Research Data Management

Queen Mary IT Services, in partnership with colleagues from the Library (me), Records Management Office, academic representatives and senior management have been working together to draft a new, improved, centralised policy on research data collection, management and curation.  The big buzz-phrase of the moment for any research institution funded by EPSRC (Engineering and Physical Sciences Research Council), and indeed any research institution with RCUK (Research Councils UK) or charitable funding that includes a policy on research data is of course... 'Research Data Management'.

No surprise that there are a number of JISC-funded RDM strand projects underway or recently completed then.  See here.

An draft policy for Queen Mary is now available on the DCC website.  With thanks to the Universities of Edinburgh, Oxford, Glasgow and Warwick for inspiration when formulating the policy.  And a personal thanks to IT Services for inviting the Library to be a part of this very important piece of work, and recognising the skills available from Library types to help get this up and running.

But having a policy is only the first step.  Helping researchers to understand their responsibilities, and how to comply with them, indeed providing/enhancing the kind of centralised infrastructure and mechanisms to support these responsibilities is something else entirely, and the focus of further work in the coming months and years.  Just thinking about the potential size of some data makes my head hurt (how much is a terabyte again?).

However, as Kevin G Ashley (DCC) recently pointed out at the Institutional Repository Manager's Workshop at Senate House, University of London (15th June 2012); 'not all data should find a home in your institutional repository' - which is a relief not only to repository manager types like myself, but to technical types who have to provide the networking and infrastructure to support such storage in the long term.  The thing that is really getting the IT people here hot under the collar?  Not the size of the data as you might expect (though this is a significant concern) - it's the length of time that access to that data has to be accessible.  10 years from the last date it was accessed in reality could be 'perpetuity' by another name and that's no small task when your data store grows with every project.

So, to prepare myself for the inevitable talk about research data - I have started to investigate some of the services of interest to the project board, and a few of my own:

DataFlow and DataStage
DSpace - our IR and looking likely to be the institutional data repository platform
DataCite
FigShare


In fact, if it has the word 'data' somewhere in the name - I'm quite interested in it at the moment!