On the whole Mark, I believe you're right =)

So far, none of the existing XSLT implementations have been too concerned
with performance, although there was a particular thread about a dynamic
optimization technique a bit back.

Two cutting-edge areas for research are the optimization issues you raise
and the feasibility of incremental processing.

I'd be happy to discuss implementation issues as I'm considering the same
things.

. . . Sean.



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