W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2003

Re: Slightly OT: Web Accessibility Certifications

From: Charles McCathieNevile <charles@sidar.org>
Date: Tue, 20 May 2003 14:59:52 +0200
Message-ID: <3ECA26C8.80205@sidar.org>
To: Kynn Bartlett <kynn-edapta@idyllmtn.com>
CC: "Section 508.US" <tagi11@cox.net>, "Dwight H. Barbour" <dbs@dbsolutions.net>, w3c-wai-ig@w3.org

Kynn,

summary: silly joke, there are risks in how NFB approaches 
certification, there are procedural ways to minimise the risk, and I owe 
an email on some technical stuff for certification.

Note that everything I write here is my personal opinion only.

you could always get a license for YASR, Screader, Speakup, Gnopernicus, 
Mercator, or emacspeak. They're free (although most of them are not 
available to Windows users) :)

More seriously, I share your concern about fragmenting accessibility 
into accessibility for one user group group or other. While I understand 
a group like NFB limiting itself to its own expertise and constituents I 
think there are long term problems created by their apparent approach. 
It seems to lead to a number of "Accessibility Certifications", each of 
which is really just access for a particular group of users. There also 
seems to be no attempt to fit this into a framework where people can 
attempt to cover all types of users.

A similar effect came about when tools like Bobby offered their own 
certification mark, which the marketplace believed was a real 
certification of accessibility. (The mark is no such thing, and in any 
case far too many users of Bobby didn't check whether they actually 
qualified for the mark - they saw it, so they copied it).

The danger is that the market will come to behave as though achieving 
any certification is enough, whether it covers all disabilities or just 
some factors for some users. The probably inevitable result of that 
would be that the easiest path will be followed, cutting out anyone 
whose needs aren't met by that certification program.

There are ways to minimise the risk - one is technical, and one is about 
how to work, and they go together. The how to work one is easier to 
explain, and even demonstrate with examples, so I'll start there, and 
discuss the technical stuff in another email.

The basic idea is to coordinate with other groups, making sure there is 
an overall understanding of accessibility for all. In that context, it 
is possible to say "we have expertise in a particular area, and can 
certify that part, but the whole field is important, so there are people 
who can provide the experise in (all the other stuff we don't do)". This 
is essentially how WAI works, as I understand it. There are experts on 
issues related to blindness, hearing impairment, intellectual 
disabilities, sensory disabilities, etc. Collectively they can work on 
identifying all the possible problems people with disabilities might 
face. If they specify those clearly enough, somebody can check a page to 
make sure it contains none of the problems. Or two people can divide the 
assessment between them, concentrating on what they know best.

When the work is divided, whether it is between two people, or 17 
organisations, it is important to check that everything did get tested 
at some point in the process. A simple approach to this is to have the 
"list of everything to check", and for each one of the testers to mark 
off the things they are testing for. If there are any potential problems 
identified that nobody is checking for, then anything tested by that 
group may or may not have a problem, and nobody knows yet.

So how does this relate to the NFB (and WAI)? Let's assume that WCAG is 
the most complete available list of things to check for. (It isn't 
perfect, but it isn't bad, and it was developed in the way I described. 
For our example it works.) I will assume that the NFB certifies people 
based on how they do tests for things known to cause problems for people 
who are blind. They have a list of the things that an evaluator is 
supposed to find in test pages.

They should be able to state which parts of WCAG are tested with their 
process. Assuming it isn't everything, there is therefore a list of 
things their process doesn't cover. If they publish this list, another 
organisation with different expertise could develop a certification for 
covering those points. (Or it could be done by combining several 
certifications - the fact that there is some overlap would be a good 
thing for the skill of the tester).

In this way, a group of cooperating organisations could provide 
certification in their area of expertise, as a stand-alone product, and 
could help raise awareness of the fact that there are different types of 
disability. They are also part of a process that can certify people as 
covering (in theory) all the problems for different types of accessibility.

People who achieve accessibility for one group of people by making lots 
of dynamic multimedia presentations, and those who achieve it for a 
different group by having well-structured text-oriented presentations, 
are helpful to organisations trying to communcate with users. People who 
can solve both problems at the same time are valuable too. But if a 
person who can only do one or the other is mistaken for someone who can 
cover accessibility completely then we will have a serious problem in 
the marketplace - fragmentation and confusion over what needs to be 
done, and people who thought they were doing the right thing failing to 
solve problems, which means the problems still exist.

In Europe there are a number of organisations certifying accessibility. 
They speak different languages, have worked with different disabilities, 
and have interpreted the guidelines in different ways. There is a desire 
to have accessibility certification be at the level of the best practice 
internationally, and to have consistency. One approach to this has been 
24 organisations (including Fundacion Sidar, which explains my 
involvement) getting together to form the EuroAccessibility  Consortium. 
In coordination with and through participation in WAI, which the 
consortium has recognised as setting the standards on an international 
technical level, we hope to develop an evaluation/certification process 
which does provide a measure of exactly what is covered, and which 
refers to the "complete" standard requirements to make the Web 
accessible to all.

Cheers

Chaals

Kynn Bartlett wrote:

> Hey, neat, I don't qualify:
>
> http://www.nfb.org/seal/wacs.htm
>
> I don't use JAWS or WindowEyes, which seem to be a requirement.
> ("Do you own licenses for at least two screen reading programs?" -- no.)
>
> I guess I'm not good enough to be an NFB "Web Accessibility
> Consultant".
>
> Based on the criteria they use -- which focuses ENTIRELY on designing
> pages which can be used by blind people, without any apparent concern
> for people with disabilities which are not visual -- perhaps this should
> be called Blind Web Accessibility Consultant?
>
> --Kynn, Mac OS X user who would have to spend over $1000 to buy
> Microsoft Windows software he'd never use in order to be certified...
>
Received on Tuesday, 20 May 2003 09:00:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:09 GMT