The Institutional Web Site and the Institutional Repository: Addressing Challenges of Integration – workshop at IWMW13

(Warning: includes more than a little musing on CRIS vs repositories!)

Last week I attended possibly the very last Institutional Web Managers’ Workshops (IWMW), my very first, at the University of Bath. From Wednesday 26th – Friday 28th June, web developers, commercial software vendors and independent consultants came together for an excellent programme that took in everything from Open Access and Open Education, which I know a little about, to Responsive Web Design, about which I now know a little more than I did. The conference asked “What next?” which, in the current climate (TM) carries a multiplicity of implicit questions for all of us working in HE and digital technology, but rather more explicit and personal for tens of colleagues at UKOLN which has been decimated by cuts and I wish those that have been made redundant all the best for the future, many of whom I have come to know through their work in the repository space.

I had been invited by Brian Kelly to deliver a workshop with UKCoRR colleague Stephanie Taylor (who is also one of those being made redundant from UKOLN) on “The Institutional Web Site and the Institutional Repository: Addressing Challenges of Integration”. Ahead of the workshop we disseminated a brief survey and I am grateful to UKCORR who plied me with case studies and have put together this resource (using the excellent Open Source software from the University of Nottingham xerte online toolkits) which hopefully people will find useful.

The final plenary before the parallel sessions was from Amber Thomas, previously a programme manager at Jisc and now of the University of Warwick, talking about “Turning our attention to supporting research”. As always, Amber’s talk was insightful – in fact, hearing it as I did while thinking about my own imminent workshop, possibly seminal! As I mentioned to Amber immediately afterwards I was struck by the gap in sophistication between what needs to happen and, frankly, what many of us are actually achieving (or, more fairly, what we are able to achieve within the strictures of our organisations and software) . Amber’s slides are embedded below and you can follow the link to a recording of her talk from http://iwmw.ukoln.ac.uk/iwmw2013/video/ – Wed 26 June (afternoon: 13:55 – 15:45) – Supporting Key Institutional Drivers (direct link).

During her talk, Amber referred to the same article as was the starting point for us, “Where are university websites hiding all their research?” (Guardian, 10th January 2013):

“When scouring through university websites in search of their latest developments and projects for the launch of our new research round-up, Research in brief (RIBS), it became increasingly apparent this information was not always easily accessible – to those outside the realm of academia at least.

Amber went on to emphasise that through activities including academic blogging, open lab notebooks, collaborative texts, crowdsourcing, citizen science, open access and public datasets a “more participatory and public scholarly discourse is emerging” and that “institutional webpages are not enough”. I can’t do Amber’s talk justice here and it’s well worth taking the time to view but the central message I came away with was that the life of research is outside the university website and we need to effectively aggregate and make sense of that activity:

Thomas: How do you allow research to live ‘out there’ but pipe it back into the institutional website #iwmw13 #P4

— iwmwlive (@iwmwlive) June 26, 2013

Our workshop, by contrast, was starting from the premise that while institutional webpages might not be enough, they are certainly a start and with the help of UKCoRR I have assembled case studies from the Open University, the University of Glasgow, Leeds Metropolitan University, Northampton University, London School of Economics, University of Sussex and the University of the West of England which you can read more about here.

A question on the initial survey was “Do you know what repository/research management software your institution uses?” and by way of introduction to the workshop I posed the same question. Most – though not everyone – did know which software their institutions  were running and had more or less involvement in reusing data from those systems. I had deliberately avoided the use of the term “CRIS”, assuming that it might not be familiar to academics or web developers – in any case I would argue that terminology can be problematic and that “CRIS” generally refers to specific commercial systems with comparable infrastructure described simply as a Research Management System (RMS) perhaps based on an existing repository appropriately linked with HR and finance systems, for example.

I have suggested previously, though possibly not in public, that the dichotomy between repositories and CRIS is at best unhelpful, at worst specious and the great thing about bringing different people together is that it suddenly throws into relief the functional and semantic obsessions that I think tend to develop in any specialised community…so, UKCoRR, what do you make of this tweet during the workshop from the CMS vendor @TERMINALFOUR:

eprints & pure seem to be the most common repositories used so far #iwmw13 #A4

— TERMINALFOUR (@TERMINALFOUR) June 26, 2013

But EPrints is a repository and PURE is a CRIS I hear you cry! Yes and perhaps I should have been clearer on that point but I’m not sure the lady that tweeted it would really care and nor should she!

Research management at UK institutions increasingly comprise a combination of software ranging from DSpace and EPrints (and their ever more sophisticated ecology of plug ins) to commercial CRIS like PURE and Converis which may be linked to a repository or, increasingly, subsume the functionality of a repository, managing full text and other digital assets and supporting interoperability standards like SWORD and OAI-PMH. The other popular commercial software is Symplectic Elements which shouldn’t perhaps be regarded as a fully fledged CRIS (though can certainly become part of an integrated research management infrastructure) and needs to be linked to a repository in order to manage full text (disclaimer, this is the system we are currently implementing at Leeds Met and with which I am most familiar.)

To my mind, the alleged dichotomy between these various systems fails to take proper account of the functional components of what an integrated research management infrastructure should comprise and even perhaps, to some extent, reflects the entrenched infighting between the Green and Gold Open Access camps (apologies to James Toon for capitalisation!)

It could be broken down further but in essence I think this functionality comprises 4 components:

• Research grant and award management
• Easy / effective workflow to capture institutional research outputs (including data)
• Dissemination/discoverability of those research outputs (and data!) on the open web
• Support for Open Access (both Green and Gold)

I am aware of the potential strengths of the CRIS model, especially in the context of research grant and award management, though I don’t think it effects the general argument that disparate institutional systems can be integrated.

The main difference is arguably one of scope which, in turn, is reflected in different data models, repositories tend to be based on Dublin Core metadata with specialised CRIS software, by comparison, supporting CERIF (Common European Research Project Information Format) to facilitate a more sophisticated model of relationships between data, and for greater flexibility when sharing local data with another institution or funding body. There is a great deal of variety across institutional research management infrastructure however and a specific implementation will depend on individual requirements and the underlying software.

Another key difference tends to be the service which manages the respective systems, with repositories traditionally being located in the library and tending to emphasise archival, preservation and dissemination of research outputs. A CRIS, on the other hand, is likely to be overseen by the university research office with a focus on projects, proposals and other funding information. This distinction, of course, is a generalisation and the boundary is increasingly variable, especially with the renewed emphasis on Open Access resulting from the Finch report and the new RCUK policy that came into force on 1st April 2013 meaning that Open Access is likely to become more relevant for research administrators. Accordingly it is ever more important for research administrators to liaise with academic libraries, in terms of policy, expertise, software and systems.

And if anything, for me, this was the point of iwmw13. I would not typically have the opportunity to attend a conference that is arguably largely outside the narrow remit of my day job. I learnt plenty and made some valuable contacts. Like our data, people tend to be in functional silos, and it’s impossible to know what serendipity might arise until we mix it up a bit.

Aside from technical and system considerations there is an overriding need for effective communication channels amongst the full range of stakeholders at a given institution in order to more effectively integrate the disparate systems of institutional research management infrastructure.

IWMW has been running since 1997. 16 years. I hope that the community finds a way to run another.