Laurabelle's Blog

The other day, a KCLS patron emailed in and requested Amazon-like features for iPac (the web catalog), such as people who checked out this author’s work also checked out these other authors. Now, of course we can’t do that right now because it’s not one of the features available in iPac, but the question got me thinking about all the reasons why one wouldn’t want to do something like that in a library system. All I said to the patron at first was that it was contrary to KCLS’ privacy policy. He wrote back, asking if it were possible to remove the patron data, so that privacy wouldn’t be violated. Since he asked, and since he referred to the so-called Patriot act, I decided it was probably safe to talk to him honestly.

There are some technological barriers to this kind of feature. KCLS licenses its OPAC software from other companies; this means that it doesn’t have to develop, maintain, or support the software itself, but it also means that there’s less flexibility and autonomy. We are dependent on what Dynix decides to develop. (I have issues with Dynix, both as a company and as a product, but of course it’s really hard to move everything to another software system, once the data is there and the staff trained etc. Thus administrative inertia and the ability of Dynix to say Screw you, haha because it’s really hard for us to leave once they’ve got us. But I digress.)

There are a couple of ways I could think of to implement this kind of functionality. One is fairly simple, complete, and horribly wrong. The other isn’t as wrong but doesn’t actually provide the functionality it promises. Let me elaborate.

The first solution is to save patron circulation data and use that to generate book suggestions. This would be simple to implement and would be able to link items that a single person checked out over time. However, it’s also gross, horrendous, and horribly wrong, because there’s absolutely no way to sever the patron from the circulation data. Without the patron connected to the bibliographic data, there is no patrons who, which is the whole point of the feature. Not only is all that against KCLS’ privacy policy (KCLS’ system connects patrons to items only while the book is checked out and up to 4 days after it’s checked in, or sooner if someone else checks it out before that time), but the mere notion gives any librarian nightmares.

The other possibility would be to maintain data only for individual checkout sessions. This would require keeping another (kind of redundant) file of data and would link only items checked out in the same checkout session. This implementation would sever patron information and might even have a greater likelihood of linking only relevant works, but it might also miss out on a lot.

The other problem with implementing either of these options, whether there actually is a privacy violation or not, is that it looks like it. Patrons don’t know how this stuff is implemented. There are a lot of people who are concerned about personal circulation data, or about the idea that it could be subpoenaed, and we don’t want to give the impression that we’re violating privacy. We also don’t want to give law enforcement professionals the impression that we have data that could be subpoenaed. (But you have it in your catalog…)

I am reminded of the power of information, and this attempt by our government to control people’s access to it. (Filtering too… bah. But that’s another rant entirely.) Knowledge is power. The mission of the library is empowering people through information. No child left behind, my foot. What about the adults? What about self-learning? Mutter grumble ultra-conservative pigs.