Rating models

I wouldn't object at all to the application of different rating models in
addition to the existing priority scheme, though it would no doubt give
rise to additional complexity and may confuse readers if included in the
documents. I agree with Ian that openness and transparency are of vital
importance, and that criteria need to be applied consistently. This is why
I have argued (and I take it that Ian is not disagreeing here) that in
applying the existing priority scheme, the criterion of impact is the only
relevant consideration, taking account of the current state of technology
and web-related standards. In deciding among different requirements, any
of which would alone be sufficient to remove an access barrier, we have
indeed considered cost (time and effort required on the part of a content
provider), among other factors.

In setting out the requirements document for the next version of the
guidelines, it would be useful to agree more explicitly upon the factors
that ought to be considered in deciding (1) what requirements should be
included in the guidelines, as checkpoints; and (2) how these are to be
assigned priority levels.

There are several reasons why I would argue that cost (to use a general
term) should not be taken into consideration:

1. What is time consuming and difficult for an individual web site
developer may not be unduly onorous from the perspective of, for example,
a large organisation that already invests substantial resources in content
development.

2. Cost is significantly reduced if and when certain tasks are automated
by an authoring or repair tool.

3. Costs will vary depending on the design of a web site and the tools
involved.

4. The problem of balancing the need to provide equality of access against
the expense of doing so, is a political and, in many jurisdictions, a
legal issue, which this working group is not well equipped to prejudge and
the precise outcome of which varies according to individual circumstances,
as already mentioned. Regulatory authorities, Courts, and web developers
themselves are in a much better position to balance these requirements and
to determine whether or not a particular requirement should be followed.

5. The best that the W3C can achieve is to provide as accurate an
assessment as possible of the extent to which satisfaction of a particular
checkpoint will remove access barriers, leaving it to content developers
and, where necessary, regulatory bodies to examine the policy implications
(after all, such bodies are in a much better position than the W3C/WAI to
do so in their own specific contexts).

This is why I like the existing priority definitions, and the conformance
requirement (namely that conformance is not mandatory; this is a W3C
Recommendation and not a governmental regulation, but if any individual or
organization wishes to assert conformance, it must do so in the manner
prescribed by the document). I also think the guidelines provide valuable
input to regulatory processes in as much as they offer the best available
summary of access requirements, means whereby they can be met, rated
according to impact upon the accessibility of the web to people with
disabilities.

Received on Thursday, 16 December 1999 22:24:10 UTC