XML/XPath is definitely a rational solution. CSS is yet another textual
language which needs a specific interpreter. For that, I don't like
either Javascript or JSON : it remains me that too many programming
languages are still using a C-like syntax, created so many years ago. I
still think XML as the new ASCII for machine-to-machine transfers :
whereas a structure exists, it should be expressed using an XML
notation, that's all !



With an XML notation ("CSSX" ?), programs could generate, modify,
compare them, and so on, much more easily.



For one of my customers, I have build a small CSS to CSSX application
to be able to patch some CSS files using an XML patch notation before
reconvert them to CSS via XSL-T.



On the other hand, I still think it is much easier for human beings to
read and understand CSS than CSSX but the problem should be just an
editor or a rendering problem.



Alain COUTHURES

<agenceXML>

http://www.agencexml.com



Thomas Lord a &eacute;crit :
485FF7AC.6020704@e..." type="cite">
  
Robert Koberg wrote:
  
    On Mon, 2008-06-23 at 12:04 -0700, Thomas Lord wrote:
  
    
      An XML/XPath solution could [....]
    
    
    <!---->

the kids are calling it ajax
  
  
  

Sorta yeah.  Makes sense.   Scrubs clean some old messes

with its powerful abrasive power or something like that.

Far more effective on caked on messes than mere SOAP.

  

Still, there's some beautiful austerity and maturity of thought

to what's emerging in the XML / XPath / XSLT / XQuery

world that draws on a lot of experience and perspective 

"the kids" clearly lack.    It'll all get a lot more interesting 

when things like the ajax world start to get coalesced and 

rationalized and standardized a bit more.  "The kids" will be

shocked :-)

  

Thanks,

-t

  

  

  

  
    :)