W3C home > Mailing lists > Public > xml-uri@w3.org > May 2000

Re: Irony heaped on irony

From: Dan Connolly <connolly@w3.org>
Date: Fri, 19 May 2000 05:16:29 -0500
Message-ID: <3925147D.C0859468@w3.org>
To: Ron Daniel <rdaniel@metacode.com>
CC: David Megginson <david@megginson.com>, xml-dev@xml.org, xml-uri@w3.org
Ron Daniel wrote:
> 
> I strongly agree with David Megginson when he says:
> 
> > The schema
> > for schemas (and others) should reference an XML schema for the xml:
> > Namespace using the xsi:schemaLocation attribute, as in
> >
> >   xsi:schemaLocation="http://www.w3.org/XML/1998/namespace
> >
> > http://www.w3.org/XML/Schemas/xmlschema-20000518.xsd"
> >
> >
> [Ron Daniel]
> 
> I very commonly construct DTDs (and assume I'll be doing
> Schemas RSN) that mix and match elements from multiple
> namespaces. They regularly feature elements from the
> RDF, Dublin Core, and XHTML namespaces, plus some
> custom elements from customer-specific namespaces.
> A particular mix is chosen to meet a customer's needs,
> and the DTD enforces the constraints of a particular
> mix.

Very well... I don't see how this conflicts with
the owner of the namespace name publishing definitive
material about their namespace.

> The Dublin Core may define dc:Description, but
> that has nothing to do with whether paragraph breaks
> will or won't be allowed in those descriptions.

It may or may not; the Dublin Core may well define
a content model for dc:description that excludes
or requires paragraph breaks.

> XHTML defines a way for me to indicate paragraph breaks,
> but it can't know in advance if a particular customer
> will or won't need images and tables in their Dublin
> Core Descriptions. These specific constraints are the
> role of the DTD and schema, not the namespace.

You may use customer-specific schemas in stead of
namespace-owner-provided schemas for your work,
but again, I don't see how making namespace-owner-provided
schemas conflicts with using customer-specific
schemas.

Case in point: we haven't made an XML Schema available
at http://www.w3.org/XML/1999/xhtml yet, so
in the stuff that I'm hacking on, I use schemaLocation
to point to my hacked XHTML schema:

"
<html
 xmlns="http://www.w3.org/1999/xhtml"
 xmlns:xsi='http://www.w3.org/1999/XMLSchema-instance'
 xmlns:t='http://www.w3.org/2000/TR/smil-animation10'
 xsi:schemaLocation="http://www.w3.org/1999/xhtml html.xsd
         http://www.w3.org/2000/TR/smil-animation10 smil-animation.xsd
                "
  >
"
	-- http://www.w3.org/XML/2000/04schema-hacking/h+s.html


If there were an XML schema available at
http://www.w3.org/1999/xhtml , it would have no impact
on the output of xmlschema-check when it validates h+s.html


> Lets keep the space of names and the syntactic constraints
> on particular document classes as very separate things.
> Merging them will reduce, not enhance, the ability to
> mix and match elements from multiple namespaces in order
> to meet people's needs.

Please elaborate in detail on how it will "reduce the ability to ...".
I'm confident, based on experience with running code, that
it does not.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Friday, 19 May 2000 06:16:46 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:32:42 UTC