Open data ‘tougher’ than open access and needs ‘mindset change’

As the open access movement shifts its focus from sharing publications to the “tougher” question of sharing data, universities must shift the focus of their reward systems from publications to other modes of knowledge creation, according to the signatories to a new pact.

University of Sydney deputy vice-chancellor Duncan Ivison said a radical change in mindset about the value of academic contributions to research was needed to achieve an equally radical “opening up” of the scientific method.

“How do we reward people for putting up their data and making it accessible prior to publication – in a sense, making the journal article less important than engagement around the scientific questions?” asked Professor Ivison, who chairs the research committee of Australia’s Group of Eight universities.

“The methodologies, the semantics, the descriptions of the things you’re measuring and observing – how do we make that as valued as the actual publication?”

The influence of academic recognition is highlighted in the Sorbonne declaration on research data rights, recently signed by the Go8, Russell Group, League of European Research Universities and six other research-intensive university networks.

The nine groups vowed to help their members in North America, Europe, Africa, Asia and Australia develop “appropriate recognition” for researchers who made their data “findable, accessible, interoperable and reusable”.

UNSW Sydney environmental scientist Richard Kingsford said open data was “one of the big challenges” for society. “Governments and universities have spent enormous amounts of public money collecting data which sometimes you can’t find,” he said.

“As we get better computational ability to crunch large datasets, significant insights can come out of having access to that data, perhaps never thought of by the people who collected it.”

But he said it would be difficult to resolve the academic recognition issues, notwithstanding initiatives to ensure that people who harvested data were properly cited. “Universities live and breathe on publications,” Professor Kingsford said.

“If you look at universities’ reward structures in terms of careers and promotions, collecting data is not really going to curry much favour on a selection panel.”

The new pact asks governments and funders to develop consistent policies to support the sharing and reuse of research data and to prevent commercial platforms from “locking in” data on their own servers, as well as asking for additional financial support for data management.

Sorbonne University president and Leru chairman Jean Chambaz said data-sharing multiplied its societal and economic impact. “Universities can’t afford to watch from the sidelines as the data revolution occurs,” he said.

But the benefits come at a cost, Professor Chambaz stressed. “It requires infrastructure for data curation and storage, and highly skilled professionals. This cannot be supported by research stakeholders on their own. They need extra funding.”

The statement reflects demand for a specific data-sharing framework beyond existing copyright and intellectual property legislation, Professor Chambaz said, arguing that the nine signatories’ “political weight” would help persuade policymakers that the “extraordinary” potential of research data warranted special attention. “Many declarations on open science or open data are drafted by specialists or activists,” he said.

“This was drafted and signed by nine networks representing more than 160 prominent research universities all around the world. Their message will be heard. We expect more stakeholders to realise that research data is a cross-cutting, political issue – not just a technical one – and endorse the declaration. This should push governments and funders to act.”  

Professor Ivison said the economic benefits of data sharing would also help sway funders. “Enabling the outcomes of investment in research to get out into the world in the most efficient and effective way, helping governments and industry tackle problems – ultimately that’s the best argument.”

He said the declaration was particularly timely in Australia, where discussions about open science had been “more fragmented” than in Europe. “The worst-case scenario is that the world moves ahead in terms of the compliance agenda, and Australia is left behind,” Professor Ivison said.

Dominique Roche, a behavioural ecologist at Carleton University in Ottawa, said the declaration would also help shape government actions in countries with more developed open science policies – like Canada, where the 2018 unveiling of a Digital Research Infrastructure Strategy included up to C$375 million (£218 million) for data management, research software and advanced research computing.

“They’re figuring out how they’re going to invest that money right now,” said Dr Roche, a Marie Curie postdoctoral fellow working on open data. “Now is a good time to participate in consultations and wave that declaration around.”

He said the statement demonstrated the need for extra funding not just at the institutional level, but potentially within laboratories. “Open data’s good for society, but it is an added burden on the researchers. It needs to be supported.”

Professor Kingsford highlighted the risk of perverse consequences from mandated data-sharing arrangements. One was that projects reusing other people’s experimental results became “more competitive” than proposals involving the harvesting of fresh data.

john.ross@timeshighereducation.com