RDF-in-XHTML Task Force

<Steven> There is a national ADSL breakdown here (3 days long alread), so the only internet connection I have is via the same mobile phone I will be phoning in on

<Steven> so expect poor sound

<msporny> no chance you're going to be able to see Test Cases to review them, then, Steven?

<Steven> I'll try

<Steven> As soon as I hit a link, the sound quality deteriorates

Shane: I raised issue-120 and I believe it's [easy to] close

<msporny> ActionSummary

ACTION: [DONE] Manu to e-mail final Christian Hoertnagl response. [recorded in http://www.w3.org/2008/05/08-rdfa-minutes.html#action13]

ACTION: [DONE] Manu to review current on hold test cases and e-mail list on what we should do with them. [recorded in http://www.w3.org/2008/05/08-rdfa-minutes.html#action11]

ACTION: Michael to determine which useless-triples test cases to remove and which to add. [recorded in http://www.w3.org/2008/05/08-rdfa-minutes.html#action12] [CONTINUES]

ACTION: Manu to reach out to Slashdot and attempt to get RDFa integrated into Slashdot. [recorded in http://www.w3.org/2008/05/08-rdfa-minutes.html#action10] [CONTINUES]

Manu: I've sent email, awaiting a response

ACTION: Ben followup with Fabien on getting his RDFa GRDDL transform transferred to W3C [recorded in http://www.w3.org/2007/11/15-rdfa-minutes.html#action01] [CONTINUES]

Manu: it's in progress

ACTION: [DONE] Ben to respond to issue 87 [recorded in http://www.w3.org/2008/02/28-rdfa-minutes.html#action09]

Shane: issue 87 is closed

ACTION: Mark to move _:a bnode notation to normative section [recorded in http://www.w3.org/2008/04/03-rdfa-minutes.html#action05] [CONTINUES]

ACTION: Michael to create 'RDFa for uF users' on RDFa Wiki [recorded in http://www.w3.org/2008/03/13-rdfa-minutes.html#action12] [CONTINUES]

On Hold Test Case Review

<msporny> "Resolving ON HOLD Test Cases"

-- test 4; xml:base

Manu: we don't process xml:base, so I suggest we reject test 4

<msporny> rdfa-test-harness

PROPOSE to reject test #4

<ShaneM> +1

RESOLUTION: to reject test #4 xml:base

-- test 17; Related blanknodes

Manu: I suggest we rewrite test 17 to cover explicit bnode relationships

<msporny> "Test Case #17 (v2): specifying named bnode relationships"

Manu: note that the SPARQL intentionally uses different bnode names just to insure that the implementation didn't hardcode the names

PROPOSE to replace test 17 with 2008May/0096.html

Shane: this is the case that Mark's action refers to
... to take some non-normative text about bnodes and CURIEs and make it normative
... I just talked with Mark and am preparing to fix the spec now

<ShaneM> I think the new test case is fine.

<Ralph> +1 to proposal

<Steven> +1

RESOLVED to replace test 17 withhttp://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2008May/0096.html

RESOLVED test 17 approved perhttp://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2008May/0096.html

-- test 28; @xml:lang and @datatype

PROPOSE to reject test 28 @xml:lang and @datatype

<ShaneM> +1

RESOLVED to reject test 28 @xml:lang and @datatype

Manu: tests 107 and 108 replace 28

-- test 39 @rev - @src/@href test

Manu: this passes Ivan's parser and my parser
... this test checks that implmentations don't use @href to override @src

<ShaneM> looks good.

<Steven> is there a similar test for @resource?

<ShaneM> yes its next

Ralph: looks good to me

<Steven> good

RESOLUTION: test 39 approved

-- test 40; @rev - @src/@resource test

<ShaneM> that one seems fine too.

