- From: Kashyap, Vipul <VKASHYAP1@PARTNERS.ORG>
- Date: Fri, 31 Mar 2006 13:33:31 -0500
- To: "Jim Hendler" <hendler@cs.umd.edu>, "Danny Ayers" <danny.ayers@gmail.com>
- Cc: <public-semweb-lifesci@w3.org>
- Message-ID: <2BF18EC866AF0448816CDB62ADF65381033C3267@PHSXMB11.partners.org>
Jim, Contrary to the impression you may have got based on our current exchange, we are actually in "violent agreement". I am in complete concurrence to the idea of "little semantics" and incremental ramping up of semantic sophistication. My only concern is that this "little semantics" be characterized accurately and the value proposition should be evaluated So far I have not seen any work in this direction. the "sameas" type information (expressed through same URI names) is very powerful, even before you start worrying about the next levels of semantics. [VK] This is the sort of characterization I was looking for. So what the Semantic Web provides is the ability to explicitly specify and exploit the "same as" and other "little semantics" relationships. What would those other relationships be? Even more important, once the data is on the Web in RDF, it can be INCREMENTALLY extended, by the original provider or by third parties, in ways that do add the expressivity - something not doable when the datasources are not Web accessible. [VK] At an intuitive level, this is fine, but what the SW community needs to provide the world is with a proof and demonstration that this actually happens and what are the cost benefit trade-offs involved. I believe this is one of the things the BIORDF task force is investigating in the context of HCLS.. Here's another way to think about it - on the Web my documents can point to your documents. However, my databases (or their schemas) cannot point at elements in your databases. my thesaurus cannot point to words in your thesaurus, etc. The Web showed us that the network effect is unbelievably powerful, and we need to be able to use that power for data, terminologies, ontologies and the rest. [VK] We have seen proof of the network effect in the context of web documents. However, we are yet to see it in the context of the semantic web. And may be a characterization of the network effect may be viewed as a transitive closure over the "same as" relationship... So whereas, intuitively, I agree with you, we are still to see a proof or demonstration of these things... Cheers, ---Vipul
Received on Friday, 31 March 2006 18:33:41 UTC