Medical libraries

February 20, 2007

Back to business. I’ve been meaning to post this for some time, but I’ve simply been unable to do so (I’ve been willing, it just wasn’t possible). Several weeks ago, I spoke at the University Health Network in Toronto on the rather general topic of “Emerging Trends in Libraries”. Quite remarkably, the organizers of a strategic planning exercise had found me through my web page at the U of Guelph … this seemed extremely unlikely, as I have a fair amount of difficulty finding that page myself. However, I made my way down there, and found the entire experience to be fascinating.

I’ve always loved hospitals (I’m not implying that I want to be a patient, I just love hospitals). I love hospitals in just about the same way that I love airports. Neither of them are all that great if you’re there for the standard reasons. If you’re sick, that’s no good, if you’re visiting someone sick that’s not so good either. If you’re flying, you’re usually tired and/or bored and somewhat stressed about getting everything worked out. I love being at either of these places, though, when I’m not a customer. I love the busy-ness, and the huge number of people who seem to have very important yet mysterious things to do (I always try to imagine what they’re up to). Needless to say, both places are also full of constant drama … people coming and going, being reunited, saying goodbye … hugging, crying, kissing, living out some of the most intense times of their lives (not all of them, just a bunch of them).

So, I got to be in the crazy environment of a huge downtown hospital complex without having to worry about any of the usual hospital stuff. I was able to observe everything, without having to go through the hard parts. In the middle of all of this, I spoke to hospital people about information. Their environment has very clear qualities about it … doctors ask very important, time-sensitive questions and librarians perform research and get them the information that they need. This is very much an environment of demand and response, and while most librarians would suggest that they are in the same type of environment, the intensity here was fairly unique. I very much appreciated what these people do and their expertise is appreciated by the medical staff.

There was one rather huge thing that struck me, however. Each transaction was a discrete event. The question was answered and then everyone moved on to the next question. It was as though there was no institutional memory, and anything that might have been learned by answering the one question, would only be known to the two parties involved. My talk was on web 2.0 with a particular focus on social navigation … in particular indirect social navigation. This is a type of social navigation in which those receiving advice get it from a community of people and that community doesn’t necessarily know that they are giving advice. Google is a fine example of this. Google ranks articles based on how web pages link together. By linking to other pages, people generally suggest that those other pages are somehow significant and worth linking to. By looking at all of the link patterns out there, google analyzes all of these suggestions of value, and decides which pages are important by how many people point to them. By doing this, the collective value judgements of the community are used to suggest to you which pages are most important to your search. This is going on all over … Amazon tells you which books you might like based on what others “like you” have bought, Netflicks or the equivalent tells you what others like you like in the way of movies.

It seemed to me that doctors could benefit from the same type of thing (in fact, almost anyone can). If a database was created of the questions and answers generated daily in the hospital, a recommender system sould be built. This doesn’t mean that the system would answer questions for people based on previous answers. This means that if a document was found in an answer to a specific query, that one would also be pointed to documents that tend to be used when that document is used (get it?). This isn’t artificial intelligence (and it doesn’t take away anyone’s job), but it begins to provide linking not only between people, but between documents as well. Interestingly, this type of practice is far from new to academics. They have always looked at citations to detect patterns of document use. The best way to research a topic is to find someone who has done that topic, or one like it, before and to see what they read. By then going to those documents and seeing what others have read one eventually builds up a network of use (and you can feel comfortable that you’re on the right track when you start seeing the same documents repeatedly). This is nothing new … it’s just a lot easier now.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: