- From: Pierre-Antoine Champin <pierre-antoine.champin@liris.cnrs.fr>
- Date: Thu, 19 May 2011 19:47:32 +0200
- To: "Höffernig, Martin" <Martin.Hoeffernig@joanneum.at>
- CC: "public-media-annotation@w3.org" <public-media-annotation@w3.org>
On 05/19/2011 12:45 PM, Höffernig, Martin wrote: > Dear all, > >> Note also that I therefore renamed ma:isRelatedTo to >> ma:hasRelatedResource, in order to be consistent in our naming >> scheme. > > In the latest MA ontology version (rev 34), ma:isRelatedTo is now > defined as the inverse of ma:hasRelatedResource. Indeed. This is now consistent with the naming scheme of other properties. It used to be an exception, justified by the fact that this was the only symetric property in our ontology. As this is no longer the case, I aligned it with the others. > ma:isRelatedTo is > only defined as ObjectProperty without domain/range assertions even > though the definition of ma:hasRelatedResource includes a domain > assertion, namely ma:MediaResource. From my point of view, missing > corresponding domain/range assertions between these two properties > could lead to confusion about the correct usage. Therefore I suggest > to add range ma:MediaResource to ma:isRelatedTo. Again, this was a choice to keep the RDF/XML file small enough and more readable. Furthermore, editors are usually smart enough to automatically infer domain and range of inverse properties (at least, Protégé does). > Furthermore I would > rename ma:isRelatedTo to ma:hasRelatedMediaResource, which would > describe the semantics better. It would, but again it would also be inconsistent with the naming scheme adopted for other properties... pa > > Best, Martin
Received on Thursday, 19 May 2011 17:47:57 UTC