Re: Does XSD/RNG Based RDF in XHTML need a new FPI?

Joseph Reagle <reagle@w3.org> wrote:

> > Unless there is a normative schema for RDF/XML, we are not going to
> > define a new XHTML host language document type.
> 
> But what of your (1) XHTML foreign lax schema,

It's not defining a new XHTML host language document type at all, nor
does it attempt to use XHTML Modularization Framework.

> or (2) a XHTML+RDF RNG?

Probably we cannot define a reasonable schema with RELAX NG alone.

> If I 
> write a profile spec that builds on top of the concept of a lax XHTML 
> document with embedded RDF -- and also specify how to extract the RDF -- 
> what would I have to do to [1] with respect to FPIs, the media type served, 
> or anything else?

XHTML Modularization is just a tool to address certain type of needs.
It doesn't try to solve all the problems of mixing namespaces.
Specifically, the XHTML host language document type conformance
defines a set of rules in order to be a member of the "XHTML Family"
document types.  It is intended to be used by language designers
who wish to define their own document type by subsetting and/or
extending XHTML *and* want it to be called an XHTML Family member.
The latter part is the key - they may use the term "XHTML" in their
proprietary document type so long as it adheres to the set of rules
defined in the XHTML host language document type conformance.
This is primarily for "tight" integration, to define a specific
vocabulary based on XHTML.

I don't think RDF fits into this category.  RDF is not a specific
vocabulary, but a framework.  Maybe we could define XHTML+DC or
XHTML+CC, but defining XHTML+RDF is like defining XHTML+arbitraryXML.
We could write such a schema in some schema language, but I don't
think the XHTML host language document type conformance is worthwhile
here.  It was not designed for this task.

> Or is [1] find just as it is?
> 
> [1] http://www.w3.org/2003/06/xhtml-rdf/test-cc.html

If you stick to the DTD, it's just invalid, regardless of the XHTML
host language document type conformance.  You cannot claim it's
XHTML 1.0 Transitional document type (which is a specific XHTML Family
member with fixed set of elements and attributes).

Regards,
-- 
Masayasu Ishikawa / mimasa@w3.org
W3C - World Wide Web Consortium

Received on Friday, 4 July 2003 06:51:32 UTC