- From: Carlos Iglesias <carlos.iglesias@fundacionctic.org>
- Date: Tue, 9 May 2006 13:13:57 +0200
- To: "Johannes Koch" <johannes.koch@fit.fraunhofer.de>, <public-wai-ert@w3.org>
Hi Johannes, > as far as I understand the "Understanding" document, there > are sufficient techniques and common failures for each > success criterion. If a subject passes a test for a > sufficient technique, the subject meets the success criterion > (sufficient to pass). If it fails a test for a common failure > (fail meaning a problem is found), the subject does not meet > the success criterion (sufficient to fail). > > On the other hand, if a subject does not pass a test for a > sufficient technique, no pass or fail statement can be made > about meeting the success criteria, because the subject may > pass another sufficient technique, even one not listed in the > "Understanding" document. And if a subject passes a test for > a common failure (no problem is found), again no pass or fail > statement can be made about meeting the success criteria. > > So in EARL we need to record whether a test (case) is > passSufficient or failSufficient for a higher level test > (requirement). I don't think so, EARL records Test Results at whatever "level" you're testing (call it success criterion, sufficient techniques, advidsory techniques...), and it shouln't care about the relations between TestCases (This is why we need a Test Description Language) If you test a sufficient technique you pass or fail this technique (although you could pass also a succes criteria) If you test a Success criteria you pass or fail the Success criteria Regards, CI.
Received on Tuesday, 9 May 2006 11:15:44 UTC