See also: IRC log
http://www.w3.org/WAI/ER/EARL10/WD-EARL10-Schema-20081106
MS: objective for today is to go through some
of the open questions
... primarily the "specific questions", then the "editorial questions"
... may need to leave the others for future calls
http://lists.w3.org/Archives/Public/public-wai-ert/2008Nov/0025.html
[Definition of Outcome]
CV: philosophical question, how we want design this
http://lists.w3.org/Archives/Public/public-wai-ert/2008Nov/0023
http://lists.w3.org/Archives/Public/public-wai-ert/2008Nov/att-0012/earl.rdf
<earl:Outcome rdf:ID="pass">
<rdfs:label xml:lang="en">Pass</rdfs:label>
<rdfs:comment xml:lang="en">Test passed</rdfs:comment>
<rdf:type rdf:resource="http://www.w3.org/2002/07/owl#Thing"/>
</earl:Outcome>
SAZ: seems to be agreement that owl:Thing is
not necessary
... want to have custom values
... option 1: individual outcome values should be subclasses of
earl:Outcome
... option 2: is to do the approach explained in http://lists.w3.org/Archives/Public/public-wai-ert/2008Nov/0023.html
JK: preference to make the outcome values
resources rather than classes
... and to make new values new resources
<rdf:Description rdf:about="http://www.w3.org/ns/earl#cannotTell">
<rdf:type rdf:resource="http://www.example.org/my#CannotTellOutcome"/>
</rdf:Description>
<rdf:Description rdf:about="http://www.example.org/my#CannotTellOutcome">
<rdf:type rdf:resource="http://www.w3.org/ns/earl#cannotTell"/>
</rdf:Description>
CV: do we need to do this?
... can just subclass earl:Outcome
JK: what would my:probablyPass mean?
CV: would be earl:Outcome
SAZ: want to create a "probablyPass" value,
which is actually a "cannotTell"
... think that the atomic values that we provide in EARL (pass, fail, ...)
need to be classes
CV: seems cumbersome
MS: not convinced that need to be classes
JK: what was the use case? some tool creates a
"my:probablyPass" but a second tool doesn't know the "my:" vocabulary
... should still be able to infer that it is an earl:cannotTell
SAZ: workaround by JK changes meaning of
initial vocabulary
... think this is bad RDF ethics
<scribe> ACTION: shadi to ask SWIG if there other ways to do this rather than to create individual classes for each value [recorded in http://www.w3.org/2008/11/19-er-minutes.html#action01]
<trackbot> Created ACTION-64 - Ask SWIG if there other ways to do this rather than to create individual classes for each value [on Shadi Abou-Zahra - due 2008-11-26].
<JohannesK> <rdfs:Class rdf:ID="OutcomePass">
<JohannesK> <rdfs:subClassOf rdf:resource="&earl;Outcome"/>
<JohannesK> </rdfs:Class>
[Adoption of DOAP]
MK: should we adopt this, and if so how?
CV: description of a project, vocabulary is
quite simple
... offers versioning, home page, etc
... all the part of open source projects
MS: do we want to adopt this?
SAZ: do not think we need to add any terms for
DOAP
... could adopt it like we adopt FOAF
MS: what benefit would we get from adopting DOAP? we have dc:version and foaf:homepage in a software already
JK: is there a problem with someone using DOAP terms?
SAZ: would be just repeating dc:version and doap:version for example, but no compatibility issues
<scribe> ACTION: cvelasco to suggest specific DOAP terms that would be useful for us to adopt (because they go beyond what we currently have) [recorded in http://www.w3.org/2008/11/19-er-minutes.html#action02]
<trackbot> Created ACTION-65 - Suggest specific DOAP terms that would be useful for us to adopt (because they go beyond what we currently have) [on Carlos A. Velasco - due 2008-11-26].
[Conformance section]
http://www.w3.org/WAI/ER/EARL10/WD-EARL10-Schema-20081106#conformance
MS: we have conformance notes inline, do we also want a conformance section?
SAZ: think there is a level describing the individual classes and entities, and the other is about processing
MS: is there presendence?
SAZ: i think SPARQL and maybe POWDER
<scribe> ACTION: msquilla to check about SPARQL and POWDER approach on processing rules [recorded in http://www.w3.org/2008/11/19-er-minutes.html#action03]
<trackbot> Created ACTION-66 - Check about SPARQL and POWDER approach on processing rules [on Michael Squillace - due 2008-11-26].
SAZ: also CC/PP may have a similar approach
JK: there is a difference btween EARL generating tools and EARL processing tools
<scribe> ACTION: msquilla and shadi to take another pass at the conformance section [recorded in http://www.w3.org/2008/11/19-er-minutes.html#action04]
<trackbot> Created ACTION-67 - And shadi to take another pass at the conformance section [on Michael Squillace - due 2008-11-26].