xml-dev - Re: Animal-friends implemented as a pattern (Re: XML-Data:advantages ove

["Henry S. Thompson" <ht@cogsci.ed.ac.uk>]

Note that as written Rick's solution lacks a feature of the XML-Data
proposal, namely that e.g. in the internal subset I can add a new
declaration

<elementType id="stick-insect" extends="#pet"/>
  <empty/>
  <attribute id='colour'/>
</elementType>

and non-intrusively extend the content model of animal-friends.  To
cover this Rick's solution would need place-holding empty parameter
entities in most of his existing entities, e.g.

<!ENTITY % animal-friends-content 
	" ( pet | cat | dog %extra-animal-friends-content )* "

so that you could do

<!ENTITY % extra-animal-friends-content '| stick-insect'>

[Note this is not valid XML, I don't think]

This I think completes the reductio -- the point is not that you can
do things with schemata that you can't do in XML, but that you can do
them in ways which are vastly more transparent and maintainable.  Just
because we CAN write all logical formulae using only Shaeffer stroke
and constants doesn't mean we SHOULD do so.  Occam didn't say "Don't
proliferate", he said "Don't proliferate beyond necessity".

Note also that I argued at the XML day in Montreal that to avoid the
dangers of multiple incompatible approaches to schemata, we should
always provide a semantics in terms of vanilla XML, which is how I'd
describe what Rick has shown is possible!

ht

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo@ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)