W3C home > Mailing lists > Public > www-tag@w3.org > December 2002

RE: what's wrong with using XML Schema/HTML/RDF to document names paces?

From: David Orchard <dorchard@bea.com>
Date: Mon, 9 Dec 2002 15:11:40 -0800
To: "'Williams, Stuart'" <skw@hplb.hpl.hp.com>
Cc: <www-tag@w3.org>
Message-ID: <000501c29fd8$55fd2140$6a0ba8c0@beasys.com>

Oh sure, point out the facts from a canonical source.  Ouch. :-)

As I raise myself from the mat after a five-count after that body blow, I
would say that SOAP implementations don't deref these URIs at run-time.
There are many, potentially thousands of, other applications that might make
the mistake of doing so however.  Further, SOAP implementations are often
used in conjunction with WSDL, which defines what the receiver considers
allowable schema for the SOAP messages.

I'm thinking more of the generalized use case of vocabulary creation and
publishing, rather than the clear understanding that some of the arguably
best xml protocol folks around came up with.  The use case I see needing
solved is the problem of where a human finds an instance of a document with
an xmlns declaration, and they have to figure out how to build their
software around it.  I don't think that really applies to SOAP
envelope/header implementors (the soapbuilders group interop testing alone
probably ensures that this problem doesn't occur), but is highly likely in
specific message formats.  And there may (hopefully will be?) thousands of
vocabularies of messages flying around that devs have to implement to.

Just becuase XMLP is a good citizen (in oh so many ways :-) and puts
reasonable documents available at the URI, doesn't mean others will be good
citizens.

If you are looking for a stronger argument against my position btw, it would
be that ws implementations will always use WSDL (rather than any xml
schema ) for defining types and information.  But I have a good
counter-argument to that one as well ;-)

Cheers,
Dave

> -----Original Message-----
> From: www-tag-request@w3.org
> [mailto:www-tag-request@w3.org]On Behalf Of
> Williams, Stuart
> Sent: Monday, December 09, 2002 2:07 PM
> To: 'David Orchard'
> Cc: www-tag@w3.org
> Subject: RE: what's wrong with using XML Schema/HTML/RDF to document
> names paces?
>
>
>
>
> Hi Dave,
>
> > -----Original Message-----
> > From: David Orchard [mailto:dorchard@bea.com]
> > Sent: 09 December 2002 21:41
> >
> > Dan,
> >
> > I have a very big concern that most namespace names will
> become used as
> URIs
> > to XML Schema documents.  That trend is already occuring.
> > And XML Schema documents are, how shall I say it, not very human
> > understandable.  The XMLP group has provided wonderful
> rationale for why
> required machine
> > understandable references (DTDs) in a document are a bad thing.
> Widespread
> > adoption of XML Schemas for namespace name documents will
> cause the same
> > problem(s).  XML Schema documents or DTDs, the logic is the same.
>
> Hmmm.... dereferencing all of:
>
> 	http://www.w3.org/2002/06/soap-envelope
> 	http://www.w3.org/2002/06/soap-faults
> 	http://www.w3.org/2002/06/soap-upgrade
> 	http://www.w3.org/2002/06/soap-encoding
>
> yield what the current SOAP WD describe as normative XML Schema [1].
>
> >
> > Cheers,
> > Dave
>
> Cheers,
>
> Stuart
> [1] http://www.w3.org/TR/2002/WD-soap12-part1-20020626/#notation
>
>
Received on Monday, 9 December 2002 18:13:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:14 GMT