WebAIM - Web Accessibility In Mind

E-mail List Archives

RE: flash satay firefox bug

for

From: Robinson, Norman B - Washington, DC
Date: May 24, 2005 10:44AM

John,

You bring up an interesting point, but I think your email
ignores the problem of testing and developer acceptance.

If only one screen reader provided accessibility for Flash, and
all other screen readers did not provide that accessibility, which
screen reader would you use to TEST accessibility? There are
accessibility standards, but as you well know, you can make a 508
compliant application or web page and it still not be accessible. I
think this is one reason the word "accessibility" wasn't in the section
508 standards. There is simply "508 compliance". The empirical, hands-on
proof many seek are often driven by testing using the same assistive
technologies used by the end-user. It is end-user simulation.

I think that if I had to test I would test with as many user
scenarios as is important to success. I certainly don't test my
applications with the two previous versions of MS Windows. I use what I
know my users are using.

If you can access it with a specific assistive technology does
that mean it is accessible? Probably. Does it mean it is 508 compliant?
Probably. Are their exceptions? Definitely! I don't think the fact one
vendor has provided more access because of figuring out non-standard
ways to give that information to the user means that the whole industry
will flock to that vendor. I can testify to how slow the industry moves.
I probably could pull an example of a specific vendor's software that
used non-standard video hooks and reliance on system information that
wasn't documented. Did that mean the other vendor's software should be
treated differently? No. It meant they figured out a way to provide
access. Nonstandard, but desperately needed access. If a standardized
way did exist, I'm sure they would have used it. Standards help
everybody!

That said - I'm with you! Developers should never target a
specific assistive technology, with the exception of scripting or
something equivalent that help the assistive technology use the
standards. And if the vendors are doing their jobs, they should have
little needs of scripting support for accessing standard technology. I
think having all the assistive technologies tested against a standards
test, similar to the ACID2 test (http://www.webstandards.org/act/acid2/)
is a desirable goal. It would promote standards and let developers know
when to better trust the results of their hands-on testing.

I would like to see the latest numbers on specific user agents,
including combinations of user agents and assistive technologies. I
think it enhances our understanding and encourages vendors of all
software products, specific to an operating system or web based, to
create accessible products.

Regards,


Norman Robinson