>  as would a node equality predicate so you
> > could test
> > whether two nodes were equal (or one node was in a node set) without
> > resorting to generate-id() or count(.|$ns) = count($ns).
> 
> Less sure on this one, but I think you are almost confirming
> the approach I suggested. Bring the common ones into mainstream XSL
> and keep out scripting.
> 
The XSLT 1.1 approach is to consider features that have been commonly
implemented, but what I'd like to see for this would not be a legal
implementer extension in 1.0: how about "==" for a "same-node" operator,
complete with usual node-set comparison sematics?

Francis.

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