I’ve been thinking about serendipitous discovery in library catalogues following discussion in Adrienne Cooper’s cataloguing and classification session at Library Camp. I am also much indebted to Owen Stephens for continued discussion about this over lunch.
If you’re unsure about this subject, I recommend Elizabeth Watson’s masters dissertation on serendipity in LIS (PDF). Read it now!
I think the previous and current generations of library catalogues are not supportive enough – or at all in many cases – of serendipitous discovery while browsing. During my masters interviews I found readers at Senate House Library loved that style of discovery far, far beyond what I expected and were eager to talk about their experiences in our library. No-one mentioned using the catalogue for such a thing.
- Using the classification to browse nearby shelves, the initial catalogue search used as a rough indication of where to start “proper” browsing.
- Chancing upon things in unrelated areas of the library. Just wander in and see what’s there. Moving stock around for our refurbishment actually worked in favour of this!
- Opening a book on the reshelving area. The argument is previous borrowing is some indication of usefulness.
- Journals too: I know of a prof. who prefers print journals because you can flick through a current issue and chance on unexpected things.
Sadly I had nowhere near enough words available in the dissertation to expand on this and it was somewhat off-topic anyway.
Katie Birkwood said, commenting on a post at the “What’s the point?” cataloguing blog:
“I had a thought recently, sort of on the back of mentions of ‘serendipty’ of discovery […] what if all or some of this perceived ‘serendipty’ is nothing of the kind, but is instead the result of careful classification?”
I think that’s true, and moreover I think the major benefit of classifying items rather than just bunging them on the shelf and numbering them. I think we lose something in relegating a lot of stock to closed stores as this style of discovery diminishes too.
It’s an unsolved problem to support serendipitous discovery online. In theory an online catalogue should be good at this because it isn’t really limited by space and includes the possibility of hyperlinking things together…
There have been attempts to replicate the experience of print items. For example a “shelf browse” type of view – but that’s missing the obvious: you can’t pick the thing up and flip through it, plus you limit yourself to the view of the collection imposed by your classification, as good as it is, you’re not encouraging random connections.
I would love to see a library catalogue with some awareness of what those funny numbers and letters in classmarks actually mean. Offering a browse view of classmarks in a list is one thing, and useful for power users, but what a step forward if the catalogue had some understand of what say 303.48330904 means and how that could be related to other classes.
An approach I find very interesting is recommender systems. This was discussed in Helen Harrop’s and Dave Pattern’s session on the same at Library Camp where Dave linked them to serendipity – a recommender being one of the ways of reproducing the ‘joy’ of accidental discovery online. Helen argued there is a wealth of data in library management systems that could be used to create these connections, the basic idea being “People who borrowed X also borrowed Y” or similar.
I remember being very impressed with the LibraryThing for Libraries recommender when I first saw it make a connection between Charles Mackay’s Extraordinary popular delusions and the madness of crowds and a book on Dutch tulips, the former having a chapter unknown to me about the tulip craze. Years on I just take that sort of thing for granted on LibraryThing, but relatively few academic libraries have started using recommenders in earnest.
This is one reason why I’ve made ‘patron reading history’ opt-out in our catalogue for the academic year just started.
Watson, E.A. (2008) Going fishing: serendipity in library and information science. Unpublished MS dissertation. University of North Carolina at Chapel Hill (Online). Available at: http://etd.ils.unc.edu/dspace/bitstream/1901/487/1/elizabethwatson.pdf