Re: Leventhal's challenge misses the point (long)

["Paul Tchistopolskii" <paul@xxxxxxx>]

>There are not yet 100% CSS implementations in released products, which
>Michael also claims is partly due to XSL. And suppose there are:
interactive
>web page means DOM and scripts so what do we need XSL for?


For making different transformations from one (XML) document
created in one company to another (XML) document that would
be more convinient for another company.  XSL at least provides
some ways to do it  in resanoble way. FO may help in layout part.
(To be honest, I don't think it was the original idea of FO. I feel that
transformation part seems to be in better shape than FO, but FO
part improves fast).

Of course, you can do it in perl.  I tried both. perl for years and
XSL for days. I like XSL and think that it'l survive, because it
is good. Yes - it's a bit strange... Actualy, perl is strange as well.
Actualy, first C compiler was (is?) also kind of strange software...

Hovewer, the biggest problem I see with XSL / CSS is that they
are not using the exactly common formatting objects  ( with  XSL
being a superset of CSS ), but as far as I understand you could
easy render CSS into FO. Rendering FO into CSS seems
to  possible only to some degree ( sufficient for the real life, from
my point of view ). When I tried to put XSL into Mozilla, I firstly
tried just to render FO into XML + CSS.  I found the quality
of such rendering to be sufficient in  real life (especialy
comparing with the shape of existing FO implementations),
even it is actualy a 'logical hack'.

Logical hacks are bad, so I still think that XSL FO /  CSS  issue
will continue to appear in maling lists for a long time.

Rgds.Paul.





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