The Working Group has considered this comment.  While recognizing the value of
"eating our own dog food", the conclusion is that there would be insufficient
value to justify the resources necessary to generate a meaningfully-different
XSLT 2.0 stylesheet.  Therefore, the Working Group has determined that, because
the spec is not actually broken in this regard, no change to the XSLT version of
the stylesheet will be made. 

This is the official WG response. 

Please let us know if you agree with this resolution of your issue, by adding a
comment to the issue record and changing the Status of the issue to Closed. Or,
if you do not agree with this resolution, please add a comment explaining why.
If you wish to appeal the WG's decision to the Director, then also change the
Status of the record to Reopened. If you wish to record your dissent, but do not
wish to appeal the decision to the Director, then change the Status of the
record to Closed. If we do not hear from you in the next two weeks, we will
assume you agree with the WG decision.

Currently the XSLT1 stylesheet ignores the QName annotations on certain elements
and just takes the literal Qname prefix/local name pairs as names to use in the
generated Xquery expression. One way of fixing this would be to use a schema
aware XSLT2 stylesheet (and also making sure that the schema types used were in
fact, appropriate). It is however possible to make a correctly functioning XSLT1
stylesheet so I may not, in the end, object to the decision to use XSLT1.
However the resources required to fix the XSLT1 coding are similar to those
required to write the thing in XSLT2 and in either case the details depend on
the final resolution of other reports on the use of xsi:type and Qnames in the
XqueryX coding.

As such I'm declining the offer to close this report, and would rather defer
a final decision on this report until 1284 and 1286 are resolved.

David

Thanks for your additional comments.  The Working Group again discussed the
proposition to provide an XSLT 2.0 stylesheet and reconfirmed its decision that
it would not be appropriate to do so at this time.  An additional reason given
is that there are currently relatively few XSLT 2.0 processors and the WG wants
a wide audience to be able to use the stylesheet.

However, your most recent comment on this bug referenced the relationship to
bugs 1284 and 1286.  In a response just posted to bug 1284, we have indicated
that we intend to make changes along the line that you suggested.  We hope that
this will satisfy the specific requirement that QNames be treated in a
namespace-aware fashion instead of merely character strings. 

Once we have made the changes suggested in bug 1284, we encourage you to
reconsider this bug's status and perhaps mark it CLOSED. 

Thanks again for your continuing interest in XQueryX!

Thanks for keeping me informed.

David

Reopening bug for reconsideration

Thanks again for your interest (and patience). 

We believe that we have addressed your xsi:type concerns by changing the
architecture of the XQueryX schema from one based on a type hierarchy to one
based on substitution groups. 

We believe that the new schema's creation of an xqx:Qname type to be used in
lieue of the xs:Qname type used by the previous schema resolves your concerns
about QNames being handled as (xs:)QNames.  Instead, they are now treated as
pairs of string literals (the prefix and the local part), the first of which has
a relationship to a namespace declared in the XQuery represented by the XQueryX
(the "XQueryX level") and no relationsip to any namespaces that may or may not
be declared in the containing XQueryX document itself (the "XML level"). 

Please let us know if this response is satisfactory. If not, please respond to
this message, explaining your additional concerns.

Jim Melton
On behalf of the XML Query Working Group