Going LoC(o) with Zotero: Scratching the inner librarianPosted: August 17, 2012
I have always been a very messy person, especially in my work area. Here for example, is a not unrepresentative photo of my home office in 2005 (since one normally doesn’t take pictures of messy rooms, this is the only one I have: I took it to use as a slide in my 2005 Pseudo Society talk at the Kalamazoo Congress on Medieval Studies, “Using computers to improve efficiency in research and teaching”).
Perhaps oddly, however, this same messiness has never extended to my bibliography. Ever since I began university as an undergraduate in 1985, I have kept very careful bibliographic records. I still have the “Articles Card Cat.” I put together at some point during my undergraduate years.
As the first card in the photo might suggest, this was no ordinary bibliography: articles are catalogued under author name, on subject cards, and in some cases under journal titles. I remember beginning this catalogue in my second or third year when I realised I was citing the same papers in different essays. I kept it up right through graduation and the beginning of graduate school.
The next stage in my mania came when networked citation managers began to offer the ability to access Z39.50 servers. This allowed the rapid acquisition of bibliographic data and building of bibliographies and works cited lists. Previous generations of citation managers and bibliographic software were really little more than souped up card catalogues: the bibliographic manager I used to replace my card catalogue in graduate school, for example, provided little more than an incremental improvement over my previous system: it could be more easily maintained, searched, and sorted (and of course there was no longer any need to keep separate subject cards since subjects could be drawn from the item records), but the data was still typed in by hand and my records were still being prepared by an amateur based on what he had time to type out from the title page.
The ability to access Z39.50 servers brought with it another improvement: the automatic collection of call numbers. Because ProCite (the reference manager I used at the time) accessed library catalogues and other databases directly, it also had the capability of importing library call numbers with the rest of the bibliographic information. Initially, I thought this would be useful because I could use it to find books whenever I was working in a library: if I was working in the PR 1500 section of some library, for example, it might be as quick to look on the shelf to see if they had Colin Chase’s The Dating of Beowulf (PR 1585 .D37 1986) as it would be to find a terminal (in those days before Eduroam and ubiquitous wifi) and look it up in the catalogue.
This actually didn’t turn out to be as efficient as I expected. I found there was a surprising amount of variation among libraries in the precise call numbers they assigned to books, especially books published before the 1990s, and libraries rapidly became generous enough in the distribution of terminals that it was rarely required more than a short walk to consult the catalogue after all.
It wasn’t long however, before I thought of another use for this data. For years, I had tried, ineffectively, to enforce a policy of not lending personal books to students. Especially in the case of my more specialised collections, my resolution not to lend always broke down in the face of student excitement or inspiration. When an undergraduate comes to your office hour full of questions about early English philology, language change, or dialectal variation, it is difficult to let them leave without lending them a copy of Roger Lass’s Old English: A historical linguistic companion (PE 125 .L37 1994, currently on loan to a graduate student), Janina Brutt-Griffler’s World English: A study of its development (PE 2751 . B78 2002) or John Baugh’s Beyond Ebonics: Linguistic pride and racial prejudice (PE 3102 .N42 B37).
You can only do this, however, if you know where the books are, and, more importantly, get them back after you’ve broken your rules and lent them out anyway. From my first days as a graduate student, I had tried to deal with this problem by making the bibliographic database do double duty as circulation software as well: the idea was that I would record the student’s email address and the date in the book’s record in my database. When I needed the book and couldn’t find it on my shelf, then I could look it up in the reference manager and see who had it out.
This system was about as effective as a non-automatic backup. In the excitement of working with students, I usually forgot to record what books I had sent them out the door with, and, since students often returned books to me in class or in the hallway when I was far away from my computer, I always forgot to record that books were returned. The result was that my records were extremely unreliable. The fact that a book was not on the shelf and had a notation on its record saying that it had been lent to a student did not mean that it was not lying under a pile of papers on my desk: the student could easily have returned it and I could easily have forgotten to remove their name from the system. And more importantly, the fact that a book was not on the shelf and did not have a notation saying it was lent out did not mean that a student didn’t actually have it tucked away in a dorm room somewhere. The absence of a record did not necessarily mean that I should be able to find it if I just looked around my office thoroughly enough.
The call numbers offered a solution to this problem: if I organised by books by call number, I figured, then I could use the shelf itself as a way of keeping track of where my books were: if I lent a book to a student, all I had to do was leave a card on the shelf where the book was supposed to be saying who had it and how I could contact them. And when the book was returned, the call number would lead me to back to the point on the shelf where the record of the loan was kept. Even if somebody gave me the book in the hallway, I would always remember to close the record of the loan out when I returned the book to the shelf.
