This is related to questions of how the finate state machines for schema/dtd validation are generally implemented. There's a lot of SGML experience showing that it _IS_ a pain in the neck to implement "any order, limited number of instances" efficiently, and XML DTDs deliberately eliminated that capability. Schemas reintroduced a limited version of it which they believed could be coded cleanly.
So yes, it's politically a hot button... but there are legitimate technical arguments behind it. We aren't going to settle it here. Unless someone is proposing that Xalan be extended to support a new keyword so folks can write schemas that will run only on Xalan, the right place to argue this is the W3C. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
