Re: is XSLT 2.0 implementable? (was: N : M transforma

  If, as many people on this list are predicting, users don't care a damn
  about having schema support, then they are unlikely to miss it. The
  market will decide.

But they have to pay the price of vastly increased complexity in the
specification (and one assumes in the text books that will follow).


They also have already paid the price of several years delay in getting
out a usable successor recommendation to XSLT1 that standardises many of
the features that were recognised as missing even before XSLT1 was a
rec. (rtf-nodeset/grouping/multiple output etc, also user xslt-defined
functions which is another good part of the current XSLT2).

As the conformance levels are not specified it's hard to comment on them
but it would seem likely given the current XSLT draft that given a
document and a stylesheet which does not refer to a schema explictly
a schema-aware  XSLT engine will produce different results (because it
sees vastly different input) than a non-schema aware processor if the
schema-aware parser used by former happens to validate the document
against a schema of its choice. Simple expressions like
string-length(.)
will no longer inter-operate amongst processors, even amongst schema-aware
processors, some will return the length of the original character data,
some will return the length of the canonical representation of the typed
value. If simple expressions like that can no longer interoperate, what
chance of real sized stylesheets with thousands of such expressions?



David

________________________________________________________________________
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________

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