Jez Cope and Angela Davies describe a research sharing initiative in the north of England
On 30 June the University of Sheffield Library and Corporate Information and Computing Services (CiCs) proudly launched ORDA (Online Research Data), a digital research data sharing platform for research staff and students.
Powered by Figshare for institutions, ORDA provides a simple means of sharing datasets, software code, presentations, posters, grey literature and much more – a shop window for our research.
The story started much earlier though. The university library working in partnership with researchers and other professional support departments has been developing services around research data for some years. ORDA is part of a suite of services for our research communities including advisory and educative work and managing infrastructure for sharing and safeguarding.
The platform itself quietly opened for business 12 months ago. In that short period our researchers have already deposited 195 items, including our most popular dataset, “United States Commutes and Megaregions data for GIS” by Alasdair Rae and Garrett Nelson; this has already received 44,979 views and 6,570 downloads, and generated rich opportunities for further collaboration.
ORDA complements our existing research repositories - White Rose Research Online (WRRO) for publications and White Rose ETheses Online (WREO) for PhD dissertations; both shared with the Universities of Leeds and York. We are now exploring further interoperability between these systems to link together all outputs of a research project.
Meeting the need for open data
'Openness implies more than disclosure of data. All those engaged with research have a responsibility to ensure the data they gather and generate is properly managed, and made accessible, intelligible, assessable and usable by others unless there are legitimate reasons to the contrary.' — Concordat on Research Data, HEFCE/RCUK/UUK/Wellcome, 2016
Open research data is considered the next phase in the UK’s open science ambitions. For research-intensive universities, providing free access to research outputs is critical to our success. The right system can act as an effective showcase, fostering interest and excitement in research across society in general. This transparency means that not only is our research easier to discover by those with a specialist interest, it is also easier for anyone to build on, leading to greater impact, and easier to scrutinise, leading to greater trust.
Within this context, a number of large funders and publishers expect data to be made widely open, especially when it is publicly funded. An easy to use system makes compliance with these requirements much simpler for our researchers, letting them focus on what really matters - furthering their research - rather than having to navigate complex deposit systems.
Working in partnership across the institution
Building networks to harness skills and experience across the university is a key strategy for us, particularly pertinent when addressing RDM support across a large institution. The Library is a key player, drawing on our expertise in information curation, discoverability, preservation and our long established relationships within the publishing industry. To provide an agile and joined-up service we work in close partnership with colleagues in CiCS, Research and Innovation Services and directly with researchers to ensure we can meet diverse disciplinary needs.
The development of the ORDA portal is evidence of this approach in action. We wanted to quickly develop the out-of-the-box product into a shop window for the University's data that could integrate with our existing systems, reflect our visual identity and, most importantly, provide flexibility to design new functions from a user perspective. Building on the comprehensive figshare API our CiCS developer was able to produce, in a matter of months, the platform in use today combining the best of in-house development with an off-the-shelf product.
Data loss is a real risk to any institution. From October 2017 all published items in ORDA will be automatically archived and preserved in the Library’s ArchiveUS digital preservation service, built on Ex Libris’ Rosetta platform. Not only will this protect the files against loss and corruption, but it will also allow us to assess and mitigate higher-level risks such as file format obsolescence.
Visualisation of data is an important technique for effective communication of research, both with specialists and the wider public. We are developing ways to build our researchers’ data visualisation skills and showcase representations within ORDA.
Until recently, we have focused our educative efforts with research staff and students on more conceptual aspects of RDM. Now, in partnership with the University’s Research Software Engineering group, we are expanding that offer to address the nuts and bolts of day-to-day data management by delivering Data Carpentry training, with two members of Library staff certified as accredited trainers by the Data Carpentry Initiative.
We are also keeping an eye on this rapidly evolving landscape – it’s a challenging environment to work in!
Jez Cope is research data manager at University of Sheffield; Angela Davies, is head of research services and scholarly communication at the University of Sheffield Library