HI Adrian & all,
yes, I can see your point and indeed, if a rule system was
around, it
may do the trick in simple cases. Another (non-exclusive)
approach is
what I briefly mentioned at the F2F: is the development of a
kind of an
intermediate layer between RDFS and OWL; a layer that would
allow some
extra information to be added to the knowledge base without
forcing the
usage of a DL reasoner. There are several approaches for
this, and we (I
mean, W3C) may want to look into this more closely next
year...
Ivan
Adrian Walker wrote:
> Hi Ivan, Kerstin & All --
>
> A quick thought about "you are a perfectly decent
Semantic Web citizen
> even if you do not use OWL".
>
> We have found that, even in rather simple cases, it's
quite hard for a
> programmer to check that inferences over RDF are
producing correct results.
>
> An approach that we have found useful is to reason over
RDF using rules
> in executable, open vocabulary English.
>
> With this extra English semantics attached, we can have
the system
> explain, in English, at the business or scientific
level, how it derived
> a result.
>
> There's a simple example at
>
>
> https://www.reengineeringllc.com/demo_
agents/RDFQueryLangComparison1.agent
>
> You can view, edit, and run the example (and others) by
pointing a
> browser to reengineeringllc.com <http://reengineeringl
lc.com> and
> selecting the example RDFQueryLangComparison1 .
>
> HTH, -- Adrian
>
>
> Internet Business Logic (R)
> Executable open vocabulary English
> Online at www.reengineeringllc.com <http://www.reengi
neeringllc.com>
> Shared use is free
>
> Adrian Walker
> Reengineering
> Phone: USA 860 830 2085
>
>
--
Ivan Herman, W3C Semantic Web Activity Lead
URL: http://www.w3.org/Peop
le/Ivan/
PGP Key: http://
www.cwi.nl/%7Eivan/AboutMe/pgpkey.html
FOAF: http://www.ivan-h
erman.net/foaf.rdf
|