guia do novo coronavirus
Imprimir Republish

COVER

Science communication without barriers

European Commission and research funding agencies seek allies for global open-access initiative

Marcelos Cipis

In 2016, 2.3 million scientific articles were published in reputable journals indexed in the Scopus database. Access to this vast quantity of information, which allows researchers to keep up to date with the latest discoveries and innovations in their fields and to collaborate with one another to advance human knowledge, could soon be expanded and facilitated thanks to an initiative launched in September by research funding agencies from 14 countries, mostly in Europe. The logic behind what is known as Plan S is simple: if a study is in any way funded by public money—as is the case for the overwhelming majority of basic science research—its results should be published in a scientific journal or platform that is available to everyone at no cost. Through such an approach, society at large has greater access to the innovative knowledge that it helped pay for.

The initiative will come into force in 2020 and will be adopted by the European Commission through its new €100-billion multilateral research funding program Horizon Europe, as well as by a group of agencies from 14 countries, including philanthropic institutions such as the Wellcome Trust in the UK and the Bill and Melinda Gates Foundation in the USA. The institutions backing the plan together fund 3.5% of global scientific output, a modest share to attempt such a radical transformation. They are thus seeking new partners to take the plan global. They have already made a strong ally: the Chinese Academy of Sciences. “Since the plan was announced, several funding bodies and nations have expressed interest and support for it, recognizing the need to establish immediate open access. We are in a moment of global development,” Dutchman Robert-Jan Smits, senior advisor for open access at the European Commission, told Pesquisa FAPESP.

It is easy to argue that knowledge generated with the help of public money should be quickly and openly shared. But putting this idea into practice without weakening the crucial role journals play in scientific development is significantly more complex. Every paper needs to be analyzed and peer-reviewed by publishers and researchers who are familiar with the topic. These reviewers propose improvements and recommend whether or not a paper should be published. This procedure is trusted by scientists, institutions, and funding agencies to screen the quality of research results. The issue is that reviewing and publishing an article, whether in print or online, costs money. One of the long-established means of funding this process is to charge for subscriptions to journals and articles, a commercial model that has shaped science communication worldwide and led to the creation of huge media conglomerates.

In recent years, global scientific output has grown exponentially, and the advent of the internet has made the dissemination of research considerably faster and easier. The emergence of open-access journals—which charge authors a fee to make articles freely available online and are sometimes subsidized by funding agencies or scientific societies—has increased competition and put traditional publishers under pressure to freely share their content. Some have created their own open-access titles, while others have allowed authors to publish their articles online after an “embargo” period of subscription-only access—usually between six months and a year. Many are backing a hybrid model, where readers are charged a subscription fee, but authors can pay to have their papers published without restriction on the journal’s website. The number of hybrid journals has increased from 2,000 in 2009 to almost 10,000 in 2016, according to a study by Finnish researcher Bo-Christer Björk, from the Hanken School of Economics in Helsinki, Finland. Most of these publications belong to major publishers such as Elsevier, Wiley, and Springer-Nature. Plan S hopes to change that. The aim is to encourage researchers to use open-access journals, as well as to promote the creation of such journals in disciplines where few exist—chemistry is one example. There will also be criteria for defining which open-access journals and platforms will be able to publish articles, to try to steer researchers away from predatory journals that are more concerned with charging fees than conducting rigorous peer reviews.

Hybrid journals would not be usable under Plan S, because its creators consider that collecting a subscription fee while also charging authors to publish their work online is abusive. Publishers disagree. “There will be profound implications for the economics of science and the dynamics of scientific research dissemination,” Magdalena Skipper, editor in chief of the journal Nature, told Pesquisa FAPESP in December (see Pesquisa FAPESP issue no. 274). “It doesn’t just mean that all journals will have to think of a new business model—every author will have to pay or find funding to have their work published.”

Plan S is based on 10 core principles. The most important is that articles should be published without copyright restriction, under a license that permits free sharing, provided the source is cited. In November, the guidelines on how to implement Plan S were opened for public consultation. The document does offer some leeway, suggesting that the hybrid model should not immediately be brought to an abrupt end. Such journals could be used transitionally as long as they submit a plan for adopting the new model and immediately begin allowing articles to be freely published in an open repository. Many publishers of subscription or hybrid journals already publish articles in online repositories, although usually only after an embargo period. “Now, we want to ensure instant access to this content,” says Robert-Jan Smits.

