See also: IRC log
<shadi> http://www.w3.org/WAI/ER/tests/usingTCDL
<shadi> http://lists.w3.org/Archives/Public/public-wai-ert-tsdtf/2006Nov/0002.html
CS: summarizes chages done in usage document
SAZ: will update usage document according to
changes in TCDL 2.0
... wasn't there a general agreement about adoption of HTTP Vocabulary and
EARL location pointers?
<shadi> ACTION: SAZ to add the elements that were recently added to TCDL 2.0 to the "usage document" [recorded in http://www.w3.org/2006/11/14-tsdtf-minutes.html#action01]
CS: currently working on last changes to be online later this week
<shadi> s/was't there general agreement/there was general agreement
<shadi> ACTION: CS notify mailing list once TCDL 2.0 document is updated [recorded in http://www.w3.org/2006/11/14-tsdtf-minutes.html#action02]
SAZ: two more questions: first one about nesting of locations and techniques
CS: techniques are after location (original proposal has been kept)
SAZ: does this need further discussion?
CS: not from my side
... request came from error (techniques used at two locations)
<scribe> ACTION: everybody review this part about locations and techniques and make sure you agree [recorded in http://www.w3.org/2006/11/14-tsdtf-minutes.html#action03]
CI: could someone summarize the information in the mailing list?
<shadi> start of thread: http://lists.w3.org/Archives/Public/public-wai-ert-tsdtf/2006Oct/0024
SAZ: second question about usage document:
naming convention
... naming convention works for metadata files pretty well, but when you
start to name the content files, you could get problems if you want to reuse
files
... one test could be useful for more than one success criteria
... how do we deal with that? copy?
CS: some techniques are relevant to more than 1 SC
SAZ: how are you dealing with this problem in the BenToWeb project?
CS: we copy them
SAZ: reusing creates a dependency, e.g. when
you change the content for a certain reason you have to consider also the
other success criteria that are attached to this test.
... naming conventions currently only reflects success criteria
... techniques are referred in metadata
... is it a resolution to do duplication rather than directly reusing it?
... any objections to duplication?
RESOLUTION: test files used for more than one success criteria will be duplicated
<CarlosV> '/WAI/ER/tests/metadata'
<CarlosV> '/WAI/ER/tests/testfiles'
CV: there are subdirectories "metadata" and "testfiles" on first level in CVS which sould not be there
SAZ: will delete those folders
... any other questions or remarks on the usage document?
<shadi> http://www.w3.org/WAI/ER/tests/Overview
SAZ: also working on an overview page (link
"test samples" on the left hand side)
... document should give an idea about the test samples repository
... process for test samples does not yet exist - task force has to set that
up
... summarize: TCDL2.0 will use earl location pointers and http vocabulary,
but that has not been in the document when I wrote usage document
SC: currently work on http vocabulary
<shadi> http://www.w3.org/WAI/GL/WCAG20/tests/ctprocess
SAZ: have already talked about directory
structure that before.
... regarding the process, the document linked above could be a good starting
point
... have tests been developed according to this process?
CR: yes
MC: challenge was to get people to submit test
samples
... the process is ok, the problem is getting stuff into it
SAZ: this task force is not a working group and
has very specific goals - so this could be a benefit of this task force
... we have stuff available, what we need is the process
... what are the experiences from BenToWeb?
CV: BenToWeb is currently working on review
process.
... there were heavy discussions about "test mode"
when author creates test sample, the status is "draft"
Then the first validator takes a look at the documents - any requests to the author to change, then "pending bugfix" - no requests, then "accepted by first validator"
Then we still have a second validator and end user test
<scribe> ACTION: CV summarizes how process is done in BenToWeb [recorded in http://www.w3.org/2006/11/14-tsdtf-minutes.html#action04]
SAZ: the test samples cannot be more stable than WCAG 2.0 and corresponding techniques, any changes in techniques need to have another review round
CV: status of test case and status of WCAG 2.0 should be independent
SAZ: new version of techniques would need new
version of test samples.
... this task force is not a decision making group within W3C process.
... have to take that into account in our process
... another aspect: how is a test sample assigned to somebody?