I’ve posted a bit about the long tail before but figure I probably should again after a recent post on the ALA TechSource. In terms of libraries the long tail is probably easiest to understand as the library has a few books that get the most circulation while the majority of the holdings get little circulation. For most book publishers the high selling would be the new books, new to paperback, etc while older books barely sold and of course, out of print not at all. To get a better understanding I recommend reading the following:

This is one of those areas where I think it is worthwhile to think of libraries as businesses. It costs money to house books. In order to save costs many libraries become part of consortia to spread holdings out and many weed through their collection for items that can be removed to make room for new ones. The thing to note is when your removing low circ items your affecting the long tail and the question to ask is could this item be a more in demand item if people could find it. Many businesses go the route of just carrying the most popular items, but how can libraries take advantage of their already large collection. In other words, making that collection earn it’s keep.

Again I recommend reading both of Lorcan’s posts above as he puts things better than I can. There are a couple ideas that I think should be stressed though in the case of libraries. These are also important when thinking about “next generation catalogs”. I think the ALA post got it right when they stressed the ability to find things easily.

One of the main problems I have with library catalogs is that it is difficult to find things unless you know exactly what your looking for. There’s many reasons for this that I won’t get into, but right now many libraries just have an inventory system. One of the ways NetFlix and Amazon make their large holdings worth the cost is suggestions. The majority of the movies I’ve rented from NetFlix have been suggestions and it’s one of the things that keeps me coming back. I have not used the feature at Amazon to the same extent though it has come in handy quite a bit when researching products. The simple things like “people who read this also like this” are well worth the effort to add. Many libraries already have circ statistics but they aren’t being put to full use.

For research libraries this would likely be a little more difficult but there is still plenty of room for improvement. Better metadata and linking between texts would help. Unfortunately much of this would require more digitization, which leads me to:

O’reilly has probably taken advantage of this the most (with Safari), though Google Book Search has the same premise. Amazon is also beginning to take advantage of it. Many books are hard to find because of limited metadata. There’s been many times I’ve missed books in the catalog because table of contents weren’t included or the title was not descriptive. In many cases you really need some sort of full text or at least a term index for the work in order to make it findable and useful. This seems especially important in science related works but is probably important elsewhere. I think it will be really interesting to see how some of the Google Libraries put their digitized works to use.

The ideal of course is offering instant online access but when not possible this doesn’t have to be the case. I think Amazon will be one to watch as they push sales through full-text while not actually providing the full text online in all cases. Regardless this is something worth watching and it will be interesting to see if the possibility of sharing indexes comes about.