Re: [MMUSIC] Possible BUNDLE alternative syntax: explicit m-line for bundled session

[Christer Holmberg <[e-mail]>]

Hi,

>>IIUC, The same port offer may be rejected by answerer irrespective of 
>>"m=bundle" or "a=group:BUNDLE" as it is not just new SDP syntax but 
>>also, the same port offer may have impact in other layers like RTP in 
>>answerer side implementation. The bundling of the different media in 
>>the same port is not guaranteed to negotiate in the answer side with 
>>"m=bundle" mechanism. The point is that answerer will reject in case 
>>the same port offer is not accepted in the specific implementation.
>
>If the m=bundle proposal, you don't have to use the same ports. You can assign a separate port for the m=bundle media description.
>
><Partha> I could not understand your explanation here. Please explain how the separate port for m=bundle solves the problem of answerer which does not want to have the same port for its multiple media types. </Partha>

If the answerer does not want to do multiplexing, or does not support multiplexing, it sets the m= bundle port to zero, and uses the separate m=audio/video ports.


>>Apart from this, The individual attribute behavior for bundle has to be 
>>defined somewhere irrespective of "m=bundle" or "a=group:BUNDLE". Say 
>>ptime attribute applies for the audio codecs only or applies for video 
>>codecs as well in the bundle has to be defined. I'm not see much 
>>advantage with "m=bundle" here.
>
>It is already possible to specify attribute values for individual SSRCs. We could do something similar for media types if we want.
>
><Partha> In case of using the same port in all "m" line using group attributes, the specific attributes values works does not need any update </partha>

Yes, but people have raised issues with using the same port in all "m" lines - that's why we've suggested the m=bundle solution.


>>Instead, Let us discuss the generic categories of attributes and mention in this draft.
>
> I am not sure I understood.
>
> <Partha> Paul mentions about some of the categories of the attributes in http://www.ietf.org/mail-archive/web/mmusic/current/msg09467.html. The contention for same port in "m" line mechanism is that "those that are summed across all the 
> bundled media sections". Those issues shall be fixed by the semantic representation. Say bandwidth parameter in bundle group is the sum of all the bandwidth in the specific bundle group. </Partha> 

Yes. But, again, I don't think the attributes have been the main issue. The issue have been using the same port in multiple m- lines in the first place.

Regards,

Christer




-----Original Message-----
From: Christer Holmberg [mailto:[e-mail]]
Sent: Friday, September 21, 2012 10:57 PM
To: Parthasarathi R; 'Lishitao'; 'Ejzak, Richard P (Richard)'; [e-mail]
Subject: RE: [MMUSIC] Possible BUNDLE alternative syntax: explicit m-line for bundled session

Hi,

>I agree with Cullen & others that m=bundle does not have consensus. We 
>need more discussion and analysis. The current Bundle syntax is 
>designed in a
way
>to reuse lot of existing SDP mechanism (Syntax & Semantics). If 
>possible, let us try to fix the open issue with the current mechanism 
>before
proceeding
>with m=bundle approach.

Feel free to provide input on how you think we can solve that issue :)

Regards,

Christer



-----Original Message-----
From: [e-mail] [mailto:[e-mail]] On Behalf Of Lishitao
Sent: Friday, September 21, 2012 7:21 AM
To: Christer Holmberg; Ejzak, Richard P (Richard); [e-mail]
Subject: Re: [MMUSIC] Possible BUNDLE alternative syntax: explicit m-line for bundled session



> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]] On 
> Behalf Of Christer Holmberg
> Sent: Friday, September 21, 2012 3:32 AM
> To: Ejzak, Richard P (Richard); [e-mail]
> Subject: Re: [MMUSIC] Possible BUNDLE alternative syntax: explicit 
> m-line
for
> bundled session
>
> Hi,
>
> >>> m=bundle requires significant extensions to the syntax and also a 
> >>> near doubling in the size of the SDP.  It is harder to specify 
> >>> allowed combinations of codecs and other options that are normally 
> >>> associated with a single m line.
> >>
> >> Please give an example.
> >
> > To specify two types of audio flows with very different 
> > characteristics
I would
> normally have separate media lines with separate codec lists and
attributes.
> One may require DTMF and the other silence suppression in addition to 
> the selected (but different)
> > codecs.  We need to keep this characteristic.  We also need a way of
> specifying bandwidth per (old) media line, which is already supported
using
> separate media lines but requires that something new be defined for
m=bundle.
> What about RTCP
> > characteristics?  We certainly want to specify them per (old) media
line.
> >
> > I understand that this can be done with m=bundle, but it requires 
> > new
syntax.
> Why bother if we have something that can work?
>
> I am not sure it requires new syntax, if you e.g. can use the ssrc
attribute in
> order to map attributes to specific flows.

I would suggest that put the m=bundle proposal in a separate new personal draft or in the annex of the existing working draft, and make a comparison for this proposal and the existing proposal. That will make it more clear.
For the existing proposal, I believe that people are not totally against, although it has some issues. (the m=bundle proposal also has some issues).

Regards
shitao

_______________________________________________
mmusic mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/mmusic==