The Long Tail and Libraries

I just finished reading The Long Tail by Chris Anderson. He writes about how business has been transformed by the Internet and the availability of choices in a business not bound by bricks-and-mortar.

The three main forces behind the Long Tail are:

1. "Democratize the tools of production"--digital cameras, video recorders enlarge the field of content producers beyond the specialist or professional to the passionate amateur.


2. "Democratize the tools of distribution"--this is the role played by aggregators such as Amazon or Netflix


3. "Connect supply and demand" --this is done by filters such as recommendations in Netflix, customer reviews and list on Amazon. Essentially, a filter orders information so a user can find items.

A library would seem to be both an aggregator and use filters via the catalogs. Most libraries allow searching for subjects as well as authors and titles and one could certainly argue that the filters used in libraries could be more robust and some libraries are attempting that by adding patron reviews or linking to patron lists or even allowing tagging by patrons to help identify items using labels that are most meaningful to them.

In my former career in publications, we found that categorizing was, no surprise, the most difficult part of determining how to organize information and we used to have to mediate disagreements between the professional developers who would tell us the proper way something should be organized versus our need to make sure the readers would be able to find the information and the two were often in conflict. We tended to go with an information organization that the readers could understand (which tended to be task-based) because if the they could not find what they were looking for our work was not very successful.

While reading the book, although fascinated by the 80/20 rule and was reminded that we discussed the 80/20 rule in our collection development class--20% of the items circulate 80% of the time. We talked about the importance of user needs surveys to improve our collection practices and our services in general. I have been thinking a lot about user needs surveys and it is clear that while they are very important they are also extremely difficult to do well and I wonder whether we can get better user needs information from examining the circulation data of our own libraries and other libraries using the library's ILS data. Do our library systems collect data about searches that patrons make that are not successful? We could use that data to determine what items they are searching for that they cannot find and determine whether they cannot find them because the items are not available in our collections or whether there is a problem with their search terms or our categorization? That would go a long way to helping us refine our choices.

Anderson argues for the importance of the filter. Businesses such as Netflix are successful because they are able to distribute older films and less-viewed items such as documentaries. In fact, it is a large part of their success because they are able to distribute many of the older and lower-priced DVD options. Most importantly, people are drawn to Netflix because of the older and more obscure films. They simply have titles not found elsewhere and they make it easy to find them via their recommendations and terrific search capabilities. Right now, many libraries do not have the appropriate terms to search for CDs or DVDs; they do not have fields particular to those media, a director, or an actor or performer might get listed as author since that was the closest field available. I can understand that this might be necessary for older systems and older entries but I was quite shocked in a recent class, that many classmates seemed to think that listing a director or performer as author was just fine. Our patrons are getting used to the search capabilities of a Netflix or iTunes and we owe it to them to produce the best search experience that we can.