ERT WG

Updated EARL 1.0 Schema Editor's Draft

<shadi> http://www.w3.org/WAI/ER/EARL10/WD-EARL10-Schema-20070226

SZ: Some minor comments addressed in the Schema doc and now ready for review
... Any additional comments on it?

CV: Namespace changed?

SZ: We're going to request the namespace change with the last call request to publish

<shadi> http://www.w3.org/WAI/ER/EARL/nmg-strawman#

SZ: nmg-strawman in the current namespace

<shadi> http://www.w3.org/2007/earl#

SZ: We said we'd stay with this until last call and then change to http://www.w3.org/2007/earl#
... And we need directors approval for that namespace and we'll request that at LC time

<shadi> http://www.w3.org/WAI/ER/EARL10/schema.rdf

CI: In Section 2.3 testSubject - should we clarify which date to use tested or identified
... e.g. if you tested something "it should be the date tested" or "if you fetched it but didn't test immediately use the fetch dated"

SZ: There's another, which is the creation of date of a web-resource

CI: For testing proposals, I think you need the date of the test

SZ: this is the testSubject class so it's "what was tested" not information about the test

CI: creation date of a resource is unreliable, you can't rely on it so for me the test date it was run is more reliable and useful
... We need to develop something in the date property so people should know what to expect here.

SZ: We should try to pick this up on mailing list - I agree more description on the dc:date would be good, and maybe add date to the test result too.

CI: Section 2.4 earl:Test Criterion - unclear language confusion between testable and test criterion, testable is the Class but we confuse with test criterion

SZ: This is just an editorial comment?

CI: Yes
... We should be clearer on references we use

SZ: We should increase the examples

CI: I'd like to add a note to encourage the use of WCAG public URI's

CV: We discussed that before - close to Test Description language

CI: I want to avoid an incompatibility problem, by encouraging the right URI's up front

CV: I don't think we can solve this problem here

SZ: I think we could do some things - e.g. instead of using example.org uri's use real ones from the HTML test-suite or something
... Please send your comments to the list

Updated HTTP Vocabulary in RDF Editor's Draft

<shadi> http://www.w3.org/WAI/ER/HTTP/WD-HTTP-in-RDF-20070301

SZ: Substantial changes out of the F2F
... Next week want the decision on publish or not too publish

<shadi> http://www.w3.org/WAI/ER/HTTP/WD-HTTP-in-RDF-20070301#body

SZ: CV was it understandable?

CV: Yes it was clear

<JohannesK> RFC for data URI scheme is RFC 2397

CV: although DATA URI's aren't well known

SZ: So we should probably have clearer link to the DATA RFC reference

<scribe> ACTION: JK and SZ to add RFC 2397 to references, and link it from section 2.6 [recorded in http://www.w3.org/2007/03/07-er-minutes.html#action01]

SZ: Should we have cardinality restrictions on some of the properties?
... e.g. on parameters we require two things sometimes, but don't require people to record things like HTTP response codes - or with 10 response codes
... so we could require at most 1 response code, or no exactly 1
... JK had preference for exactly 1, I think you might want to report a broken HTTP response without a response code

JL: I like the idea of at most 1

JK: I favour exactly 1, as I don't think it's ocmplete but I could live with at most 1
... I feel a response without a response code would not be a response but an error

CI: I think there are several use cases where we require minimal information to make sense and responseCode should be one of those

SZ: Okay sounds like something for the mailing list

CV: what's the use case for not having a response code?

SAZ: stand-alone, server sends incomplete response, you want to be able to describe this

<scribe> ACTION: SAZ to send mail for discussing missing response code scenario to mailing list [recorded in http://www.w3.org/2007/03/07-er-minutes.html#action02]

<shadi> http://www.w3.org/WAI/ER/HTTP/WD-HTTP-in-RDF-20070301#example

SAZ: appendix D: realistic example, but only one request/response pair; should we have a second request/response pair (e.g. additional image to show data URI)?

JK: good to have more request/response pairs to show that more than one request is possible in connection

<scribe> ACTION: JK to add another request/repsonse pair (image) to appendix D [recorded in http://www.w3.org/2007/03/07-er-minutes.html#action03]

SAZ: there's an issue with document type declaration in body, you cannot use it directly with parseType="Literal"
... keep it this way

Updated EARL Introduction page

<shadi> http://www.w3.org/WAI/intro/earl.php

SAZ: introduction page about EARL, updated to latest work
... send comments about it

CI: why is EARL Guide in Technical, not Educational section?

SAZ: EARL Guide will be published as technical document
... Educational is more like presentations, ...

CI: add information about status of EARL to encourage feedback

<scribe> ACTION: SAZ to add text about encouraging feedback on EARL intro page [recorded in http://www.w3.org/2007/03/07-er-minutes.html#action04]

Check-in on Last Call publication schedule

<shadi> http://www.w3.org/WAI/intro/earl.php

http://lists.w3.org/Archives/Public/public-wai-ert/2007Feb/0068

<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2007Feb/0068

<CarlosV> I need to take off, see you next week

SAZ: nearly in time
... Jim will be able to create a new pointers draft next week
... later tonight strawpoll will be open for publication of EARL schema draft as last call
... send comments also to list first, then vote later