- From: Bijan Parsia <bparsia@cs.man.ac.uk>
- Date: Tue, 19 Aug 2008 17:59:21 +0100
- To: <conrad.bock@nist.gov>
- Cc: "'Alan Ruttenberg'" <alanruttenberg@gmail.com>, "'OWL 1.1'" <public-owl-wg@w3.org>
On 12 Aug 2008, at 20:01, Conrad Bock wrote: > Alan, > >> Hypothetically, if we had only had the object/metamodel, and >> documented the global restrictions on axioms in terms of the >> metamodel, what would we lose (other than a syntax that not many are >> likely to use). > > The metamodel by itself would appear arbitrary, I don't agree, really. > because it is designed > specifically for the functional syntax, Not exactly, they were designed in parallel. > and because it does not suit the > commonly used textual syntax (XML, see earlier discussion on > metamodels > and concrete syntax). Please remember that this is tendentious (we've had a dispute about it). I think it *does* well suit the RDF/XML and the XML and the Manchester syntax. From the point of view of OWL these all encode *axioms* though they do so in very different ways. An OWL Centric api should parse all of them to the same structure, i.e., that defined by the metamodel. If I design a new service (say, like justification finding) I can define it in terms of the metamodel by means of the functional syntax and it will be very clear what triples some interface needs to return. This is very awesome, by the way. It eliminates a lot of confusion, makes extensions and new services MUCH easier to define and implement in an interoperable way. Cheers, Bijan.
Received on Tuesday, 19 August 2008 16:56:57 UTC