[standards-jig] [jepnews] JEP-0020: Client Feature Negotiation

Iain Shigeoka iainshigeoka at yahoo.com
Mon Mar 4 17:30:17 UTC 2002

On 3/2/02 1:32 PM, "Julian Missig" <julian at jabber.org> wrote:

> On Fri, 2002-03-01 at 20:52, Iain Shigeoka wrote:
>> I can wholeheartedly agree with your sentiment.  To avoid reinventing the
>> wheel, what about using UDDI or some other existing XML lookup standard?
>> Has that been discussed?
> Have you ever looked at UDDI? You are aware one of Jabber's goals is
> simplicity, right? Again, I ask if you've ever looked at UDDI :)

:)  Yup.  But I'm thinking that there are a lot of existing libraries for
UDDI... So it may be complex if you're writing everything from scratch, but
simple if you use existing tools.  The same complexity argument could be
used against XML.  A fixed code syntax ala SMTP would be simpler to parse...
But with existing tools, XML is a breeze.

I'm becoming increasingly distressed at the "not invented here" trend that
seems to be emerging from the JEPs.  One of the major advantages that Jabber
has with its XML foundation is the ability to fully exploit all the existing
XML standards out there.  No standard will be perfect so there will always
be a temptation for reinventing it just a bit better.  However, the more we
keep diverging from the rest of the XML world, the less relevant Jabber
becomes for anything beyond IM within the Jabber network.

If Jabber uses UDDI, we can very easily plug into existing UDDI directories,
access web services via SOAP, etc, etc.  If we have our own Jabber lookup,
Jabber negotiation protocols, Jabber database access, etc we will continue
to exclude ourselves from the vibrant XML community out there.

Are we really that much smarter than the rest of the world that we can
reinvent _significantly_ better solutions to these same problems while still
meeting needs?  I think we should have to be able to justify at least a
10-fold improvement over an existing standard in order to diverge... I don't
think that exists here.  Perhaps selecting a simpler subset of the standard
would be better...


Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com

More information about the Standards mailing list