Re: [MMUSIC] NEED WG CONSENSUS: media-loopback is all-or-nothing

[Gunnar Hellström <[e-mail]>]

NO

There are unforeseeable effects of requiring all or nothing.

What if a new RTP based medium is registered, e.g. m=aroma.
Then the logic would require that the media-loopback specification would 
need to be updated before m=aroma could be used at all.

There is also an inconsistency in that 3.1 requires all m-lines in the 
sdp to include loopback payload type and the draft is valid only for RTP 
media. There are other media also declared by m-lines, e.g. m=message 
for MSRP, and many m=application are not RTP-based.
A proposal taking care of this problem, but not the one above about new 
media would be to in 3.1 in the phrase
"

each individual media description"
say
"each individual RTP-based media description"

/Gunnar



On 2012-11-20 19:42, Paul Kyzivat wrote:
> NO
>
> *Why* should the draft be limited this way if it need not be?
>
>     Thanks,
>     Paul
>
> On 11/20/12 1:16 PM, Hadriel Kaplan wrote:
>> Howdy,
>> during IESG review of the media-loopback draft, a discussion arose 
>> regarding the language around indicating media-loopback support per 
>> media description (i.e., per m-line), in sections 3.1 and 3.2.
>>
>> My interpretation/understanding was we were requiring ALL m-lines to 
>> indicate media-loopback; in other words it's an all-or-nothing in 
>> both the offer+answer, and either every media stream is looped back, 
>> or none are.  However the text currently doesn't make this clear, and 
>> it could be argued either way, with pro's/con's either way too; so 
>> I'd like the WG's consensus on whether it should in fact be an 
>> all-or-none approach, or not.
>>
>> Please respond to this email by **Tuesday, December 4th** with a 
>> "YES" if the draft should mandate ALL media be looped-back, or "NO" 
>> if it should allow a hybrid approach of some streams being looped 
>> while others are not.
>>
>> For reference, the draft is here:
>> http://tools.ietf.org/html/draft-ietf-mmusic-media-loopback-24
>>
>> -hadriel
>>
>> _______________________________________________
>> mmusic mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/mmusic
>>
>
> _______________________________________________
> mmusic mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/mmusic