[standards-jig] [jepnews] JEP-0021: Jabber Event Notification Service

Thomas Muldowney temas at box5.net
Mon Mar 4 23:34:19 UTC 2002

Julian and I were just talking about actual URL based namespaces today,
and we both prefer it.  It came up because we were talking about
rewriting avatar JEP once pubsub is agreed upon.  I want to use URLs in
my future JEPs, so I say go ahead and use one =)


On Mon, 2002-03-04 at 17:12, Robert Norris wrote:
> > Just a small comment on the protocol specifics:
> > In most other IQ requests there is a query tag which holds the namespace, 
> > and the actual parameters to the request are encapsulated in the query 
> > element. It would be nice to see that continue or the xmlns declared in 
> > the IQ. This makes for cleaner code when parsing the xml structure.
> > 
> > For example:
> > <iq id='sub1' type='set' from='subscriber-jid to='ens-jid'>
> >    <subscribe xmlns='http://xml.cataclysm.cx/jabber/ens' jid='event-
> > jid'/></iq>
> > 
> > could become:
> > <iq id='sub1' type='set' from='subscriber-jid to='ens-jid'>
> >    <query xmlns='http://xml.cataclysm.cx/jabber/ens'>
> >       <subscribe jid='event-jid'/>
> >    </query>
> > </iq>
> <query/> is a holdover from the days before <iq/>. As far as I can tell,
> most people use <query/> inside <iq/> because its what all the examples
> do. There's no real reason to do it that way, though.
> The semantics are the same either way, and I think not using <query/> is
> cleaner and less cluttered.
> > Does this make sense, or is it too petty?
> > It allows me to be stricter in my validation and to keep validation 
> > feature-local.
> What sort of validation?
> > The other comment was upon acception into the jabber standard, the 
> > namespace probably ought to be changed to jabber:iq:ens or something like 
> > that.
> Possibly. The reason I didn't do it was because of this post to the
> Council list by David Waite:
>   http://mailman.jabber.org/pipermail/council/2002-January/000167.html
> Although this never got past being proposed, I thought it made sense.
> Besides, why do we shove everything into a jabber:* namespace? No other
> XML-based system forces this. Besides, if we start making this a common
> practice, then don't we (very eventually) run the risk of running out of
> meaningful namespace names?
> My personal feeling is that jabber:* should be kept aside for the "core"
> (RFC-defined?) protocol, and that any other extensions should go
> somewhere else - and it should be left to the proposal authors to come
> up with this.
> As an aside, using http://* as a namespace name means that we can place
> information about the namespace at that URL -
>   http://www.w3.org/1999/xhtml is a good example of this.
> -- 
> Robert Norris                                       GPG: 1024D/FC18E6C2
> Email+Jabber: rob at cataclysm.cx                Web: http://cataclysm.cx/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 232 bytes
Desc: This is a digitally signed message part
URL: <http://mail.jabber.org/pipermail/standards/attachments/20020304/89860b68/attachment.sig>

More information about the Standards mailing list