W3C home > Mailing lists > Public > w3c-wai-er-ig@w3.org > April 2004

Re: EARL Testcase

From: Charles McCathieNevile <charles@w3.org>
Date: Thu, 15 Apr 2004 11:11:02 -0400 (EDT)
To: Chris Ridpath <chris.ridpath@utoronto.ca>
Cc: WAI ER IG List <w3c-wai-er-ig@w3.org>
Message-ID: <Pine.LNX.4.55.0404151101020.32709@homer.w3.org>

Well, one thing to do is to try and agree on URIs to use. For example, I have
used the URIs of the checkpoints to cover claiming each individual
checkpoint, and that is what WAINu, HERA, and Axforms all do.

If you want to declare that 2 URIs mean the same thing you should use an OWL

foo:waigl owl:sameAs bar:wcag1

which relies on tools that understand a bit of OWL to make it work. If you
only want to require tools to understand RDF Schema you can go with

foo:waigl rdfs:subClass bar:wcag1 AND bar:wcag1 rdfs:subClass foo:waigl

So long as the information is available (Hera reports actually link to a
bunch of stuff. Axforms reports include a lot of information in the report)
whether a particular tool understands it or not is a problem for the tool

But having this kind of OWL stuff is probably useful. I recently wrote a
piece for the WCAG group [1] explaining how to use some other OWL stuff to
describe the fact that meeting some checkpoint was the same as meeting some
set of sub-points - for example WCAG double-A is the set of WCAG level A plus
all priority 2 checkpoints, but it applies equally for the more detailed work
people are doing.

The alternative is to make up special EARL magic which people would still
have to implement - it seems to me easier to sell the idea that implementing
this stuff is in fact doing stock-standard work for your basic parser, and
you can expect to find the right piece of basic parser off the shelf if you
look, rather than some once-off code useful only in EARL tools.

[1] http://lists.w3.org/Archives/Public/w3c-wai-gl/2004AprJun/0025



On Thu, 15 Apr 2004, Chris Ridpath wrote:

>EARL can be used to state that a particular resource passes or fails a
>particular test. The test can be something subjective like the WAI
>guidelines. We can use the earl:testcase element to make these sort of
>Our checker program makes a statement that a particular page passes or fails
>the WAI guidelines and uses our URI as the definition of the WAI guidelines.
>rdf:resource="http://checker.atrc.utoronto.ca/wcag-1-0-aa.xml" />
>Other programs will generate a similar statement and will reference their
>definition of the WAI guidelines. Example:
><earl:testcase rdf:resource="http://accessibility.tester/WCAG-AA.html" />
>How can a program collect the EARL results from various checking tools and
>tell if the page passes/fails the WAI guidelines?
>To put it another way - How can you tell that both programs are testing the
>same guidelines? (Perhaps using the earl:testcase rdf:about attribute?)

Charles McCathieNevile  http://www.w3.org/People/Charles  tel: +61 409 134 136
SWAD-E http://www.w3.org/2001/sw/Europe         fax(france): +33 4 92 38 78 22
 Post:   21 Mitchell street, FOOTSCRAY Vic 3011, Australia    or
 W3C, 2004 Route des Lucioles, 06902 Sophia Antipolis Cedex, France
Received on Thursday, 15 April 2004 11:11:13 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:01:35 UTC