- From: Sean B. Palmer <sean@mysterylights.com>
- Date: Mon, 7 May 2001 04:18:39 +0100
- To: <w3c-wai-er-ig@w3.org>
Just a few issues that I've spotted since publication:- 1) Currently we have the following:- earl:asserts rdfs:domain earl:Assertor; rdfs:range earl:Assertion . This is a bit harsh because it means that you can't reuse earl:asserts in other schemata (which I tried!), whereas you probably should be able to - it's a very vague term. Therfore, we could replace it with the following:- earl:Assertor rdfs:subClassOf [ a daml:Restriction; daml:onProperty earl:asserts; daml:toClass earl:Assertion ] . Meaning that earl:Assertor is a sub class of some (anonymous) class that is a daml:Restriction, where the particular restriction is on earl:asserts to the class earl:Assertion. This is like saying that if something that is an earl:Assertor has the property earl:asserts used upon it, then the range of earl:asserts is earl:Assertion. Compare and contrast that with the current method which simply states that whenever earl:asserts is used, its subject must be an instance of earl:Assertor, and its object(s) must be an instance of earl:Assertion. 2) The next problem is that:- earl:date rdfs:domain earl:TestResult; rdfs:range earl:Date; = dc:date . But this is incorrect because (once again) of the restrictive semantics, and the fact that these ranges and domains don't apply to dc:date. It would probably be O.K. to say that:- earl:date rdfs:subPropertyOf dc:date . Or alternatively:- earl:TestResult rdfs:subClassOf [ a daml:Restriction; daml:onProperty earl:date; daml:toClass earl:Date ] . And then one could state that it is equivalent to dc:date, which would give dc:date the same semantics (why not just use dc:date?). Also, we could use daml:samePropertyAs, but the cyclic properties get in the way (a noted issue, which I expect to be resolved in the recommendation for RDF Schema, and future versions of DAML). Anyhow, I think changing it to subPropertyOf would be the best thing to do in this situation. 3) We already have:- earl:Evaluation rdfs:subClassOf rdf:Statement . But we forgot to also state that:- earl:Assertion rdfs:subClassOf rdf:Statement . Which would make the overall reified architecture of EARL something like:- [ a earl:Evaluation; rdf:subject [ a earl:Assertor ]; rdf:predicate earl:asserts; rdf:object [ a earl:Assertion; rdf:subject [ a earl:TestSubject ]; rdf:predicate [ a earl:TestCase ]; rdf:object [ a earl:TestResult ] ] ] . All minor issues, but just raising them so that they're archived somewhere, and if pointed out again we can refer back. -- Kindest Regards, Sean B. Palmer @prefix : <http://webns.net/roughterms/> . :Sean :hasHomepage <http://purl.org/net/sbp/> .
Received on Sunday, 6 May 2001 23:18:31 UTC