Re: XML Schema becomes a W3C Recommendation

[Rick Jelliffe <ricko@allette.com.au>]

From: Eric van der Vlist <vdv@dyomedea.com>

>RSS 1.0 can't be described by W3C XML Schema (nor by a DTD).

>A W3C XML Schema on RSS 1.0 would have been incompatible with 2 of our
>requirements:

>1) Simplicity: W3C XML Schema would have imposed an order on elements.
>2) Modularity: you can't say that you want to allow any element from any
>namespace from undeclared modules with W3C XML Schema.

This brings up the old question of whether a schema defines or models (the
"document type declaration" versus "document type definitions" issue.) If we
say that the schema models the requirement (i.e., there may be some
constraints which are not checked) then an XS schema certainly can made for
RSS.

It is like the early claim that a DTD could not be made for RDF. However, a
DTD fragment certainly could, and the ANY declared content type model some
elements nicely.

I am not up-to-date with RSS recently: can RELAX and TREX cope with its
model OK?  I believe Schematron can.

Cheers
Rick Jelliffe