- From: Robin Berjon <robin.berjon@expway.fr>
- Date: Wed, 29 Jun 2005 11:53:46 +0200
- To: noah_mendelsohn@us.ibm.com
- Cc: www-tag <www-tag@w3.org>
noah_mendelsohn@us.ibm.com wrote:
> For the element example
> you give, I'm less sure. Shouldn't I be able to write a URI identifier
> for {XHTML,p} without first writing a schema? Not sure.
Is there any reason why doing so wouldn't be entirely orthogonal to
whether a schema (or multiple, possibly using multiple schema languages)
exists?
It's more than very likely that one would like to identify xhtml:p
elements in order to make statements about them (eg. this element is
used to structure text into paragraphs. Paragraphs can have a gazillion
variations of semantics and intents depending on language, culture,
time, author, manual of styles, etc.) and to do so without even pausing
to think about which profile or version of XHTML 1.x such a 'p' element
occurs in and even less about who wrote the schema, whether it's
normative, whether it uses DTDs, WXS, RNG, whether it's been recombined
with other schemata to create a compound language, etc.
--
Robin Berjon
Senior Research Scientist
Expway, http://expway.com/
Received on Thursday, 30 June 2005 01:20:57 UTC