Re: [Atoca] Requirement D2: "Large Audience"

Mainly I was just suggesting one way of looking at the "how large is 'large'" question.  Although it does highlight some of Martin's questions about how to adapt a session initiation protocol to a one-to-very-many application, if that's what we have in mind.  As he rightly points out, responses don't only happen at the application level.

Thanks!

- Art

On Jan 16, 2011, at 11:22 AM, Hannes Tschofenig <hannes.tschofenig@gmx.net> wrote:

> Hi Art, 
> 
> that's an interesting comment particularly since we do not have a requirement for responses by receivers (or Recipients) of alert messages. I am not sure to whom they would sent their response to. To the originator, to some gateway, etc? 
> 
> Ciao
> Hannes
> 
> On Jan 15, 2011, at 11:58 PM, Art Botterell wrote:
> 
>> One difference in "broadcast" delivery is that it's frequntly impractical and/or undesirable to try to track individual deliveries or process individual responses.
>> 
>> Thanks!
>> 
>> - Art
>> 
>> On Jan 15, 2011, at 11:30 AM, Hannes Tschofenig <hannes.tschofenig@gmx.net> wrote:
>> 
>>> Hi all, 
>>> 
>>> during the ATOCA meeting at IETF#79 Igor raised a question regarding requirement D2:
>>> "
>>> Req-D2:
>>> 
>>>    The protocol solution MUST allow delivery of messages
>>>    simultaneously to a large audience.
>>> "
>>> 
>>> He was asking whether there is a need to put a number for large audience there.
>>> 
>>> Maybe Igor has a specific number in mind that would matter from a protocol design perspective but I do not have such a number that would provide a meaningful distinction. 
>>> 
>>> There is, however, the question what types of problems we expect with very large number of alert message receivers and how we document this aspect in the requirements document. 
>>> 
>>> Suggestions? Thoughts? 
>>> 
>>> Ciao
>>> Hannes
>>> 
>>> _______________________________________________
>>> earlywarning mailing list
>>> earlywarning@ietf.org
>>> https://www.ietf.org/mailman/listinfo/earlywarning
> 