Telstar

Mobile Libraries

On Thursday I blogged the presentation of the Erewhon project based at Oxford University. I promised that I would follow up with a post on how I thought the project was relevant to the work TELSTAR is doing – so here goes.

One of the services that Erewhon is looking at delivering is something like being able to answer the question ‘where is the nearest (Oxford University) library to my current location where I can get this book/journal/other?’. This relates very much to questions that we’ve been discussing for TELSTAR and I’ve already blogged about – given a reference, what is the route of access to the resource being referenced.

The TELSTAR project is focussing on answering that question with respect to the online resources offered by the Open University via the Library service. This is clearly a great starting point, and as we deliver more and more resources electronically, it becomes more likely that for any particular resource we can deliver it electronically.

However, what about those things that aren’t delivered as part of the course materials, and we can’t provide electronically via the library? This is where we want to start being able to respond more to a particular users ‘context’ – and where they are is a good place to start. It would be great to be able to answer the question “where is the nearest library to me that I can get this book?”or perhaps more generically “where is the nearest library to me that I can access this resource?”. The latter question is important because you may be able to access electronic resources by going into a library – often licenses enable libraries to offer ‘walk-in’ access to electronic resources to visitors.

There are some services that already try to answer these questions. Worldcat can answer this question for libraries that have their records in Worldcat. Talis have setup a directory of libraries, including location details, which might be used to generate this type of service (as Chris Keene demonstrated at the Mashed Library 2008 event). If we could combine this type of service with the type of Geo-location work Erewhon is doing then we could start to answer these questions. It’s a shame that the SCONUL Access scheme database doesn’t seem to have an API to get details of local libraries.

However, this is focussing on the physical collections – I’m also interested in the electronic. Within TELSTAR we are pushing all our links via our SFX OpenURL resolver – this allows us to work out the most appropriate route of access (termed the ‘appropriate copy’) to an electronic resource. The OpenURL Router could point the way for this type of service – this service allows you to push OpenURLs to a centralised service, which then tries to route the OpenURL to your ‘local’ resolver. I believe it can do this either based on IP range the request is coming from, or via the specification of a parameter which is linked to concepts of Federated Access Management.

Perhaps with some further work, this could use Geolocation information with ‘nearest library with resolver’ information to push the OpenURL to the relevant resolver, and so offer walk-in access. Potentially this could be combined with information about printed collections if it was supported by the local link resolver, and so you could actually answer both ‘do you have this in print’ and ‘do you have this electronically’ in a single query.

Of course one of the problems is that the user is actually in multiple ‘contexts’ at one time. That is to say, if I’m sitting in one of the Oxford libraries, and I’m a student at the Open University, I have access to all of the Oxford library physical collection, the electronic resources they offer walk-in access to, and all the Open University’s electronic resources.  I may even have access to other sets of resources via membership of another library or organisation completely separately. In an ideal world, what the user gets access to is informed by bringing together information from all these contexts, and delivering the resource to the user by whatever route is available.

This is clearly complicated – I sometimes wonder if the solution is for users to have their own OpenURL resolver that can draw on all the information relevant to them, as they are probably the only person who really understands their full context. This seems a rather unlikely scenario to be honest, although this recent blog post on e-books made me wonder if there was a market for an individual  resolvers.