PLAN S

What is it?
Initiative launched by agencies from 14 countries proposing that all research funded by public money should be published in open-access journals

Start
January 1, 2020

Aim
To create an international alliance and establish a global open-access system

What’s at stake

  • For funding agencies and universities
    Costs will increase as they will need to cover publication fees charged by open-access journals. Any deal to be agreed with publishers will depend on more countries adopting the plan.
  • For publishers and scientific societies
    Business models based on subscriptions and open-access charges are under threat. Plan S could put some journals out of business.
  • For researchers
    Will be unable to publish articles in high-impact journals that have not adopted Plan S. However, they will have greater access to scientific content without having to pay for it.

Two of the plan’s principles address publishing costs. One of them dictates that article processing charges (APCs) should be covered by universities and funding agencies whenever possible, not by individual researchers. The other states that these fees should be standardized and capped. The aim is to control costs. APCs, which cover the costs of reviewing and publishing articles, currently vary widely—they can range from US$1,500 to US$5,000 per paper. The value of the proposed cap has not yet been decided, but current discussions suggest it will be somewhere between US$1,500 and US$3,000, depending on the average number of citations (the impact factor) of the publication.

There are concerns that the changes will be costly for funding agencies, which will be responsible for paying most APCs. This has already proven to be the case in the UK, which in 2014 adopted an open-access strategy for research funded by 107 institutions linked to Research Councils UK (RCUK). A study published by British consultancy Jisc in 2016 found that the cost of APCs in the UK doubled between 2013 and 2014. According to the same study, it rose again—by 6%—between 2015 and 2016.

Marcelos Cipis

Defining these costs is essential to encouraging new countries to adopt Plan S. “We’re not ready to commit if the costs are too high,” Véronique Halloin, secretary general of the Belgian National Fund for Scientific Research (NFSR), told the journal Science. It would be an even greater obstacle in poorer countries, which would find it difficult to channel more money toward publication costs. “Plan S was born in Europe and cannot be simply transplanted to developing countries,” said Arul George Scaria, a researcher at the Center for Innovation, Intellectual Property, and Competition at the National Law School of India University, in an interview published on the blog of British journalist Richard Poynder.

Costs, however, will inevitably increase at first, observes Abel Packer, coordinator of the Brazilian online scientific library SciELO. The Max Planck Society in Germany estimates that scientific libraries worldwide spend almost €7.6 billion a year on subscriptions to journals for their students and researchers. In Brazil, there is increasing investment in the periodicals website run by the Brazilian Federal Agency for Support and Evaluation of Graduate Education (CAPES), which provides free access to more than 45,000 titles. In 2016, the agency spent R$357 million on journal subscriptions; in 2018, that figure rose to R$402 million. CAPES has said in a statement that it has no plans to end these subscriptions. The costs will disappear if Plan S succeeds, because research results will be openly available online. During the transition, however, APCs and subscription costs are still a major expense. According to Robert-Jan Smits, the money saved on subscriptions can be redirected to APCs. “Libraries could renegotiate contracts, guaranteeing that funds previously intended for subscriptions will be diverted to open access,” he suggests.

Abel Packer believes that Brazil should consider setting up a fund specifically to pay APCs. He estimates that if all journals indexed in the Web of Science database were open access and charged a processing fee, the national annual cost would range from US$75 million to US$100 million. His calculations were based on an average cost of between US$1,500 and US$ 3,000 per article. Brazil, however, is well positioned to adopt Plan S. A study by consultancy Scimago found that in 2016, 82.3% of the country’s scientific journals were open access—the global average was 13%. This is largely thanks to the SciELO library, a program funded by FAPESP since 1997 with support from CAPES and the Brazilian National Council for Scientific and Technological Development (CNPq), which currently provides open access to 293 Brazilian journals online. For Packer, the plan could encourage foreign researchers to look for quality open-access journals in Brazil. “Brazilian journals indexed on the Web of Science and Scopus databases and published in English could begin to receive more articles from authors abroad,” he says.

