[Standards-JIG] Re: roster synchronization

James Bunton james at delx.cjb.net
Wed Sep 15 22:06:23 UTC 2004

Peter Saint-Andre wrote:
> The consensus from this morning's Jabber Council meeting [1] is that we 
> are not comfortable with the presence-based solution described in the 
> Roster Subscription Synchronization proposal [2], and that we would 
> prefer to move forward with a solution based on JEP-0093 [3] or, more 
> likely, a revised version of JEP-0093 (which we would publish as a 
> separate, standards track JEP to supersede JEP-0093). We would also 
> prefer to find a solution that will address both roster synchronization 
> with gateways and the desire for "shared groups" as described in 
> JEP-0140 [4]. Obviously, integrating this with JEP-0100 [5] for gateways 
> would also be required. Since I am author of JEP-0093, JEP-0100, and 
> JEP-0140, I have volunteered to write a suggested solution here (i.e., 
> the replacement for JEP-0093, the modifications to JEP-0100, and a 
> replacement for or serious modification of JEP-0140).
> Peter
> [1] http://tinyurl.com/5fbsp
> [2] http://www.jabber.org/jeps/inbox/rostersync.html
> [3] http://www.jabber.org/jeps/jep-0093.html
> [4] http://www.jabber.org/jeps/jep-0140.html
> [5] http://www.jabber.org/jeps/jep-0140.html

Ok. I'm waiting with great interest to see what you guys can come up with! =)
Just a note, the current revision of JEP0140 really isn't suited to solving 
the problem with gateways. Firstly because there aren't any pubsub 
implementations (client or server) that I know of, and also because pubsub 
doesn't seem to be the right choice for something that's only going to have 
one subscriber.

Any chance roster-subsync could become informational? Because I'll probably 
still end up implementing it in PyMSNt, at least unless/until another 
protocol comes out to supersede it. I can always distribute the JEP with the 
transport, but it's probably nicer to have it all in one place (namely 



More information about the Standards mailing list