RE: Ontology editor + why RDF?

> I
> don't see what this toy does for us." If we sincerely believe we have a
> better mousetrap, then shouldn't we be catching mice instead of worrying
> about skeptics?

[VK] But wouldn't it be a nice idea to show that the better mousetrap catches
more mice, or catches them more conveniently at a cheaper $$/mice caught?

The spreadsheet example is important. However, if the idea is good, the value is
demonstrated soon enough. The SW concept has been around quite a while and there
is a need to demonstrate the value proposition. We want to make sure that the SW
doesn't make tall promises (like AI did once upon a time) without showing that
it works!

> I suggest that, especially in the community represented on this mailing
> list, our time is best devoted to *DOING* this work. The most effective
> "proof and demonstration" is a working prototype that solves a specific
> need -- and a workable solution to one or more of the use cases already
> presented here as an excellent starting point. 

[VK] I agree completely... But one also needs to show that the workable solution
was achieved at a smaller cost and showed better performance along some
dimension.

>Rather than debate
> buzzwords
> and acronyms, won't it be more helpful for us to solve the problem, using
> whatever technologies get us to a workable result? Those of us who are
> convinced that RDF is the answer will, in this paradigm, start with RDF.
> Those of us who are convinced that RDB semantics are the answer will, in
> this paradigm, start with RDB. It seems to me that we could then have a
> well-grounded and informative discussion among multiple groups of
> developers
> who have each attempted to solve the same problem using different
> technologies -- I'm reasonably certain that each will ultimately provide
> insight that helps the other, and the solution we're all seeking will
> emerge
> from the dialog. I'd rather see us discussing what we've *done*, and what
> specific challenges we face, then what we're going to do and what *might*
> come up.

[VK] I agree! It is then we will be able to demonstrate that RDF is better or
worse than RDF for a certain class of applications. For instance as a community,
we should also be willing to identify use case where SW technologies are NOT
appropriate.

Cheers,

---Vipul

Received on Tuesday, 4 April 2006 13:26:27 UTC