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

Re: [July F2F]] Proposed wording on Conformance and Conformance Claims

From: Jason White <jasonw@ariel.ucs.unimelb.edu.au>
Date: Thu, 15 Jul 2004 13:00:28 +1000
Message-ID: <16629.62284.529410.479913@jdc.local>
To: "John M Slatin" <john_slatin@austin.utexas.edu>
Cc: <w3c-wai-gl@w3.org>

Turning to the interesting part:
John M Slatin writes:
 > 
 > Conformance claims
 > 
 > 1. All conformance claims must include at least the following
 > 
 > information: 
 > 
 > a. The version of the guidelines to which the conformance claim is
 > 
 > made.
 > 
 > b. The URI of the authored unit for which the claim is being
 > 
 > made.
 > 
 > (Editor's note: the phrase "authored unit" should be linked to a
 > glossary definition based on the definition of this term in the Device
 > Independence Working Group's Glossary of Terms at
 > 
 > http://www.w3.org/TR/di-gloss/ <http://www.w3.org/TR/di-gloss/> ) .

The definition says:

   Authored Unit
          Some set of material created as a single entity by an author.
          Examples include a collection of markup, a style sheet, and a
          media resource, such as an image or audio clip.

This is rather vague. How am I supposed to know what is and isn't
covered by the conformance claim; all I am given, under the above
proposal, is a URI. The definition is no help in defining the scope of
the claim. Consider a typical Web page with raster-based images - do
we interpret this as comprising a single authored unit, or as one for
the XHTML, one for the style sheet, and one for each of the images? If
so, we will need separate conformance claims for each; if not, we need
a principle that determines in general and across a wide variety of
scenarios what is included in the authored unit and what is not. My
suspicion is that we will end up requiring the author to specify the
scope of the claim explicitly.

 > 
 > c. The level of conformance being claimed. 
 > 
 > 2. The conformance level for an authored unit that contains other
 > authored units is equal to the lowest conformance level claimed for the
 > authored unit and any of the units it contains.
 > 
 > Editor's Note #1: A question has been raised as to whether the
 > information required in items 1(a)-1(c) above should all be transmitted
 > in the HTTP header or in some other way.
 > 
 > Editor's Note #2: We are currently looking at how to handle unknown or
 > community-contributed, authored units that are created using an
 > aggregator supplied tool. If the aggregator-supplied tool conforms to
 > 
 > ATAG, can ATAG conformance be used to imply that the aggregated content
 > conforms to WCAG?

No. If the original source of the material makes claims with respect
to its conformance level, then it should be permissible to rely on
this in making a claim regarding the combined content that result from
performing the aggregation. If the aggregated content is not subject
to human and machine testing sufficient to warrant a conformance claim
(e.g., it can be submitted directly to the Web site and automatically
posted without such human/machine verification being done) then it
should be excluded from the scope of the conformance claim.

This is where the principle that the conformance level of the totality
is the lowest attained by any of its components, breaks down. That
some content (e.g., aggregated material) has to be excluded from the
scope of conformance claims is another reason why the author has to be
able to specify scope, and why doubts can be raised regarding
"authored units" as an over-all framework for addressing scoping
issues.

The alternative is simply to decide that if there is aggregated
content that may not conform to level 1, no conformance claim can be
made for any larger unit containing the aggregated content. That
result, one could say, is fair enough and accords with the
above-quoted principles - i.e., if the conformance level of some of
the aggregated content is 0, then no conformance claim can be made for
any larger unit containing it.
Received on Wednesday, 14 July 2004 23:01:43 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:17:58 UTC