W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > April to June 2004

Regrets and comments Re: Agenda April 15th

From: Charles McCathieNevile <charles@w3.org>
Date: Wed, 14 Apr 2004 02:45:02 -0400 (EDT)
To: Gregg Vanderheiden <gv@trace.wisc.edu>
Cc: w3c-wai-gl@w3.org
Message-ID: <Pine.LNX.4.55.0404140236500.28932@homer.w3.org>

>Agenda for this week
>1 - Discussion of proposal for Conformance Scoping

Seems sensible on the whole. I'll sendd specific comments (including showing
how to define different guidelines in terms of each other in a way that lets
you show that they are or are not subsets of WCAG)

>2 - Definition of Levels?

From a semantic Web perspective I have no opinion on this. From Sidar's
perspective we concluded that the old WCAG 1 thing was as good as anything,
and that anyway discussing the levels now is premature since we don't even
have the requirements and user benefits of each laid out in detail.

>3 - Consensus continued
>          (primary goal is to collect all ideas - identify consensus
>           if there are any easy issues.)
>    a) - Is sampling OK for large sites or not?  Why?  If so -
>           any restrictions?  Resampling?
>    b) - How about aggregated content?

I don't see any reason why we would authorise conformance claims unless
people have grounds for considering them to be true. I think we should be
encouraging people to avoid claiming aything unless they are sure it is true
- this is an incentive to actually implement proper checking systems that can
keep track of it instead of guessing, or accepting the results of a tool that
gets close to doing an assessment.

With the Semantic Web methods for claiming conformance I discussed last week
people can readily maintain and work with information that says what they do
conform to (on a point by point basis as well as on a levels basis) and use
that to match for various needs - planning future improvements, determining
whether a given user with a particular profile will be able to use the
content or not, etc. If people make blanket claims because they think they
are overall true, then these possiblities disappear through lack of accurate


Received on Wednesday, 14 April 2004 02:45:03 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:48 UTC