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

Re: CSS 2.1 WD and non-CSS presentational hints

From: Stuart Ballard <sballard@netreach.com>
Date: Wed, 28 Aug 2002 17:24:56 -0400
Message-ID: <3D6D3FA8.4060903@netreach.com>
To: Chris Lilley <chris@w3.org>
CC: www-style@w3.org, Lachlan Cannon <luminosity@members.evolt.org>

Chris Lilley wrote:
> SB> This would cover <center>, align="left", <br>, <pre>, <font>, etc. All 
> SB> the usual suspects.
> The trouble with those is that their rendering semantics are often not
> clearly defined. So getting "the same rendering" is not always easy.
> But yes, it would cover those,and it would cover for example stuff
> produced by XSLT.

Legacy elements and attributes aren't sufficiently well-specified in 
general to make it possible to get "the same rendering" in all 
circumstances, I don't think. I don't think that's a problem, either: 
we're only trying to specify the precedence of their behavior with 
respect to CSS, not define their behavior in detail.

For these purposes, it doesn't matter exactly how <center> is rendered 
by default so long as "text-align: left" in a regular user stylesheet 
doesn't override it and "text-align: left" in an author stylesheet does. 
And for the purposes of the definition, all that matters is that 
<center> explicitly indicates a preferred rendering (regardless of 
whether the details of that rendering are perfectly-specified), and that 
it doesn't carry any other semantics.

"Stuff produced by XSLT" is a little bit too broad for me. As I said in 
a previous post, I can imagine domain-specific XML being transmitted for 
the web, with XSLT to translate it to *semantic* HTML, accompanied by a 
reference to a regular old CSS stylesheet. Just because an <h1> element 
was generated by XSLT from a <chapter-title> element in some other 
namespace doesn't make it any less semantic. And it still shouldn't 
qualify as a non-CSS presentational hint.

> SB> We might need an explicit exemption for the style= attribute, which 
> SB> otherwise would qualify based on this definition,
> good catch, all the more so since CSS 2.1 proposes to give it the
> highest possible specificity thus disabling all possible restyling
> further down the line.

Ugh, overriding user-stylesheet !important? I don't like that very much; 
the user, not the author, should have the final say, imho.


Stuart Ballard, Programmer
NetReach - Internet Solutions
(215) 283-2300, ext. 126
Received on Wednesday, 28 August 2002 17:25:02 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:03 UTC