W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2009

Re: accessibility supported questions

From: Phill Jenkins <pjenkins@us.ibm.com>
Date: Tue, 31 Mar 2009 13:08:24 -0500
To: Ramón Corominas <listas@ramoncorominas.com>
Cc: w3c-wai-ig@w3.org
Message-ID: <OF93B4CD6C.8687A0B9-ON8625758A.0055AF82-8625758A.0063A5DF@us.ibm.com>
> . . . it doesn't matter for compliance if there is assistive 
> technology that is able to interpret that content. So the
> content can comply, but nobody can access it? . . . 

Of course it matters if something works with assistive technology or not. 
Most all of us want documents, web sites, and web applications to be 
usable by all users. 

The original post in this thread was about testing a specific PDF form 
with two specific screen readers using the "success criteria" in WCAG 2.0. 
 The reported results was that sometimes it worked and sometimes it 
didn't. Unless I'm mistaken, the PDF form itself wasn't changing, so if it 
worked sometimes and not others, its seems to me to be a problem with 
either the PDF renderer (not sure what they were using, was it Adobe 
Reader 8?), or the screen reader (they didn't have the latest versions), 
or the user configuration (not discussed), and/or the user operations (not 
discussed).  In other words, the testing was revealing that the author of 
the PDF Form probably didn't need to change anything, it just that further 
debugging will be needed (by the testers?? by someone to determine the 
root cause of the inconsistency.  If the screen reader never worked we 
would be having a different discussion.

The above dialog is *not* about law, regulations, and other disability 
rights stuff.  I'm talking about simple software engineering concepts 
here.  We can have a disability rights discussion, but let's try not to 
mix the two - please, it makes my head hurt.  The PDF form was either 
tagged correctly or not per the specification, or the spec is consistent. 
It is a simple technical conformance (or technical compliance - See Note 
1) questions with PDF.  There are in this case at least two assistive 
technologies, screen readers, that have some level of support for PDF.

Note 1 - Some people or cultures or languages seem to have a problem with 
using conformance vs compliance.  Use whichever term fits you best, but I 
am talking about software and web specifications here, not disability 
rights legislation.

> For me, "Accessibility Supported" should be understood as "there are 
> enough ... that have no additional cost . . . 

Thanks for sharing your opinion, but "enough" and "cost" is a policy 
discussion in my pinion, not a technical software engineering discussion. 
"Costs" and "enough" is only mentioned WCAG 2.0 (see Note 2) as one of 4 
choices (2d), but I do agree it should be part of the policy discussion. 
Policy to me relates to the laws and regulations like ADA in the U.S., DDA 
in the U.K., or a enterprise policy for its employees.  The purchasing 
policy in Section 508 of the U.S. Rehab Act for example, which has nothing 
to do with the ADA, does NOT have any policy provisions for purchasing 
only "free AT", or only purchasing ICT that works with "free AT".  The ADA 
policy, administered by the Dept of Justice, does not mention Internet, 
WCAG, or screen readers - so it is being left up to everyone's 
interpretation.  So do you see how different each policy can be?  And how 
we can still have a discussion (although somewhat technical) on whether 
the PDF document itself is tagged correctly or not, independent of the 

Note 2 - 

Phill Jenkins, 
IBM Research - Human Ability & Accessibility Center
U.S. Access Board

Ramón Corominas <listas@ramoncorominas.com> 
03/31/2009 02:54 AM

Phill Jenkins/Austin/IBM@IBMUS
Re: accessibility supported questions

Hi, Phill.

If I understand your answer, it doesn't matter for compliance if there 
is assistive technology that is able to interpret that content. So the 
content can comply, but nobody can access it? If there is no screen 
reader capable to consistently render the content, it can not be 
considered compliant, I think.

For me, "Accessibility Supported" should be understood as "there are 
enough user agents (including assistive techonolgy) that can render the 
content, and there is enough variety of user agents of this kind that 
have no additional cost for a person with a disability, compared to a 
person without a disability". Of course, in this context, perhaps 
"enough" can be only one unique free user agent that works on a free 
operating system.

This has the following consequences:

- Adobe PDF, Adobe Flash and WAI-ARIA can not be considered as 
"accessibility supported" until they are supported by user agents and 
assistive technology.
- Any technology that is only supported in Windows can not be considered 
as "accessibility supported", because Windows is not free, so if users 
with a disability are forced to buy Windows, there is an additional cost 
for people with disability. The same could be aplicable to Mac platform, 
of course.
- Any technology that is only supported by JAWS and/or Window Eyes can 
not be considered as "accessibility supported", because these programs 
are really expensive, meaning that there is an additional cost for a 
disabled person.

Perhaps I am missunderstanding the concept of "accessibility 
supported"...? What do you think about it?

Ramón Corominas.

Phill Jenkins escribió:
> Seems like your are testing the screen reader's support of the form, 
> not the form itself.  Did you report your inconsistencies to the 
> developers of the screen reader?
Received on Tuesday, 31 March 2009 18:09:27 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:39 UTC