Topic: Session Policy Framework and Data Set Format Discussion led by: Volker Hilt Relevant documents: draft-hilt-sipping-session-policy-framework-00.txt draft-ietf-sipping-media-policy-dataset-00.txt draft-hilt-sipping-policy-package-00.txt The chairs will add a milestone date for the session policy framework to the charter. There was consensus that using an SDP format for session specific policies would be OK. Topic: Consent Framework Discussion led by: Gonzalo Camarillo Relevant documents: draft-ietf-sipping-consent-framework-03.txt There was consensus to remove the operations part from the permission documents. Furthermore, there was consensus on having permissions that are as simple as possible, given that they will be stored by servers that do not any relation with the users uploading them. Topic: Conference Bridge Transcoding Model Discussion led by: Gonzalo Camarillo Relevant documents: draft-ietf-sipping-transc-conf-00.txt The author will clarify how transcoders create their offers. Topic: IPv6 Transition Discussion led by: Vijay Gurbani Relevant documents: draft-ietf-sipping-v6-transition-01.txt draft-gurbani-sipping-ipv6-sip-01.txt The topic of using TURN for IPv4/IPv6 transition will be further discussed in the BEHAVE WG. Agreed that the draft should not contain examples of network APIs. It was noted that some interactions between DNS and dual-stack SIP nodes could also belong to a future update of RFC 3263. It was suggested to remove the general discussions on IPv6 transition strategies from the draft. The WG is interested in working on this, but remains undecided as to whether to schedule this work immediately or defer it for a future schedule. Topic: SIP-unfriendly Functions in Current Architectures Discussion led by: Jani Jautakorpi Relevant documents: draft-camarillo-sipping-sbc-funcs-02.txt It was noted that this draft is not only useful for SBC implementers, but also for UA implementers. Therefore, it is a good thing that the draft is not only focused on SBCs. It was suggested to document the requirements for features that SBCs currently implement, how they are currently implemented, and how those implementations break SIP. With all these things documented, the draft would be useful to start new work that meets those requirements in a more SIP-friendly way. The chairs will meet the authors to align terminology and to determine the scope of the draft. A few people agreed to send comments to the authors and to the list on the examples in the draft. There seems to be agreement that this document is useful and that should eventually become an RFC. Topic: Max Forwards Issues Discussion led by: Robert Sparks Relevant documents: draft-lawrence-maxforward-problems-00.txt It was noted that loop detection would eliminate this problem. There could be ways of implementing loop detection that require less state information than the current proposals. It was noted that RFC 3261 specifies loop detection for proxies, but does not require its usage. It was suggested that other solutions to this problem could consist of limiting arbitrary registrations of contacts or limiting inter-domain forking somehow. It was concluded that his discussion should continue in the SIP WG since this is a bug in RFC 3261. Topic: Multi Transcoding Discussion led by: Tae-Gyu Kang Relevant documents: draft-kang-sipping-transc-scenario-00.txt The draft deals with the following use case: if a call is placed through a transcoder, and the transcoder cannot complete an offer/answer exchange with the next hop, then a 488 response would indicate a problem with the transcoder, not with the "far side". This may justify a new response code. The author will need to clarify what is the use case that gets us into a situation with more than one or two transcoders when using the conference bridge transcoding model (which is intended for simple scenarios). The first SIPPING session ended. WEDNESDAY, November 9, 2005, 1300-1500 Topic: Redirection Issues Discussion led by: John Elwell Relevant documents: draft-elwell-sipping-service-retargeting.txt The draft was not presented. The chairs informed the group that the authors have realized that their requirements can be met by slightly modifying draft-jennings-sip-voicemail-uri-04.txt. There was a discussion on the path for this draft, which will be AD sponsored informational. Topic: Extending the SIP Reason Header with Warning Codes Discussion led by: Jani Hautakorpi Relevant documents: draft-hautakorpi-reason-header-for-warnings-00.txt The authors clarified that the main purpose of this draft was to trigger discussions on the usage of Warning header fields in SIP. At this point, not many implementers use them, and according to RFC 3261, their scope is limited to reporting problems related to session descriptions. It was noted that external SDOs are planning to propose SIP extensions in the near future and they need to know whether proposing new Warning codes to meet their requirements is appropriate, or if they should propose new response codes instead. There were discussions on whether or not response codes are enough. The conclusion seemed to be that they are enough and that we should not encourage new Warning code definitions at this point. In any case, this discussion will be continued on the list. There were discussions on whether the response code address space is running out. Robert Sparks agreed to look into all the new response codes being proposed (otherwise, they are not registered in the IANA until they are documented in an RFC). Topic: GRUU Reg Event Package Discussion led by: Paul Kyzivat Relevant documents: draft-ietf-sipping-gruu-reg-event-00.txt Consensus to go with the draft as is and schedule its WGLC. Topic: Requirements and Possible Mechanisms for File Transfer Services Within the Context of SIP Based Communication Discussion led by: Markus Isomaki Relevant documents: draft-isomaki-sipping-file-transfer-00.txt There were discussions on whether or not this work is useful and, if it is, where should this work proceed. The conclusion was to write a requirements document and discuss it in SIPPING. Topic: SIP Identity Usage within Enterprise Scenarios Discussion led by: Steffen Fries Relevant documents: draft-fries-sipping-identity-enterprise-scenario-01.txt There were discussions on the difference between device certificates and user certificates. The topic needs to be clarified so that everybody can understand it. Topic: Calling Party Category Discussion led by: Rocky Wang Relevant documents: draft-rocky-sipping-calling-party-category-01.txt There were discussions on whether it would make sense to do this with SAML. The author will talk to the folks working on SAML. Topic: Implementations of PRACK Discussion led by: Rohan Mahy In the MMUSIC session, some folks mentioned that people were trying and avoiding implementing PRACK due to its complexity. There were some discussions on what to do without a clear conclusion. Topic: Payment for Services in SIP Discussion led by: Jason Fischl Relevant documents: draft-jennings-sipping-pay-03.txt There were discussions on whether this work is useful for things that are not SPIT or SPAM. It was proposed to use SAML, but not many people were up to date with SAML. Topic: URN for Services Discussion led by: Henning Schulzrinne Relevant documents: draft-schulzrinne-sipping-service-01.txt It was concluded that this work should continue in the ECRIT WG. The second SIPPING session ended.