RE: [Simple] Change to prescaps XML schema

[mikko.lonnfors at nokia.com]

> > > The softphone will be able to send and receive audio but 
> > only receive
> > > video. If I understand the prescaps draft correctly, this is not
> > > possible to express. Either the whole service is full 
> duplex or half
> > > duplex.
> > 
> 
> In this case video is not even half duplex but simplex into 
> one direction. The question is whether we can ever get to 
> this kind of detail within presence. On this list I have seen 
> people saying that details such as supported codecs etc. are 
> outside the scope of presence and will be negotiated with SDP 
> once the session establishment is in progress. You can 
> similarly find out the limitation you explain below through SDP.

Right. As <duplex> is defined right now it applies to *all* media elements like <voice>, <video> and etc that are listed in a tuple. Now the question is does that make sense? If it doesn't then we have two option: remove it altogether or define it so that it can appear under media (voice,...) elements.
 
> But the main point I wanted to say was that instead of 
> half-duplex the correct way would be defining something like 
> "receive-only". 
> 
> > Yes, that is correct. Basically this 'limitation' comes from 
> > caller preferences because prescaps draft tries be very close 
> > to the model presented in caller preferences. This is why you 
> > can now only specify duplex for a whole contact. However, in 
> > presence it actually might make sense to be able to specify 
> > different duplex modes for different medias in a single tuple. 
> > Other way to solve this could be to split the service into 
> > multiple tuples which map into single device (softphone). One 
> > tuple would have full duplex audio and other one half 
> duplex video.  
> > 
> 
> That's not right either, since the service capability is 
> really the combination of those two medias, and there are no 
> two separate services, as that kind of tuple splitting would suggest.

Yes, I think you are right.

- Mikko


> 

_______________________________________________
Simple mailing list
Simple at ietf.org
https://www1.ietf.org/mailman/listinfo/simple