Re: [MMUSIC] RID update (draft-pthatcher-mmusic-rid-01,> )

[Paul Kyzivat <[e-mail]>]

* Section 7.1:

In:

  1. It MUST generate a 'rid-identifier' that is unique within a media
     description

IIUC, the rid-identifier must be unique within an RTP session, including 
all the media descriptions pertaining to that RTP session.

* Section 7.2.2, item 2:

Restating my earlier comment, why not simply check the PT against those 
listed in the m-line? There could be default PTs in use that have no 
rtpmap or fmtp. Those are only defined for audio, but there *could* be 
rid constraints applied to audio. (I think max-br would potentially be 
applicable. Or new ones might be defined that apply.) At least it seems 
harmless to define it this way.

* Section 7.4, item 4:

I think there should be a check that the PTs in the answer are a subset 
of those in the offer.

(And, while it can be inferred, it might be good to state somewhere that 
in cases where the answerer m-line PT values differ from those in the 
offer, the PTs in the answer rid use the offerer's PT values.)

* Section 10:

As somebody noted, it is confusing to have ";" separate multiple PT 
*values* within rid-fmt-list, and also to separate rid-params.

Possible solutions:
- use a different separator for fmt values
- use a different separator for params
- use a "pt=" for *each* fmt.

	Thanks,
	Paul