Minutes 15 Oct 2001 telecon

Present: Chris, Katie, Sean, Harvey, Wendy

Summary:
- SBP will create 1.0 schema with changes he sees should be made to make it 
easier to use. He will show the differences between 0.95 and 1.0 and create 
some examples.
- SBP Will also send thoughts about a canonical version of EARL to the list.
- We decided to meet every other week instead of every week.
- Our next meeting is 29 Oct 2001.
- CR will send info about the meta info in the test files to the list.
- WC is putting together slides for the UAAG mtg next week that could also 
be used at the QA meeting to educate people about EARL.


EARL 1.0

SBP Yes, RDF.  TestMode seems to be in the wrong place. e.g., if you want 
to say that you manually validated a page, then put that in the 
testCriteria bit.  Think it belongs in a statement by itself. When the page 
is passed/failed, it is a manual test.

HB Any way to standardize assertions of a manual test? If manual, subjective.

SBP Good point. In other words, why need the info anyway?

HB We need it, but we need to give guidance on how it can be used.

WC Important for conformance. e.g. "as clearly and simply as 
possible."  Need to know if person just "think it is" or ran a fog index on it.

HB if text equiv makes sense.

WC That's where test specification languages come in.  CR, what if move to 
1.0 right now?

CR Fine to me. Not sure if current EARL that ATR generated is correct. How 
do we determine if it is correct?

SBP I wrote an RDF validator in May that could be used to check 
schemata.  Should be able to play with to see if EARL is correct or 
not.  Libby argued that canonical form of EARL would be easier for people 
to follow.  Easier to follow DTD than schema.

WC Really like idea of validator. What takes to get it into a canonical form?

SBP decide  on structure. subset of RDF.  mainly aesthetic.

CR Could we have 1/2 dozen examples of how it could be used?

SBP that would be useful.

CR Implementations?

WC http://www.w3.org/WAI/ER/#earl List of tools we want from F2F. Including 
Nick Kew agreeing to work on Site Valet. How's that going?

SBP Well.  There's a message in www-archive.


Test Spec Languages

WC CR - your own dialect for test files?

CR Yes. Certainly need human readable results from the tests. It 
fails/passes because of...

SBP For EARL you need a URI for a test, like CR's test files have a URI.

HB Meta info, such as dublin core, good to expose if it is present on 
docs.  Buried info, not part of WCAG assessment. Good meta data.

SBP In the schema for dublin core, have info in a comment rather in schema.

WC DTD for test file info?

CR no, i'll explain.

Action CR: explain meta info used on test files and dtd.


Test suites

WC DOM working on test suite. UAAG also.  I'm attending mtg next week. 
Thoughts and questions?

HB EARL is an unknown in UAWG.  I've encouraged them to use it.  Not much 
has happened.

WC Will create a mini EARL tutorial slide set to use, first test will be 
with UAAG WG.  Part of ongoing action item to work no primer.  If we get 
this to work for UAAG, then should have a tool can be used for DOM suites 
or others. Pass along to QA.


QA F2F

WC

CR Just machine readable


Meeting schedule

Next meeting 29, going to every other week.


SBP What about 1.0?

HB Is it a final thing?

SBP It's the first of final. Nothing to stop us to going to 1.0.

HB What about 0.96?

SBP Have status that it's not really ready yet.

WC Have you heard that?

SBP Not sure, maybe just CMN

WC Sounds like you have several ideas about this already. How long would it 
take?  I'd be interested to see what it looks like.  Particularly, if the 
differences were explained.

Action SBP propose 1.0 schema. Show differences between 0.95 and 1.0 and 
examples.
Action SBP write up ideas about canonical form of EARL.
--
wendy a chisholm
world wide web consortium
web accessibility initiative
seattle, wa usa
/--

Received on Monday, 15 October 2001 10:51:30 UTC