Manu: 40 similar to 39, checking that @resource doesn't override @src
... I'm not sure this one is correct
... if we're checking that @resource doesn't override @src then we shouldn't use @about
... if we remove @about then the SPARQL uses the @src value

<msporny> <img src="http://sw-app.org/img/mic_2007_01.jpg"

<msporny> rev="foaf:depicts"

<msporny> resource="http://sw-app.org/img/mic_2006_03.jpg"

<msporny> alt="A photo depicting Michael" />

<msporny> ASK WHERE {

<msporny> <http://sw-app.org/img/mic_2006_03.jpg> <http://xmlns.com/foaf/0.1/depicts> <http://sw-app.org/img/mic_2007_01.jpg> .

Ralph: I believe this is a more useful test but the semantics are wrong (for foaf:depicts) now

Manu: foaf:alternate?

<Steven> Can't you use depicts for a photo within a photo?

Manu: ??:alternate ?

<Steven> html:alternate

Shane: the definition of html:alternate isn't quite right for this

Manu: html:next works

<msporny> }

<msporny> <img src="http://sw-app.org/img/mic_2007_01.jpg"

<msporny> rev="previous"

<msporny> resource="http://sw-app.org/img/mic_2006_03.jpg"

<msporny> alt="A photo depicting Michael" />

<ShaneM> I think we should change the definition of "alternate" in our vocabulary for what its worth. This is dumb - "alternate designates alternate versions for the document."

<msporny> ASK WHERE {

<msporny> <http://sw-app.org/img/mic_2006_03.jpg> <http://www.w3.org/1999/xhtml/vocab#previous> <http://sw-app.org/img/mic_2007_01.jpg> .

<msporny> }

<Steven> http://farm1.static.flickr.com/15/22341209_48084a909b.jpg?v=0

Ralph: foaf:thumbnail ?

Manu: these two particular images are different; one isn't a thumbnail of the other

Shane: html:alternate is fine, as is html:next

<msporny> <img src="http://sw-app.org/img/mic_2007_01.jpg"

<msporny> rev="alternate"

<msporny> resource="http://sw-app.org/img/mic_2006_03.jpg"

<msporny> alt="A photo depicting Michael" />

<msporny> ASK WHERE {

<msporny> <http://sw-app.org/img/mic_2006_03.jpg> <http://www.w3.org/1999/xhtml/vocab#alternate> <http://sw-app.org/img/mic_2007_01.jpg> .

<msporny> }

<Ralph> +1 to html:alternate

<Steven> 22341209_48084a909b.jpg foaf:depicts 06-steven-goteborg/guido.jpg

PROPOSE: change test 40 to drop @about and use html:alternate

<Steven> +1

RESOLUTION: test 40 approved after changes to drop @about and use html:alternate

-- test 42; omitted @about

Manu: the SPARQL in the test harness is still wrong
... no triples should be generated
... this should be turned into a negative test
... testing that @src is not a target
... this was a change to our processing rules

Ralph: I agree; no triples should be generated

PROPOSED: test 42 approved after changing to a negative test and removing triples from the SPARQL

RESOLUTION: test 42 approved after changing to a negative test and removing triples from the SPARQL

-- test 43; @src/@href test with omitted @about

<msporny> Test #43: REJECT Duplicates TC#34, TC#35, and TC#90

<msporny> Test #44: REJECT Duplicates TC#32, TC#37, and TC#40

<msporny> Test #45: REJECT Duplicates TC#32, TC#36, TC#37

<ShaneM> +1

Manu: 43, 44, and 45 are obsoleted by 34, 35, 90; 32, 37, 40; 32, 36, 37

Ralph: I trust you

<Steven> +1

RESOLUTION: tests 43, 44, and 45 dropped

<ShaneM> updated vocab document is at xhtml-vocab-20080515.html - Ralph, can you please install it in /1999/xhtml/vocab as appropriate?

Unreviewed Tests

-- test 95; No triples with two nested @rel

Manu: I think tests 95, 96, and 97 should be rejected
... we changed the processing rules to retain "useless" triples
... so these tests are obsolete

PROPOSE: reject tests 95, 96, 97

<ShaneM> +1

<Steven> +1

RESOLUTION: reject tests 95, 96, 97

-- test 104; rdf:value

Manu: the objective of 104 is to have an example of a value with a unit

Ralph: this doesn't really test anything new about RDFa; it's more demonstrating a use case

Manu: whomever raised this was concerned specifically about rdf:value

Ralph: I don't mind duplicate tests

<Steven> where is 104?

