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 11:35:18 -0400 (EDT)
To: Anne Pemberton <apembert45@yahoo.com>
cc: <w3c-wai-gl@w3.org>
Message-ID: <Pine.LNX.4.30.0110111122490.8682-100000@tux.w3.org>
I think your perception is accurate, that P3 items were ignored by many
people becuase they were focussed on the high priority, high impact stuff (as
assessed in the priority scheme we have).

If there are needs that are very important, but which are listed in WCAG 1 as
P3 then we have simply given them the wrong priority.

The reality is that there are 68 checkpoints in WCAG 1, and the priorities
based on impact are only one of that factors that is taken into account in
developing any particular implementation plan for accessible content
production. The less we distinguish the priority of requirements based on
user needs, the more that other factors (cost of implementation, perceived
value for known target audience, who happens to be working that day, etc) are
goingto influence the priorities in what gets done. (Everyone is aiming to
get them all done. But some things happen before others).

On the other hand, the more complex the scheme we provide, as you point out,
the harder it will be for people to understand, and the more likely that
people will settle for something in the middle.

Personally, I think that it is easy enough to sell the idea that getting 2/3
of the way there is really important to make a high priority, but that if we
had an even number of levels people would settle for halfway, and if we had 5
or 7 people would be likely to settle for whatever was one step above
halfway. That's just my feeling, but it is based on giving a presentation on
this about every 3 weeks. If you can say "must, should, may" or "must,
should", the first one gets the shoulds included more often, so the net
result is better. (Actually, I think that the long term result will be that
we get most of the "may" stuff too, but that is some time away yet
undfortunately.)

Which is one reason I am in favour of three levels.

Another is that I think the existing priority _scheme_ is right, although I
think there are existing checkpoints whose priority is not assigned
correctly.

The final one is for convenience. In developing the Authoring Tool
Accessibility Guidelines specification we depend very heavily on WCAG and its
structure has a big influence.

cheers

Charles

On Thu, 11 Oct 2001, Anne Pemberton wrote:

  Charles,

       Unless my perception isn't accurate, the version
  1.0 plan of three levels didn't work since few if any
  people bothered with level three. So why have three
  levels. It seems that two are enough. It's not that P3
  was ignored because it was the cognitive needs, but
  because the indication was that few people were
  supposed to be left out at the P3 level. Whoever's
  needs are in P3, unless they are purely fluff (gee it
  would be nice's), are going to be ignored.
Received on Thursday, 11 October 2001 11:35:36 GMT

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