The solution was attractive for another reason: it required library supplies! If I was going to find the card indicating that a student had borrowed a book, then I needed a way to indicate where the book should be shelved: in other words, spine labels. And since the system would break down if I needed to do anything other than have the student write their email down and put the card on the shelf where the book was supposed to go, I would need to have a prepared card for every book: meaning borrowers cards and pockets. As Hunter S. Thompson once said in a different context: “once you get locked into a serious… collection, the tendency is to push it as far as you can” (Fear and loathing in Las Vegas, PN 4874 T444 A3 1989).
This system actually turned out to work. It really is quite easy to remember to record that a book has been lent out if there’s a card in the back, and students seem to take borrowing more seriously when they’ve actually signed the book out by putting their email address down: I rarely have to chase books down nowadays, and don’t think I’ve lost one since I began using the system.
Moreover, cataloguing and shelving books in a personal library by Library of Congress Classification (LCC) also produces some interesting juxtapositions: Elisabeth A. Lloyd’s The case of the female orgasm: Bias in the science of evolution (QP 251 L56 2005) shows up on my office shelf between Halliday, Resnick, and Walker, Fundamentals of physics (QC 21.3 .H35 2005) and First on the scene: The first aid activity book (RC 86.8 .F592). The proximity of Jennifer Niederest, Web design in a nutshell (TK 5105.888 .N54 2001), Edward Webster, Print unchained: A saga of invention and enterprise (TK 7887.7 .W43 2000), and Hieatt and Butler, Curye on Inglysch (TX 705 .C87 1985), likewise, is a useful reminder that we did not invent technology and change any more than the baby boomers actually invented sex.
While the LCC typology is neither perfect nor the only way of understanding the knowledge a library contains, adopting an external typology such as the LCC for your personal library is also interesting because it causes you to think outside your own intellectual boxes and categories: before adopting the LCC for my office library, I used to shelf books by genre (fiction, drama, poetry, biography, history, criticism, etc.), by period (Old English, Classical, Victorian, and so on), and by author name; under my old system, Curye on Inglysch would almost certainly have been classified as non-fiction medieval history or culture, rather than as having something to do with technology or home economics.
In recent years, a new generation of cloud-enabled citation managers and bibliographic resources has made collecting references even easier. By taking advantage of massive aggregated collections like World Cat and Google Books, these managers are able to find books far more quickly and reliably than the first generation of networked systems. With ProCite I frequently had to search several individual libraries before finding one that contained a copy of the book I had on my shelf, with Zotero, Google Books, and World Cat, I now rarely come across a volume in my collection that I cannot find an existing record for on my first attempt. While neither Google Books nor World Cat return records with a Library of Congess classification, finding the number is also relatively easy: most books intended for the U.S. or Canadian markets have LoC Catalogue in Publication data (including the call number) printed on the copyright page. For those that do not, Zotero and World Cat offer one click access to records in circulating libraries that do have this information. With Scanner for Zotero, an Android App that scans and searches for ISBN numbers from the bard codes found on the back of most books, nearly the whole process has become automated. If I could figure out a way of getting World Cat to return a LCC with its output, I would be able to catalogue most of my newer books by doing little more pointing my phone at their back covers.
Even having to find the LCC by hand does not slow me down too much: most new books can be added to my collection in less than three minutes–including the time it takes to write up the spine label and borrower’s card for the back cover. Even older and more rare books seldom take more than 5 minutes to find a record for somewhere in the world. Every morning now I catalogue 5 books from my collection before beginning other work, something that usually takes me between fifteen and forty-five minutes. This is peaceful work, and, since it involves reminding myself about potential references and sources I have on my shelf, I also find it is a good way to get started: there’s nothing more inspiring for your own research and writing than reminding yourself of what others have done (the same reason I find the DH twitter chatter to be so inspiring). And when ideas strike a catalogued reference is far easier to find than one that isn’t.
I wouldn’t recommend cataloguing this way if you don’t have an inner librarian that is itching to get out. But if you do have an inner passion for call numbers and bibliographic order, organising your library by LCC does offer some advantages. Whenever I’ve had to move office, my bookshelves have been easy to recreate–books can be boxed by size and then easily resorted once you are in the new office. But perhaps most importantly, I find it almost encourages me to lend books to students. Students really do seem to treat books they have signed out, however informally, better than they do books they have just taken with them, and having the books organised by a formal system means that you are more likely to discover similarly useful books on your shelf that you might otherwise have missed. And you get to buy library supplies!