It's not a bug - it's an undocumented feature.
More On Dyslexia
A number of questions were raised following Does W3C Get Its Contrasts Wrong?. The following was orginally posted as a comment follow-up, but, since it’s rather long, I’ve re-posted it here.
Does the problem of too high a contrast prove to be a barrier to all people with dyslexia?
As long as you are clear what you mean by ‘dyslexia’, yes – as far as I am aware. The difficulty arises because the term ‘dyslexia’ is often used simultaneously to mean a specific reading difficulty and a group of learning difficulties of which reading is one.
For example, someone might suffer from dysgraphia (writing), dysorthographia (spelling) and dyscalculia (numbers) as well as the common short-term memory problems but be able to read reasonably well. However they will probably be ‘labelled’, and think of themselves, as dyslexic as these are learning difficulties that are specific to written symbols of one type or another.
The issue with contrast is sometimes referred to as ‘Scoptic Sensitivity Syndrome’. Those who suffer from this sensitivity find high contrast text difficult or impossible to read. But because most ‘generic dyslexics’ suffer from more than one specific difficulty to varying degrees, it’s difficult to find accurate figures as to how many ‘generic dyslexics’ suffer from specific Scoptic Sensitivity.
The purely anecdotal evidence that I’ve seen suggests that, as soon as you reduce the contrast (for example, print worksheets on yellow paper rather than white), the vast majority of reading problems are alleviated to some degree – which suggests, to me, that a significant number of people who have a reading problem suffer from some form of contrast sensitivity.
Do all colour schemes that meet W3C’s colour contrast present a barrier for people with dyslexia
I never like hard and fast rules, so I’d draw the line at ‘all colour schemes’ creating a barrier. I think it depends on a number of other factors as well – such as:
- How much text is present on a page?
- Is the text broken down into small, well-defined chunks?
- Is good use made of headings to sub-divide text?
- Is the language used plain and simple?
All of these go some way towards making pages easier to read and could lessen the impact created by high contrasts. Where some of the above haven’t been, or cannot be, implemented (a technical page, for example), the contrast issue becomes far more significant, in my opinion.
One of the concerns I do have is that there is no upper limit on the colour contrast. No warning that says “
Look – if you go above this level, you could be creating problems for some people“. Everything seems geared towards ‘more contrast, higher contrast’. I can fully appreciate why this is an issue for people with visual impairments but, like everything else in web design, there are two sides to every coin and I feel that the W3C recommendations only address one side.
Do we need to always have a supplementary stylesheet that consistently needs to fail the colour contrast tests?
Yes – I think it should be considered. Other options are:
- Re-examine the W3C recommendations and see if there is room for improvement.
- Consider allowing bold font colours to fall below current thresholds.
Keep as close to the lower end of the thresholds as possible.
Is a high colour contrast making text too sharp, and thus creating a situation where the text itself seems to move/wobble/shimmer
Yes – as I understand it. The phrases I’ve heard are “
the text keeps moving” or “
the words seem to dance on the page“.
I also suspect that the kind of eye-tiredness everyone suffers from when trying to read high contrast text hits dyslexics doubly so – probably because they already have to expend a lot of effort in order to read at all. Add the fact that we’re talking about a group of people who tend to have significant short-term memory problems and then consider that reading from a screen (as opposed to paper) is hard work for everyone and you end up with a recipe for extreme tiredness and frustration very quickly.
Is the problem caused by the spaces between the text seeming to dominate more?
I’ve not come across that as an issue with dyslexics specifically – or at least no one has ever described to me in that fashion. That said, I do think justified text, with its variable word spacing, creates problems all of its own and I tend to avoid using it for that reason.
One of my friends finds that anti-aliased/shadowed text causes her problems
I’ve not heard of this before but it makes sense. Presumably the demarcation between letters and background is ‘blurred’ which means that the brain has problems recognisng the word/letter shape. Some of the evidence I’ve read in the past suggests that dyslexics actually suffer from poor lexical memory (ie the ability to recognise letter/word shapes).
Ever done a crossword and you think you have the answer to one of the clues but can’t quite remember how it’s spelt so you scribble it out on a piece of scrap paper to see which version ‘looks right’? That’s lexical memory in action. I don’t think contrast is the issue here but if it is, then it would suggest that there is a lower contrast limit. However, I suspect that this would so low that it would impact on all readers, so for all practical purposes, could be ignored.