[JDEV] Voice over IP

[Greg Boulter]

>I have skim read this draft. Generally I like it, but there are things
>I dont like about it.

Whereas I just want something I could implement now. Given that there 
doesn't seem to have been a lot of discussion about it .....

>- Why this draft is so hidden? If you didn't sent links here I will
>   probably not know it exists. Thats why I would prefer it as JEP.

.... and that so far I too have just skim read the draft I can't really 
comment on how much I like it or not. My question is very practical. I have 
to come up with something now. I'm perfectly capable of doing it my own way 
so that my client can recognise when it has connected to another of its kind 
and then as my clients can have different features activated or not the two 
would need to negotiate which features they share that are compatible and 
therefore determine amongst other things the layout of their respective 
graphical interfaces. No use having a "See Contact" button if contact 
doesn't have a camera, etc.

My client, and I don't know what to call it, my "inhouse" name has been 
JabberBot, but I'm sure someone must have called their client that, so I 
thought of "JabberMX" as it is very tightly integrated with the various 
Macromedia MX stuff, but I see someone once said they were releasing a 
JabberMX so anyway, this client can coordinate up to 12 cameras, 12 
microphones, play mp3 streams, load graphics or flash swf files, be 
connected to a bot, view video, transfer documents, interact with vxml, 
recieve telephone communication and respond via speech synth or 
microphone(s).

Or, if the client provider (a website administrator) may have set up the 
configuration files so that my client has access to none of the above, or 
the enduser may have further modified the starting configuration to deny 
access to services that it is capable of but doesn't wish to tell contacts 
how many cameras they have, or even that there is access to a camera.

For my own use it's pretty simple, I can easily transfer an array of info 
concerning each client's capabilities and that will take only a short period 
of time. My client also loads modules though, so it is quite possible that 
one client will ask the other to recieve a voice message and the address of 
the module if the recieving client doesn't have the capability at this 
moment.

I figure that it all works out until I come across some client that says 
sure, they can play mp3's, subscribe me to the stream please, or send the 
url, or whatever, my client needs to let it's user know the capabilities of 
each contact's client so I can send a mp3 to contact1, have a video 
conference with contacts 2,3,4 and 5 or translate client 7 into German and 
so on .... which is where I'd rather be, so if it's at all possible to set 
things up as close to the recommended format as I can, I'd like to do that.

I'm no real expert on these protocols and JEPs, I've skimmed numerous 
documents in the course of coding my client, but I'm much more of a copier 
than anything, so I guess I'd like a little direction, it's not as if I 
can't rewrite code later or anything, I'm just wondering if I should just do 
things my own way and keep it as modular as I can so that later I can 
conform to protocol?

I'll make another attempt to look at all those documents a bit more closely 
and appreciate any comments as to how I should go about this. I realize that 
maybe that there is something I've missed, but yes, I am trying to become 
familiar with the protocols but no, it isn't the main thing I do so don't 
hit me if I'm slow.

Greg.

_________________________________________________________________
The new MSN 8: advanced junk mail protection and 2 months FREE*  
http://join.msn.com/?page=features/junkmail