W3C home > Mailing lists > Public > www-rdf-interest@w3.org > October 2002

RE: (SeWeb) Re: KAON - KArlsruhe ONtology and Semantic Web Infrastructure

From: Raphael Volz <rvo@aifb.uni-karlsruhe.de>
Date: Thu, 10 Oct 2002 14:22:27 +0200 (CEST)
To: Danny Ayers <danny666@virgilio.it>
cc: "Arthur T. Murray (by way of \"Ralph R. Swick\" <swick@w3.org>)" <uj797@victoria.tc.ca>, <www-rdf-interest@w3.org>
Message-ID: <Pine.LNX.4.44.0210101421160.26547-100000@aifbshark.aifb.uni-karlsruhe.de>

On Thu, 10 Oct 2002, Danny Ayers wrote:

> 
> 
> > > Interesting to reflect on the fact that ontologies were (partially)
> > > introduced to enable interoperability between application systems.
> > > Now we are seeing interoperability-problems between ontology systems.
> > > Do we need to recurse once more, to solve this problem?
> >
> >No, we need to create truly artificial Minds such as
> >http://mind.sourceforge.net/mind4th.html -- AI in Forth.
> 
> I'm a big fan of 'artificial mind' approaches, but it doesn't actually help
> very much in this case: the problem just moves from AI <-> ontology or AI
> <-> existing data interoperability. (not to mention web hacker <-> Forth
> interoperability ;-)
> 
> Personally I don't think the interoperability problems anything like as bad
> as suggested - on the RDF level at least the language is common, and if the
> domain of discourse is the same between different applications (even if it's
> the ontologies themselves) then interoperability shouldn't be difficult.
> 
> If there is a common, higher-level language such as OWL available that isn't
> blindingly complex, then tool builders will tend to try and support it. This
Unfortunally it is blindgly complex, hence very few tool builders will 
fully support it. Rather everyone will implement the features they wish to 
support and leave many others out. This actually decreases 
interoperability instead of increasing it.

> is what has happened with XML, and with ontology tools has been happening
> with RDF, and to some extent with DAML+OIL. There will undoubtedly be other
> (perhaps 'higher') levels of incompatibility between internal models, but
> the overall interoperability & utility will be increasing.
> 
> Cheers,
> Danny.
> 

-- 

Raphael Volz
Institut AIFB
Uni Karlsruhe(TH)

Tel:  +49 (721) 608 7363
Mail: volz@aifb.uni-karlsruhe.de
Received on Thursday, 10 October 2002 08:22:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:56 GMT