OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

docbook message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [docbook] Whatever happened too CSS+XML?

On Monday 07 November 2005 01:20 pm, Dave Pawson wrote:
> On Sun, 2005-11-06 at 17:05 -0500, Steven T. Hatton wrote:
> > A few years back I was looking at this stuff, and never understood why
> > people weren't more interested in it. Basically, you should be able to
> > server a DocBook document on the web without every creating XHTML.  When
> > I consider the effort that has been put into creating XSLT stylesheets to
> > generate XHTML, it seems the same amount of effort (or considerably less)
> > could produce XML + CSS that would have as much functionality as any
> > XHTML produced from the DocBook XML.
> Rubbish.
> When CSS gets implemented
> When CSS is accepted
> It still won't have 10% of the capability of XSLT.

Well, I suspect it may never be fully implemented.  There aren't very many 
people who appreciate its potential.  There is at least one major player in 
the industry which seems reluctant to fully implement CSS (or any industry 
endorsed standard recommendation). CSS is not intended to transform, or 
manipulate the original document.  It is specifically intended as a means of 
rendering it.  I don't believe there are very many CSS functions which can be 
achieved by XSLT without the assistance of something similar to CSS.  The 
current practice of transforming DocBook into HTML is merely hiding the fact 
that it's relying on what amounts to CSS for the final rendering.  In my view 
XSLT should play a far less significant role in the DocBook to browser 
rendering process.  Web browsers are ideal tools for processing XML.  The DOM 
is basically a formalization and enhancement of the Mosaic document 
structure.  Combining that with server-side document awareness it is possible 
to produce the same end-user experience (and more) using XML technologies 
directly, without the intervening translation (mutilation) into HTML.  

There are roles for XSLT.  Many of its capabilities are probably, as yet, 
unimagined.  But many of the current applications I've seen seem to be 
workarounds to address the lack to good tools supporting the existing W3C 
recommendations for CSS and XML.  Quite surprisingly, one of the most capable 
challengers to XSLT that I am aware of is Mathematica.  What it can do with 
pattern matching and expressions transformmation is simply amazing.


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]