LibraryThing, I think I love you
I’ve been aware of the existence of LibraryThing for over a year, but I only got around to cataloging my own library a few weeks ago. It was so much fun that I cataloged the libraries of four other friends within the next two weeks.
In addition to allowing you to organize and sort your personal collection, LibraryThing provides a wide array of statistics, based on the catalogs of its individual members.
You can see how many people own a book and how many books are more popular. You can even see the usernames of all 17,542 members who own a copy of Harry Potter and the Sorcerer’s (/Philosopher’s) Stone, if you really want to. One clever thing that LT does is allow its users to combine different editions of a book into a single “work,” which tidies up the statistics for books which have been published many times. (A disadvantage of this system is that all translations are grouped together with their original language editions. I’d kind of like to see info on just the people who own Harri Potter a Maen yr Athronydd – the Welsh version of the 1st Harry Potter book – but I haven’t figured out how to do that.)
Like Amazon, LT offers suggestions on an individual book bases. However, I think that LT’s book suggestions are superior to Amazon’s. For one thing, Amazon only has information about books you bought through them – if you acquired the book before Amazon came into existence, or from another source, they don’t know about it. Another issue is that Amazon is in the business of selling books, so they’re not going to recommend a title that’s out of print, even if it was very popular in the past. Lastly, LT takes into account statistical obscurity when making its recommendations. (If this wasn’t the case, LT would just recommend out a list of bestsellers for every book entered.) Every time a new Harry Potter book comes out, Amazon recommends it in addition to almost every book I purchase from the site. (Of course, I don’t know if this is due to a statistical issue or if they’re just trying to move copies of Harry Potter.)
The “similar libraries” feature is by far my favorite statistical toy. This feature looks at your entire library and finds other users with the most similar libraries. Again, it takes library size and book obscurity into account, which means that someone with a very small library who also owns Supposition Error will rank as more similar to me than someone with a large library who also Jane Eyre.
As I started cataloging my books, it was interesting to see the clusters of similar libraries that appeared. Not many people own Jim Krause’s graphic design reference Color Index, but many of those who do also own his Layout Index and Idea Index. (They have separate ISBNs, but they’re also available as a boxed set.) A lot of people who have Jim Krause’s books also have Stop Stealing Sheep, and the most similar book to Stop Stealing Sheep is The Elements of Typographic Style, which I also own. The Art of Looking Sideways is another popular book with the graphic design / typography crowd. (Speaking of which: Thmazing, if Lady Steed doesn’t have that last book, I think she’d like it.)
Another cluster of people own library-science related books like The Organization of Information, Reference and Information Services and Essential Cataloguing. (These were actually required textbooks for my classes here at GSLIS, so some of the other people who own them are not just fellow librarians, but classmates of mine here at the U of I.)
I started out by cataloging the books I have with me in Illinois, which include all the books and textbooks I’ve bought here, as well as books I brought with me from Utah. (This latter group mostly includes dictionaries and other reference books.) After adding just these books, Melyngoch showed up on my “similar libraries” list; we have a number of linguistics books in common, as well as some other books.
Another LT feature is the “author cloud.” It’s a list of all the authors represented in your library, the more books you own by a given author, the larger their name appears.
When I looked at my author cloud, I realized that a lot of my favorite authors weren’t represented, either because I didn’t have their books with me in Illinois or because I’ve only checked out their books from the library.
I decided to add books I’ve read but don’t own (or don’t have with me) starting with what I’ve recorded in my book diary, and then just adding any books I remember reading.
This had the effect of making my author cloud a more representative record of my actual favorite authors, including Kurt Vonnegut, Salman Rushdie, and Douglas Adams. Unfortunately, this also had the effect of bumping Melyngoch off my “similar libraries” list.
Statistically speaking, this isn’t surprising. For all that she and I having in common, I haven’t read much Adrienne Rich or Marianne Moore, and I don’t own a single copy of Beowulf (let alone five).
Nonetheless, I was bothered by this event. I mean, how do I know we’re really friends if we don’t have statistically similar libraries? Maybe the last seven hears have just been a fluke . . .
The obvious solution, then, is to read a bunch of books from Melyngoch’s library so that I can add them to my own. Ideally, I’ll start with the more unusual ones, since that will have a greater statistical effect. (Melyngoch’s reaction on my explaining this scheme to her: “I love that you’ve found a way of quantifying our friendship, and now you’ve figured out how to mess with the statistics.”)
All this is by way of explaining why I was reading Shakespeare, Monty Python, and Renaissance Drama two weeks ago on my flight to Maine.