W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2006

Re: What does WCAG mean by "a set of Web units"

From: Jason White <jasonw@ariel.its.unimelb.edu.au>
Date: Fri, 18 Aug 2006 12:16:06 +1000
To: WAI WCAG List <w3c-wai-gl@w3.org>
Message-ID: <20060818021606.GA7826@jdc>

On Thu, Aug 17, 2006 at 06:29:14PM -0700, Loretta Guarino Reid wrote:
 
> We started a discussion at the end of today's call about what we mean,
> in these SC, by a set of Web units. Some of the ideas that were
> discussed included:
> -	the web site
Ambiguous, since the term is not well defined, with no commonly agreed upon
criteria with which to decide what is included and what isn't.
> -	the contents of the conformance statement
This has the advantage of clarity. The conformance claim, after all, must
specify a set of Web units to which it refers, and these could easily be taken
as the "set of Web units" for purposes of the relevant criteria. The result
would be that a separate conformance claim would have to be given for each
such set.
> -	a set of Web units that is intended by its author to be used
> together
If it isn't clear to a potential evaluator from reading the cofnormance claim
and the Web content what the set is supposed to contain, then the conformance
can't be properly assessed.
> -	all of the Web units within a site that can be reached from the
> graph starting at some root Web unit.
Subject to the same problem, as above, in defining what a "site" is, i.e., how
its limits are established.
> 	
> More thoughts or suggestions?
I would prefer the set of Web units to be defined as comprising all of the Web
units within the scope of the conformance claim. Thus, if the claim is made
for an entire "site" (whatever that means) then that's the set of Web units;
but if the author decides to make the claim with respect to something smaller,
for whatever reason, then that's the set of Web units for purposes of the
success criteria.

I would also maintain that the set of Web units can never be larger than
whatever is specified in the conformance claim as lying in the scope of
conformance. If the set of Web units extends beyond the scope of the
conformance claim, then the problematic situation arises whereby the
conformance of Web units within the scope of conformance depends on the
conformance of Web units outside of it. Yet, the whole idea of specifying the
scope of a conformance claim is to circumscribe the material which is subject
to the guidelines.

Thus, if I make a claim that applies only to a single Web unit, whether the
claim is true should not depend on other Web units to which mine is linked;
the latter may well be outside my control or subject to other constraints,
which is why they weren't included in the conformance claim in the first
place.
Received on Friday, 18 August 2006 02:16:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:46 GMT