[standards-jig] Presence priority finetuning

Julian Missig julian at jabber.org
Tue Jul 2 20:13:05 UTC 2002


On Tue, 2002-07-02 at 15:29, Sami Haahtinen wrote:
> the first message was to test if the server supports it, the client
> could just dump the request to the server if it's not critical for the
> client to know that the request has been noted.
> 
> if the client is just a normal client it would ofcourse just dump the
> data, normal clients couldn't care less if the server accepted it or
> not. and without that first test you get 2 packets, which would be the
> case with any other implementation too.
> 
> The main thing is that if the client has to worry about not sending the
> first <presence/> before the server has a clear view of the needs of
> client, then the client should worry about sending the initial request
> too.
> 
> Sami

At this point, I don't feel like arguing about this any more:
1) I'd like to have the browse/disco thing decided on
2) I'd like it if they could be modify to present this priority
information
  - if not, fine, but
3) I don't want *FOUR* new steps to the login procedure. browse/disco
would add 2. If you invent a new protocol then please try to keep the
login procedure uncomplicated.

That's all I'm saying because we're really wasting people's mailbox
space at this point.

Julian




More information about the Standards mailing list