<shadi> Scribe: Andrew K.
<karl> scrunch scrunch lunch
sa: test suites - who has seen it?
test suite status http://www.w3.org/WAI/GL/WCAG20/tests/checkstatus.html
sa: helpful for eval tools?
<chrisoula> hello IPcaller should probably be me, but I cannot connect through the phone
<Zakim> karl, you wanted to ask about tests, test suite and test assertions
kd: must be in EARL
... distintion between assertions and test cases
... useful to see work flow that can be used
... more than test cases needed
... EARL should be independant of test cases
sa: yes EARL should be seperate from test
cases
... 3 layers of WCAG
... OWL or RDF might also be used to define relationships
... EARL is other component
... should look at test case description language TCDL too
wc: working group work focus on WCAG clear
enough to get to rec status
... need to get through process to rec status
... still trying to figure out some details, looking for help from ER
gb: EARL too general, not specific or
complete
... likes checks, uses in ht/checks
sa: is EARL too general? What is missing?
gb: finds that EARL is not very usable - start with concrete case
sa: EARL needs more concrete examples
sa: make sure that test suites are what we need
to produce good evaluations
... what do test cases lack?
... example, measure for completeness
... should be helping tool developers
... keep eye on test suite to see how we can support it.
<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2005Feb/0015.html
sa: complete?
ma: complete
gb: seems OK for me
... how to make assertion in Italian?
sa: internationalization is important
... use name attributes, multiple messages in EARL
... if description from test case, that can be used, no need to generate your
own
... somewhere translation must exist
... multilingual issue raised elsewhere, important
ga: have translated some, can provide
sa: need to look at structure of test case
within WCAG, use TCDL?
... we need spec
sp: can we specify what EARL does?
<karl> http://esw.w3.org/topic/TestCaseMetadata
<karl> http://esw.w3.org/topic/TestSuiteArchitecture
sp: how to define need for EARL?
<karl> RDF Test Case Schema http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema.n3
sa: need to say what EARL can't do
<karl> http://esw.w3.org/topic/TestDevelopmentMethodologies
sa: would like to produce some sort of
requirements this quarter
... asks SP if he could write up his ideas
sp: trying to figure out what EARL's really
gonna do
... what can't we do that EARL will do?
sa: use cases on list demo what EARL does
... uniform way of describing test results - will discuss with SP off line
ma: senarios are OK but don't cover everything so more senarions needed
gb: useful to define requirement documents
<karl> chrisR this time it's Karl ;)
wc: do senarions first will clarify requirements
sp: ack
sa: generate requirements first or use cases first?
kd: if cycle is short enough then can do
both
... after creating user senario then draw out requirements
sp: both are related - without user needs can be no requirements
sa: ack
... ask for discussion on list - use cases seem to be complete now so on to
senarios and requirements
ma: at FTF to take ideas from list and discuss
sa: will have draft doc that can go back to
group
... other topics?
<shadi> http://www.w3.org/WAI/eval/
sa: will have joing meeting with EO where we
can discuss
... where does EARL fit into this process, tools too
... another topic is test suites, Friday meeting will have observers from
Techs Task group
<scribe> ACTION: everyone will review user senarios and identify any gaps [recorded in http://www.w3.org/2005/02/22-er-minutes#action01]
<scribe> ACTION: brainstorm and produce requirements to list for all [recorded in http://www.w3.org/2005/02/22-er-minutes#action02]
sa: next meeting Mar. 8 (no meeting next week)