- From: Damian Steer <pldms@mac.com>
- Date: Tue, 18 Nov 2008 00:27:40 +0000
- To: Chris Bizer <chris@bizer.de>
- Cc: public-lod@w3.org, Semantic Web <semantic-web@w3.org>
<...trimming reply list...> On 17 Nov 2008, at 23:31, Chris Bizer wrote: > > But what does this mean for WEB ontology languages? > > Looking at the current discussion, I feel reassured that if you want > to do > WEB stuff, you should not move beyond RDFS, even aim lower and only > use a > subset of RDFS (basically only rdf:type, rdfs:subClassOf and > rdfs:subPropertyOf) plus owl:SameAs. Anything beyond this seems to > impose > too tight restrictions, seems to be too complicated even for people > with > fair Semantic Web knowledge, and seems to break immediately when > people > start to set links between different schemata/ontologies. Hmm. My sympathies are in this direction, having had unhappy experiences in the past (akt:Person was very restrictive, IIRC). However I don't think your conclusion is quite right. Firstly sameAs has the nasty habit of infecting your data with somebody else's cruft. In a web context it seems more useful in conclusions than in premises. seeAlso inverse functional, and some of the things in OWL 2. Secondly it's worth revisiting OWL's allValuesFrom, which gives you the benefit of rdfs:range (to describe data structure) but in a more limited way. For example it could allow the range of dc:contributor to be foaf:Person when used with foaf:Documents, but not infect other uses of the property. Thirdly hasValue is pleasingly structural. Being able to determine types not because they have been stated, but because of the features of the data, has its virtues on the web. Damian
Received on Tuesday, 18 November 2008 00:28:29 UTC