[Sipping] Re: Comments on draft-roach-sipping-callcomp-bfcp-00.txt

Gonzalo Camarillo wrote:
> a few comments on draft-roach-sipping-callcomp-bfcp-00.txt.
>
> The draft proposes to to establish a SIP session for the BFCP
> connection. Once the floor is grated, the callee establishes another SIP
> session for the actual call.
>
> The draft should explain how the callee correlates both SIP sessions.
> That is, how does the callee know that the user calling is the same one
> as got the floor. The draft should deal with such correlations for
> anonymous callers as well, where the caller's identity will not be
> available for the correlation.

The intention is that the GRID on the GRUU that the caller uses to 
complete the call will uniquely identify the caller with regards to the 
CCBS service. (That is, it can't be used to necessarily identify them, 
but it does correlate them).

> A question: the draft talks about the establishment of one or more BFCP
> sessions. What do you have in mind? In which cases would UAs establish
> more than one BFCP session?

That paragraph could be made clearer. The intention is: if the INVITE 
forks to, say, three terminals (all of which support call completion), 
then the calling party will set up three BFCP sessions (one with each 
terminal) for the call completion service.

/a

_______________________________________________
Sipping mailing list  https://www1.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-implementors@cs.columbia.edu for questions on current sip
Use sip@ietf.org for new developments of core SIP