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

Re: conformance to functionality classes Re: Proposals: Priority and Conformance schemes

From: Charles McCathieNevile <charles@w3.org>
Date: Sun, 9 Sep 2001 16:06:22 -0400 (EDT)
To: Jason White <jasonw@ariel.ucs.unimelb.edu.au>
cc: Web Content Guidelines <w3c-wai-gl@w3.org>
Message-ID: <Pine.LNX.4.30.0109091551330.29794-100000@tux.w3.org>
If all we are doing is enabling developers to state which checkpoints they
meet, then we have nothing to do and need not write anything about
conformance to the document as a whole (I already argued this in another
sub-thread).

I agree that this document should not be attempting to describe policy
setting priorities. The existing scheme attempts to explain how important
various checkpoints are to users, without differentiating users into
different groups.

I think there are two important parts of that.

 One is that it doesn't say anything about which type of people is more or
less important. In the real world of implementations being developed with
associated cost, someone may have to make such decisions. But we don't have
any authority that I can imagine for doing so, since in the abstract it is
merely an arbitrary decision based on factors related entirely to individual
situations.

The other is that it isn't based on how easy something is, but how important
it is - what does it mean to users.

I believe that there is plenty of value in the statement that we should
critically re-examine the particular priorities we have assigned, as well as
the checkpoints we have assigned them to.

I am much less convinced that the idea behind the current scheme is flawed in
ways that the various new proposals would resolve.

The one really sound argument I have seen is that people may not attempt to
meet WCAG at any level except those of particular conformance levels, and
that therefore people will not d othings that are valuable and easy because
they aren't contributing to achieving a particular target level.

This seems a very foolish approach to accessibility, and we may be able to
help. It should be clear from the existing WCAG that every checkpoint is
helpful to people, and meeting any checkpointis a good thing. It should also
be clear that conformance levels represent basic minima rather than ultimate
goals. The ultimate goal is universal accessibility regardless of disability,
and getting to triple-A is at least a good step alolng that path, although it
may be insufficient. Settling for less is in any case a compromise between
reaching the goal and saving development effort for something else. But
that's what happens in the real world, so we need to acknowledge it.

The goal is that this works for the people who read Web content. A minor
objection is that the proposal for functionality classes fails to identify
who wins and loses in each combination. A major objection is that this
approach seems to me fundamentally counter-productive when reaching for our
ultimate goal of universal accessibility, by making arbitrary divisions of
people.

Cheers

Charles

On Sun, 9 Sep 2001, Jason White wrote:
[snip]
  If the purpose of the conformance scheme is to enable developers to make
  assertions, accurately and verifiably, regarding which checkpoints, and
  therefore which needs, have been met, then such a consequence is
  unobjectionable.

  However, if the purpose of the guidelines is to set a policy (snip) then
  Charles' concern becomes a valid objection to the kind of proposal that I
  have outlined.

  One of Kynn's criticisms of the WCAG 1.0 conformance scheme, as I
  understand his argument, is that it is overly prescriptive in the way
  in which it sets priorities and requires developers to privilege some
  needs over others. The concept of a multi-dimensional conformance
  scheme is intended to avoid this shortcoming.
[snip]
Received on Sunday, 9 September 2001 16:06:24 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:12 GMT