W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 2001

Re: conformance

From: Charles McCathieNevile <charles@w3.org>
Date: Thu, 11 Oct 2001 07:17:34 -0400 (EDT)
To: Anne Pemberton <apembert@erols.com>
cc: W3C Guidelines <w3c-wai-gl@w3.org>
Message-ID: <Pine.LNX.4.30.0110110702290.27108-100000@tux.w3.org>
The problem of most accommodations for cognitive disabilities being only P3
will not be solved by removing the priority level, it can only be solved by
changing the priority of individual requirements. This is not to say that we
should not consider a two-level scheme, just that the argument you give here
doesn't justify it.

You then identify one need for a reporting scheme - the desire to start with
a base level, and support for known audience (e.g. there is one blind student
in my class so it is more important to add the things that student definitely
needs than things that someone might need one day).

Of course, in a perfect world, all material will be "completely" accessible,
but we don't live there yet - and won't until the tools people use to make
Web content do most of it for them.

Which leads to another audience for a reporting scheme. People in the process
of producing content, especially where there is more than one person involved
in the production, as is the case for many large sites, need to know what
they have done and what still needs doing. A reasonably standard method would
enable people producing content to more easily use different tools, each for
what they do best.

I don't think this group needs to develop a reporting mechanism (in part
because other groups in W3C are working hard on it, including the people who
produce tools like Bobby, aprompt and the various plugins for assessing
accessibility in authoring tools, who all believe they have an audience). But
it is important that the guidelines document at least enables such a scheme.
Fortunately this is not hard, and it would be hard to produce a document that
met our own requirements and did not do this.

(Technically, what it requires is that each checkpoint have a URI, and with
Xpath or Xpointer that means that each checkpoint is a seperate element,
although ideally it would have an ID attribute. We need that to make the
techniques documents useable anyway, and already have it in all the drafts
we have produced to date. So in many ways it is a moot point.)

Finally, given a decent reporting scheme we will enable people to find
content that works for them, whether or not it is "completely accessible" - a
need that you have often identified.

Charles McCN

On Thu, 11 Oct 2001, Anne Pemberton wrote:

  In the triple-layer scheme, the accommodations for cognitive disabilities
  were only required at the top level which conveyed a sense that they were
  basically unnecessary, barring a lot of people from using sites. Reducing
  the conformance to two levels eliminates the level no one ever strives for ...

  Using the purpose of a site to determine what conformance level to use
  makes the most sense to me. If a site has a limited audience, go for
  minimum and add only what you need to serve your known audience .... but if
  your site is for the general public, then it must comply such that everyone
  can use it and none have the empty plate. To do anything less is to consign
  some users to only using entertainment sites and never getting at any meat.

  I see no need for a reporting system since there is no audience for the
  reports. It's a waste of time.

                                                   Anne
Received on Thursday, 11 October 2001 07:17:34 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:16 GMT