RE: Normalize / Simplify HTML-Tables with row-span / c

> but I don't 
> think turning a syntax error into a "feature" is a mild
incompatibility.

First point: the only way to extend a language in a backwards compatible
way is to turn constructs that were errors in 1.0 into features in 2.0.

Second: If we were only talking about syntax, then it might be
reasonable to ask a 2.0 processor to reject new 2.0 constructs. But for
an XSLT 2.0 processor to enforce the 1.0 node-set restriction we would
need to make changes to the data model to distinguish two different
kinds of document node, and we would need to make changes throughout
XPath to define exactly which constructs are and are not permitted on
the two different kinds of document node. That's too high a price to
pay.

If you want a processor that imposes the restrictions in the 1.0 spec,
you've got plenty to choose from and they aren't going to disappear
overnight.

Michael Kay


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list