W3C home > Mailing lists > Public > www-style@w3.org > December 2002

Re: XBL is (mostly) W3C redundant, and CSS is wrong W3C layer for semantic behavior *markup*

From: Shelby Moore <shelby@coolpage.com>
Date: Sun, 29 Dec 2002 18:43:52 -0600
Message-Id: <4.1.20021229182525.009e3950(null)>
To: David Woolley <david@djwhome.demon.co.uk>
Cc: www-style@w3.org

One more point about the .PDF

I am philosophically against the B&W (all or nothing) argument that says
use .pdf if you want to have some control over presentation and use
semantic markup if you want benefits of semantic abstraction.  The whole
point we have layers is so that it isn't a "all or nothing" design decision.

This is the reason CSS was so widely adopted (besides and because of being
orthogonal and optional).  You do not give up all semantics by using CSS
with semantic markup and you don't give up all presentation control by
using semantic markup with CSS.

The primary design argument against thin orthogonal layers appears to be
that presentation agents are empowered with semantic understanding.  I
would like to point out that this (semantic level standards) is *NOT*
mutually exclusive with having optional (orthogonal) layers in between the
UA (user agent, i.e. presentation) and the semantic layer.  Whereas,
merging layers *IS* mutually exclusive with fine orthogonal gradations.

If you code all your widgets in XBL, then later you want your widgets to
degrade gracefully on non-DOM, non-CSS, different events model,
non-proprietary window model, etc, then you will have morph XBL since it
relies on all these things *BY SYNTAX*.  With XSLT implementation, you can
morph your code only (and hopefully localized as well as you have designed
(abstracted) your implementations), without any need to morph the
underlying syntax of XSLT or implementations of XSLT.

Okay I think I have a made very compelling case for why orthogonal,
general, yet thin objective layers are preferrable over merging layers for
narrow objectives, even without providing any examples yet with XSLT.  So
unless someone provides a very factual rebuttal, then I think my work now
has to proceed in the implementation area, so eventually I will have many
examples to present when (if ever) XBL comes up for standards review.

Regards,
Shelby Moore
Received on Sunday, 29 December 2002 19:42:59 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:18 GMT