Matthew Cockerill wrote:
> I am coming to this (and I think Alf too) from the perspective of wanting to develop practical services about actual stuff, as opposed to operating in domain neutral RDF world.
>
> These services would build on top of all the abstract RDF stuff.
>
> But ultimately, we need to boil things down to a few practical things that we know how to deal with.
> i.e. the infinite regress has to bottom out somewhere, on some leaf nodes, which are taken as atoms to actually mean something.
> info URIs are one way to provide this.
> This may be philosophically inelegant, but it seems to me pragmatically invaluable.
>
>
> In terms of the suggestions that have been made for alternatives:
>
> If I have came across 2 URIs:
> http://en.wikipedia.org/w/index.php?title=Special:Booksources&isbn=0393060349
> and also:
> http://isbn.org/madeupresolver?039-3-06034-9
>
> Using those alternatives, how would I as an application developer know that they were corresponded to the same entity?
Matthew,
here's the thing: say you trust the info-uri people and I trust the
wikipedia people, we'll use different URIs for the same things.
How is that going to help?
Having the URI and having people using it are two different things,
especially in such a controversial space such as identification and
dereferencing.
We must develop systems that are more flexible than that and both Leight
and Rickard point it very well: using a URI as a URL is only one of the
possible ways to dereference something.
At the end of the day, it boils down to the fact that it really doesn't
matter what schema you use or what URI you use, as long as they are
consistent in your own infospace and globally unique.
Unlike what people normally think, the semantic web will emerge only if
we allow to encode and deal with disagreement, indipentent reinvention
and misalignment.
And this is why OWL equivalences are way more important than the whole
entailment, description logics and all that stuff.
So, people let's agree to disagree: you use what you like, and somebody
else will link the pieces together.
--
Stefano Mazzocchi
Research Scientist Digital Libraries Research Group
Massachusetts Institute of Technology location: E25-131C
77 Massachusetts Ave telephone: +1 (617) 253-1096
Cambridge, MA 02139-4307 email: stefanom at mit . edu
-------------------------------------------------------------------
Received on Tue Nov 15 2005 - 13:56:52 EST