W3C home > Mailing lists > Public > www-validator-css@w3.org > November 2007

Re: Bug in CSS-validator?

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Mon, 19 Nov 2007 16:07:45 +0200
Message-ID: <01e501c82ab5$91434050$0500000a@DOCENDO>
To: "Simon Wiles" <simonjwiles@gmail.com>, <www-validator-css@w3.org>

Simon Wiles wrote:

> Hi, I think there's a bug in the CSS validator at
> http://jigsaw.w3.org/css-validator/.

I tend to agree, but it's a documentation bug rather than program bug.

> cjk-ideographic is not a list-style-type value : cjk-ideographic

That's because the default version ("profile") used is the CSS 2.1 draft, 
which is treated as the de facto standard these days, despite its status and 
the fact that CSS 1 and CSS 2 are nominally "W3C recommendations".

> yet clicking the very link that is referenced shows cjk-ideographic
> as a list-style-type value.

That's the documentation error. The W3C CSS Validator's reports refer, via 
links, to the CSS 2 recommendation, even though it does not actually check 
against it by default (as it used to do). Worse still, even if you manually 
select (via "More Options") a different version ("profile"), the reports 
still refer to the CSS 2 specification.

I wonder whether it would be better to remove the links (if that's possible 
with some simple coding change), until they can be fixed to give correct 
information only: only link to a specification or a draft if you actually 
checked against it.

If you manually set the version to CSS 1 (official, but really obsolete) or 
CSS 3 (collection of drafts), then list-style-type: cjk-ideographic passes 
the check. Of course, this would not make your page work any wider than it 
does, or the exact numbering scheme any better defined than it is now 
defined in an obsolete specification and in a draft.


Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/ 
Received on Monday, 19 November 2007 14:05:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 June 2012 00:14:19 GMT