BROWSING THE INTERNET'S SHELVES
You know how, when you head to your stacks looking for a specific book, you find that it doesn't do what you wanted at all, but you still walk away with a stack of books that were sitting next to it, that DID do what you had hoped that one would do? That's the beauty of a dual purpose cataloging system: one that assigns subject headings, and one that physically places the book on the shelves near books on the same subject. What that means is that, if we don't find what we want with one of those systems, we have a backup system that gives us a fighting chance of finding it anyway. In other words, we've built a system that allows us to find things by a combination of precision searching and damn fool luck.
Neither cataloging nor classification works perfectly. In the Library of Congress system, a subject like "women in the arts" inevitably scatters the books you want across the N's for female painters and sculptors, the M's for women composers, the TR's if they were photographers, the PR's if they were English women writers, or the PS's if they were Americans, unless they're in the PN's because what they wrote was plays. You need a good subject catalog with a good set of cross-references to even think of all the subject headings, let alone find the books.
But because catalogers don't all assign the same subject headings to similar works, and searchers might use different terms altogether, the classification system often keeps people from finding something that is sitting right there on the shelves. My favorite example of a cataloger's insanely specific subject heading is the one assigned to Edmund G. Love's The Situation in Flushing, a charming memoir of his small home town in Michigan at the beginning of the 20th century. The book dealt with small town life in America, the delayed, reluctant adoption of electricity, and a young boy's love affair with trains. The sole subject heading assigned to it was: "Flushing, Michigan -- Social life and customs," which guaranteed that virtually nobody would ever look for, or find, this book. (A clear case of bookicide by cataloger.) At least in the Library of Congress system, you could find this book in the F's, beside other books on midwestern history.
The Internet presents even more of a challenge: there are no helpful call numbers, the subject headings, if any, are assigned by the people who create the pages, and there are no helpful librarians right at hand rushing to assist the bewildered.
True, there are a number of helpful subject catalogs to the net, if people would only use them. Unfortunately, the preferred strategy of users is to look for a blank box, type in what they want, and hit enter. Sometimes it even works: If they spelled it right. If they thought to mention what it was they wanted to know about the topic: "abortion laws," "muscular dystrophy treatment." If they were patient enough to look past the first ten results.
That's why my favorite search engines imitate libraries by allowing a similar combination of precision searching, good luck, and some sort of quality assessment. My current favorite, Google uses collaborative filtering to rank search results, which means that people have endorsed the top ranking sites by linking to them. The "more like this" feature then allows you to shuffle the deck and find sites with more explicit resemblance to the sites you've liked -- the equivalent of sitting on the shelf right next to the book you chose in the first place. (Alta Vista allows both extremely precise searching and "related pages," but I don't see any sort of quality assessment in its results.)
HotBot also allows a highly specific search -- "walk the dog" as an exact phrase, must contain the word "yoyo" -- and then its Direct Hit collaborative filter allows you to see what sites others have most often chosen for that topic.
Most search engines will also refer your search to commercial partners like Amazon, which serves as a useful reminder that there are still such things as books out there, and you might want to consider using one.
But since we know all the pitfalls naive users encounter on search engines, and we know how much the amount of trash on the net exceeds the good stuff, we'd really like to guide them, and increase their chances of finding high quality material on, and by way of, the net. What can we do on the net to provide the same fail-safe combination of precision searching and sheer dumb luck that we offer for our physical collections?
The Internet Public Library (http://ipl.org/) and the Librarians' Index to the Internet (http://lii.org/) both offer a combination of carefully chosen sites, organized by subject. The keyword search box looks through those selected sites. The search retrieves links to selected sites, describes them, AND makes the subject headings assigned to them clickable so that you can find more sites like them. Only if users find nothing this way are they invited to search through the entire net rather than through the cozy confines of a quality collection.
There are two other things we might do. We could program each results page to include the question: "Would you like to search this topic through our online magazine collections or our catalog?" And we could also include this: "Did you find what you needed? If not, your librarian can help you," adding our phone number and/or our clickable reference desk chat line or e-mail address.
The search engines are getting better all the time. But we still know more than any software program about what users want and how to compel search systems to deliver it. We've just got to do everything possible to convince people of that: by giving them the good stuff, giving them multiple ways of finding it, and providing the ultimate fail-safe back-up system: a friendly librarian on the other end of the line.
* * * * * * * * *
COOL QUOTE
Luck is what happens when preparation meets opportunity.
Darrel Royal, quoted in James Michener's Sports in America
* * * * *
A HALLOWEEN IDEA
A few years back, when we got to the point where we needed to carve out paths through our books to make it to the bathroom in time, I made my son sort through his book collection and decide which ones he no longer needed. (Better him than me.) There were a couple of boxes of really neat kids' books, which I might normally have taken down to the public library for their book sale. But the next night was Halloween, so I decided instead to give away a lollipop and a book to every trick or treater.
Much to my surprise, the kids' reactions were, "Cool!" (and the parents' were even happier). I still had plenty of books left, so I did the same the next Halloween, and found that I had repeat customers, kids who were specifically hoping "the book lady" would have something neat for them. I've been doing this for ten years now, and the kids still keep coming. To augment the collection, I've taken to buying children's books for 10 cents apiece at library sales and yard sales and Salvation Army stores.
Try it yourself, why don't you? Lollipops don't last, except in the form of cavities. Books keep on giving. And the future of libraries depends on raising a new generation of readers.
* * * * * * * * *
Note on schedule: There will be an ExLibris and Neat New November 2, but not November 9, since I'll be at the Internet Librarian conference, where I'll be speaking on the topic, "If You Give It Away for Free Will He Still Marry You? or, Building a Business by Giving Away Free Content on the Net." I hope to meet some of you there.
* * * * *
You are welcome to copy and distribute or e-mail any of my own articles for noncommercial purposes (but not those by my guest writers) as long as you retain this copyright statement:
Ex Libris: an E-Zine for Librarians and Other Information Junkies.
http://marylaine.com/exlibris/
Copyright, Marylaine Block, 2000.
