Re: [rtcweb] Making progress on the signaling discussion (NB: Action items enclosed!)

[Magnus Westerlund <[e-mail]>]

On 2011-10-05 04:15, Matthew Kaufman wrote:
> On 10/4/2011 9:01 AM, Ted Hardie wrote:
>> At today's Chairs call, Cullen, Magnus and I had a discussion of how 
>> to make progress on the signaling discussion.  We feel the mailing 
>> list discussion needs to have more concrete proposals in order to make 
>> progress, and so we're putting forward the following:
>>
>> 1) If you plan to put forward a draft proposing a concrete solution in 
>> this space, please send your name to the mailing list with that intent 
>> by October 7th *THIS FRIDAY*.
> 
> I'm going to nominate myself here.
> 
>>
>> 2) Please have a -00 draft out for discussion by October 14th (the 
>> following Friday).  This is to allow for a discussion and update prior 
>> to the -01 deadlines.
>>
> 
> I believe that defining a signaling protocol between web browsers and 
> web servers for RTCWEB is out of scope, so I believe there is no draft 
> required... unless you really want a -00 draft that says "do nothing".
> 
> I believe that defining a signaling protocol between web servers and 
> other web servers for RTCWEB federation is out for scope for "version 
> 1", so I believe it is not appropriate to submit a draft at this time... 
> unless you really want a -00 draft that says "do nothing for now, use 
> any of the existing protocols until then."
> 

I think also this proposal do require an Internet draft which discusses
the requirements put on the API. What of the underlying protocols that
are this WG task needs to be exposed in different forms?

My understanding of your idea is that you have both capabilities and
knobs that can be set. I also believe it requires some state
transitioning events. For example to kick of ICE negotiation between the
peers when the right things has been set.

Thus, I personally do see a need for draft from you that explains the
details of how things are expected to work. I don't see you needing to
define how the API should really look, but what the requirements on that
API are based on your solutions. But without that your proposal is still
not clear and open for a lot of interpretation.

Cheers

Magnus Westerlund

----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM
----------------------------------------------------------------------
Ericsson AB                | Phone  [phone]
Färögatan 6                | Mobile [phone]
SE-164 80 Stockholm, Sweden| mailto: [e-mail]
----------------------------------------------------------------------