Weeding is controversial. So I’m told, at least.

One of the first things I was asked to do in my new job was to trim down my collection. Space needs were a part of the reason, but more importantly, we were trying to clarify what was on the shelf: do we have what we need? Do we need what we have? Can people find what they need?

That said, weeding always comes with resistance. A library is for books, and look, you have all these books; why would you get rid of anything? Well, there are reasons, happily described elsewhere, but I want to key in on a particular objection: if we weed based on past use, how do we avoid getting rid of the very thing someone might need next month? Just-in-case is a pretty important argument. About 40 percent of the books in my library have never been checked out, according to our best records, and I’m certainly not going to discard 3 out of 5 books just because no one has stumbled on them.

With such a debate on the tip of every tongue, I decided to start with a much easier battle: duplication. Admittedly, there are good historical reasons for most of it: formerly-huge reserve collections, branch libraries consolidating into central collections, or donation-grasping. But now that we have reliable ILL and the Internet, I feel pretty comfortable putting an ultimatum on my collections: we only need more than one copy of something in two cases:

  1. More than one person is likely to want the book at the same time.
  2. The book has serious historical significance, so a copy should be kept for historical purposes, plus one more for circulation.

That’s it. If neither of those things apply, we only need one copy.

But how much of a problem could duplication really be? A really big problem. After a bunch of Excel-based ado, I have some pretty good estimates on duplication in my library’s QBs and QCs (my liaison areas), and it’s not a pretty picture. Out of around 48,000 items, there are about 5,000 twin (or triplet, or dodecatuplet…) items. More than one in ten.

Thence came a project. Since September, my staff and I have been withdrawing dupes left and right. By February when (I really hope) we’ll be done, I think we will have gotten rid of around 2,000 duplicate items.

Then, yeah, I’ll have to do proper getting-rid-of-the-last-copy weeding. But until then, I’ll happily focus on the easy stuff.