Re: First Stab at Set of Principles for 'Minimum Conformance'

"Anne Pemberton":
> Jim,
>
>          You test for illustrations the same way you test for alt
text....
> You can only test that it is there, not whether or not it is adequate.

You can't test it is there, the ALT attribute exists on an IMG, therefore
it's trivial to test if it's there (and some common mistakes can be
identified of course.), with information contained in text, there's no
way of associating that text with a particular image, therefore there's
no way of testing that for every piece of textual information there is a
corresponding image - or are you just suggesting, well if there's an
image somewhere on the page that is okay?

>          As for the ease of illustrating, it is really no harder than
> writing alt text or long descriptions?

I disagree, I understand neither of your examples in
http://users.erols.com/stevepem/Guidelines/Cognition.html
without the alt attribute on the image telling me what it was for.  Also
this is a trivial example, consider providing a similar set of
illustrations on an "Introduction to Cosmology page".
Then there are cultural and societal references, which such images (have
to?) rely on - you've got the lightbulb and questionmark symbols - are
these known throughout the world?

I think a more real world case would be needed for me to be convinced
that it was simple, and I'm definately not convinced it's in any way
testable.

How have Charles (and some of the repliers) points
http://lists.w3.org/Archives/Public/w3c-wai-gl/2001JulSep/0766.html
been addressed on the "clear and simple" issues been addressed?

>          Last time the argument of illustration content came up, back
in
> the spring if I remember right, Wendy took the "obtuse subjects",
searched,
> and found numerous sites illustrating these subjects.

I can't find this in the archives - do you have a link?

Jim

Received on Wednesday, 24 October 2001 07:41:33 UTC