[Standards-JIG] media conferencing, flow segmenting

Ulrich Staudinger us at die-horde.de
Fri Mar 12 15:50:25 UTC 2004


CORVOYSIER David FTRD/DMI/REN wrote:

>I don't understand what you mean by "Related to SIP/H323": do you mean
>you will actually use these protocols ? Or is it just that this step
>corresponds to an XMPP equivalent of these protocol (I do agree with
>that).
>  
>
I'd prefer to use a XMPP equivalent. "Related to SIP" means the 
functionality of this step is the same as the SIP functionality.

>If you really mean using SIP or H323, then I don't understand why you
>need a separate SDP step, because both protocols already have it (SIP
>actually uses SDP).
>  
>
I thought SIP just does the calling (RINGING, ACK, EOT) and transmits 
upon accepted call an SDP chunk?

>Another thing: what about NAT issues ? It is often desirable to know
>whether users can reach each other before actually trying to start the
>RTP flow. I think this step may be included between the feature
>negotiation and the calling step. 
>  
>
NAT issues are adressed during "data exchange method negotiation".

>There is another use case too: when you start a 1-2-1 call and switch to
>a conference when someone else joins (a bit tricky actually, perhaps not
>supporting it at all is a valid option).
>
>  
>
No protocol support for directly moving from 1-2-1 to MMUC, though a 
client may support automatic renegotiation.





More information about the Standards mailing list