W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2003

Re: [TECH] Colour Difference Algorithm

From: Joe Clark <joeclark@joeclark.org>
Date: Tue, 26 Aug 2003 12:23:34 -0400
Message-Id: <a06001facbb7132db6f82@[]>
To: WAI-GL <w3c-wai-gl@w3.org>

>The current WCAG2 draft [1] checkpoint 1.6 is still missing an 
>algorithm for determining when "foreground content is easily 
>differentiable from background" regarding the use of colour.

Because it can't be programmatically determined?

Because, if we exclude colourblindness from consideration for a 
moment, the extent and variation of vision deficiencies are much too 
vast to be summed up in a nice formula?

>The AERT document [2], technique 2.2.1 for the WCAG1 proposed an 
>algorithm that worked reasonably well but was not perfect. Some 
>colour combinations were passed, though many people would find them 
>unacceptable, and other colours were failed though many people would 
>find them acceptable.

That's not a good sign, is it?

>Note: I'm suggesting that the general term "colour difference" 
>includes, but is not limited to, a difference in brightness and a 
>difference in colour (separation along the colour spectrum). Other 
>factors such as saturation may also come into play.

I believe the Working Group continues to:

1. misunderstand the nature of colour deficiency
2. mistakenly overgeneralize and project visual *impairment* onto 
issues of colour
3. ignore authoritative published research on colourblindness and accessibility

When I see the Working Group getting past point 3 above, we can start 
talking about the fact that its emphasis on "contrast" (and now 
"brightness") is incorrect, is not going to help people with visual 
impairments read the Web better, and ignores the central issue: 
Colour. (And here's a talking point: Why isn't this an issue of user 

The paragraph quoted above demonstrates this well, placing hue in the 
same category as brightness and saturation. It isn't.

>Listed below are several sites dealing with colour on the web:

And of course the WG continues to pretend that the weeks of research 
I put into Chapter 9 of my book, "Type and Colour," amounted to 

I spent days on end at the library. I talked to and E-mailed actual 
experts in colour deficiency and colour mapping. I had them vet my 
chapter before publication. (That chapter was proofed seven full 
times.) I sent the experts the published book and asked them to 
correct any errors that remained. No corrections were offered. I of 
course found some typing errors myself later on.


The book's been out since October 2002 (with HTML version on the 
included CD-ROM), and that chapter was online in (not very 
accessible) PDF since, if memory serves, December. It's been online 
in HTML for months.


I can also forward the text to the list. In any event, use it to 
improve your guidelines.

Free advice: Delete the current wording altogether until you do some 
more research.


     Joe Clark | joeclark@joeclark.org
     Accessibility <http://joeclark.org/access/>
     Expect criticism if you top-post
Received on Tuesday, 26 August 2003 12:23:37 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:45 UTC