RDF and the semantic web
Dare Obasanjo points to this piece by Elliotte Rusty, which questions the value of RDF:
I guess the RDF model is a little simpler. It's all just triples, that can be automatically combined with other triples, and thereby inferences can be drawn. Does this actually produce anything useful, though? I don't see the killer app. Theoretically a lot of people are talking about combining RDF and ontologies from mulktiple sources too find knowledge that isn't obvious from any one source. However, no one's actually publishing their RDF. They're all transforming to HTML and publishing that.
That's part of it. The other part is that this is a lot like the problem with OO and the holy grail for a generic "Customer object". The problem is that your definition of a customer is likely to be different than my definition of a customer. In the same vein, the RDF triples defined by one group of developers is unlikely to match those defined by another one. The basic problem is vocabulary - go read this book on the creation of the OED to see how hard a problem that is :)
Comments
Except...
[Rich] May 22, 2004 21:16:18.000
... that it's quite straightforward to define mappings between ontologies; e.g. if you have a CustomerA class with a relationship mapping a CustomerA to an order, and I have a CustomerB class being mapped from an order ("hasOrder" and "hasCustomer", say), they can be simply declared as inverses, and everything runs like clockwork.
Differences in ontology have been expected, planned for, and researched from the start: see the TAP project, Akahani's paper on ontology translation, and Gruber's work, amongst others.
Whatever's useful can be combined; it's not very much like OO, because it's property-centric with an open-world and no unique names assumption.
Rich
Re: RDF and the semantic web
[Michael Lucas-Smith] May 22, 2004 22:55:31.000
Comment on RDF and the semantic web by Michael Lucas-Smith
A topic relative to this is TopicMaps and Ontology Web Languages (OWL). See http://www.topicmaps.org - it's an XML topics interchange language that entirely deals with the concept of merging two topic maps together which are of differing ontologies. OWL is a relatively new beast for defining your ontology.
my favorite page about the semantic web
[Dan] May 22, 2004 22:58:47.000
A fascinating perspective on the semantic web is Paul Ford's August 2009: How Google beat Amazon and Ebay to the Semantic Web (from July, 2002).
Dan