Sent by Mauricio Samy Silva on 30 March 2007 12:12
Apart the 'probably' typo (grey instead of gray) on the David answer the
If you specified a color name (one of the 17 valid colors keywords on the
Specs) browsers can render
it slightly different (i.e. red (or other color name) is more ou less darken
according the browser).
This can "broken" the contrast the same way as:
#008083 provides a good contrast over #fff
#099 (slightly different from #008083) doesn't provide sufficient contrast.
In my opinion, if I'm not missing something, the main point is #008083 (or
other valid number color) is the same in all browsers and gray (or one of
the 17 valid colors keywords on the Specs) isn't the same across browsers.
Number color CSS value is consistent across browsers and colour values
Maurício Samy Silva
----- Original Message -----
From: "Jukka K. Korpela" [EMAIL-REMOVED]>
Sent: Friday, March 30, 2007 10:35 AM
Subject: Re: [css-d] CSS color names values versus accessibility
> On Fri, 30 Mar 2007, Mauricio Samy Silva wrote:
>> David White said:
>> "...The point about using numbers (I.e. Hex values) instead of names is
>> purely so that there can be no misunderstanding when parsing on the
>> browser. Some browsers render "grey" (for example) differently but if you
>> Hex there can be no ambiguity. ..."
> There is no color name "grey" in CSS specifications, so the argument is
> relevant to nonstandard color names only, and they were not under
> discussion. They are of course to be avoided on the same ground as any
> other nonstandard constructs (including color codes without "#" - they too
> "work" on some browsers and make the declaration ignored on other,
> conforming browsers).
>> It makes sense cause sometimes a slightly color difference crashes the
>> threshold for contrast.
> I don't see how this could be a matter of a slight difference. The name
> "grey" is incorrectly recognized as a synonym for "gray" on some browsers,
> correctly treated as malformed on some. If there are browsers that accept
> it and treat it as denoting something _almost_ identitical to "gray", then
> I'd be delighted to hear about such a monstrosity, but this has nothing to
> with the difference between "gray" and "#808080", which is no difference.
> By the way, if your contrast is so near to the threshold (as defined by
> the W3C or some other party) that a _slight_ change makes you cross it,
> then you were already too near. Crossing the threshold has an extremely
> small impact in such a situation on real accessibility, even if it may
> change some technical status by some _measure_ of accessibility.
> Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
> css-discuss [EMAIL-REMOVED]]
> IE7 information -- http://css-discuss.incutio.com/?page=IE7
> List wiki/FAQ -- http://css-discuss.incutio.com/
> Supported by evolt.org -- http://www.evolt.org/help_support_evolt/
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/