Re: [rtcweb] Referring to 5245bis or 5245?

[Christer Holmberg <[e-mail]>]

Hi,

...

>A lot of this comes down to what bundle says.  Now, I see that bundle depends on both 5245 and its -bis, which seems pretty inconsistent. 
> I don't immediately see any strong reason for bundle to refer to the -bis, though it does refer to the ice-sip-sdp draft, which might be 
>sufficiently implicated as to make the change necessary.  We should ask Christer to confirm this.

I don't think BUNDLE as such requires 5245bis. The references to ice-sdp-sdp could also be replaced with a reference to 5245.

However, the text in section 11.1 says:

   "When an offerer associates a shared address with a bundled "m=" line,
   the offerer MUST associate SDP 'candidate' attributes (and other
   applicable ICE-related media-level SDP attributes) with the "m=" line
   following the procedures in Section 8.1."

...and later:

   "NOTE: The following ICE-related media-level SDP attributes are
   defined in [I-D.ietf-mmusic-ice-sip-sdp]: 'candidiate', 'remote-
   candidates', 'ice-mismatch', 'ice-ufrag', 'ice-pwd', and 'ice-
   pacing'."

The 'ice-pacing' attribute did not exist in 5245, so if we change the reference we would have to remove that.

>I think that if we clarify that either way, then the reference in -dualstack-fairness seems less of a concern; that document
>doesn't need to reference 5245bis, though it would be nice if it could.

I think the document people have been talking about is ice-trickle. Some people seem to think that it requires 5245bis.

Of course, by referencing 5245 we will still have things like aggressive nomination around. That obviously doesn't break RTCWEB, but I understood RTCWEB was the main reason we wanted to get rid of those things...

Also, at IETF#97 we agreed on a number of things that will be defined in draf-ice-sdp-sip, e.g., when a transport change requires a new offer/answer exchange, non-supported m- lines in offers/answers etc. Those things may not be important for RTCWEB, but they could be important for other, SDP O/A-related, RTCWEB dependencies.

Regards,

Christer