Paul Stainthorp Schrödinger's cataloguer

As promised, here’s the first of three blog posts about this week’s trip to the University of Glasgow, sponsored by the RSP for Open Access week 2011, on the theme of Repositories and REF preparation.

Three of us made the trip north: myself, the Library’s Repository Officer, Bev Jones, and the University of Lincoln’s new REF Co-ordinator, Melanie Bullock. We were received and looked after very generously by the team at Glasgow, including Susan Ashworth, Marie Cairney, Morag Greig, Valerie McCutcheon, Robbie Ireland and William Nixon. Thanks to them all for making our visit pleasant as well as useful.

Over the course of a morning, we discussed many aspects of research information management, the REF, and developments to our own institutional repositories (and repositories in general).

I made copious notes, and reading them back I thought it might be useful to identify and list some of the factors that seem to be necessary (or at least desirable) in successfully placing the repository at the heart of an institutional research information/administration system – what makes it possible for the repo to play its part?

Before started: this is my own interpretation, filtered through my brain, notes, and prejudices. It doesn’t necessarily bear any relationship to the real situation at the University of Glasgow. Nor, for that matter, at the University of Lincoln…

Here’s a checklist of what’s making repo-REF integration work for Glasgow:

1. Good data. Because it’s not about your component systems, it’s about your data. Decide what data you have/what data you need and what you need to do with that data – then any tool that matches those requirements is the right system for you. You can always change your systems, but your data are here to stay. Design your system around your data, not the other way around. It’s necessary also to decide early what information to store, then to defend that decision vigorously – best to store the complete record of a publication or a project, NOT the filtered, controlled, ‘for-public-consumption’ version of it.
2. Good relationships: between the library, research/enterprise, ICT services, schools/faculties, etc.: but not only at an operational/service/development level; it’s essential to have joined-up thinking about research data and systems at a management–strategic level. Glasgow seem to have this in spades.
3. An idea of where you’re headed. Glasgow have received JISC funding to do interesting development work across a number of projects (the most notable in 2009/10 being the Enrich project), but haven’t let the funding distort their overall plan – they haven’t lost sight of the overall aim. While the outside world sees separate projects [until our visit I was personally bewildered about how it all fit together…], Glasgow have the bigger picture in mind! It’s the Research and Enterprise Operations Manager‘s job to make sure it all hangs together, working closely with the repository manager and the head of ICT services (see 2).
4. A good development culture. The way Glasgow manage their development depends on the bit of the system in question. They have developers in each bit of the university (and centrally as part of ICT services). It’s important to recognise the [occasionally attractive] danger of rushing off and building something to meet a local need, while at the same time jeopardising the bigger picture for research administration.
5. Taking your users’ needs seriously. Glasgow have a rigorous approach to stakeholder analysis and ‘workload modelling’. Quite often, people working in universities aren’t used to being asked what they actually need a system to do. Genuine user engagement has paid dividends.
6. Mandatory data processes – not just mandated deposit of the final publication. Achieved through diktat of the research strategy committee; the attitude of senior management is “…if it’s not in Enlighten, it doesn’t exist!”. High-level advocacy win! The respository/research information system plays a part in the staff appraisal process and for SMT planning. “Advocacy is beating people with a big carrot.”
7. Internal miniREF-type exercises. Glasgow had a big internal drive, and more than 1,200 staff responded. Suddenly, people became much more interested in the quality of their own data(!) and in the completeness of their publication record. Having information about all known publications  in one place has increased interest in metrics from the repository. Publication “healthcheck” exercises – informing a university-wide publication policy.
8. Useful reporting tools – make it as easy as possible for your users to get data out of the system, via intuitive, meaningful export tools, and in useful formats (Excel output is always good!). Basically, reduce the temptation for people to build their own local silos of data by making it more attractive for people to invest in the repository/institutional research system, safe in the knowledge they can always get the data displayed and/or exported the way they want it.
9. A secret agent in every faculty. Offer training and additional administrative rights to research administrators in academic departments – encourage a culture of devolved/outsourced deposit, advocacy and administration. Allow administrators to ‘impersonate’ academic authors for deposit/editing. Use bibliographic services (e.g. the Web of Knowledge) to send alerts to schools as a trigger to initiate deposit; allow schools to use these alerts/feeds to create records en-masse through filtering. Learn to talking in a language appropriate to different subject areas. Let the schools/faculties add value to the repository!
10. Time and a head start. Glasgow’s overall research information infrastructure is well-established. Probably 90% of the system was in place 2 years ago. While we don’t have a time machine(!) we should at least recognise that proactive, consistent, ongoing development is far better than a reactive approach (“Quick! Build me something to deal with the REF!”). Invest in the repository/research information system now, and you’ll reap the benefits when an information need does arise in future.

That’s it for now – except for some links:

• Enlighten (the University of Glasgow’s research repository)
• Enlighten blog
• Research & Enterprise at the University of Glasgow
• Enrich JISC-funded project (2009/10) – improving the integration of Enlighten with institutional research systems
• The RSP’s blog – with posts from the OA week visits