XPath Transform, 3rd version

Hi all,

We must be getting close to completion because it only took an hour to tweak
the document this time, including the changes suggested today by Petteri
Stenius and TAMURA Kent as well as prior feedback from Jonathan Marsh about
removing the serialization as an addition to the function library.

TAMURA-san: You posted a question about order of namespace nodes.  The
paragraph about serializing the namespace and attribute axes states that the
primary key for lex order of nodes on these axes is the namespace URI, so I
don't know what the problem is.  Do you see a problem with using namespace
URI (e.g. are you thinking we need to always add the default definition for
xmlns to the initial evaluation context)?

All: Is it an urgent issue that the parsing requirements still specify the
need for retaining the namespace prefix from the input document?  We can
actually change the serialization process so that this is not required, but
I'd rather not do that if there is wide availability of parsers that retain
the original QName of elements, etc.

Editors: Please consider this new draft of the XPath transform specification
for the candidate recommendation.  The changes are based on last-call
feedback from the WG.

Thanks,
John Boyer
Software Development Manager
PureEdge Solutions, Inc. (formerly UWI.Com)
Creating Binding E-Commerce
jboyer@PureEdge.com