How do we make the case for research data centres?
How do we make the case for research data centres? The RIN-JISC event at the Wellcome Collection on 17 November sought to answer this question by exploring some of the issues raised by the recent, jointly-published report, ‘Data centres: their use, value and impact’. A panel of ﬁve speakers gave short presentations, reﬂecting on their own perspectives and experiences as researchers, funders, policy makers and data centre managers.
One important theme of the evening was about making the hidden more visible. Matthew Woolard, of the UK Data Archive, described the many activities, including selection and validation of data, which are never seen by data centre users but which are absolutely crucial to the centre’s ability to provide a productive service. These activities need to be recognised in any assessment of a data centre’s value, even though they may be completely invisible to researchers who are asked to assess the centre’s usefulness.
At the other end of the research data re-use spectrum, two users of research data identiﬁed some other important ‘hidden’ values of data centres. Graham Stark of Virtual Worlds described the work that his company has done using ESDS data (provided by the UK Data Archive) to explore how the legal aid system could be improved. There is some evidence that the changes made following his work have saved around 10% of the legal aid budget, a sum which he estimated would fund the ESDS for ﬁve years. Similarly, Professor Matthew Collins, of the University of York, used Archaeology Data Service datasets to test a debated hypothesis about the relationship between collagen degradation and temperature in dinosaur bones. He had previously spent several years trying to test this relationship in a laboratory, with much less conclusive results. In both cases, time and money has been saved by reusing data, but the link between those savings and the data centre isn’t always traced and made obvious to funders. The usage case studies developed by the UKDA are a positive step forward in this respect, and the ESRC have responded very well to them.
The other main message that I took away from the evening was that the case for research data centres must be just one part of a wider case for research data. Kevin Ashley, of the Digital Curation Centre, posed an important question when he asked why so many disciplines don’t have data centres. Is it a lack of impetus on the part of funders? Or is it because data centres simply aren’t appropriate? Furthermore, how do data centres meet the needs of highly-specialised researchers in a particular sub-sub-discipline while, at the same time, supporting inter-disciplinary research? Mark Thorley, of the Natural Environment Research Council, talked about the need to separate the storage function from what he called the ‘intellectual component’ of data centres. He proposed a future where data might be stored in institutional repositories or even in the cloud, but where data centres provide an overlay service to pull in such data, to promote it, to provide access to it and to support researchers in creating new datasets which can, in turn, be deposited locally for the ‘data centre of excellence’ to pick up.
The lively question and answer session which followed suggested that two hours were not really enough to answer the event’s titular question. We’re looking forward to opportunities to continue the conversation.