Re: [Sipping] WGLC for draft-ietf-sipping-reason-header-for-preemption-00.txt

At 08:14 PM 3/1/2004 -0600, Dean Willis wrote:
>James M. Polk wrote:
>>how exactly is informing the UA translate into telling the user why 
>>something has happened always (which is the way I read your question)?
>
>Basic spook information theory.
>
>1) A is talking to B.
>2) A gets pre-empted and knows it.
>3) A reasons that B must have gotten a call from someone more important 
>than A, as Mike assert that's the only reason to get pre-empted. A calls 
>his Evil Master and reports "something's up, B got a pre-emption call".

All above is accurate (though I question the Evil Master angle will apply 
in all cases), but to direct at Mike's comment: He wants the preempted user 
to hear a chime that they have indeed been preempted - he's 
fearing/believing that if the Reason header indicates the type of 
preemption - access or RSVP based - the user will know this difference too 
(from the BYE message) as some unique indication of the two possibilities. 
I contend that just because a UA receives a more specific message doesn't 
mean that translates directly into a UI indication of that level of detail.

Put simply, regardless of the Reason Header value, the user hears the same 
generic chime (thus doesn't know which type of preemption event occurred, 
which is his preferred UAS behavior).

>Issue: There are other reasons why A-B might get pre-empted. For example, 
>C might need to talk to D, requiring a circuit used by A-B, so BOTH A and 
>B get preempted. This helps only slighly.

this is a network (or RSVP) based preemption

>--
>Dean

cheers,
James

                                *******************
                 Truth is not to be argued... it is to be presented

_______________________________________________
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