[Standards-JIG] Client Capabilities (rant)
chris.mullins at coversant.net
Mon Nov 21 21:37:36 UTC 2005
JD Conley wrote:
>> The assumption that Ian made about a single disco#info
>> request certainly would make things simpler... Making
>> separate requests to random contacts for each of the
>> EXT's just makes things way too complicated.
> Yes, but AFAIK it isn't necessary even as the JEP stands. :-)
> Perhaps I'm missing something?
As it sits now, it is necessary. The entire "I've seen this node+version before, so I don't need to disco for features" in fact relies on this fact.
Let's suppose in Exodus 0.9.9.1 that PGM makes XHTML an optional feature. I, as a user, can turn it on or off. He would then need to add "xhtml" into the EXT node in the presence capabilities. If you disco the root caps node (Exodus#0.9.9.1) you MUST NOT get the xhtml feature. It's only when you disco "Exodus#xhtml" that you would get the xhtml feature. This is key to the entire jep.
Let's say you had two clients in your roster with the fictional Exodus 0.9.9.1 - one of them has XHTML on, and one of them has XHTML off. You disco the root node of the 1st client, and discover that xhtml is supported. Because the client node and version of the 2nd user matches that of the first you don't bother doing the disco. You have now incorreclty identified user 2 as having XHTML messaging enabled. This is bad.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 4403 bytes
Desc: not available
More information about the Standards