Namespace URI address resources

[Murray Maloney]

At 10:42 AM 6/14/99 -0700, Tim Bray wrote:
>At 12:40 PM 6/14/99 -0400, Murray Maloney wrote:
>>There is nothing preventing another spec -- say XML Schema -- from
>>leveraging namespace identifiers and adding mechanisms to allow 
>>resolution of associated resources, is there?.
>
>Yes.  I think it would be completely inadmissable for XML schemas to
>try to claim ownership of the resource at the end of the namespace URI.
>(Note that RDF kind of does so, which is a grievous error). 

I think that you may have misunderstood me, so let me try again.

I am not suggesting that XML Schema *claim* ownership.

I am suggesting that a schema-aware processor is allowed to resolve
the URI to an XML Schema, if possible, and use that schema to 
inform its processing.

	<muzmo:foo xmlns:muzmo="http://www.muzmo.com/">
		<html:body xmlns:html="http://w3.org/...">
		...
		</html:body>
	</muzmo:foo>

If there is an XML Schema at www.muzmo.com and there isn;t one at w3c,
then a namespace-aware processor will work just fine on this document.
An XML Schema-aware processor could, additionally, use the schema
to validate the document, notwithstanding that <html:body> is not 
known to the schema at www.muzmo.com, or even that there is no schema
available by resolving the w3.org URI.

>Here's why.  A schema is an interesting external resource that you 
>often need to aid in processing an XML resource.  But it's just one
>of them.  You also sometimes need
> - a DTD
> - stylesheet(s)
> - some related java classes or COM objects
> - a set of documents involved in an extended hyperlink group
> - a topic map
> - some graphics 
> - some other multimedia resources
>None of these are more important than all the rest, and none of them
>has the right to pirate the namespace URI.

Again, I am not talking about claiming ownership. But if I choose
to use URIs that resolve to an XML Schema, then a schema-aware
processor cannot be forbidden from taking advantage of that.

I guess that that would make it fair game for any processor to 
resolve the URI and test whether or not it returns a document
that is interesting to it.

It is interesting that some, but not all, of the external resources 
that you listed share the characteristic of defining 'names' and thus 
'namespaces'.

The DTD can, of course, be included in the internal or external subset.

Other resource types listed above seem unlikely candidates for namespace
URIs. For example, while it is certainly legal, it would probably
be a bad idea to use the URL that actually resolves to a GIF file.
>
>It is becoming painfully obvious that we need a general-purpose 
>packaging mechanism to deliver an arbitrary number of related 
>whatevers along with a piece of XML payload.  There has been a lot
>of discussion about this around the W3C.  It may be the case that
>multipart-mime provides a general solution for this problem (don't
>understand it well enough myself to have an opinion), or perhaps
>we need an XML Packaging Language to use for this purpose. -Tim
>
Yes, that is almost certainly true.

And, perhaps XML Schemas can leverage the namespaces spec.


----------------------------------------------------------
Murray Maloney, Esq.          Phone: (905) 509-9120
Muzmo Communication Inc.      Fax:   (905) 509-8637
671 Cowan Circle              Email: murray at muzmo.com
Pickering, Ontario 		Web:   http://www.muzmo.com
Canada, L1W 3K6    		

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)