Thomas Baker’s analysis of “Libraries, languages of description and linked data” was both helpful and frustrating. Since my group is doing linked data (I think?), I have put some research into it and found that this video:
to be more helpful in explaining how linked data actually works, while Baker’s article on occasion veered off the road onto the academic, jargon shoulder: “An ontology should require the minimal ontological commitment sufficient to support the intended knowledge sharing activities.” A word should not be used in its own definition.
Nevertheless, he made a few solid points, such as “The usability of today’s linked data in future decades will depend in part on the preservation of its vocabularies by memory institutions.” And “The viability of linked data in the long term will depend on the preservation of vocabularies across generations,” and begins with a good definition of RDF (“RDF is fundamentally a grammar for a language of data. It is a language designed by humans to express human thoughts in a form amenable to processing by machines.”) The first two sentences explain the divisions and bickering among metadata specialists on how to name things. Linked data takes it one step further and needs a vocabulary that expresses not only how things are named and stored for retrieval, but how they can be connected to other things. The article I downloaded did not include the tables, so perhaps there were neat tables showing examples of the all important “triples.” Sometime diagrams are clearer than language.