I have way too many end-of-term projects to finish in the next few weeks, so in an attempt to force myself to write my upcoming conference paper on time, I’m going to put some of the points in my head into a post. With any luck, it’ll help me over the getting-started hurdle. Those of you who are here for the opera reviews or the random poems or the minutiae of my daily life, feel free to ignore this post — there are more of all of the above coming before long, I promise.
So the conference theme is "Questioning Authority," and I’ve been wanting to write something about LibraryThing for ages, mainly because LibraryThing fascinates me. In part because it’s kind of counterintuitive that a bunch of website users could generate not only a classification system (the tags), but also a way to disambiguate authors with the same name, as well as a way to link up multiple editions and versions of works (the "combine
works" function, which approximates what libraries are trying to do
with Functional Requirements for Bibliographic Records, a.k.a. FRBR).
So my angle on the "authority" theme is to play on a couple of different senses of "authority": authority as in "authority control" (librarian-speak for the work of creating consistent, "authorized" forms of names and subject terms to avoid confusion in the catalog), and authority as in who’s authorized to perform this kind of work: professionals at the Library of Congress? Or, in the case of LibraryThing, regular people cataloging their own book collections?* I’m using the phrase "distributed authority control" to describe the phenomenon of lots of people each contributing bits and pieces of knowledge to create a kind of home-grown bibliographic authority system.
One of the things I’m curious about is what motivates people to do this. It makes sense that LibraryThing’s tag classification emerges out of people’s desire to tag their own stuff. But the work-combination and author-name disambiguation seem like an extra layer of user effort: and yet the "Combiners" group is among the most active on the site, and people are contributing to the "Common Knowledge" wiki, which lists information about both books and their authors, at a great rate. My hunch is it’s a combination of a lot of factors: a desire for accuracy (no, wait, that’s not the right Alexandre Dumas!), a certain degree of fannish enthusiasm (I’m thinking of those who contribute detailed character lists for their favorite novels), and — probably the key thing — a preexisting community of interest.
Because if you throw a bunch of random library catalog records at people who aren’t interested in the books in question and say "want to help us improve this data?", they’re not going to care. But get a group of Patrick O’Brian fans** to catalog and describe and disambiguate their favorite author’s work, and they have a stake in it. The Library of Congress (speaking of) put some of its photo holdings on Flickr, and found that a horde of enthusiastic Flickr users promptly started tagging them.
There are more things I want to say: how this isn’t an ideal model, how distributed and professional authority control might coexist, how all of this compares with Wikipedia, etcetera etcetera — but I only have 12 minutes to say it, so a lot of these points will have to get edited down (and, hopefully, re-elaborated in the Q&A session afterwards).
Whew. If you’re still reading this, feel free to weigh in!
* This isn’t an either-or, of course. LibraryThing pulls in that structured library data automatically. And it helps that a fairly large number of LibraryThingers are librarians.
** Example chosen because I’m a Patrick O’Brian fan myself, and there was some interesting conversation in the LT O’Brian group about adding character names to Common Knowledge.
