[
Lists Home |
Date Index |
Thread Index
]
On Wed, 9 Jul 2003, Simon St.Laurent wrote:
> At 10:26 AM 7/9/2003 -0400, Mike Kozlowski wrote:
> >My point was, and is, simply that XSLT and CSS are solving different
> >problems entirely, and don't directly compete in any way. Is that a
> >statement with which you can agree?
>
> No. XSLT was designed for formatting-by-transformation, while CSS was
> designed for formatting-by-annotation.
> That may not feel like competition to you, but to people who are primarily
> interested in the formatting, it's competition.
I just can't see that. Historically (in both XSL and DSSSL),
transformation and formatting have been viewed as complementary to each
other; and practically, the things that XSLT does aren't the things that
CSS does.
Consider again the example page I mentioned before
(http://www.klio.org/mlk/example.xml). In that page, the most useful
contributions of XSLT are to:
1. Sort the output
2. Produce summary data
3. Apply header/footer-style templating
The most useful contributions of CSS are to:
1. Control the positioning of elements
2. Control the properties of text blocks (color, spacing, margin, etc.)
The things that XSLT is doing simply can't be done by CSS, at all, ever.
The competition to XSLT would be PHP/Perl/JSP/JavaScript and XML
libraries.
The things that CSS is doing can't be done by XSLT alone, either. You
could use XSLT to target inherently-styled vocabularies like HTML 3.2 or
XSL-FO, yes, but in that case it's XSL-FO and HTML 3.2 that are replacing
CSS.
If you want to convince me that XSLT and CSS compete directly, you need to
show me how CSS could do what XSLT is doing, and vice versa.
--
Mike Kozlowski
http://www.klio.org/mlk/
|
