US orders publicly funded research be made free to access immediately

A Biden administration order that published results of federally funded research should be made immediately and freely available to readers worldwide has been hailed as a crowning victory for advocates of open science.

Government agencies will have until no later than 2025 to make its grant recipients comply with the administration’s order, which includes companion provisions that will also make related data openly accessible.

“The American people fund tens of billions of dollars of cutting-edge research annually,” Alondra Nelson, the director of the White House Office of Science and Technology Policy, said in announcing the decision. “There should be no delay or barrier between the American public and the returns on their investments in research.”

On one level, the requirement extends decades of efforts across multiple administrations of both parties in the US, and by governments abroad, to prevent journal paywalls and subscription fees hindering basic scientific cooperation on fundamental pursuits of human health, economic prosperity and societal well-being.

Yet the sudden extent and finality of the US decision – covering a powerful industry with estimated annual revenues of $20 billion (£17 billion) to $30 billion – shocked both sides of the debate.

The imposition, complained the industry’s main lobby group, the Association of American Publishers, “comes without formal, meaningful consultation or public input during this administration on a decision that will have sweeping ramifications, including serious economic impact”.

Open access campaigners, in turn, expressed unmitigated delight. “It is huge,” said Brian Nosek, co-founder and director of the Center for Open Science, a decade-old venture to push sharing and integrity in scientific research. It is “an enormous leap forward”, said Heather Joseph, executive director of the Scholarly Publishing and Academic Resources Coalition, which has spent two decades working on the problem.

It brings academia very close to ending the long fight over publisher paywalls, said another leading combatant, Jeff MacKie-Mason, head librarian at the University of California at Berkeley and a top negotiator in recent years for pushing major publishers to accept subscription-fee arrangements.

“This, combined with Plan S in Europe,” Professor MacKie-Mason said, referring to the continent-wide campaign begun in 2018, “really says the game is going to be over.”

“It's not over right today, but it’s going to be over, and they’re going to have to figure this out pretty darn fast,” he said of the world’s academic publishers.

The US spends about $180 billion a year on scientific research and development, with a little over half of that amount coming from outside the Pentagon, which could exempt large chunks of its work.

The academic publishing industry – including many journals produced by small scholarly societies – has long insisted that it could not survive without making some profit from the process of compiling, editing and organising peer-reviewed research papers.

Biden administration officials acknowledged that some of those fears are legitimate and will not necessarily be solved overnight. But the three-year implementation window – as well as the plan’s explicit authorisation of scientists spending some of their federal research money on author-pays publishing models – should help, they said.

The government is committed to helping “publishers and scholarly societies of all sizes with the change”, OSTP officials said.

Administration officials also rejected the industry's complaints about lack of consultation, saying that they conferred with a range of affected parties. Yet, the Biden team said, persistent cases of discrimination and structural inequalities in science, combined with the data-sharing problems made clear by the fight against Covid, gave the changes a greater sense of urgency.

With both Covid and the emerging monkeypox outbreak, Ms Joseph said, government officials found themselves publicly pleading with publishers to make all available research and data immediately available to all scientists. “How many times do you have to do that before you kind of go, ‘We need it for everything,’” she said. “To me, it was a long time in learning that lesson.”

The move comes nearly a decade after the Obama administration took its own major step in the direction of open access by requiring that the published results of federally funded research be made openly available within 12 months of publication. The Obama rule also exempted federal agencies that provide less than $100 million in annual research funding.

As with that rule, the new Biden version will allow journals to keep their finished copies of an article behind a paywall, as long as some final peer-reviewed version is publicly posted somewhere.

The existing 12-month policy means that the larger federal agencies already have the necessary processes in place, raising expectations that much of the shift to the immediate public posting of government-sponsored science could occur far faster than the 2025 deadline, Professor MacKie-Mason said.

While the Association of American Publishers said that it was especially concerned for the fate of smaller journals, some of the world’s larger publishers – including Elsevier, Springer Nature and the American Association for the Advancement of Science – offered statements more clearly indicating a willingness to work with the government on implementation.

"By opening up all outputs of research, underlying data as well as published articles, we all benefit from a faster and more trusted, effective, and equitable research system, able to deliver solutions to global challenges,” said Carrie Webster, Springer Nature's vice-president for open access.

That reflects an evolving shift, Professor MacKie-Mason said, after the UC system and other major US universities made increasingly clear in recent years that they will not continue to sign contracts with journals that put any limits on the global sharing of taxpayer-financed scientific discovery.

“It used to be, we told the publishers we want to talk about a transformative agreement, and we had to sort of drag them to the table kicking and screaming – in Elsevier’s case, definitely kicking and screaming,” he said. “Now, it's flipped, from us pressing and pushing the publishers, to the publishers pushing us” to join in, he said.

The new data-sharing requirement, meanwhile, will put a heavy burden for change on universities and their scientists, Professor MacKie-Mason said. Challenges there include defining how much data will be included, what machine-readable formats will be acceptable, and how researchers can be persuaded to abandon old habits.

“Data publishing is not something that is very standardised or uniform, and a lot of people have no experience with it,” Professor MacKie-Mason said. “It’s a hard slog on our side to make this happen, and make it happen effectively.”

The Center for Open Science specialises in creating online platforms for data-sharing, and yet Professor Nosek acknowledged that it remains “far from a perfected practice”.

“Millions of data files are shared yearly,” he said, “and they vary in quality and usability based on training, complexity and motivation of the sharing researchers.” To comply with the Biden plan, he said, academia will need ongoing improvements in infrastructure, training and community norms.

paul.basken@timeshighereducation.com