Re: xsl:param may not be used here

On Tuesday, August 28, 2001 9:20 PM
Wendell Piez wrote:

> Hi Manny,
>
> At 03:18 PM 8/28/01, you wrote:
> >
> >My interpretation of everything I read was that the namespace
> >was only providing a way to distinguish between similar tags and that
> >there was not inherent "functionality" associated with the namespace.
>
> Ah. Well, that's more-or-less true except when it comes to XSL.
> [...]

Or, to put Wendell's point in apparently the opposite way, it's true and
really bites *especially when* you come to XSL. The namespace
identifier string determines what you mean by your local use of the
namespace
prefix "xsl" (or any other legal prefix string that you feel like
associating with that identifier) and therefore what the syntax and
semantics of all elements with that prefix is supposed to be. So it is
indeed a way of distinguishing between "similar tags". It  indicates
whether eg <xsl:apply-templates/> is to be acted on according to the
XSLT specification, or the significantly different draft XSL spec used
in the earlier MS processors. Which is why, if your use of elements
prefixed with "xsl" doesn't match the specification behind the namespace
indentifier you've associated with the prefix, you get no or apparently
"wrong" results.

I have a horrible feeling that namespace problems are set to join
encoding and whitespace issues as FAQ-resistant FAQ's here, especially
when schemas come into wider use and future versions of XSLT are able to
access them fully. And that the secondary, local, arbitrary status
of namespace prefixes, which many tutorials skate over when
explaining how namespaces work, is going to be chiefly to blame for the
confusions.

Michael
---------------------------------------------------------
Michael Beddow   http://www.mbeddow.net/
XML and the Humanities page:  http://xml.lexilog.org.uk/
---------------------------------------------------------



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