Evaluation and Repair Tools Working Group Teleconference

Welcome our guest Ivan Herman

Ivan introduces himself

MikeS introduces himself

JohannesK introduces himself

CarlosI introduces himself

Unresolved comments on ERT drafts

saz: EARL is based on RDF because it may fit into the semantic web world and more tools may be able to process it
... so it's not just an XML format

<http://lists.w3.org/Archives/Public/public-wai-ert/2008Sep/0013>

<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2008Sep/0013

<shadi> http://www.w3.org/TR/HTTP-in-RDF/#terms

saz: RDF guys don't like restrictions; but we want to have required properties to define a "proper" EARL report

ivan: OWL has cardinality restrictions

saz: we did this for EARL schema

jk: and in HTTP-in-RDF as well

ivan: a general RDF processing environment don't know any restrictions

saz: we should make a disticntion between general RDF processing and higher-level EARL/HTTP-in-RDF processing

<ivan> p rdf:range Q

ivan: range and domain are not restrictions but inferences

<ivan> a p z

<ivan> z rdf:type Q

saz: for us the restrictions are sort of conformance requirements

<ivan> h rdf:type http:Connection

<ivan> h http:connectionAuthority x

<ivan> ht http:connectionAuthority y

<ivan> x owl:sameAs y

ivan: if x and y are URIs then x and y are the same thing
... if they are literals, you may get an error message
... you could use SPARQL queries to describe expected results

jk: there were comments about URIs or authoroties not being literals

jk. what we have here is just strings conforming to a URI scheme

<shadi> http://www.w3.org/TR/EARL10-Schema/#schema-rdf

ivan: you should make clear that you talk about URI datatyped literals not URIs for resources
... in the semantic world some people don't like bags, seqs and alts

<shadi> <rdfs:Class rdf:about="http://www.w3.org/ns/earl#SingleAssertor">

<shadi> <rdfs:label xml:lang="en">Single Assertor</rdfs:label>

<shadi> <rdfs:comment xml:lang="en">One person or evaluation tool that claims assertions</rdfs:comment>

<shadi> <owl:oneOf rdf:parseType="Collection">

<shadi> <owl:Thing rdf:type="http://www.w3.org/ns/earl#Software"/>

<shadi> <owl:Thing rdf:type="http://xmlns.com/foaf/0.1/Agent"/>

<shadi> </owl:oneOf>

<shadi> </rdfs:Class>

<shadi> <rdfs:Class rdf:about="http://www.w3.org/ns/earl#CompoundAssertor">

<shadi> <rdfs:label xml:lang="en">Compound Assertor</rdfs:label>

<shadi> <rdfs:comment xml:lang="en">Group of persons or evaluation tools that claim assertions</rdfs:comment>

<shadi> <rdfs:subClassOf rdf:parseType="Collection">

<shadi> <owl:Restriction>

<shadi> <owl:onProperty rdf:resource="http://www.w3.org/ns/earl#mainAssertor"/>

<shadi> <owl:minCardinality rdf:datatype="http://www.w3.org/2001/XMLSchema#nonNegativeInteger">1</owl:minCardinality>

<shadi> </owl:Restriction>

<shadi> </rdfs:subClassOf>

<shadi> </rdfs:Class>

ivan: tools may not have implemented all things defined for alt
... rdf:type on owl:Thing should rather be rdf:about

Draft agenda for upcoming f2f

<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2008Oct/0006

MS: I
... will be available on phone during F2F
... we may discuss HTTP and Content if there's still time left; otherwise EARL schema/guide and Pointers have priority

saz: on friday we may talk about pointers in the first morning session
... in the second morning session we could discuss HTTP and Content
... postpone Guide discussion to friday afternoon
... fri session 1: HTTP/Content, session 2: Guide, session 3: Pointers
... no meetings next two weeks

saz. see you at the F2F

saz: friday session 4 should be open