Re: ANNOUNCE: Petition to withdraw xsl:script from XSL

On Thu, 1 Mar 2001, Jeni Tennison wrote:
> >> How would you like to see the easy distribution of extension
> >> functions being addressed?
> >
> > For me, I'd like to see things like *plucks examples from this
> > morning's thinking-in-bath session*: RDF parsers (RDF has multiple
> > elements or attribute serialisation options) or - for XSL FO or SVG
> > - exsl:text-depth($text, $box-width, $font-name) distributed as exsl
> > libraries.
> 
> Sure, but what about distributing extension functions that do things
> like returning the current date or working out whether a directory
> exists on the file system? Things that can't be done with EXSLT?

I had in mind a RDDL [1] like mechanism where one could ask for a
"your-favorite-language" implementation of the "some-uri-reference" 
extension function.  Certainly this means going out to the 'net
for implementations.  I include by reference all of the caching
discussions relted to RDDL.  

> Or is your position that such things should only be allowable through
> (community-based or implementer-based) extension functions?

I also think that community-based "extension libraries" would 
be a complementary mechanism.  I view this as a subset of 
the more dynamic full-fledged RDDL case, as the 'cache is
static... ie, you can use RDDL to fetch only those implementations
already found in cache.

Anyway, this was my thought on the item.  I'm sure it's not
perfect by any strech.  But the point was not added to the petition
without thought on its ramifications.

;) Clark

[1] http://www.openhealth.org/RDDL/


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