Every Identity, its Ontology

Or: The shared identity of the concept of the fictional person Dr Strangelove: How I learned to stop worrying and love inconsistency.

The notion of Linked Open Data is that there are identities that can be shared and reused. The cultural heritage sector demonstrates a common challenge for this ideal in its multitude of "authority" lists from Library of Congress, to OCLC, to the Getty Vocabularies. Even when mapped into Linked Open Data, these existing identities are inconsistent, even unintelligible, as we lack a common ontology to serve as the means of expressing the fundamental particulars of the individuals we claim to have those identities. The long established tradition of collecting names of things and blessing one over others does not translate successfully into an ecosystem predicated on shared identities. For example, some systems distinguish pseudonyms just as alternate names, some as personae that can act separately, and others as actual individual people. Given this challenge, which is both philosophical and practical, what paradigm shifts are possible to bring about sufficient confidence in the distributed system that we can have a true web of cultural data? This seminar will address some of the philosophical difficulties of identity within Linked Data, some of the practical improvements that we can make, and some of the processes that will be improved by doing so.