[jdev] Re: presence stanza processing with group chat

[Kevin Smith]

On 31 Jul 2006, at 19:42, Magnus Henoch wrote:
> Hm... if the requirement in JEP-0045, section 7.1.2, were loosened
> (currently the MUC service MUST accept groupchat 1.0-style join
> requests), you could have one server for both accounts and chat rooms
> - I think there is no occasion when a client should send presence with
> <muc:x/> to something else than a chat room, so the server could
> bounce MUC presences to existing accounts, and create a room when the
> account doesn't exist.

I think this is possible anyway isn't it? The server knows what  
accounts it has and can determine that way if something is directed  
presence or a muc/gc join? Indeed, one could feasibly make some  
artificial jid-limiting if they wanted (all jids starting "muc-" are  
muc rooms) or somesuch.  I think the seperation of components is  
mostly a matter of convenience, rather than protocol (of course, I'm  
about to be proven wrong).

/K

-- 
Kevin Smith, PhD Student, Computer Science, Exeter University
http://www.dcs.ex.ac.uk/people/kismith
Psi XMPP client project leader - http://psi-im.org