RE: Testing Setup for Accessible Web Sites/Applications

AL,

I am in total agreement that there is nothing better than actual users to
get feedback on the accessibility of a site, but there should be some
baseline that developers can use as a starting point.  It is not just a
matter of passing Bobby, that does not give any sense of whether a sites is
"functionally accessible".

Of course, this "baseline" is a moving target. But, right now, should it be
Lynx, Opera, IE 5 with JFW 3.1, or something else?

That definition would be helpful.

dc

-------------------------------------------------------------
David M. Clark
David's Access Page
16 Harcourt Street, #2I, Boston, MA 02116 Phone: 617-859-3069
Email: david@davidsaccess.com  Web: http://www.davidsaccess.com

-----Original Message-----
From: w3c-wai-ig-request@w3.org [mailto:w3c-wai-ig-request@w3.org]On Behalf
Of Al Gilman
Sent: Tuesday, January 04, 2000 5:01 PM
To: Reidy Brown; 'EASI-ED3 EASI Online Workshop: Creating Accessible HTML'
Cc: disacc@onelist.com; w3c-wai-ig@w3.org
Subject: Re: Testing Setup for Accessible Web Sites/Applications

At 09:09 AM 1/4/00 -0500, Reidy Brown wrote:
>I am setting up a computer as an accessibility testbed for the web
>developers in our company. What hardware and software do I need? What are
>the most common setups for people with various disabilities? (I know this
is
>a complex question, but I'd appreciate any input.)
>

Why are you setting up one client computer to be your lab?

For evaluation, you would be better off if your core infrastructure were a
network of cooperating evaluators.  In particular, the only way you are
likely to get decent coverage of the diversities of AT is by fanning out to
collect evaluation services from remote participants.  After all, it's web
content you are evaluating.

There are things, like running Bobby, that the content developers should do
themselves on their own computer before the serious evaluation activity
sees the proposed web content.  Then there are other checks where you
really want to get a gestalt assessment by people in diverse situations,
and ask them "what are the top five positives and negatives about this
realization of the content?"  Talk to Eric Hansen of ETS about how they did
their "quick and dirty" study to find the key issues in their intranet
setup.  They call it quick and dirty because they are trained in the Nth
degree of statistics, but it was a most elegant hack and very effective
communication.

What you most need for your own site is a prototype website exposure area
with some access control so the world doesn't boo your first attempts, and
some groupware for sharing between the evaluators with their varied
functional impairments and the developers with their conceptual
underdevelopment.

This is a topic which bears an organized, in-depth discussion.  The most
natural homes in the WAI are in the Evaluation and Repair Interest Group,
for methods, and the Education and Outreach Working Group, for networking
with evaluators.

If your content developers are in academic settings, get the same
institutions to recruit qualifying individuals as evaluators from some of
their enrolled students and employed staff.  Pooling across institutions
will help them all get the coverage each cannot provide alone.

Your job, at the hub, is a) getting the material scrubbed once for obvious
gaffes before it goes to the qualifying indiviuals, b) supporting the
groupware and facilitating author/evaluator dialog, and c) creating macro
or template libraries and development process scripts that your content
developers use, based on the results of prior evaluations.

Al

>I suspect that the software version number is as important and the software
>type... how far back do I need to go? (Most of our users are in the higher
>education field.)
>
>Reidy
>
>_________________________________________
>Reidy Brown
>Accessibility Coordinator/
>Senior Web Application Developer
>mailto:rbrown@blackboard.com
>http://www.blackboard.com
>____________________________________________
>

Received on Wednesday, 5 January 2000 12:24:05 UTC