RE: [Fwd: Upgrading Saxon XSLT 1.0 support addressing

> To be honest, I'd be a lot more interested in doing the work myself or
> as part of a team than in paying somebody else to do it.

Most of the work, of course, is not in fixing the small number of known bugs
(the code changes are already applied) but in resurrecting all the
regression tests, running them, and identifying which of the differences in
results are significant, and which are an accident of all the environmental
changes that have happened over the last couple of years like changes in DOM
implementations - and then deciding which of those accidents needs further
changes made. 

I'm not sure that a release that simply built a new JAR file without
achieving that level of quality assurance would be of much value to the
community - otherwise I'd happily donate the few hours that it would take.
There'd be a strong risk that the new version might introduce more problems
than it solves.

There are also lots of little things to be done like checking the changes
that have been made to the JAXP specifications that have been made over the
years (for example, there is now more precise definition of "errors" versus
"warnings") and seeing which of these changes could or should be reflected
in a new release - which depends on a rather fine judgement of the benefits
of conforming to the latest specs versus the benefits of leaving things as
they are.

Michael Kay
http://www.saxonica.com/