XHTML and the Three Namespaces

Sat Sep 25 04:11:34 BST 1999

Andrew Layman wrote:

> I believe that you at one point use 'definition' to mean 'mapping' in the
> sense of an architectural form 'definition' providing some projection and
> transformation of the elements in one language to those of another.  When
> I
> write "Anything so identified can have, at most, one definition" I am
> using
> 'definition' in the sense of 'specification of the characteristics of
> something, including constraints and relations that are intended by its
> creator to apply to all instances.'
> 
	Even with that definition the needs to be some clarification. Do
'relations and constraints' include content model, i.e. if an element has a
different content specification or is contained differently in one DTD
versus another is that part of the definition?

	For instance, the <html> element in frameset has different content
than in strict and the <p> element may be contained in a <body> element or a
<noframes> element (indirectly thru body element). If containment matters
then using an XHTML fragment in another XML document changes its definition.

	If I assume containment doesn't matter then why does it when an
XHTML element is contained in another? What's the difference otherwise
between a frameset <p> and a strict <p>?

	The point with architectures is well taken, I could label the same
<p> element as a headline in one architecture and a title in another. The
creator decided different subsets of characteristics applied to the same
physical element depending on architectural view. If I used architectures
and never meant the document to be rendered, then many of the XHTML
characteristics of the element are 'not meant to apply in all instances'

	XML is supposed to be a content language, not a presentation
language. By introducing separate namespaces for the basic different
rendering models of XHTML you have made a presentation dependency. There's a
bit too much of a 'presentation is the meaning' flavor as opposed to
'content is the meaning'. That's fairly natural for HTML, but is not the
view of XML.

	Perhaps a simpler question would help. Why would anyone care whether
an XHTML fragment is strict, transitional or frameset? 

	It would seem only to be the case if the application had limited
processing capabilities, for instance no frames handling. But what would
happen in that case? The application would either be guaranteed never to get
such a fragment (in which case it is moot) or should fail in the graceful
way of HTML browsers. Remember that there is far more commonality than
difference, If I include a <frameset:p> should that cause a failure in an
application that handles only strict?

> Marc B. McDonald
> Principal Software Scientist
> 
> Design Intelligence, Inc.
> 1111 Third Avenue, Suite 1500
> Seattle, WA  98101
> marc.mcdonald at design-intelligence.com
> Ph: 206.343-7797
> Fax: 206.343.7750
> 
> http://www.design-intelligence.com
> 
> 
> 

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)