W3C home > Mailing lists > Public > www-html@w3.org > July 2005

Re: Unstyled Content (was Re: Draft suggestion: "normal" needed)

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Sat, 16 Jul 2005 09:30:36 +0300 (EEST)
To: www-html@w3.org
Message-ID: <Pine.GSO.4.63.0507160911510.8435@korppi.cs.tut.fi>

On Fri, 15 Jul 2005, Al Gilman wrote:

> The way I read Appendix H [1] in the current Working Draft of XHTML 2.0,
> browsers are not required to implement CSS but they are required to
> make the default appearance as described in the provided CSS stylesheet
> available to the user.

It seems to say that very clearly, and stronger:
"This Appendix defines a normative [CSS2] style sheet for XHTML 2. While 
visual user agents implementing XHTML 2 are not required to support CSS2, 
they are required to behave as if the following CSS2 styles are in 
effect."

(And authors will now have more justification than ever in thinking and 
saying "blockquote means indent".)

This would mean great progress towards the goal of making XHTML 2
a simple formatting language, with an extension mechanism called CSS.
The specification would dictate the exact visual appearance,
in a manner that can be overridden in CSS only. Although this would 
theoretically permit different browser style sheets, visual user agents 
would be required to use the normative style sheet as the default.

Undoubtedly, the normative style sheet will continue the tradition of Web 
browsers in nullifying 500 years of typography, with Mosaic-style 
paragraphs, equal margins before and after a heading, etc.

The "sample" style sheets in HTML specifications have been a failure, in 
their attempt at being both descriptive and normative, failing to be 
either. I don't think the solution is to make it explicitly normative, 
especially if it will be little more than a presentation of the way
browsers have behaved, with all the mistakes now hard-wired into a 
normative formalization.

The realistic approach is to leave presentation out, and not include any 
style sheet into the XHTML 2 specification. A better approach would be to 
include at least two essentially different sample style sheets, naturally 
saying that they are not normative or even suggestive, just illustrations 
of feasible possibilities. Maybe there should be a third sample style 
sheet too, more or less aimed at capturing the "tradition".

-- 
Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Saturday, 16 July 2005 06:30:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:03 GMT