David Carlisle wrote:
> So what is needed is not conformance levels by which to grade
> implementations but "processing modes" by which to control
> schema use (or more often useful, non-use).

Yes, that makes sense, and such a processing mode could go hand-in-hand with
a conformance level. This would allow PSVI-ignorant processors to output an
error when it comes across the "psvi-aware" attribute (or whatever) in the
stylesheet, rather than simply outputting different results. I agree that
this is important and should be heard by the working group. (Of course, when
you marry a processing mode with a conformance level, you've essentially got
a different version of the language, interesting thought...)

However, I think that in practice, a conformance level can still help the
potential for multiple interoperable implementations at *any* level. Having
multiple implementations that behave the same (which seems far more likely
among PSVI-ignorant processors) will certainly impact users in the end. I
could have just requested that users send comments requesting that XSLT 2.0
be implementable so that they'll be able to use it, but I figured that was a
little on the vague side.

Evan Lenz
certainly speaking only for himself


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