Scientists and subscription journals tussling for power

A recent controversy triggers long-standing debate over subscription vs open- access journals. Fatima Arkin reports.

Scientists are getting more and more interested in open-access journals. They see the advantages: more citations of their work, and a favourable eye from publishers, especially those based in Europe, who face political pressure to veer towards open access.

But two years shy of a deadline set by the European Union to make all publicly funded scientific papers in the region open-access, some publishers continue to introduce new subscription-based journals — and in the process, face pushback from the global scientific community.

The Nature Publishing Group, which is part of Springer Nature, caused uproar recently when it announced that its forthcoming Nature Machine Intelligence (NMI) journal will be subscription-based.

“It’s the scientists who hold the power,” Tom Olijhoek, editor-in-chief of the Directory of Open Access Journals (DOAJ), a global directory based in the UK, tells SciDev.Net. “The seller of services depends on the customer, not the other way around. So if scientists or funders choose open access, publishers will have to comply. You see this happening now.”

In light of the NMI standoff, this may be about to be tested.

An access revolution

Open access turns the traditional business model of journal publishing on its head. Normally, subscribers pay a fee to read journals, leaving authors to pay nothing to publish their work. Non-subscribers can still access the material, but only if they pay for the article that they want to read.

With the newer model, authors are the ones who pay — either through their own pocket or via their institutions — and the published papers are freely available to anyone. This means their work can reach the widest audience possible. And this, ideally, means it has the best chance of stimulating understanding and innovation.

The open access publishing model has existed since the 1990s, but began in earnest in 2002 following the Budapest Open Access Initiative, which is an international effort to make research articles in all academic fields freely available on the Internet.

The movement has grown steadily since then. And now, many institutions and funders, such as the Bill and Melinda Gates Foundation and the Welcome Trust, require that the research they fund is made freely available. According to the DOAJ, the number of open-access peer-reviewed journals has increased by 17 per cent in the last year alone.

Usage of the DOAJ website has increased by over 100 per cent in that same year. Science, technology and mathematics journals have been growing their open-access presence most, particularly medical journals and articles. And of the 120 million documents in the academic journal search engine BASE, approximately 60 per cent are now open access.

Some of the biggest names in publishing, including Elsevier in New York, United States, and Springer Nature in Germany, are investing heavily in new open access journals, while also acquiring existing ones based in other parts of the world.

But, as the latest Nature Publishing Group controversy suggests, the old subscription-based model is not about to disappear.

Seeking viable options

Thomas Dietterich, a professor at Oregon State University and past president of the Association for the Advancement of Artificial Intelligence, has gone so far as to create a petition calling for a boycott of the journal, which is set to be launched next year. At the time this article was published, the petition has garnered nearly 3,300 signatures from around the world.

“My colleagues and I hope that Nature Publishing Group will decide to make NMI an open access journal,” Dietterich tells SciDev.Net. “We also want to discourage other publishers from creating fee-based journals in artificial intelligence and machine learning.”
 
A Springer Nature spokesperson tells SciDev.Net that NMI involves substantial editorial development, offers high levels of author service and publishes informative, accessible content beyond primary research — all of which requires considerable investment.

“At present, we believe that the fairest way of producing highly selective journals like this one, and ensuring their long-term sustainability as a resource for the widest possible community, is to spread these costs among many readers — instead of having them borne by a few authors,” the spokesperson adds.

And that is just one of the problems with the open-access publishing model: the article processing charges (APC) or publication fees sometimes charged to authors to make their work freely available will exponentially rise in order to keep profit margins high and shareholders happy, warns DOAJ’s Olijhoek.

He says that this is highly likely if the publication is large and noteworthy. This is because scientists, including those from developing countries, still want the status associated with a big publisher, believing that their journals are worth the enormous price tag because they have a high impact factor.

For instance, Nature Communications charges up to US$5,200 per paper, excluding local taxes, compared with the average APC, which Olijhoek pegs at roughly US$800.

Matt McKay, director of communications at the International Association of Science, Technology and Math (STM) Publishers, cautions that open access will always cost publishers more — especially gold open access, which is when the final, published version of an article is made freely available to the general public as soon as possible after it is accepted. This is "to offset the costs that are presently sustained through subscription publishing,” he adds.

But, Olijhoek argues that open access doesn’t have to be so costly. “65 per cent of the journals in DOAJ have no APC, so the business model of funders or foundations paying for publishing is a viable option, especially in the Global South,” he says.