This is the last session of the day by Dave Pattern (http://www.daveyp.com/blog/), the Library Systems Manager at the University of Huddersfield. Dave has been very active in doing cool stuff with their Horizon OPAC.
The first question – Does your OPAC suck – this was something that surfaced in quite a few library weblogs a little while ago. This was summed up by a posting that said ‘my opac needs more cowbell’ – you have to watch http://www.youtube.com/watch?v=EVbAuMr5eac to understand apparently.
The title of the talk refers to a quote from Roy Tennant, and one from Robert Heinlein:
"you can put lipstick on a pig, but it’s still very much a pig." (http://www.libraryjournal.com/article/CA516027.html)
"Never try to teach a pig to sing; it wastes your time and it annoys the pig." (http://thinkexist.com/quotation/never_try_to_teach_a_pig_to_sing-it_wastes_your/218581.html)
So, in preparation for a talk about the OPAC, Dave decided to do an online survey – he was expecting a handful of responses, but got over 700. This asked questions like ‘how happy are you with your OPAC’ – the questions are here http://www.daveyp.com/blog/stuff/opac.html and there is some initial analysis of the results here http://www.daveyp.com/blog/index.php/archives/239/
As well as this, Dave started looking at how people were using the OPAC at Huddersfield, and looked at things that he could add. One of the issues they noticed was that a large percentage of searches end with no results, and students would give up. They had already added a spell check, which definitely helped, but this dealt with zero results where they’d misspelt, not when they had used (for example) a phrase that was too specific. So they added something that looks up the search term the user has entered against ‘answers.com’ and pulls out related links from the Answers.com webpage – they call the ‘serendipity’, as they acknowledge they have no control over the terms returned.
Dave also found that the library system had been collecting data for several years about library usage, but they hadn’t been used. They started to mine the data for ‘people who borrowed this also borrowed’.
They introduced an ability to ‘rate’ a book (star rating) to see if anyone would use it. And one day, someone did. They then added the ability to comment – it hasn’t been used very much, but more by the academics than the students – they can do so anonymously (story about an academic leaving an unflattering anonymous review of a colleagues book!)
Dave started to use the xISBN service from OCLC and thingISBN from LibraryThing to link together all editions of a book owned by the library.
Used the MetaLib ‘saved search’ which it will do on a regular basis. Also added RSS feeds for results – not clear if this was using MetaLib functionality or directly from their OPAC – must ask him, as we have MetaLib but not Horizon.
One of the points that Dave makes is that they did these things in a completely speculative way – they were just trying things out and seeing if anyone used it – I really think this kind of approach (a bit like Google Labs) a great idea. They’ve found the most popular service is the spellcheck. The ‘people who borrowed this’ service wasn’t popular initially, but has increased in popularity (a 300-400% increase since they first introduced it).
However, despite all these improvements, there is a worry that all we are doing in Roy Tennant’s words are ‘putting lipstick on a pig’.
One of the issues that Dave encountered was resistance from other librarians – so had to introduce staff to the ideas, and sell the ideas. Overcome the fear that ‘sudden changes’ might confuse the users – but the changes were small, and users are used to websites making these subtle changes overtime.
- Dave suggests we need to do the following:
- Encourage suggestions from library staff
- Include users in decision making process
- Encourage play and experimentation
- Don’t be afraid to make mistakes
- Build crappy prototypes – a prototype is worth 1000 words
Dave is showing some of the other ideas that haven’t yet reached the light of day – a search that presents books by colour (based on the covers – possibly from Amazon?); search visualisations that show what people are searching for at the moment on the catalogue as a tag cloud; cover shots of the last 50 books borrowed from the library – some of these may not be useful ideas, but lets experiment.
Dave is mentioning the work at Ann Arbor Library; the Endeca powered catalogue at NCSU; Librarything for Libraries; Scriblio; Talis Platform; Ex Libris Primo; Innovative Interfaces Encore
Dave’s view of OPAC 2.0:
- spell checking (did you mean)
- relevancy
- imrpove serendipity
- expose hidden links between items
- APIs and Web Services to expose data
…lots more stuff – you can find quite a few slideshows at http://www.slideshare.net/daveyp/ which has these slides
In the responses to Dave’s survey, the majority of responses came from the US. When looking at the UK responses there is a gap, although the general trends are in the same direction. The biggest gap was on ‘faceted browsing’ which seems to big in the US, but not so much interest in the UK.