RE: SVG accessibility - how capture the reading order of text

At 12:59 PM 2001-09-07 , William Loughborough wrote:
>At 10:32 AM 9/7/01 -0400, Al Gilman wrote:
>>A linear discourse is not obvious from a graphical design.  Unless the
author
>>thinks about how to couch their message as a narrative, you don't get a
>>narrative flow by assembling the pieces that appear in a graphical
exposition
>>of an idea.  Fluent narrative has glue in it that the layout does without,
for
>>one thing.
>
>Have you thought about plunging this hot sword into the freezing water on 
>GLWG's never-ending raves about 
>illustration/graphics/multi-media/text-equivalents? somehow it seems 
>pertinent to their "effort"?
>

AG:: I've said it before and I'll say it again.  

But there is a draft SVG techniques module on the table for next week's
meeting
and I haven't read that.  So I am not up to date with where the group is on
this.

We need the graphical design stuff to have enough object by object
semantics so
we can guess a feasible order, and the graphical composition process to put
the
author through a few natural language questions to confirm or correct the
likely guesses, and we can then have a rote linearization that will stand
up to
widespread use by amateurs.

To make this work we have to develop a model that is cross-media-capable, that
is to say a model in which is is possible to compare and contrast the
rhetorical form of a poster and an essay on the same topic, and build a
database so that both views are generated by generic algorithms from what's in
the common database.  [Does this sound like the Device Independence scenario?]

Developing the model is a lab course in comparative analysis of media.  Like
comparative literature, except that the literature one is comparing is oral
vs.
visual vs. video game media.

See who all is interested in such an exercise.

Al

>--
>Love.
>EACH UN-INDEXED/ANNOTATED WEB POSTING WE MAKE IS TESTAMENT TO OUR HYPOCRISY
>

Received on Friday, 7 September 2001 14:27:40 UTC