RE: [RDFa TC] Review and Approval of XHTML Test Case partition

Ben,

Thanks for your comments. At least one who replied ...

Before I go into detail, let me shortly report on 
the status of the XHTML+RDFa validation of our TC:

I ran all the XHTML-TC source files through the new validator
(http://validator-test.w3.org/) and beside the following,
all are valid XHTML+RDFa:
  
 + 0004, due to '... there is no attribute "xml:base".'
 + 0012 and 0013 due to ' ... there is no attribute "about".' (at
<head>)

Shane might be able to chime in?

Now, back to your comments - as always very valid and helpful.
Please find my answers inline:

>A few points on the test cases: I think they need some reviewing before
>they're approved.

I think this is the plan, right. Review them, then approve them.

>I know you've been swamped, so please call out if you
>need help, because I fear we're putting too much work on your 
>shoulders!

Though I do appreciate any helping hands, especially from 
RDFa implementers regarding feedback on the TC, I do not 
share your concerns, here ;)

>OVERALL
>
>- about CC: can you change the namespace in all examples to
>http://creativecommons.org/ns#? We made the change a while ago, and the
>tests should reflect them ASAP. Thanks!

DONE.

>- a number of bad subjects throughout: it looks like you're using the
>old URIs of the test cases as subject, e.g. testcases/000106.html
>instead of xhtml1-testcases/0023. There are a number of these, 
>and a few.ttl also instead of .html.

DONE. Note, it is '.xhtml' throughout; remember, we are in the XHTML TC
partition ...

>- a number of examples use XMLLiteral instead of plain literal when the
>content has no markup in it. As per our agreement on the resolution, we
>should change these. It's mostly when you have "Mark Birbeck", but it
>happens a dozen times.

DONE.

>SPECIFICALLY:
>
>0002: We'll need to discuss whether we keep META in the body.
>0003: same comment as 0002.

The same applies to 0004. To be discussed today; I'd propose to remove
them.

>
>0005: CURIE in the @href in XHTML 1.1... I'm very worried about this, I
>guess we should bring this up for discussion.

To be discussed today; I'd propose to remove them.

>0015: a LINK or META on the HEAD of a document resolves to the current
>URL as subject, not a bnode.

DONE.

>0019: is a mailto: URI really a good foaf person?

Probably not. To be discussed today.

>0028: datatype should be "" instead of "plainliteral"

DONE. Still I'm not sure what should take precedence (xml:lang or
@datatype)
when @datatype is present, but empty (cf. also
http://www.w3.org/2006/07/SWD/wiki/RDFa/LiteralObject)

>0029: should the xsd:string really trigger the stripping of markup? I
>guess so, just wanted to make sure that's what we want.

This was my guess as well. Any objections?

Cheers,
	Michael


>Hausenblas, Michael wrote:
>> 
>> All,
>> 
>> As the responsible for the RDFa Test Cases, I'd like to invite you
>> to have a look at the current state of the RDFa Test Suite [1].
>> 
>> Ben pointed out earlier today [2] that we plan to review and approve
>> the XHTML Test Case partition at our tomorrow's RDFa telecon. 
>> Though the remaining time is very limited, I'd very much appreciate
>> it if you spent some minutes testing your RDFa extractor with the
>> currently proposed XHTML TC partition, available at [3], along with
>> the test manifest document [4].
>> 
>> Any feedback is welcome!
>> 
>> Cheers,
>> 	Michael
>>  
>> 
>> [1] http://www.w3.org/2006/07/SWD/RDFa/testsuite/
>> [2]
http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2007Jun/0144.
html
>> [3] http://www.w3.org/2006/07/SWD/RDFa/testsuite/xhtml1-testcases/
>> [4]
http://www.w3.org/2006/07/SWD/RDFa/testsuite/xhtml1-testcases/rdfa-xhtml
1-test-manifest.rdf

Received on Thursday, 28 June 2007 07:34:29 UTC