Because good research needs good data

Jisc's Research Data Spring - how you can get involved

Kevin Ashley | 04 December 2014

The Research Data Spring is a new programme from Jisc which is intended to support the development of new tools and services for research data management in UK universities. This news item describes how you can get involved. Don't delay - the first phase ends on 12 January 2015.

It's different in a number of ways from the funding programmes of JISC in its previous incarnation. Two changes are worth highlighting up-front. There's an opportunity for the next few weeks for anyone to help prioritise ideas that have been submitted, so you can get involved even if you don't have a project to propose yourself. Another striking difference is the iterative nature of the review process, with projects being successively weeded as they move through a series of phases, including pilot implementation, with funding being injected at each stage.

Before examining those points in more detail, let's look at the areas that Jisc want proposals to cover. There are five:

  1. Research data deposit and sharing protocols and tools development to meet a wide set of requirements in a range of disciplines and/or data curatorial requirements
  2. Data creation, deposit and re-use by discipline – tackling the improvement and take-up of commonly used or innovative tools and improving the researchers experience thereby enhancing research data management and use (this will support open research practice and methods where appropriate) 
  3. Research data systems integration and interoperability, in particular to focus on solutions to improve the interfaces and seamless working between and across systems for research and its management
  4. Research data analytics - develop and test ways to use big data analytical methods for the benefit of research or to develop ways to analyse research data activity and associated metrics
  5. Shared services for research data that Jisc or others could deliver to improve research data management and use in the UK university sector

Those descriptions are taken directly from Jisc's site. That site also says that area 5 - shared services - should be considered for all of the other four areas.

Areas 1 and 2 can be considered as looking at data management from either end, area 1 being about making it easier to capture research data (if you are a data custodian) or to hand it over (if you are a data creator.)  Area 2 reminds us why we curate data - it's all about enabling reuse. As worded, this area is looking for work on existing tools and services rather than the development of new ones, and it is expecting there to be a greater discipline-specific focus. Area 3 is about joining up services within and outside the institution to make workflows smoother and removing the all-too-familiar experience of having to provide the same information to yet another system. Linking research information management systems, publication repositories, data repositories, working research environments and external services are just some of the things that would be appropriate here. Area 4 really covers two quite different topics - the use of big data analytical methods using our research data, and the analysis of activity data about our research data. Area 5 is both a catch-all and an umbrella. It allows a good case for a shared service proposal to be put forward that doesn't fit into one of the other areas and it encourages us to think of all the areas in the context of shared services.

There has been less work on shared tools or services for research data than one might expect after so many years of successful domain research data centres existing. Although there are many examples of excellent subject data centres, along with evidence that they provide remarkable value, there are very few examples of the reuse of tools and services between such domain data centres. The same can be said for data curation within institutions - reuse and shared services are the praiseworthy exceptions rather than the rule. The picture is changing, with provision by national organisations such as the DCC, ANDS, DANS and CSC of some shared services, more from projects such as EUDAT and DataONE, and better support for data in repository systems such as Eprints, DSpace and Fedora.

Enough of the what - back to the who and when. The process is show diagramatically in the picture below ((c) Jisc & Liveworks, licenced CC-BY-NC-ND.) Step 1 is to register on the Ideascale site - you need to do this to comment or vote on other people's proposals or to put one in yourself. At the bottom of that first page are the ideas already put forward. You can vote them up or down, comment on them, and suggest that some be joined up or offer to join in yourself. This step closes on January 12th, now only a few weeks away. There's also a workshop at IDCC on February 12th where ideas can be explored further and partners sought. Up to 30 ideas will be invited to participate in a sandpit workshop exercise on 26 & 27 February - attendance will be free. They'll pitch their ideas to judges and those that succeed will receive their first funding tranche of £5,000 to £20,000. They then have 3 months to work up their ideas further with that funding before returning to another workshop and another selection process in July 2015.  Another tranche of funding, this time of £5,000 to £40,000, is then on offer. This supports another 5 months of development before a third workshop and selection process in December 2015. Only the very best then receive a further £5,000 to £60,000 to finish development of their work before the final showcase in July 2016.

The initial phase - up to and including the first workshop - is intended to be organic and collaborative. Jisc isn't looking for fully-formed teams and fully-costed proposals and project plans, but ideas that promise benefit and are implementable. The Ideascale platform and accompanying workshops are intended to allow you to identify partners and hone your idea. Collaboration between new partners is explicitly encouraged. That means, amongst other things, not waiting until 2 hours before the deadline to submit your idea. We have some ideas here at the DCC and we'll be looking to work with others to develop and improve theirs.

That timeline is represented diagramatically here.

RDS timeline

Research Data Spring is part of a broader set of activity which Jisc calls the 'Research At Risk' co-design challenge. There's a hashtag - #dataspring - that I expect you'll see more of in the coming months. If you have any questions that aren't answered here, they may well be covered in the FAQ.

Finally, an observation and a declaration. The FAQ uses a definition of research data which appears at first sight to exclude digital research material from many areas of the Arts and Humanities. I don't believe that exclusion is intentional, so please feel free to submit ideas that would benefit those disciplines. I should declare that I'm likely to be involved with at least some steps in the judging process; I'm looking forward to hearing about the ideas you will be putting forward.