RE: Personal reply to Edd Dumbill's XML Hack Article wrt W3C XML Schema

["Simon St.Laurent" <simonstl@simonstl.com>]

At 09:06 AM 3/12/01 -0500, David E. Cleary wrote:
>Given that most XML Schema alternatives have adopted XML Schema Datatypes,
>is there anything in the PSVI that would harm alternative schema languages
>from creating one? By agreeing to support the PSVI, other specifications as
>well as alternative schema languages can coexist in an interoperable
>fashion.

I'd suggest taking a closer look at _how_ those specs have adopted XML 
Schema Datatypes.  RELAX, for example, supports only built-in data types, 
and identifies them _without_ the use of namespace prefixes - "string" 
instead of "xsd:string".

That's a lot of what brought me to suggesting a URI-based approach rather a 
QName approach last week, though Jonathan Borden is now suggesting some 
interesting ways of handling QName/URI conversion.  (I'd let QNames go, 
personally, but maybe there's a softer approach.

There's also Eric van der Vlist's notes on datatypes in a previous thread:
http://lists.xml.org/archives/xml-dev/200103/msg00478.html

In general, I'd suggest that problems with XML Schema Part 2 have been 
obscured by the general focus on more obvious difficulties with XML Schema 
Part 1.  Part 2 doesn't read as complicated, setting off fewer warnings, 
but I'd suggest there's still a lot of discussion to be had before we use 
XML Schema Part 2 as a foundation for anything we genuinely care about.


Simon St.Laurent - Associate Editor, O'Reilly and Associates
XML Elements of Style / XML: A Primer, 2nd Ed.
XHTML: Migrating Toward XML
http://www.simonstl.com - XML essays and books