<Steven> Oh found it

<Ralph> +1

<Steven> +1

<ShaneM> Harmless - +1

RESOLUTION: test 104 approved

-- test 105; inner @rel neither CURIE nor LinkType

Manu: the point in test 105 is that the inner @rel doesn't chain because "myfoobarrel" isn't a valid CURIE

Shane: mmmmm
... there was a private discussion about issue 120 that resolved this
... Mark has changed the editor's draft
... I'm not sure if the current interpretation matches 105

Ralph: propose to passover 105

Shane: passover 106 too

-- test 107; plain literal with datatype=""

Manu: 107 is to replace test a test we rejected
... test 28 (rejected)
... test 107 tests that whitespace is preserved
... both Ivan's parser and mine pass 107

Ralph: looks good to me
... do we believe all the N3 and RDF/XML variants of these tests?

<ShaneM> +1

Manu: all the N3 comes from Ivan's parser, so if Ivan's parser passes the test then the N3 is OK. But for now, continue to ignore the N3 and RDF/XML

<Steven> +1

RESOLUTION: test 107 approved

-- test 108; plain literal with datatype="" and xml:lang preservation

Manu: I believe there's an error in the SPARQL for test 108
... I believe @el is missing from the end of the SPARQL

Ralph: yep, SPARQL should preserve the lang too

PROPOSE: accept test 108 correcting the SPARQL to specify the language

<ShaneM> +1 with changing sparql to include @el

RESOLUTION: accept test 108 correcting the SPARQL to specify the language

-- test 109; xml:base should be ignored

-> "Test Case #109: xml:base should be ignored in XHTML+RDFa 1.0" [Manu 2008-05-13]

Shane: Michael's question was "should invalid markup generate any triples?"

<msporny> Michael's reply

Shane: the answer should be 'no'

Manu: we don't have a way to test invalid markup

Shane: parsers are likely to generate triples when they don't validate the input

<msporny> ?

Shane: I think it's fine to include the test
... looking at the specifics;
... @xml:base on the root and on the div

Manu: we're trying to test that @xml:base is ignored and the @about value is the subject

Shane: we're insuring parsers ignore @xml:base

Shane: we have another test for @base

<msporny> [ TEST ] Test #72 (approved): Relative URI in @about (with XHTML base in head)

<msporny> [ TEST ] Test #73 (approved): Relative URI in @resource (with XHTML base in head)

<msporny> [ TEST ] Test #74 (approved): Relative URI in @href (with XHTML base in head)

Ralph: I fall on the side of the concern that Michael is expressing
... we don't specify behavior for invalid input

Manu: this question about @xml:base has been raised on the list several times
... realistically, a lot of parsers will deal with invalid XHTML

Ralph: how about we add a big XML comment saying this isn't valid XHTML

Shane: I'm not offended by the current form, a comment is fine

PROPOSE: to accept test 109 with an added XML comment noting this is not valid XHTML

<Ralph> +1

<Steven> +1

<ShaneM> +1

RESOLUTION: to accept test 109 with an added XML comment noting this is not valid XHTML

<ShaneM> PROPOSE: Add a comment to rdfa-syntax that conforming parsers are NOT required to generate triples from invalid input.

<Ralph> absolutely +1

<Steven> +1

<msporny> +1

Shane: this tacitly encourages document authors to make valid documents

RESOLUTION: Add a comment to rdfa-syntax that conforming parsers are NOT required to generate triples from invalid input.

Shane: I'm about to make an updated editor's draft

<Steven> Regrets for next week

[adjourned]

<Steven> I am chairing a conference

Ralph: regrets for next week too

<Ralph> http://www.w3.org/1999/xhtml/vocab is now updated

<Ralph> Shane, any objection to adding $id$ or $revision$ to the vocab namespace document?

<Ralph> (a visible version identifier)