Re: Illustrating Guidelines

"Anne Pemberton" <mailto:apembert@erols.com> wrote:-

> If by "repurposable" you mean the fact that if a user has some
> fancy equipment, they can have the text read to them as well
> as reading it for themselves, this is interesting, [...]

Similarly, I think it would be interesting if equipment could one day
scan an image an somehow relate it to a meaning - but this brings us
both into the realm of AI and "reductio ad absurdum". Needless to say,
there are few documents that need to be accessible outside of a
certain scope, viz. I'd probably rather have 100% of documents conform
to WCAG-A than 90% to WCAG-AAA, but even that is being a tad glib -
some documents (such as WCAG itself) are going out to a wide range of
people, whereas others (e.g. a learned text on jurisprudence) probably
won't be.

Back to the illustrations, I'm not sure I agree that "images convey
concepts more widely than text" *or* vice versa. Sometimes a picture
(of a chair?) is more recognizable than the equivalent text, and
sometimes a picture (of mother cradling a newborn?) can say more than
words ever can... but there are also many concepts (e.g. in advanced
mathematics, quantum physics) that are impossbile to "draw" - you have
to reduce them to component parts and speak in some special langauge.
An example would be William's rant on synthesis/analysis - I doubt
that could ever be expressed in pictures, although I challenge someone
to prove me wrong. Also, I wonder how many Dilbert cartoons would make
sense without words (and some without the pictures, although I know
for a fact that some do).

Interestingly though, when words (mainly nouns) are used, people very
often have some kind of "visual pattern" (Platonic ideal) in their
mind of that object - get someone to think about a tree and you'll get
a wide response of images from individual trees, to just a generic
picture of a tree (we tried this experiment in a philosophy class
once...); I wonder what blind people (from birth) "picture" as being a
tree, i.e. how does one associate the label "tree" to the concept of
"a tree"? Are the associations distinct from the representations?

On to illustrations for WCAG... what I find is that higher order
concepts are very difficult to represent as pictures. Anne has made
some valiant efforts to do so, but I'm not sure they really help - I
think it would be a good thing to query some more people on this
though, because it's not good practice to have just a handful of
people's opinions on the matter. Certainly, the text in images is one
thing that puts me off, the second being that without the text, I'd
have no idea what the pictures are supposed to represent (Bruce
already pointed this out).

What I *did* find excellent though was the simple example of a picture
of George Washington next to his name. That is just so clearly an
example of how an visual aid associated with a run (grr... I pursued
that on PF... should have CC'd to www-archive) of text can remove an
accessibility barrier for some, and remain a non-nuisance for others.
So, in summary, I think that WCAG has to be very careful to include
scope and context into GL3.

P.S. Is there any chance that you could change the emphasized text
from blue with underline? That's just too close to many people's
defaults for link text for comfort... it's difficult even for me to
tell which is a link and which is just a bit of text (if it wasn't for
the cursor). One way of looking at it is that you could have used <em>
for the emph bits, and I could have changed it myself, but the other
way is that I can just change my default link text color... it's part
of the author vs. user thing again.

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://webns.net/roughterms/> .
:Sean :hasHomepage <http://purl.org/net/sbp/> .

Received on Thursday, 10 May 2001 12:48:57 UTC