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 19:03:32 -0400 (EDT)
To: Shadi Abou-Zahra <shadi@w3.org>
Cc: 'Chris Ridpath' <chris.ridpath@utoronto.ca>, 'WAI ER IG List' <w3c-wai-er-ig@w3.org>
Message-ID: <Pine.LNX.4.55.0404151852330.32709@homer.w3.org>

The problem with this, if you look at the graph of a couple of assertions
about hte same point, is that the rdf:about on the eral:Assertion means that
you are defining the meaning of that URI. In other words, you are claiming
that http://vendor-a.org/tool/tests/#check-for-alt actually represents an

If you continue down this path you start claiming that it represents a single
assertion about a handful of subjects, which simultaneously has different
results, and different testcases, rather than what you seem to be after,
which is using it as the identifier for a particular test. If we used OWL to
define that something couldn't have two simultaneous results that differed
then you would have a logically inconsistent assertion - effectively one
that is not valid. Since we don't, all we can say is that your statements
won't mean what you want them to mean...

You could copy the description mechanism used in axforms to provide
information about the Testcase and declare it as a an rdfs sub-class if it is
totally subsumed by one checkpoint, or just give it a description, or make
some statements about how several of your test relate to several others using
OWL. I'll try to write up some examples next week.



On Thu, 15 Apr 2004, Shadi Abou-Zahra wrote:

>i've also been playing around with this type of code:
>  <earl:subject rdf:resource="#subject"/>
>  <earl:result rdf:parseType="Resource">
>    <earl:validity
>    <earl:confidence
>    <earl:message>ALT Tags Exist</earl:message>
>  </earl:result>
>  <earl:mode
>  <earl:testcase
>  <earl:assertedBy rdf:resource="#assertor"/>
>...which makes aggregation very comfortable but kind of turns EARL
>inside out (as i understand the current spec).
>to me it makes sense as an assertion is really about the specific test
>which the tool conducts, the earl:testcase should be optional if this
>test really maps to a test suite. comments welcome...
>-----Original Message-----
>From: Chris Ridpath [mailto:chris.ridpath@utoronto.ca]
>Sent: Thursday, April 15, 2004 20:36
>To: Shadi Abou-Zahra; 'WAI ER IG List'
>Subject: Re: EARL Testcase
>Works for me.
>Although I'm not sure about the use of the "rdf:about" and
>attributes. Should these both be "rdf:resource" attributes?
>Charles - Any comments?
>----- Original Message -----
>From: "Shadi Abou-Zahra" <shadi@w3.org>
>To: "'Chris Ridpath'" <chris.ridpath@utoronto.ca>; "'WAI ER IG List'"
>Sent: Thursday, April 15, 2004 2:15 PM
>Subject: RE: EARL Testcase
>> hi,
>> currently i'm just using this hack for my prototype:
>> <earl:testcase
>> rdf:about="http://www.w3.org/TR/WCAG10/#tech-text-equivalent">
>>   <earl:testId
>> rdf:resource="http://vendor-a.org/tool/tests/#check-for-alt" />
>> </earl:testcase>
>> regards,
>>   shadi

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 19:03:32 UTC

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