Sharing research data for the benefit of all society

Click on the image above to access videos from the RDA Fourth Plenary Meeting. Image courtesy Inge Angevaare.

Last week, iSGTW attended the Research Data Alliance (RDA) Fourth Plenary Meeting in Amsterdam, the Netherlands. The RDA, which is supported by funding bodies from Australia, Europe, and the US, seeks to build social and technical bridges in order to facilitate better sharing of research data. The organization’s motto is ‘research data sharing without barriers’.

The RDA has grown rapidly over the 18 months for which it has existed and now boasts over 2,400 individual members from over 90 countries. Last week also saw the release of the first outputs by four of the RDA’s working groups (Data Foundation and Terminology, Data Type Registries, PID Information Types, Practical Policy).

Open minds open science

Neelie Kroes, vice-president of the European Commission (EC) responsible for the Digital Agenda, spoke via video at the event. She emphasized the importance of the RDA’s work for the future of knowledge and research, saying “open science depends on open minds” and arguing that “openness is empowering researchers to choose how they communicate their findings.” Kroes also gave her support to the bottom-up, community-driven approach the RDA is taking to tackle the challenges related to data sharing, adding: “The RDA has grown beyond all expectations ... that is a fantastic achievement.”

A series of cartoons were created for the RDA Fourth Plenary Meeting. Click above to see more. Image courtesy Auke Herrema.

Other representatives of the EC at the event included Augusto Burgueño Arjona, head of the e-infrastructure unit within the EC’s Directorate General for Communications Networks, Content, and Technology (DG Connect), and Robert-Jan Smits, head of the EC’s Directorate General for Research and Innovation. Their comments echoed those of Kroes on the importance the EC ascribes to research data sharing. “In recent years, the EU has stepped up its interest in data,” says Smits. “The issue of data is now high on the political agenda ... we’ve called on all EU member states to define policies for open access to data generated by publicly funded research.” However, while Smits acknowledged that policymakers and funding bodies have vital roles to play in helping the RDA achieve its goals, he also highlighted the key responsibilities of researchers themselves: “Scientists need to treat their data as they do their publications: with great care.”

Enabling and encouraging researchers to share

Christine Borgman, professor and presidential chair in information studies at the University of California, Los Angeles, US, gave an excellent keynote speech at last week’s event. Borgman also spoke about her new book, ‘Big Data, Little Data, No Data: Scholarship in the Networked World’, which is due for publication early next year, and discussed at length the complexity of real-world knowledge infrastructures. She argued that to enable the successful sharing of research data, scholars need to have all of the following in place: tools, services, skills, resources, incentives, repositories, governance models, provenance models, and data stewardship workforce.

Following Borgman’s presentation, discussion continued from the RDA Third Plenary Meeting in Dublin, Ireland, on the nature of the incentives used to encourage researchers to share their data in an appropriate and accessible manner. Most participants in this discussion agreed that both positive ‘carrots’ and negative ‘sticks’ have a role to play in bringing about the necessary behavioral and cultural changes. While Borgman agreed with this general consensus view, she also cautioned that overreliance on ‘sticks’ may risk encouraging researchers to follow the letter, but not necessarily the spirit, of the law and thus only do the bare minimum required of them to share their data. Borgman argues for a much improved system of rewards for researchers who share their data, which was an argument also made by Barend Mons in his keynote speech at the event.

Play FAIR with your data

Mons is an associate professor in biosemantics at the Erasmus Medical Centre, University of Rotterdam, and at the Leiden University Medical Centre, both in the Netherlands. Since 2010, he has also served as a scientific director of the Netherlands Bioinformatics Centre. Mons contends that 90% of the challenges associated with data sharing are social, rather than technical, and goes so far as to say that the technical issues have all — at least almost — been solved. During his talk, he cited research published in the journal Current Biology showing that as time passes following the publication of a scientific article, the likelihood of the raw data associated with this article being available for reuse drops dramatically. 

Research data, says Mons, needs to be ‘FAIR’, meaning that it must be ‘findable’, ‘accessible’, ‘interoperable’, and ‘reusable’. He also argues for a major shift in how researchers think about their own data: Instead of datasets being considered as ‘supplementary information’ related to a given academic paper, Mons believes that the datasets researchers produce should themselves — along with nanopublications, each consisting of individual scientific assertions made on the basis of this data — be considered the primary output of research, with narrative papers regarded as ‘supplementary documents’ describing this data.

Throughout his presentation, Mons argued passionately in favour of professional data publishing, pointing out that researchers aren’t expected to publish their own papers, so why should we expect them to do this with data? “If data is as valuable as we’ve all heard, why isn’t there an industry? If, as Neelie Kroes says, data is the new oil, why don’t we have pipelines? Why don’t we have a market for all this?”