Re: Re : [Linphone-developers] The possibility of expanding the program.

[Simon Morlat]

Hi,

>From my point view, implementing a kind of Push To Talk feature requires
important changes and thus cannot be done through a plugin mechanisim.
Currently there are two plugin framework:
- the mediastreamer plugin: to add sound support, webcam support, new
codecs
- the liblinphone plugin: whose goal is to provide additional features
such as remote address book, easy registration, but not to interfere
with core media behaviour.

Simon

Le jeudi 14 octobre 2010 à 15:59 +0400, Ogogon !!! a écrit :
> 14.10.10 14:24, Aurelien Bouin wrote:
> > Hi,
> Good afternoon, Aurelien!
> 
> > It is a very very interesting feature
> I hope that is not irony. (I, unfortunately, not very well feel the 
> nuances of the English language.)
> 
> >   but you are talking about sending or receiving the sound(oRTP) 
> > considering SIP signalization(which one ? SIP MESSAGE ?)
> I believe that the half-duplex control channel must be carried out not 
> by means of RTP, but the means of SIP.
> First, the SIP has mechanisms for receiving information about the 
> connection when it was set, and RTP, as I recall, no. And in any case, 
> when connecting to the peer, we need to know it is a duplex or half 
> duplex. Depends on it, how we communicate with him.
> Secondly, SIP, by definition, an extensible protocol connectivity and 
> make it add a conceptually more correct and technically simpler.
> Third, if the extension of SIP commands could not be avoided, it is 
> hardly worth upgrading yet, and RTP. In addition, by means of SIP, we 
> can fully realize all the required functionality to us.
> 
> > I don't think that you can do that with a plugin ... or you will have to 
> > develop your own audio codec that could handle message + audio payload that 
> > come from oRTP...
> Here is another argument in favor of the control half-duplex channel 
> need to be implemented by means of SIP instead of RTP.
> 
> Now, several American companies produce equipment for the control 
> half-duplex radios via VoIP, but control signals to pass through it 
> retouched RTP, but in my opinion, it is fundamentally wrong. These 
> manufacturers have started to adapt to specific needs of the most 
> similar existing developments, in failing to follow existing standards.
> 
> I and a group of colleagues to create a solution for amateur radio.
> We believe the correct observance of standards and an architectural logic.
> Such a decision requires client software, and it should work as a 
> full-duplex, as well as for half-duplex connections.
> 
> > What do you think simon ?
> Excuse me, but what is simon? (Unfortunately, my English is far from 
> perfect.)
> 
> > Aurelien
> Ogogon.
> 
> _______________________________________________
> Linphone-developers mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/linphone-developers