- From: Jan Richards <jan.richards@utoronto.ca>
- Date: Mon, 19 May 2003 17:17:02 -0400
- To: WAI-AUWG <w3c-wai-au@w3.org>
Minutes for AUWG Conference Call (May 19,2003) JT: Jutta Treviranus TB: Tim Boland KB: Kynn Bartlett JR: Jan Richards PJ: Phill Jenkins ACTIONS: JT: ACTION: classify checkpoints accoring to whether external tester is required and send to list TB: ACTION: to draft some test cases TB: describes purpose of test suite principles - intended to stimulate discussion. Not a testing process document. Comments are welcome - this is a draft. Intended to be consistent with QA activity documents - not replace them. JT: Would you see the test suite as a series of test cases? JR: Do you see anything specific to tools opposed to content? TB: Not really... JT: Would structure be that user would perform a task and then there be an evaluation of the task. TB: More or less. JT: That would work well for our Rel Priority checkpoints. But what about integration, etc. that goes beyond a single tester. JT: For WCAG, one person could evcaluate. JT: For some ATAG checkpoints , we may need to separate author and tester. JR: Could be an automated tester. KB: We might need to be more explicit about types of author. Checkpoints apply to 3 types of authors [naive (check, correct), advance (i.e. allow accessibility), author with disabilities]. This will have an impact on the design of the test suites. TB has a good doc. TB: What's the best way to move forward? JT: The priciples are general enough to take account of the various types of authors. KB: That comments applies more to ATAG doc. TB: Should we start on a sample test case. JT: Would be good to select some representative checkpoints for that. KB: Will this be a vague set of instructions for the author? ATAG is about a process. @@Phill Jenkins joins.@@ JT: AERT exists to check correction and repair. JT: We can use instructions in some cases. The more difficult cases will be the guiding and support type guidelines. TB: could you classify the checkpoints JT: ACTION: classify checkpoints accoring to whether external tester is required and send to list TB: ACTION: to draft some test cases PJ: Question on some of the terms. "Edge testing"? TB: To determine the behaviour at limits. PJ: Test suite is more like WCAG suite, and then we have other authoring tool specific requirements. JT: We reached a balance in which we don't explitly say what needs to be done to prompt. We are not prescriptive when it comes to UI. KB: Difficult to make absolute judgements whhen users differ so much. JT: we are trying to get around that with "average author of the tool" JT: can we start tailoring it for ATAG. TB: It began generic, but would like to add ATAG specific wording. KB: What about requirement for validity? Will this rule out exception tests? TB: Did not mean to rule this out. JT: We can make a note of this. All: discussion of validity of test files. Moved on to 250 HTML test files - may or may not be public. WCAG1.0 files are public. =============== E-GOV Discussion (PJ has a potential BOF opportunity during E-Gov) Next meeting June 2nd - we will discuss (1) Kynn's proposal, (2) Janina's structure checkpoint concern and (3) finalize F2F agenda.
Received on Monday, 19 May 2003 17:58:35 UTC