Re: Checkpoint 3.4 again

At 13:20 -0400 2001.07.28, Anne Pemberton wrote:
>         Joe, all elements on a page need an equivalent, not just the 
>non-text elements. Text is an element and it needs an equivalent, 
>like all the other elements. 3.4 is the only place this is 
>addressed, and it doesn't do the deed.

There is no parallel between adding an alt, title, or longdesc to an 
image, applet, or frame and attempting to draw a picture for every 
single extended paragraphs of exposition.

We are all in favour of increased accessibility for learning-disabled 
persons. However, it is impossible to create access equivalent to 
that enjoyed by blind and visually-impaired people. It is easy and 
simple to make a page with graphics accessible to the blind. It is 
taxing and in many cases impossible to make a page with text 
accessible to the dyslexic.

The two tasks are not parallel or equivalent. The intent certainly 
is. The principles underlying accessibility remain universal. There 
is nothing especially bad or good about the blind or the dyslexic, or 
any other disabled group, or nondisabled people. But it is intrinsic 
to the nature of the written word that it cannot always be rendered 
in a picture. The converse is true, and that's why accessibility for 
graphics is possible.

Text does not need an equivalent. Text is the basis of written human 
communication. Graphics, images, and drawings are a different form 
(whether antecedent or subsequent is beside the point). Text is a 
*primitive*. The concepts expressed by the written word are often 
irreducible.

In the grand tradition of leftist movements, I will now immediately 
be accused of a textist bias. It is not as though I have no knowledge 
of learning disabilities and their accessibility requirements. Nor am 
I allowing the fact that I am a better writer than illustrator cloud 
my judgement.

Text is fundamentally different from graphics. You cannot use exactly 
converse techniques to make both accessible (add graphics to text; 
add text to graphics). The media are asymmetric.

This asymmetry has the effect of rendering learning-disabled people 
at a disadvantage that the Web Content Accessibility Guidelines 
cannot correct by demanding that every author in every case include a 
picture. Among the vast litany of reasons why this is a bad idea is 
the simple fact that authors will ignore such a requirement. It is 
unrealistic, onerous, and laughable on its face. It will bring the 
entire enterprise of Web accessibility into disrepute.

Further, adding graphics means you have to add more words. Every 
graphic at least has to carry an alt text.

WCAG members must realize that there will never be agreement on 
requiring the use of graphics. There will never be a time when 
everyone can go along with it despite reservations they may hold. It 
is not as though this disagreement is small, the sort of thing that 
educated, well-intentioned people can agree to disagree on and live 
with whichever faction prevails.

*Requiring* graphics is the most sweeping, excessive, ill-advised, 
and overblown suggestion ever countenanced by the Web Accessibility 
Initiative. A substantial number of people very strongly committed to 
accessibility everywhere (including the Web) will always be opposed 
to it.

The idea itself won't work in practice and a lot of people will 
oppose it forever. It's a recipe for disaster. It's the sort of thing 
that accessibility advocates themselves will either disregard or 
actively campaign against.
-- 
         Joe Clark | joeclark@joeclark.org
         Accessibility articles, resources, and critiques:
         <http://joeclark.org/access/>

Received on Saturday, 28 July 2001 13:43:34 UTC