Tuesday 1 June 2010

Libraries and Games

At the start of the year, the NB column in the Times Literary Supplement ran a "Literary Anniversaries" series "for the benefit of aspiring scribes in search of a subject to tempt a literary editor". Noting the trend for biographies covering connections between apparently unconnected subjects, two notable anniversaries were picked out each week and explored for their combo-blockbuster-biog potential.

There is a real value to the chance juxtaposition of subjects. Collegiate systems such as Cambridge's have long been lauded for encouraging cross-fertilisation. Browsing in a library can be similarly productive - books are pulled from shelves, connections are made, inspiration strikes and great ideas are born (new books sections are particularly fruitful for the browser - the only thing the items have in common being their "newness").

Online library catalogues (or OPACs) are often accused of ironing serendipity out of the system. You approach an OPAC search with a specific need (I want a copy of this book which appears on my reading list/is cited in an article). That need is either fulfilled or not fulfilled. Few OPAC searches are made in expectation that the item will not be held, so the best outcome is that your expectation is met. And the worst is that you are disappointed. But you are seldom surprised or delighted.

One of the topics of conversation at Mashed Libraries Liverpool (as well as the subject of a lightning talk I have since lost my notes to!) was libraries and games. Or, loosely, how to apply techniques from the computer gaming industry in library interfaces. Musing on the subject with Tony Hirst, I was reminded of an interesting and entertaining debate - "Is Google making us Stupid?" - held at Wolfson College last September. At one point conversation slipped from search engines to video games (are they making us evil? are they making our children evil? and stupid?).

Ian Goodyer, Professor of Child and Adolescent Psychology and a member of the panel, pointed out that the really dangerous aspect of computer games is not violent or sexual content but a phenomenon called partial reinforcement, which means doing the same thing again and again and sometimes being rewarded for it - a bit like reverse Russian Roulette. If the action/reward pattern is (or seems) random - as in a fruit machine - the strength of the reinforcement is increased.

What can libraries learn from all this? We certainly don't want our users to become addicted to OPACs (do we?). But a little slice of serendipity and surprise might change the way people use both library catalogues and library collections.

Amazon already provides a kind of semi-serendipity with its "Other people who bought X bought Y" section. So you get what you searched for, and then a little extra which is relevant but might be unexpected. This section intrigues us because sometimes it contains something of real interest, and sometimes it doesn't. Rather than the main search result which is only, rather boringly, what we were looking for. A couple of months ago Dave Patten of Huddersfield gave an Arcadia Seminar on doing this kind of thing with library catalogues, and it's something we're looking to pursue when we get access to CAMSiS course data.

Another angle is to incorporate some kind of complete "randomness" into library searches - along the lines of "hit the search button and sometimes you'll get something really interesting". I'm currently working on the JISC funded CULwidgets project in collaboration with CARET. As well as working on the provision of core services, we said we wanted to do some fun things which illustrated important points about libraries. A while ago I wrote a little web service which provides random results from the UL database. You can try it here:

http://www.lib.cam.ac.uk/api/voyager/newtonRandom.cgi?databases=cambrdgedb

(it's in completely raw XML form at the moment so you'll just see the data - but it could form the basis of something more polished with finding instructions, book covers etc.)

Hit refresh to get a new, random result. Then hit it again. And again.

A little serendipity in searching works particularly well with a collection as huge and diverse as the UL's. Try refreshing the service until you come across something unexpected, intriguing or downright odd - it might not take as long as you think. And it's surprisingly addictive.

I'm not suggesting that a completely random search is likely to be used in anger by students and academics (though if you were writing a column in the TLS you could run an interesting line in "take three random results from the UL and write an essay on them").

But it does illustrate that the way library catalogues are searched could influence the way collections are used, and how a touch of the unexpected could help to spark ideas and open up the wealth and depth of Cambridge's library collections.

No comments: