Re: using default params?

> Maybe I can ask you some other questions, regarding design decisions
> in the XSL spec:

You can ask, and I can guess answers, but I wasn't there, I'm not on xsl
working group.


> Is there any reason why it isn't possible to handle a RTF as a
> node-set?

No. I think at the beginning they were afraid of giving xslt too much
expressive power, to make it too hard to implement efficiently, but then
they gave it more expressive power anyway and just left this restriction
by accident. saxon and xt provide a node-set extension to get rid of
this.  MSXSL just ignores RTF altogether and makes nodesets directly
(which is what the spec should have said, but which makes msxsl non
compliant) 

>  At the moment I have to use a stupid hack, like 
    <xsl:param name="foo-set" select="document('')//xsl:param[@name='foo']"/>

better to use node-set on systems that support it.
select="xt:node-set($foo)"/>

> and the same goes for use-attribute-sets: why isn't it possible to use
> a computed name for this parameter ?? 
pass


> After an initial excitement over XSLT I'm now in the state where it
> feels a bit like TeX:

You do know how many thousands of hours I've contributed to latex
support don't you?, By `like TeX' you must mean that as a
compliment... :-)

> Why didn't they simply use a full
> featured scheme or clisp for the style language...

They did use scheme. It was called dsssl.
No one used it because it had too many (

Actually some people (including me) did use it, and dsssl-list is still
reasonably active, but it's just so much easier to type
<this>...</this> than (this...) that all the hype is with XSL.

David


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