Re: [rtcweb] [MMUSIC] Fwd: New Version Notification for draft-uberti-rtcweb-plan-00.txt

[Harald Alvestrand <[e-mail]>]

On 05/08/2013 01:07 AM, Bernard Aboba wrote:
> Roni said:
> "If the purpose of this draft and Adam’s one are to describe how to do 
> things in RTCweb it should be discussed in RTCWEB. And we will need a 
> similar one in CLUE probably in draft-ietf-clue-rtp-mapping."
>
> [BA] There are RTP issues to discuss (e.g. draft-ietf-clue-rtp-mapping 
> relationship to draft-ietf-rtcweb-rtp-usage), and there are SDP 
> issues.  To date, I believe we have somewhat neglected the RTP 
> discussion, which is unfortunate, because if all the time is spent 
> talking about the SDP Hammer, then the inclination is to use it to 
> pound every nail, and if there is no SDP, then we have no tools at all.

I felt that the discussion of draft-ietf-rtcweb-rtp-usage was largely 
uncontroversial in the Stockholm interim, and that not much further 
discussion of that level of the stack was needed - so I have not been 
worrying much about it after that.

It's more like we've got the RTP nail in place, and are hammering on it 
with the SDP hammer... if we were to use an axe instead (hopefully the 
back end of it), not much in RTP would change.

>
> The place for the RTP discussion is probably not in RTCWEB (more like 
> AVTCORE).
>
>
I think we've already done most of the discussion needed in AVTCORE - we 
seem to have the dictum of "don't send audio and video on the same 
5-tuple" changed to "if you send audio and video on the same 5-tuple, 
here's what you need to consider", which was really the only critical 
*change* I saw in RTP-land.

>