Evaluation and Repair Tools Working Group Teleconference

Publication timeline

http://www.w3.org/WAI/ER/timeline

MS: want to do a full Last Call publication
... need to confirm timeline and plan

CI: think we can do it
... maybe the Guide is a little behind
... but otherwise seems reasonable

SAZ: same concern about the Guide
... need to confirm with CarlosV

Pointers in RDF

http://lists.w3.org/Archives/Public/public-wai-ert/2009May/0007.html

MS: do we need "XML"? isn't it clear from the conext that ptr:Namespace is for XML content?

CI: property could be reused in other contexts
... then it becomes ambiguous

MS: even if we flip the order, we may be running against conventions

CI: prefer "NamespaceInXML" because it is clearer
... the second option "NamespaceXML" runs into the issue MS raised

SAZ: we have same issue in Content-in-RDF
...http: //www.w3.org/TR/Content-in-RDF/#xmlContentClass
... need more generic solution

MS: don't object to "NamespaceInXML"

JK: for Content we could use "WellFormedContent"

CI: not sure if this is accurate

JK: it has to be well formed

SAZ: are there other definitions of namespaces?

CI: yes, think so

http://en.wikipedia.org/wiki/Namespace

SAZ: only used in XPathPointer

CI: and XPointerPointer because it is a subclass

JK: wonders if this is not a namespace mapping

CI: brings the same problem, because we are only focusing on XML

MS: but in the context we are in it is restricted by XPointer

JK: then we define it to be an XML namespace mapping

MS: can also add an editor's note

<MikeS> +1 for NamespaceMapping

<JohannesK> +1

Resolution: rename the class to "NamespaceMapping" and add an editor's note in the document for feedback

CI: next issue is about usage of pointers
... propose to add a paragraph explaining that different pointers are more appropriate to different types of content
... don't recommend that we define restrictions

SAZ: agree with this proposal

MS: me too

Resolution: accept proposal to add a new paragraph explaining the usage of pointers

CI: removed examples using abstract classes, as per previous discussions
... next issue is about reusing examples

MS: maybe just say something like "referring to example X.Y.Z"

http://www.w3.org/TR/EARL10-Schema/#example-6

SAZ: Schema doc uses that approach

MS: think we had already resolved the XPointerPointer issue

CI: don't suggest we should rename
... we'll respond politely

Resolution: keep current naming of XPointerPointer because of naming convention

MS: any other issues for this document?

CI: will provide an editor's draft before the next meeting
... once draft is up, will send responses to commentors

Aggregation

http://lists.w3.org/Archives/Public/public-wai-ert/2009Apr/0032

MS: distinguishing three types of "aggregation"

CI: "combination" is already provided by RDF
... the other two relate to definition of rules
... which we decided is out of scope for EARL

JK: agree

MS: don't get "combination" entirely for free
... some tools may provide parts of test results

SAZ: there is combination on a level of test results, then on the parts of test results

MS: if we only mean combination, do we need a requirement for aggregation?

CI: can keep the requirement, but don't think we need to address it any further

MS: so tweak wording towards combination?

SAZ: it supports aggregation given other parts

CI: think should only talk about EARL, so tweak towards combination
... and explain aggregation in the Guide

MS: will send one or two wording suggestions

Resolution: eliminate aggregation from requirements but mention it as part of combination, and to describe it further in the Guide

EARL 1.0 Requirements

MS: please have a look at that and provide feedback

Next and future meetings

JK: will be out next two weeks, will prepare something on HTTP-in-RDF and Content-in-RDF