DCCP - IETF 76 Hiroshima
Tuesday, November 10, 0900 - 1000
Chair: Pasi Sarolahti
Note taker: Colin Perkins

15 attendees

Status:

Pasi presents slides.

re draft-ietf-dccp-rtp: Magnus notes that the rtcpssm draft has been updated, and is back with the IESG

re udp encapsulation 
- Magnus Westerlund: issue will be discussed in tsvarea on Friday. Magnus is split on the idea, he sees the benefit, but also the concern

- Colin Perkins: we have an implementation of the nat encapsulation

- Murari Sridharan: lots of people are trying to implement, but the lack of firewall traversal is stopping people. some udp encap would be good.

- Markku Kojo: don't need a spec, just fill out the iana form

- Magnus: yes, but a spec is importent.

- Colin: why don't we just publish Tom Phelan's draft as experimental? Ask Tom to resubmit it (or I'll do it...)

next steps
- Colin + Jukka Manner: the group likely doesn't need to meet, but keep it alive
- Magnus: suggests closing the group, but leaving the list alive



DCCP user guide:

Jukka presents slides

- Pasi: suggests book chapter, articles, etc., as an alternative
- Jukka: needs to be more easily accessible than that

- Yoshifumi Nishida: should do a sales pitch, highlight applications such as VLC that already support dccp

(unclear that there's interest to move forward with this)

- Magnus: would be good to have a document, even if not necessarily complete or final. is there the energy to get it done?

- Colin: problem is that the previous one had some controversial sections (e.g. MUST pad); are we just going to rathole on the same issues, since we don't have any more expertise

- chair: who has read? couple of hands. who's interested in working on this? couple of hands



MulTFRC:

- aiming for experimental

- Colin: aim of ccids was to allow experimentation; this seems to fit well. congestion control algorithms are done in tsvwg/iccrg; once they're happy we can publish the ccid

- Jukka: does it need to be a wg item? 

- Magnus: congestion control algorithms need to be reviewed by iccrg to be sure they're safe to deploy; given that the algorithm is okay, doesn't see any real problems publishing another experimetnal ccid

- Magnus: notes that new ccids need IETF consensus to publish

- chair: who has read? no hands

End of meeting.