Re: XSL proposal

[Derek Denny-Brown <ddb@xxxxxxxxxx>]

I (finnaly!) sat down and read through the XSL proposal, after leaving it
at the office and spending too many days a walking chemical daze (thanks to
the joys of flu medicines).  I would like to join my voice with the others
congratulating the team for coming up with what look to me like an
excellent compromise between CSS and DSSSL.

Amusingly, I have recently been batting around an idea for a DSSSL like
stylesheet language which outputs HTML+CSS but where the core language was
JavaScript rather than Scheme.  The XSL team beat me to it... and did it
better.

I have one problem which would make it difficult for us to use as a direct
replacement for our current (Perl based) scheme.  A 'bug' in the authoring
software causes it to insert paragraph tags with no content unbeknownst to
the author (i.e. there is no visual indication in the authoring environment
that these tags are there).  In order to render properly, we need to ignore
paragraph tags with no #PCDATA content.  I see no way of doing that with
the current XSL approach.  A while back there was talk about creating a
faux element type for pseudo-elements, using "#PCDATA" as the element-type.
 I think XSL would be an excellent place to introduce such a concept.  (and
it would allow XSL to be used in our system, without having to pre-process
the SGML)

-derek

     Derek E. Denny-Brown II      ||   ddb@xxxxxxxxxx
     "Reality is that which,      ||   Seattle, WA USA
  when you stop believing in it,  ||  WWW/SGML/HyTime/XML
 doesn't go away."  -- P. K. Dick || Java/Perl/Scheme/C/C++

 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist