Re: [rtcweb] RTCweb signalling overview

9 sep 2011 kl. 15:08 skrev DRAGE, Keith (Keith):

> With respect to question 3 in this set.
>  
> As I said on the call, the requirements for what needs to be standardised between the two web servers depends on whether web server A needs to know anything about user B, and whether web server B needs to know anything about user A. I believe this goes beyond SDP, because it may need to be information beyond the media contents, e.g. it may need to include information about each user’s capabilities and preferences.
>  
> I actually have two slightly inconsistent views about this interface.
How good it feels to not be alone and confused :-)

>  
> Yes it does need to be standardised. I don’t like the idea of fragmentation being forced on the market because an appropriate standardised solution has not been identified.
Exactly.
>  
> No RTCWEB should not standardise it because it is out of scope of RTCWEB.
Yes, agree.

>  
> Surely this is also the interface by which support of interworking with legacy systems has to be attained?
Well, if the app developer wants to interface legacy systems. The important thing for rtcweb is to understand whether access to legacy systems affects the rtcweb architecture if the signalling is removed from scope.

>  
> Perhaps the easiest way out is to identify that full blow SIP is a solution for this specific interface, and RTCWEB identifies to SIPCORE as to whether there are any additional requirements that SIP cannot meet.
In addition, I see XMPP to be a solution for that specific interface too... 

/O
>  
> Regards
>  
> Keith
>  
>  
>  
> From: rtcweb-bounces@ietf.org [mailto:rtcweb-bounces@ietf.org] On Behalf Of Harald Alvestrand
> Sent: 09 September 2011 08:26
> To: rtcweb@ietf.org
> Subject: Re: [rtcweb] RTCweb signalling overview
>  
> On 09/08/11 20:48, Olle E. Johansson wrote:
> For those of you that did not participate in today's meeting, there was an excellent overview presented by Martin Kaufman.
>  
> It gives you an overview over the issues with signalling - to sip or not to sip - and other issues. Do read it.
>  
> Use the file rtcweb-3.pptx in http://www.ietf.org/proceedings/82/slides/
> 
> Seconded.
> I liked the presentation even though I don't agree with the conclusions (I prefer Cullen's set).
> 
> 
>  
> /O
>  
>  
>  
>  
> _______________________________________________
> rtcweb mailing list
> rtcweb@ietf.org
> https://www.ietf.org/mailman/listinfo/rtcweb
>  
> _______________________________________________
> rtcweb mailing list
> rtcweb@ietf.org
> https://www.ietf.org/mailman/listinfo/rtcweb

---
* Olle E Johansson - oej@edvina.net
* Cell phone +46 70 593 68 51, Office +46 8 96 40 20, Sweden