Upto: Table of Contents of full book

Linked Data

Linked Data allows us to (possibly) answer questions such as "Who was the mistress of the author of Childe Harold's Pilgrimage?". The possibility arises that there are sufficient RDF documents detailing the relevant predicates, that are known to a suitable search engine.


Linked Data

We have seen that both resources and predicates can be identified by appropriate URIs as RDF documents. If there is a rich enough set of such resources and predicates, then we can begin to follow them to derive more complex relationships than is possible with, say, Google.

Linked Data is about situations where the available RDF data allow us to derive information simply by following links: it does not include deep deductions based on logic. That is the role of the Semantic Web.

What would be involved in the finding "the mistress of the author of Childe Harold's Pilgrimage"? We would need a bibliographic record that allows us to find the author from the title of the work. Hopefully, that author would be described by a URI. Whatever, we would need to find the author's URI in some way. Hopefully that would be, or would point to, an FOAF URI for Lord Byron. That URI would list the multitude of Byron's friends, lovers, casual acquaintances, etc. Perhaps one of the extensions to FOAF would be more explicit about these relationships, even to the extent of labelling Lady Caroline as (one of) his mistresses, and giving the URI of her FOAF document.

There is a lot of wishful thinking in this! How feasible is it now, or with thousands of cataloguers and FOAF enthusiasts would it be in the future? And are there any other RDF resources that would add to this particular web of information?


Copyright © Jan Newmarch, jan@newmarch.name

If you like this book, please contribute using Flattr
or donate using PayPal