XHTML 1.0 returned to HTML WG

[Arnold, Curt]

Tim Berners-Lee wrote:

>>Subject: Fw: XHTML 1.0 returned to HTML WG
>>In summary, W3C Members wanted the HTML working group to revise
>>the XHTML 1.0 specification to utilize a single namespace.

I think the motivation for the three namespaces and the fact that there
hasn't been a clean, popular resolution to the underlying problem is a
result that there is not a clean way to distinguish between what a
DTD/schema is capable of representing and what an application is capable of
processing.  I know namespaces aren't equivalent to a DTD or Schema, but I
think that in this case they are felt to be the moral equivalent of a DTD.
I've tried to expand a little on this in comments on the Schema drafts, but
I think it is an important distinction.

For example, a well defined address schema should be able to handle non-US
addresses.  A particular mailing label printer might only support US
addresses.  It would be nice if the XML infrastructure would allow a routing
program to recognize that a particular address did not conform to the
capabilities of the US mailing label printer and route it to an non-US
capable printer.  One option would be to cripple the schema, but this
inhibits those applications that are non-US capable.

Basically, what this would require is the document to be able to declare
what DTD/Schema that it governs it (either through a <!DOCTYPE or through
whatever mechanism the Schema group decides to associate a schema with a
document) and optionally what usage profiles to which the document conforms.


To me, declaration of usage profiles fit the definition of a processing
instruction (I don't know what the implications of using a PI with the
existing HTML browsers is however).  It is a hint to the processor that the
XML document that follows only uses a subset of the available constructs.
The processor would eventually determine if the XHTML uses frames when it
encountered the <FRAME> element, the PI would just let it know before hand.

A question would be should we try to develop a standard way of designating
document profiles or provide an XHTML specific manner.

If we were just addressing, XHTML we could do something line

<?xhtml frameset="false" strict="true"?>

If we were trying to be more generic, could we do something like

<?xml profile="uri uri uri"?>

To list the profiles that the document adheres to.  For XHTML the uri's
would be urn's that identify the frameset, strict and transitional profiles.
For other applications, these could be url's that point to DTD's or a schema
profile.


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 unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe 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)