With its vast disciplinary and geographic scope, Horizon 2020 is the ideal tool for making research fully open and interconnected. What’s needed now is vision and boldness, says Mike Galsworthy.
In 17th-century Europe, scholars took some bizarre steps to claim priority while protecting content, such as circulating their discoveries encoded in anagrams. Such practices led to calls for a more open sharing of knowledge and a better system for apportioning credit.
Thus, in 1665, England’s newly formed Royal Society took a bold step towards scientific transparency. The Philosophical Transactions of the Royal Society was the world’s first purely scientific journal. Its model of scientific exchange has stayed in place until now.
As 2015 approaches, however, today’s practices are no longer fit for purpose. Revolution is in the air, with people demanding more transparent and accessible research. The emerging era has been labelled Science 2.0—with the hat-tip to Web 2.0 an acknowledgement of the digital flavour of this new world.
This year, the European Commission ran a consultation, Science 2.0: Science in transition. It covered the topics of open access, open data, peer review, research assessment and alternative metrics to journal impact factors and citations. Where, if at all, the Commission asked, is there “a need for policy intervention”?
Some respondents said Science 2.0 was a grass-roots movement that should develop organically. However, the popularity of this view is generally on the wane.
The agitations of visionary academics for open access to journal articles would have gone nowhere without funders getting on board and demanding open access to the results of the research they pay for. Similarly, despite being given resources and encouragement for years, only some patches of open data have grown organically.
Everyone agrees that open data are the future, but scientists contemplating casting their precious results into the shifting sands of open-data technology have their concerns about credit, the risk of exploitation and possible exposure to confusing legalities.
As was the case when they were encoding their results in anagrams, scientists are reluctant to open up until everyone has to and the terms are clear. We need a top-down framework that responds to bottom-up pressure from the more visionary scientists.
When open data are demanded, aided and credited as an industry standard, scientists can participate with clear guidelines. Horizon 2020 is ideally placed to deliver such a framework and become the global hub for Science 2.0.
A healthy Science 2.0 would look something like this: when a science project that involves data collection is completed, the data are delivered to the funder for independent checking. After a negotiable grace period, this database is placed, with clear metadata, in a repository. The database has a unique, citable ID and is easily found through a global search interface for databases of all disciplines. The database ID is linked to the IDs of the contributing scientists, the project, open-access papers and, where appropriate, software. Peer reviews of papers may also be published with citable IDs.
The whole network of scientific activity, then, would be linked. Scientists’ CVs would comprise publications, databases, other outputs and reviews. Altmetrics would come not only from social media but also from policy impact, with government papers citing their evidence base.
An increasing number of projects would focus on secondary science that reuses data, alongside research that collects new data. Data cartographers would map the rich and ever-changing data landscape. Parts of academic papers would be designed to optimise text mining, meta-analysis and literature-based discovery.
This vision requires infrastructure. The €70-billion Horizon 2020 programme has the disciplinary and geographic scope to enshrine a global shift.
The Commission is already putting the pieces in place. All papers from Horizon 2020 projects must be open access within 6 months of publication. These papers are being mapped by the EU’s OpenAIRE project, which is linked to the particle physics laboratory Cern’s data repository, Zenodo. There is a pilot of open research data under way, a €2.5bn public-private partnership on big data and an EU portal for open government data.
However, the assembly of this jigsaw is happening far too slowly. The results of the data pilot will only emerge this time next year, whereas other funders are already demanding databases from projects as standard. The government data portal could also be used for science, with the Commission buying up databases from Framework 7 and funding scientists to map and reuse their content.
Horizon 2020 and Science 2.0 are a natural match, but the Commission must be bolder about driving infrastructure, data deposition and data reuse if Europe is to set the global standard.
Something to add? Email comment@ResearchResearch.com
—
Mike Galsworthy is a science policy analyst and consultant, and a visiting researcher at the London School of Hygiene and Tropical Medicine.
This article also appeared in Research Europe