SIPCORE NOTES

IETF-76


NOTES BY HISHAM KHARTABIL


Agenda and Status


Agenda Bashing

- No comment

Chair Update

- Charter summary was presented by chairs. Chair noted that we are running behind on milestones

- Invite Transaction Handling Correction is ready for WGLC according to authors and chairs

- Example security flows needs reviewers

- Etage extension: Looking for new editor to take of IESG comments

- IPv6 ABNF Fix: current plan is to publish as AD-sponsored draft


Event Rate Control

Salvatore Loreto


- Recap presented

- Chair asks if anyone understands what we are doing. A few thumbs up.

Chair volunteers to explain to people who don't understand this draft

- Note that this draft allows inclusion of Event header in a response.

Something not clear in RFC3265

- Author suggests that draft is ready for WGLC

- Author finished presentation early and chair requested for him to sing


INFO Method

Christer Holmberg


- Summarised activities since last IETF meeting including update that incorporated WGLC comments

- Request for reviewers since it was a major rewrite

- Open issue: what is required to register an info package. Alternatives presented.

- comment by Eric Burger that the point is to make it easy for someone to register a package with a balance for interoperability

- Jon Peterson: we should consider IANA requirements. He asks the question if we are going to increase or decrease

interop with this draft. Issue of interoperability was discussed vigorously.

- Comments from Eric and Hannu Hietelahti that battles for standards in IETF take too long

- suggestion by Hisham Khartabil to have AD sponsored drafts towards RFCs to speed things up

- Robert Sparks suggests that we instruct IANA to only take a package name when specification around package name has been submitted

- Hum called for specification required. Consensus was for specification required

- Open issue: Do we mandate Recv-Info in re-INVITE response + ACK

- Eric and Adam suggest that it should be included

- This was compared to offer-answer. Adam clarifies that it is not quite the same since it is not a negotiation

- Hum was taken to include recv-info. consensus was to include

- Robert asserts his concern that we don't have actual packages to guide us to the right answer. Discussions would go smoother if we do


History-Info

Mary Barnes


- Presenter summarised current status and changes since last update to draft. One update since last ietf meeting

- Presenter asked the question if this draft is ready to adopted by the WG. Hum was taken. Consensus to take draft as wg item.


NOTES BY ALAN JOHNSON


Agenda and Status


* Agenda Bash - no changes

* Status - Chairs

Example Security Flows still needs more review Etags in conditional notification looking for new author

IPv6 ABNF fix - moving forward as AD sponsored instead of adopting in SIPCORE


* Event Rate Control - Salvatore Loreto


A new mechanism to update rate control parameters mid-dialog described - no objections encountered Ready for WGLC - watch list for announcement Make sure GEOPRIV participants agree this meets their requirements


* INFO - Christer Holmberg


Major rewrite based on WGLC comments

Open Issue: What is required to register an INFO package?

Lots of discussion on namespace vs interop, specification required vs expert review

Hum: Specification Required or Different alternative.

Consensus call by chairs is to use Specification Required.

Open Issue: Require Recv-Info in re-INVITE for 3pcc Discussion supports inclusion in target refresh

Hum: Put info in any message can change it plus responses or Disagree.

Consensus to include this in every message.

Robert: Would be useful to have actual packages - will this actually get used?

Discussion about which packages might be standardized.


* History-Info and Target URI - Mary Barnes


Revised once since last IETF

Use cases from draft-rosenberg-sip-target-uri-delivery moved into document Appendix Discussion about how voicemail server knows which one to use.

Hum: Adopt as working group item as charter milestone for URI parameter delivery? Consensus to adopt.