Plan S asks publishers to be transparent about the costs and criteria used to determine APCs. “This could be more efficient than capping prices because it would encourage competition between publishers,” says Belgian linguist Johan Rooryck, a professor at Leiden University in the Netherlands and director of the Fair Open Access Alliance (FOAA), an organization that helps journals make the switch to an open-access model. But most publishers reject the idea. Tom Reller, vice president of global communications at Elsevier, warns that no company will provide data on the costs of its products. “It’s private and confidential information,” says Reller, emphasizing that Elsevier only releases annual reports, which include operating expenses and revenue.

An online petition with more than 1,600 signatures, including a number of Nobel Prize winners, deems Plan S “a risk to science.” “It could restrict researchers from publishing in more than 80% of journals,” says biochemist Lynn Kamerlin, from Uppsala University, Sweden, who started the petition. This figure is based on a 2017 study by Universities UK, which is formed of various higher education and research institutions from the country. The findings showed that the proportion of hybrid journals jumped from 36.2% in 2012 to 45% in 2016—while subscription-only periodicals declined from 49.2% to 37.7% over the same period. According to Kamerlin, Plan S ignores the fact that many publications rely on subscriptions and could not exist without them. This is especially true for journals produced by scientific associations and societies, such as Physical Review Letters, which is published by the American Physical Society. In a statement, the institution explained that its publishing and peer review expenses are primarily covered by subscriptions. Of its 12 journals, only three are fully open access—the rest follow the hybrid model.

The Genetics Society of America (GSA), meanwhile, predicts that Plan S will reduce its income by a third. Two journals account for 65% of its total revenue. This income is used to fund other initiatives that generate no earnings, such as a program to help researchers at the beginning of their career. A recent article in the journal Science described the disadvantage scientific societies face compared to commercial publishers. “The largest company is Elsevier, which publishes more than 2500 journal titles; scientific societies each publish at most a few dozen,” the report said. Science‘s publisher Bill Moran suggests that quality periodicals published by scientific societies should be exempt from Plan S due to their “unusual circumstances and roles in scholarly communications.” Published by the American Association for the Advancement of Science (AAAS), Science is one of the most prestigious scientific journals in the world. “We wouldn’t be economically sustainable if author fees had to cover all publication costs,” Moran said.

Marcel Swart Swedish biochemist Lynn Kamerlin: “The plan poses a risk to science”Marcel Swart

According to Kamerlin, Plan S will probably have less of an effect on mega journals—periodicals that make a lot of money from APCs by publishing a very high number of open-access papers online, such as Nature Communications and Scientific Reports. “There are many financially stable journals with reasonable APCs,” says geneticist Michael Eisen, from the University of California, Berkeley, and one of the founders of the Public Library of Science (PLOS), which publishes a prestigious collection of open-access journals—most notably PLOS ONE, which charges an APC of US$1,595. “We know that Plan S will impact subscription-based publishers, but we want the research we fund, such as studies on climate change and epidemic preparedness, to be freely accessible and usable,” says Robert Kiley, head of open research at Wellcome Trust. The UK agency has a budget of €1.14 billion, of which €9 million is designated to open-access publication fees.

While governments, funding agencies, and publishers discuss the costs of Plan S, the scientific community is focusing on its possible repercussions for researchers. For many, there are concerns about the right to choose to publish work in highly cited journals. “Open-access journals in the humanities are new and most do not yet have a high reputation,” said anthropologist Birgit Meyer, a member of the Royal Netherlands Academy of Arts and Sciences, at a recent event. A professor at Utrecht University, Meyer’s lecture hinted at the tension that the initiative is generating between PhD students and their advisors. “Should I suggest, as I always have done, that they publish in high-impact journals, even if they are incompatible with Plan S?”

Many scientists want to maintain their freedom to publish in titles such as Nature, Science, Cell, and The Lancet, which all use a hybrid model. “When a researcher has made an important discovery, they will choose the most prestigious journal possible, even if that means paying a high fee for open access,” says Dr. José Eduardo Krieger, a professor at the School of Medicine of the University of São Paulo (USP). “Publishing in major journals not only satisfies the researcher’s ego, but also earns them recognition among their peers and is a factor in career progression. I try to submit papers to traditional publications, such as the American Journal of Physiology, which is a hybrid journal that charges up to US$3,000 to publish an open-access article,” he explains. Krieger believes that Plan S could provide an impetus to replace evaluation parameters such as the impact factor with alternative metrics.

