Metadata provides structure and organization, both enabling and constraining searches, Big Data analysis and the use of other tools to look at what it tags. It’s sometimes invisible to users, often ignored by them in any case, and yet can heavily influence the materials they consider as well as the results they receive when applying analytic tools. There’s also a deep tension between having clear and consistent tags which convey information in a brief and precise manner, and tags which capture provisional or uncertain information or permit for the range of fuzziness which often arises in non-computing spaces. For example, the metadata on EEBO-TCP texts whose dates are conjectural defaults to the beginning of the century they were likely written in, meaning that various sorts of analysis will find spikes in 1501 or 1601 or 1701 because those dates match the metadata entries for these texts.

I’d like to invite a conversation between participants which considers both the practical, on-the-ground realities of making metadata for search and designing search tools to draw out the meaning of metadata, as well as the broader theoretical issues involved in placing a definitive tag on material which may be quite indefinite. (Does Shakespeare’s [i]King Lear[/i] receive a tragedy or history tag, for example?)

As recent massive leaks of classified data by NSA contractor Edward Snowden revealed, the U.S. government is taking advantage of previous court rulings that metadata shared with a third party carries no reasonable expectation of privacy to conduct upstream taps of telephone and web traffic metadata. With computer storage costs plummeting and processing speeds continue to curve upward, it has become feasible to capture massive amounts of metadata and to search it for patterns useful in anticipating terror attacks, locating enemies of the state overseas, and prosecuting computer network related crime. This process only works well if everyone’s metadata is available and if those being investigated are unaware of the metadata they are producing. This creates a legal system and by extension a society with asymmetric transparency. Is this desirable? Digital humanities is an emerging discipline that is heavily invested in the creation and use of metadata. More than that, digital humanists just grok metadata. That makes us qualified to participate in the conversation about metadata privacy. This talk would be an opportunity to consider what role digital humanities scholars could or should play in the transparency policy debates.

Databases are my friends. Every application that I use and care about has a database built into it somewhere: digital audio workstation, video editor, address book, calendar, to-do list (don’t forget online shopping…). Important research is predicated on the analysis of databases. Increasingly, many of my favorite works of art are driven by databases. I’m inspired by civic hacking projects made possible by the open data movement.

So why is it such a pain to create and work with databases? How come we still reach for a spreadsheet when we know darn well a database would be better? Would the world be a better place if everyone databased? Has the NSA made database a dirty word? What tools do we need to make databasing more like sledding?

Come share your love/hate experiences with databases. Teach us about your favorite tools for making life with databases more fulfilling. Let’s imagine together what will replace databases as we know them.

I’m interested in discussing how Library Science’s bibliographic theories intersect with the Humanities’ conception of creative works. In particular I would like to examine Library Science’s FRBR concept in this light.

FRBR (Functional Requirements for Bibliographic Records) is a conceptual framework that has had a large impact within the Library and Information Sciences world, particularly in how bibliographic entities are conceived within a catalog system. Among other things, FRBR proposes a system of four inter-related abstraction that are designed to describe different parts of a creative endeavor’s whole. Work describes the creative idea behind a bibliographic entity, Expression describes the specific artistic form that realizes the work, Manifestation describes the physical form that that expression takes (a specific edition of a book), and Item describes one single instance of a manifestation (the particular book that you have in your hand).

For example: Shakespeare conceives of a Work he will call King Lear. He writes a manuscript of this work for performance. This is one Expression. Shakespeare’s original manuscript is, of course, lost, and our modern text derives from the text of the First and Second Quartos and the First Folio. As the versions in the quartos and the folio are significantly different, they too represent different Expressions of the same work. The version created by conflating these two versions would be yet another Expression. The Norton Critical Edition of the play would represent one Manifestation of it. And then, the specific copy of the Norton Critical Edition that sits on your bookshelf, and that you’ve marked up and spilled coffee on, is one particular Item.

As you can imagine, these concepts can become fairly muddled, and often create significant questions about any particular bibliographic entity (especially when dealing with something that has as much of history as King Lear): Does the Norton Critical Edition represent a whole new Expression considering the editor primarily uses the Folio text, while also integrating passages from Quarto I? Yes, it probably does. What about the critical material that accompanies the text, do those have their own Works and Expressions? Yes… well… possibly?

How do library systems built around these ideas work with scholars’ expectations for information retrieval systems? How extensive should these systems be? Could systems built around these concepts impose too rigid of a structure on the critical history of a creative work?

FRBR is a rather large and unwieldy topic within Library Science, and it would be unreasonable to expect a tremendously in-depth discussion about it within the context of THATcamp, but I believe thinking about it within the concept of humanities could yield an interesting discussion.

Wikipedia has a somewhat comprehensible explanation on the concept of FRBR and, as usual, is a good place to get a broader picture of it.