Scholarly publishers are turning to tech to overcome their trust gap with researchers
Academic publishers have proved resilient in the face of technological change, but have yet to overcome their most persistent problem: a lack of trust in their industry. Some are now pointing to technology as a potential solution.
At the APE 2020 open science conference in Berlin this month, publishing representatives talked openly about what is often the elephant in the room.
“I have been surprised at the extent of mistrust in our industry, including our own company,” Elsevier chief executive Kumsal Bayazit told attendees largely from the publishing world on 14 January.
Bayazit, who took the helm at Elsevier in February 2019, highlighted several kinds of negativity among researchers, such as the belief that publishers are failing to add value or hampering the adoption of open access.
Regardless of the extent to which these perceptions are true, “we need to do a better job” of winning people over, Bayazit said.
Part of the solution means adopting the FAIR principles—findable, accessible, interoperable and reusable—for data, thereby facilitating the replication of studies, she said. But another key ingredient is developing technologies such as artificial intelligence.
Publishers could use AI to identify fraud, or use analytics to promote diversity by creating more gender-balanced editorial boards and peer review panels, Bayazit suggested.
They could also enter initiatives such as Credit, which uses metadata to better reflect the specific contributions of researchers to papers on which they are co-authors.
Ed Gerstner, head of editorial strategy at Springer Nature, warned that a failure to address the issue of trust could cost publishers their incomes.
“Our reputation is our business, so we have to go beyond merely the legal; we have to also look to the moral, to the ethical—otherwise we lose our reputation,” he told Research Europe.
Publishers are also looking to anticipate and forestall future problems around trust.
Gerstner, who has been tasked with setting data policies at Springer Nature, said publishers must pay attention to growing technologies such as AI and follow their morals, in addition to the law, if they wish to avoid future crises.
He cautioned against moving too fast on data harvesting in the near term, especially in sensitive research areas where there are questions of consent.
“In the longer term, I think it leads to real problems of public trust, which then leads to calls for legislation, which can then be counterintuitive to the advancement of these technologies.”
European Commission open-access envoy Jean-Claude Burgelman praised efforts to address the trust deficit in publishing, but warned that any such attempts need to include input from researchers and consider the entire research process.
“In the old days publishing an article was the end of the process,” he said. “Whereas in the modern day of open science, it will be a continuous stream of knowledge and that can only work if that stream can be trusted.”
Asked whether publishers’ efforts to solve the problems with AI could help, Burgelman was positive, but with the caveat that efforts must be co-created with scientists.
Publishers’ efforts would “most likely not” succeed if they ignore this, he said.
This article also appeared in Research Europe