Michael Kay wrote:
>Just don't use the tools that don't conform.
>
>It's not actually that hard to write a conformant schema processor, or at
>least to get to the 98% mark where the only non-conformances are in areas
>where the spec writers themselves still have debates about the exact
>meaning. I did it in less than six months with Saxon. If people stopped
>using non-conformant tools, suppliers would stop producing them.
>
>
Give Michael the RMS award for DIY-ism!
Err, less than six months for a great programmer steeped in XML and
involved in discussions
about XQuery/XPath2/XSLT and its type system, and attending conferences,
and
who is on personal terms with many people involved in designing the
language.
It may take a couple of extra days for someone else :-)
(I have heard rumours that some people find the XML Schemas specifications
a little hard to follow in a couple of places.)
Cheers
Rick Jelliffe
