See also: IRC log, previous 2008-02-14
Ralph: Shane has been working on the /vocab
namespace document
... he's given me a draft
ACTION: [DONE] Shane send response to Diego and Ed review comments when new editors' draft is up [recorded in http://www.w3.org/2008/02/14-rdfa-minutes.html#action06]
ACTION: Ben to add status of various implementations on rdfa.info [recorded in http://www.w3.org/2007/10/04-rdfa-minutes.html#action06] [CONTINUES]
ACTION: Ben to email mailing list to think about last substantive issue on tracker: http://www.w3.org/2006/07/SWD/track/issues/6 [recorded in http://www.w3.org/2008/02/07-rdfa-minutes.html#action07] [CONTINUES]
ACTION: Manu write 2 new tests for img[@src] as subject [recorded in http://www.w3.org/2008/01/31-rdfa-minutes.html#action08] [CONTINUES]
ACTION: Michael to create "Microformats done right -- unambiguous taxonomies via RDF" on the wiki [recorded in http://www.w3.org/2007/08/23-rdfa-minutes.html#action06]
-> tests
<msporny> just FYI, #78-#83, #87 PASS using Ivan and my latest implementations.
<msporny> so you don't have to worry about syntax issues in those test cases.
<markbirbeck> #78 is fine.
<markbirbeck> Minor points that they are easier to read if everything is in the same order in all blocks (HTML, N3, RDF and SPARQL) but that's my only comment.
Ralph: Ben was happy with test 78 on 7-Feb
RESOLUTION: Test 78 accepted
<msporny> Ben's replies to test cases 85-87
<markbirbeck> #79 is good, too.
<markbirbeck> Again, minor point...
<markbirbeck> ...don't see what the super-abbreviation gains in the N3 block. :)
<markbirbeck> Makes it slightly harder to read when comparing with the HTML.
<markbirbeck> But minor point again.
Manu: we don't believe the N3 yet; focus on the SPARQL
<msporny> N3 is generated by Ivan's parser, plus it's informative, no need to pay attention to it yet.
Manu: N3 in test 79 looks correct, though it's unreadable as is
RESOLUTION: Test 79 accepted
<markbirbeck> #80 looks lovely. :)
(Ben was happy with 80 on 7-Feb too )
RESOLUTION: Test 80 accepted
<markbirbeck> #81 is very smart...and seems right to me.
<markbirbeck> #82 also looks ok to me.
<markbirbeck> Minor point is that in the SPARQL in other tests, the square bracket syntax has generally been used, whilst here we're using "_:a" and "_:b".
<markbirbeck> Minor point again, though.
[some discussion of what "substantive change" means]
<markbirbeck> (Compare to #83, for example.)
<msporny> mark, be sure to skip #84, #85, and #88
<markbirbeck> #83 is good.
<markbirbeck> ok
Ralph: Ben was happy with 81 and wanted to check his implementation
RESOLUTION: Test 81 accepted
Ralph: Ben was fine with 82 after typo fixes
Manu: Ivan has written some SPARQL using the '[...]' syntax where we've been using bnode syntax
Manu: we might want to update test 83 to use bnode syntax
RESOLUTION: Test 82 accepted
Manu: let's rewrite 83 to use named bnodes
... skip 84 and 85 for now due to syntax errors
<msporny> correct, mark - #86 is our first FALSE test for the SPARQL.
<markbirbeck> On #86, we might want to make the query check for { <http://www.example.org#somebody> ?p <mailto:ivan@w3.org> }
<markbirbeck> The rule is that @rel="foobar" should generate nothing, not just that it shouldn't generate anything in the XHTML vocab namespace.
<markbirbeck> (And obviously in #86 we're testing for 'no matches', as opposed to a match.)
<markbirbeck> I'm not quite following #87...is there a similar test without the colon prefix? I.e., is this to test that that we support two means of expressing the same predicates?
<ShaneM> there is a similar test without a colon
Ralph: did we want our tests to allow implementations to produce "additional" triples
<markbirbeck> Either way, minor point would be that in general the items are in alphabetical order, which makes it handy for double-checking...so perhaps the few stray ones could be put into order.
Manu: yes, but not in the default graph
... we haven't said that we test [only] for triples in the default graph
Ralph: in that case I like Mark's suggestion to
use ?p in the SPARQL
... so there's NO relationshp between #somebody and ivan
Manu: that makes sense to me
<markbirbeck> (Still on #87...) Secondly, we might consider using <base> in the test, so that we're not tied to where these tests are being run from, and to abbreviate the SPARQL.
<msporny> SPARQL for test #86:
<msporny> ASK WHERE {
<msporny> <http://www.example.org/#somebody> ?p <mailto:ivan@w3.org> .
<msporny> }
<msporny> That test should return FALSE.
<markbirbeck> And third (on #87) we might consider adding an item that is not in the list of vocab values, since this is *allowed* when using the ":foo" syntax.
RESOLUTION: test 86 accepted with change to ?p
<msporny> Mark, shouldn't that be a separate test?
<markbirbeck> That's all from me...I'll try to keep an eye out if you need me for anything else.
<msporny> testing ":foo" syntax?
<msporny> thanks :)
<markbirbeck> Good point.
<markbirbeck> Yes, could be.
<msporny> I think we should make that a separate test...
<markbirbeck> Basically, whilst @rel="foo" should *fail*, @rel=":foo" should *succeed*.
Shane: remember, :foo _always_ uses /vocab#; there's no way to change the CURIE prefix
Manu: should we test @rel="foo" to insure that
the code isn't using the same branch as @rel=":foo" ?
... i.e. to test a misunderstanding of the document
ACTION: Manu create unit test to make sure that rel=":foo" generates a triple. [recorded in http://www.w3.org/2008/02/21-rdfa-minutes.html#action07]
Ralph: do we have a test for @rel="foo" ?
Manu: yes, test 86
... test 87 is missing stylesheet
... as do tests 76, 77
ACTION: Manu let Michael know that Test 76, 77, and 87 are missing stylesheet. [recorded in http://www.w3.org/2008/02/21-rdfa-minutes.html#action08]
Ralph: for those cases in test 87 where the
reserved word plausibly has a reasonable value, we should use it
... thought that's a nit
... so how about just a comment noting that these test values are not
semantically reasonable
RESOLUTION: test 87 accepted, after realphabetizing
-> Michael re: implementor's call
Ralph: the "call" that I think Michael is
asking about is the "Candidate Recommendation" transition
... CR is a "call for implementation"
... we should document how we'd like implementers to report their
implementation experience
... this can go in the SOTD of the CR draft
Ralph:I believe issue 6 has a specified answer but I want to gather definitive pointers before declaring it closed.
-> issue 7
ACTION: Manu write a response to Christian Hoertnagl for issue 7 [recorded in http://www.w3.org/2008/02/21-rdfa-minutes.html#action09]
-> issue 8
Ralph: we've chosen to defer RDF Container support
RESOLUTION: issue 8 is POSTPONED
-> issue 63
Ralph: we resolved this last telecon
[adjourned]