[
Lists Home |
Date Index |
Thread Index
]
> > If you read the WebOnt use cases and requirements document,
> > that is explicitly not the case -- so perhaps what you say is
> > true for "Semantic Web proponents" who haven't been involved
> > with the actual development of semantic web standards.
>
> See above for the pointlessness of trying to argue with "We have use
> cases and requirements that say this should be possible".
Well, if it's an explicit goal of the specs, you can demand a
demonstration that it really works. So I think we just end up looping
back to Elliotte's request for a concrete example. IMO, SNOMED is a
good enough example, but I think people would probably appreciate a much
smaller, narrowly-scoped example. The examples I have seen so far are
either very complicated, or else delve into more "theoretical" aspects
of RDF, like the calendar scheduling app. Here is a thought:
A) Show three different RDF schemas that make assertions about people.
One used by a PIM for contact info, one derived from RSS, one used by a
buddy list in IM
B) Show how you can discover relations between all of this information
by keying off of a URI (no need for OWL).
B1) Do this with simple code; load the three graphs with Drive
RDF parser and grab the info you want
B2) Do this with an RDF query engine like Jena
C) Now show how you can import some owl:sameas assertions to get even
better data
The basic idea would be to show that you can have three different
schemas maintained by three different people, yet the data could be
related meaningfully without need for any mapping layers. I suspect if
people saw this demo, they would realize that RDF is useful for some
scenarios, but is not a panacea either.
|
