July 2, 2009 - in defense of Dewey

Back on June 8, 2009, the Denver Post ran a front page story, above the fold, about the Rangeview Library District's decision to abandon the Dewey Decimal System.

The day the article came out, an out-of-state friend was visiting one of our staff members. The visitor asked, "Is this actually a hot issue here in Colorado?" Our staff member kept a stern and straight face. "Absolutely," she said. "Dewey. Anti-Dewey. There will be blood."

My thought was this: I think this is a fine and interesting experiment for a library. Rangeview's director and her staff are trying all kinds of things lately. Fun.

But almost immediately and somewhat to my surprise, I got an impassioned defense of the Dewey Decimal System via email from one of our Douglas County patrons. Her main issue was that although the bookstore style classification adopted by Rangeview might work well in smaller collections, it tends to break down for larger ones.

For instance, "Cooking" (the bookstore or "WordThink" classification) makes a lot more sense than Dewey's "Cookery."

But when you have larger collections and thus more subjects, it's important to have finer gradations. Take subject headings for Korean cooking. In WordThink, it's "COOKING/RegionalEthnicAsian," or just "Cooking International." A label for the book might be "COOKING INTNTL." In Dewey, the subject heading is "Cookery, Korean." The number corresponding to that subject is 641.59519. It is specific to Korean cooking, which means that all the Korean cookbooks will wind up next to each other on the shelf. Numbers make finer breakdowns easier to label -- but harder to find (because you have to look up the subject to find the number, and the subject isn't always the words ordinary people use).

So some people do have strong feelings about this issue. But they're part of that relatively small subset of the population that actually understands Dewey.

I know from talking to the folks at Rangeview that making such a change isn't easy. It took them at least a thousand hours of thinking and planning and procedural design. They're about to open a bunch of new libraries, so this would be the time to make the change. I'm not sure that makes sense for us, though.

Here's why.

First, almost half of our checkouts (48%) are children's materials -- and a big percentage of those are picture books. Their labeling is absurdly simple -- the letter "E" (for "Easy Reader" or "Juvenile Picture" book), then by author. Dewey doesn't usually come into play. (Although you could make an argument that sorting picture books by subject would come in handy sometimes. But not most of the time.)

Second, about 15% of our checkouts are books placed on hold through our computer catalog, then picked up later. Since it's our staff that pulls the books, it almost doesn't matter how they're labeled, so long as we can still find things.

Third, a lot of our collection -- and percentage of checkouts -- is adult fiction. Last year, it accounted for about 14% of our business. Fiction is just shelved by author.

Fourth, of the remaining non-fiction materials, many zoom off the shelf because they're on display -- usually on one of our subject-oriented "power walls." In fact, for these materials, what we're doing already feels a lot like a bookstore. But we build displays on the fly, based on immediate use.

So what's left is just the older non-fiction materials. As an administrator, I ask myself: does it make sense to relabel hundreds of thousands of materials when we have a system that seems to be working right now?

I don't think so.

I've worked with Dewey, with the Library of Congress system, and with BISAC (the Book Industry Standards and Communication classification system that forms the basis for WordThink). None of them is perfect. All of them are usable.

But I still think it's great that libraries are shaking things up -- and people care enough to notice.

---
LaRue's Views are his own.