I See Dead People['s Books] (LibraryThing)

LibraryThing has a group dedicated to the topic: I See Dead People['s] Books. Despite the ghoulish name, what they're chasing is important from a scholarly point of view -- and a business one, too, I think.

One of the most intriguing books I used for my dissertation was a catalogue of Carl Friedrich Gauss's library. Even though I kept coming back to it, I was never able to solve the basic historiographic problem that it posed: What sort of evidence is this? Evidence of what? What kinds of conclusions can be drawn from these data? (LT has just finished a similar project, a catalogue of Thomas Jefferson's library.)

Current academic historiographical conventions disallow, or at least frown upon, claims like "He was a wit, open-eyed and realistic, but susceptible to the epic kinds of romantic enthusiasms you find in his favorite authors, Walter Scott and Jean Paul."

But people in the business of hand-selling -- booksellers and literary agents -- make these calls all the time.

Library data encode important information. We just don't know what sort of information it is. Is it information about mentalities? Literary influences? Certainly, we learn something about what publishers think is worth publishing at any given time. But it's the decision to own that poses the problem -- and the opportunity.

Reflecting on the Jefferson project, LT's founder Tim Spalding put it this way: "Books are a sort of mental world, and shared books a shared mental space."

It's the "sort of" that gets me. Sort of this, sort of that -- we just don't know what to do with this information. We don't even know what to call it. I like Spalding's spatial metaphor, though. Instead of talking about some vague collective consciousness, we're actually talking about something real, even topographical. A shared library.

I have a feeling when big money gets behind LibraryThing (and it will), the idea will be to automate the hand-selling process. Amazon's "If You Like...Then You'll Like" algorithm isn't nearly as precise as LT's library data because the former is based on what you buy (for yourself, yes, but also for your kids, your mother, who likes how-to manuals, your neighbor who likes Tom Clancy, your nephew who really likes obscure poets, etc. etc.) rather than what you love enough to own, meaning make space in your life (on your shelf) for. In contrast, LT's got the data set that publishers really want, and it's been unavailable until now because people don't make their libraries publicly available. But LT could easily aggregate and anonymize the data, circumventing privacy concerns...