[standards-jig] JEP-0045 (MUC) number of users (extendingJEP-0030)

Peter Millard me at pgmillard.com
Fri Feb 13 15:50:53 UTC 2004

Richard Dobson wrote:
> IMO it would be far better to extend it as follows:
>  <iq type='result'
>      from='darkcave at macbeth.shakespeare.lit'
>      to='hag66 at shakespeare.lit/pda'>
>    <query xmlns='http://jabber.org/protocol/disco#info'>
>      <identity
>          category='conference'
>          type='text'
>          name='A Dark Cave'/>
>      <feature var='http://jabber.org/protocol/muc'/>
>      <feature var='muc_temporary'/>
>      <feature var='muc_unmoderated'/>
>      <feature var='muc_nonanonymous'/>
>     <roominfo xmlns='http://jabber.org/protocol/muc#room' desc='The place
> for all good witches!' subject='Fire Burn and Cauldron Bubble'
> occupants='1'/>
>    </query>
>  </iq>

Yes, this is much better.. But whatever gets added into the disco#info results
MUST be proposed as a JEP. It seems like a simple enough extension to write up.
Also consider that what you are basically doing though is reporting name-value
pairs. Stpeter and I talked about possibly using info-bits to extend the results
of disco#info. Using info-bits would provide a more generic framework which
could be re-used for lots of other meta-data.

This is a good discussion, but I think we need to consider using info-bits, and
then simply defining the info-bits which would apply for TC. The reason I think
this is better is mainly from a client's perspective (surprise, surprise :).
When I have a browsing UI which is using disco to generate information, I don't
want to have to teach that browse about 100 different extensions to the
disco#info results. It would be much better to allow it to handle something
generic (like x-data results, info-bits, etc..) and let the user figure out what
each "column" of data means based on the context of their browsing.


More information about the Standards mailing list