Neuroscientist Stevens Rehen, from the D’Or Institute and the Federal University of Rio de Janeiro (UFRJ), notes that many health research groups, especially those studying epidemic diseases, have already begun placing greater importance on the fast and open dissemination of data and results, to shorten the distance between scientific information and the public. He points out that higher-impact journals can take up to two years to publish a paper. While good open-access journals may have a lower impact factor, they can publish articles much faster. “I often choose to publish in open-access journals, such as Scientific Reports, PLOS, and PeerJ, which are all compatible with Plan S.” Rehen is also an advocate of preprints—papers shared for discussion before undergoing peer review (see Pesquisa FAPESP issue no. 254). “The most important thing is that results are validated by the scientific community, not just approved by a publisher who is not usually an expert on the subject,” says Rehen.

Plan S encourages the use of preprints, which are already common in some fields, such as physics and mathematics, but does not see them as a replacement for the traditional system. Marcos Pimenta, a professor at the Federal University of Minas Gerais (UFMG) and president of the Brazilian Physics Society (SBF), believes the Plan S guidelines are unlikely to influence the popularity of preprints, although they could have an impact when it comes to submitting papers to journals. According to him, many physicists share preprints on arXiv—a highly respected online repository—and later publish the final version in a subscription journal that charges less than open-access periodicals. Recently, Pimenta spent US$5,000 to publish an open-access article in Nature Communications, paid for by research funding he received from CAPES.

In Brazil, funding agencies cover the costs of publication charges, says Alicia Kowaltowski, a professor at the USP Chemistry Institute. “The fees are not paid for by additional funding; they are included in the research grant itself,” says Kowaltowski, whose letter criticizing Plan S was published in the journal Science. “Researchers often have to choose between publishing in an open-access journal or buying lab equipment. Plan S can compromise scientific research with tight budgets,” she highlights. For her, establishing an APC cap is unfeasible. “It is highly unlikely that major publishers will reduce their fees within less than a year. Many journals will continue to charge exorbitant prices,” says Kowaltowski, who is editor of the open-access journal Redox Biology, published by Elsevier.

Nikolay Doychinov / EU2018BG Plan S advocate Robert-Jan Smits, of the European Commission, wants to implement the plan on a global scaleNikolay Doychinov / EU2018BG

Plan S will probably not have any great repercussions in computer science, says Lisandro Zambenedetti Granville, president of the Brazilian Computing Society (SBC). “The norm in computer science is to publish in conference proceedings. The registration fee for the event covers the costs of these publications,” explains Granville. “Having an article accepted means the author can present their work at the conference, and the audience can give instant feedback,” he notes.

The plan presents a challenge to how CAPES evaluates graduate programs in Brazil. The agency uses its own journal classification system called Qualis to monitor the scientific output of students and professors of graduate programs in the country, based on indicators such as the Journal Impact Factor (JIF), which represents the average number of citations of articles in a journal. Open-access titles are often poorly evaluated and discouraged. But Plan S supports the San Francisco Declaration on Research Assessment (DORA), which in 2012 proposed ending the use of the journal impact factor to measure article quality. “Evaluation metrics influence the behavior of researchers, especially those at the beginning of their careers, encouraging them to seek out journals with a high-impact factor, which in most cases are hybrids,” says Smits, of the European Commission. “There needs to be a cultural shift to a system that recognizes the quality of the research and not the vehicle in which it is published.” The Plan S implementation proposal was open for consultation until the beginning of February. Later this year, the institutions responsible for the plan will review the rules, accept or reject suggestions, and present a new version.

Repository for São Paulo state universities

Colin / Wikimedia Commons The Maughan Library, the main university research library at King’s College London, UKColin / Wikimedia Commons

In 2010, the FAPESP Board of Trustees approved a policy stating that when an article based on research funded by the Foundation is published in an international journal indexed in databases such as the Web of Science, it must also be archived in an open-access repository after the embargo period established by the scientific journal, which is generally at least six months. The initiative resulted in the creation of the São Paulo State University Deans Council Scientific Repository (CRUESP) in 2013, where articles, theses, dissertations, and other scientific papers are published by researchers from the University of São Paulo (USP), the University of Campinas (UNICAMP), and São Paulo State University (UNESP). The repository unites the collections of all three universities, which together contain more than 1.2 million records and can be accessed at www.cruesp.sibi.usp.br.

Republish