Re: LINQ to XML versus XSLT

>>>>> "Scott" == Scott Trenda <Scott.Trenda@xxxxxxxx> writes:

    Scott> The point I'm trying to convey here is that rather than
    Scott> trying to shoehorn everything into XSLT 2.0 through the
    Scott> vendor's extension functions may not be the best way to go
    Scott> for most webserver tasks.

Well, I think that IS the best way (like Andrew just said
too). Although it doesn't necessarily have to be vendor-specific. In
many cases there is an absolute standard (although implementation
dependent) way of doing these things - that is xsl:result-document
of fn:doc()/document() with the appropriate URI scheme. 

The implementation-dependent bit is whether or not a given
implementation supports a given URI scheme. But both Saxon and Gestalt
(at least) allow you to easily write handlers for additional schemes.

So I think an appropriate way forward is to map out (on this list)
the best way of approaching these things in a standard way, and then
set up shared code repositories for any additional URI handlers
necessary. 

When an extension function is necessary, then exslt used to be the
appropriate forum.

I certainly don't think an W3C additional language is
necessary. Indeed it would be positively harmful in my opinion.
-- 
Colin Adams
Preston Lancashire