TSD TF

Announcements, etc.

CS: Happy New Year! :)
... had BenToWeb review in mid-december, it went quite well
... still need to do a few things but more or less closed
... still figuring out the future of this work

SAZ: all test cases on the W3C servers?

CS: no, not yet all

SAZ: not a top priority to transition them fast, we already have a lot of unreviewed ones

MC: tests are becoming ever more important, may be a good time to bring back to the WG group now
... might be good to send a formal mail to the WG, and maybe try to get a slot on the Thrusday call

SAZ: WG doing planning now or after the review period ends?

MC: doing planning now, just in case

<scribe> ACTION: CS send email to the WCAG WG explaining the status of the tests, and ask for a speaking slot at the upcoming Thursday call [recorded in http://www.w3.org/2008/01/08-tsdtf-minutes.html#action01]

CS: side note, at least 300 more test cases from BenToWeb in the queue

CI: we will talk internally in CTIC about future plans

Reviewing test samples for previous working draft

CS: several action items were set, but haven't seen follow-up
... seems nothing changed

<scribe> ACTION: CS send out reminder to the people with due action items, pointing them to the outstanding work and asking them for dates by which they can deliver [recorded in http://www.w3.org/2008/01/08-tsdtf-minutes.html#action02]

Test suite ID in metadata

<Christophe> http://lists.w3.org/Archives/Public/public-wai-ert-tsdtf/2007Dec/0000.html

CS: nothing in the metadata telling us which test suite a test belongs to
... would be useful, also to filter out old/unused tests etc

SAZ: is that different to the naming convention discussion that is still open?
... to make the ID more stable

CS: it relates as both impact the naming convention

SAZ: if we will reopen the ID/naming/tagging discusion then we may want to take a step back and collect all outstanding issues

MC: agree with the overall thought, should minimize the refactoring work
... not interested in how the IDs look like, they just need to work

SAZ: not so much about making the IDs more pretty, but if we have to touch them anyway then we should consider all aspects first

<scribe> ACTION: MC to have a fresh look at the ID/naming and linking/referencing of tests (especially linkage to techniques, stability, etc) [recorded in http://www.w3.org/2008/01/08-tsdtf-minutes.html#action03]

<scribe> ACTION: CS send summary of the ID/naming/referencing mechanism of the tests, and collect the open issues (or previously discussed issues that were not ideally resolved) [recorded in http://www.w3.org/2008/01/08-tsdtf-minutes.html#action04]

Discussion: rule/@primary and testCase/@complexity

<Christophe> See http://lists.w3.org/Archives/Public/public-wai-ert-tsdtf/2008Jan/0001.html

CS: discussion allowing non primary rules vs just one single rule per test

A - Allowing just one primary rule and several other informational ones

CI: most of the tests so far seem to only have 1 rule anyway
... allowing multiple rule makes reviewing more difficult and complex
... we should keep it simple and just use what we need

CS: we may have not added such information but that doesn't mean it isn't useful

CI: we should focus on what we need, and simplify the process
... don't see any impact on the quality of the samples

SAZ: my understanding is that the check for the proper use of the primary/compelxity attributes can be automated, and therefore it doesn't add too much reviewing overhead
... agree that we should keep it simple, but shouldn't disallow something that we may later need
... WCAG 2.0 has a lot of cross-references or relationships between different SC, especially at different Levels
... we may want to reflect this at some stage, so this feature may become handy

CI: exactly one primary rule, and 0 or more information ones?

SAZ: yes

CI: and other metadata such as expertGuidance only refer to the primary rule or to which rule exactly?

CS: don't see this as a real issue

SAZ: are there concrete examples?

CI: take color contrast SCs which are in two different levels

CS: good example, it fails Level 3 and passes Level 2

CI: then they are both primary, why is one informational?

<scribe> ACTION: CS send concrete examples of tests that make use of the primary/complexity attributes [recorded in http://www.w3.org/2008/01/08-tsdtf-minutes.html#action05]