RE: use role nouns (e.g. subject) not verbs (isSubjectOf)?

Hi Dan,

> I noticed skos:isSubjectOf in
>   http://www.w3.org/2004/02/skos/core/guide/
>   15 February 2005
> 
> I have found rdfs:subClassOf to be awkward in a lot of cases; in 
> retrospect,
> I wish we had called it rdfs:superclass.
> 
> I also found skos:hasTopConcept. I didn't search 
> exhuastively, so there
> may have been others.

Will note this for new proposals, but for current terms I think we're stuck with what we've got i.e. changing URIs now would be too disruptive.

> Does it really seem useful to define the inverses for these 
> properties?
>   skos:subject, skos:isSubjectOf, skos:primarySubject and 
> skos:isPrimarySubjectOf.
> I think owl:inverseOf is fine for post-hoc declaration of 
> inverses, but 
> let's
> not make up aliases, even indirectly like this, if we can help it. It 
> just makes
> dealing with this sort of data more expensive.

We had several requests for these inverses, from implementors who find them useful, which is why they went in.  

> p.p.s. The draft says "This document is the First Public 
> Working Draft" 
> but it's
> not. Always keep the SOTD truthful, please.

Changed to 'This document is a W3C Editor's Working Draft' ... that OK? 

Cheers,

Al.

Received on Thursday, 10 March 2005 18:38:30 UTC