Jenny Levine imagined a library with tags. Dave Pattern made one (read more about how). A year and a half later, I had an idea.
I was playing with Director’s Station and thought, “I wonder if I could do something like that, but with search data.” I played around and made (with the assistance of tagcrowd.com) a little cloud of search terms. I wrote about it on Twitter, and today mir_b made a little cloud of authors.
As mir_b notes, it’s a fair amount of work to do something like this, and it’s probably not feasible to do so on a regular basis, but it provides a fascinating glimpse into what patrons are actually entering in our catalog search boxes. You see a lot of titles entered starting with “the” and a lot of authors (in the keyword search terms I used, at least) entered “first name last name.” Depending on your point of view, that goes to show that libraries have a lot of teaching to do to get patrons to enter things the right way, or that catalog vendors have a lot of catching up to do to meet what have, by now, become current search standards. I’ll let you decide.
I had done a search cloud a while back when I was at AADL. It’s still collecting live data–you can see it here. You’ll notice that it needs some work, but the results are intriguing. A mouse-over will show you how many times a term was searched. As the numbers get large things start to flatten out because it really represents the spike on the long tail.
Doh! I had forgotten about yours–I wasn’t trying to exclude anybody, especially since I was fairly sure that this wasn’t exactly an original idea. I was just feeling impressed with myself for figuring out a way to do it without any actual coding skills (thanks, World Wide Web!). Anyway, thanks for reminding me about other efforts of this sort. They are fascinating–the hits, the misses, the different kinds of searches done by different libraries and communities.
Didn’t think you were excluding anyone š It’s just interesting to see how the data projects into the cloud after its been collected for awhile. I wonder if it becomes less or more useful over time, and whether data should be expired at some point. Nice job!