[Standards-JIG] Proposed: Simplified Client Caps

Trejkaz trejkaz at trypticon.org
Sun Nov 20 23:15:11 UTC 2005


> I don't think it turns the regtrar into the cache. Just like we allow
> extensibility through namespaces we would obviously allow extensibility
> through ext values. The registrar would just maintain a mapping of
> popular, standardized, features to ext values.

You would have to come up with a standard means of namespacing the ext
values themselves.  Otherwise a client might create an ext string called,
say, "avchat", and then the registry might create an ext string with
exactly the same name later on, and the two would not necessarily be
compatible.

I've seen some schemas in the past do this by requiring that all
unofficial values be prefixed by "x-", for example.

But note that even if you do that, if two *separate* clients both use
"x-avchat", then they're still not necessarily compatible, and will
basically still end up having to do a disco#info unless they're exactly
the same client (only now we don't know if they're the same client,
because that information was just ripped out.  Whoops.)

You could possibly come up with some client-name-based namespacing for
that.  This way, some proprietary encryption supported by GAIM and GAIM
alone might be called "gaim-e2ee", or some avchat protocol supported by
iChat alone might be called "ichat-avchat".

TX





More information about the Standards mailing list