RE: Language is not markup and markup is not language.

>You might enjoy writing
>
><expr operator="+"><operand>2</operand><operand>2</operand></expr>
>
>Call me old-fashioned, but I prefer 2+2.

Call me old-fashioned, I prefer writing:

<script>
if (someObject.someProperty == 10)
	...
else
	...
</script>

to the <xsl:when> (or whatever it is) that's required by the current draft.

The XSL WG chose to squeeze procedural features into XML syntax rather than
using something familiar.  Then, they created their own syntax* for
patterns rather than using XML, with all the attendent parsing benefits.  A
very good case could be made for either doing it the other way around
(leave procedural stuff in the familiar script/programatic form but express
patterns in XML, perhaps something like
http://www.publishanywhere.com/xsl/patterns-as-xml-1a.html) or putting both
in XML syntax.

*Finally there's an acknowledged effort underway to unify with XPointer
syntax, so "their own" no longer applies directly.  But, the point remains
that patterns can be pretty complicated and that there's lots of hidden
meanings in the / [] () stuff.

$0.25,

Scott



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