Re: [MMUSIC] bundle-11: Bundle Address Synchronization (BAS)

[Christer Holmberg <[e-mail]>]

Hi,

>Bundle Address Synchronization (BAS) is motivated by almost the same >arguments that led to the ICE updated offer/answer in 5245.
>
>Now, at IETF 90 the updated offer/answer procedures for ICE were discussed >and the consensus was to change the existing behavior to:
>" you do it (i.e. the updated offer/answer) or don't do it, but you indicate in >the o/a if you're going to do it. Basically want to have option to never do the >updated O/A exchange, with an indication in the signaling that you will not >be doing it. Of course can always initiate a new O/A exchange whenever you >want to."

Actually, that consensus is still to be confirmed on the list...

>I think the same discussion and obviously  its result also applies to BUNDLE >and BAS.
>Otherwise, if you use BUNDLE together with ICE, the updated offer/answer >cycle is back.
>If the WG agrees, this would require some additional parameter/attribute in >the offer that indicates if BAS will be performed or not.
>
>Some common approach for the updated offer/answer issue might also be >justified.
>An obvious thought  would be if re-using the same "thing" for ICE and >BUNDLE would be appropriate.
>... and for CapNeg albeit its lack of deployment

In general, I prefer a negotiation model, rather than the sender indicating that it is going to do something - without the receiver being able to impact that.

Because, in case of the updated offer (no matter whether it's ICE or BAS), it is often the receiver (or some intermediary) - not the sender - that knows whether it's needed or not.

Regards,

Christer