RE: a question of scripts (this time completed)

Thanks for this info David. I will discuss further with our programmers and we
will investigate some of the tools you have mentioned. Some of the requests we
have received appear to be very specific and we will probably end up resorting
back to JAVA. The real challenge though is how to translate what is happening
on screen into textual information.

Steven - thanks for your suggestion to refer to the NBA Tape Recording Manual
for some general principles about providing text alternatives to visual
material.

Some useful leads for me to now follow through on. Much obliged to everyone who
responded to my plea for guidance :-)

Denise

Dr Denise L Wood
Lecturer: Professional Development (online teaching and learning)
University of South Australia
CE Campus, North Terrace, Adelaide SA 5000
Ph:    (61 8) 8302 2172 / (61 8) 8302 4472 (Tuesdays & Thursdays)
Fax:  (61 8) 8302 2363 / (61 8) 8302 4390
Mob: (0413 648 260)

Email:	Denise.Wood@unisa.edu.au
WWW:	http://www.unisanet.unisa.edu.au/staff/homepage.asp?Name=Denise.Wood



-----Original Message-----
From: David Woolley [mailto:david@djwhome.demon.co.uk]
Sent: Wednesday, 24 October 2001 4:36 AM
To: w3c-wai-ig@w3.org
Subject: Re: a question of scripts (this time completed)


> JAVA or other scripting (could be Flash etc) that are designed to either test
> students' understanding of the impact that changes in variables have on what
is
> essentially graphic information, or that allow them to explore these concepts
> by changing variables themselves and viewing the results. We have many
requests
> like these - particularly in the engineering, satellite communications and

If the models can be expressed as closed form equations, I'd suggest
that a general equation plotting tool, like gnuplot, rather than any web
based approach, would be best.  That way the students learn some of the
maths and general principles, and can solve their own problems without having
general programming skills.

If you cannot get simple closed form equations, you need a general
programming language, and I'd say that Java was the most portable
current choice.  In my view, it introduces less security problems than
the common scripting languages.  (There are tools that may be able
to graph non-closed form equations.)

SVG might be a better way of providing the visual interface in the near
future, but, for cases that aren't suitable for gnuplot, I think you should
wait for Java language bindings to be implemented.  (You can probably do it
now using plugins for the SVG and JavaScript.

To maximise browser coverage, you should have an alternative of generating
a bitmapped image on the server side.

There are certainly specialised tools, like SPICE, for dealing with
specialised modelling areas, and I think you should consider using
the established tools, rather than trying to create your own subsets,
unless you can greatly enhance the accessibility, which seems unlikely.

I'd also note that "data visualisation" is a highly commercial area
and lots of companies would claim that their high priced products 
significantly aid people in understanding complex mathematical models.

All of these require that it is possible to learn the same information from
a textual equivalent that is provided, if you are going to comply with the
various accessibilty guidelines; the animations should be considered 
comprehension aids for those with good eyesight and easy interaction with 
the computer.

I don't know how you cater for those with "learning difficulties".

Received on Wednesday, 24 October 2001 11:15:57 UTC