I am excited that Documenting Cappadocia and I have scored an invitation to the Linked Ancient World Data Institute (LAWDI) this weekend, hosted by New York University’s Institute for the Study of the Ancient World. Linked Open Data is a method of structuring digitally published information so that it is stable and easily linked, enabling more effective links and better-networked scholarly resources.
Initial reading assignments for the event reveal that data on the web are interrelated. Among discussions of programmers, academics, librarians, and other practicioners of Linked Open Data is how those relationships are best conveyed by stable identifiers. Tom Scott has considered the difference between using identifiers such as URIs to distinguish web documents from the real-world things those documents represent. Ed Summers responded with a call for common sense–of course people know the difference between a real-world object and a document about it on the web, but it’s possible to maintain that distinction. Mike Bergman has taken the debate into the philosophical realm, pondering the responsibilities and implications of the practice of naming, of developing a vocabulary in the real world or in terms of identifying data.
The debate within the Linked Data community revolves around relationships between things and documents. There’s an ontological element to this discussion–these objects seem to have a life of their own, making their way around the internet, holding on to metadata and links that people attach to the documents before sending them on their way. From a medieval art historian’s perspective, this is similar to thing theory, the study of the life and existence of objects that have an agency of their own. Relics and icons functioned this way in Byzantium, working miracles, for instance. Pilgrimage souvenirs were thought to actively protect the pilgrim. Gems could conquer thirst, or blue charms could ward off the evil eye, without necessarily needing to be activated by a person every time. Some of these objects were made by people, but all of them had an agency that continued separately from their creators.
So this is my theory: Linked Data has a life of its own. Although humans initially put the data out into the world, it can then demonstrate relationships between objects and documents, and it can convey information through metadata without further human intervention. As practicioners of Linked Data, it’s up to us to send that data out into the world with adequate links and thoughtful, stable identifiers.