“The time for talking about open access is now past. With these agreements, we are going to achieve it in practice.” So said Dutch Research Minister Sander Dekker in Brussels at the end of May, after he and fellow EU ministers unanimously agreed that all scientific papers reporting publicly funded research within Europe should be freely available on the Internet from 2020 onwards. In other words, forget subscriptions to scholarly journals. Anyone – scientist, medic, student or ordinary citizen – will be able to download research findings at the touch of a button and free of charge.
Around for more than a decade, this vision of “open-access” publishing is supported by numerous academics, policymakers and librarians. But in the last few years the idea has been gaining ground. In fact, it may have acquired unstoppable momentum, thanks to both the opportunities offered by the Internet and anger at the rising profits of traditional journal publishers. Scientists argue that the work they carry out and peer-review should not be hidden behind publishers’ pay walls.
Ralf Schimmer, a sociologist at the Max Planck Digital Library in Munich, argues that it is unrealistic to think that all papers could be published using open access as soon as 2020. But he does believe that by then scientific publishing will have reached a “point of no return” where complete open access becomes inevitable.
From brains to particles
One prominent open-access project is Frontiers, a purely online set of journals set up in 2007 by Henry and Kamila Markram, both neuroscientists at the École Polytechnique Fédérale de Lausanne. Kamila Markram says that they embarked on the venture after getting fed up when a significant number of what turned out to be highly cited papers were rejected by top-tier journals after peer review. This “rejection cascade”, she argues, wastes scientists’ time and harms the economy.
With Frontiers, peer review is used only to assess the technical quality of a paper. Judgement of a paper’s importance is instead left until after publication; it is measured through quantitative “article-level metrics”, such as how many times a
paper is downloaded, cited or even mentioned in a tweet. Markram maintains that this overcomes the problem of a reviewer’s subjective bias, which might prevent perfectly valid, perhaps even groundbreaking, research from being published. “We are moving very far away from the model where one almighty editor decides these things,” she says. Meanwhile, a project known as SCOAP3 – the Sponsoring Consortium for Open Access Publishing in Particle Physics – is showing how to apply the open-access model to an entire field. It requires university libraries to put their subscription budgets into a central pot that researchers draw on to publish their papers and make them freely available. Hosted by the CERN laboratory in Geneva and backed by some 3,000 libraries, funding agencies and research centres, the project was launched in 2014 and was recently extended until the end of 2019. It involves eight journals, although the largest in the field – Physical Review D, published by the American Physical Society – is absent. Other fields have yet to follow suit. Currently only about 13% of all papers published in scholarly journals are freely accessible. Schimmer says that part of the problem is a perception by some librarians that open-access publishing will cost them more. However, an analysis he carried out with two colleagues at the Max Planck Digital Library last year showed that this is unlikely. The €8 billion a year spent in journal subscriptions worldwide divided by the roughly 2 million papers produced yields an average price per paper of about €4,000. In contrast, they found, open-access publication typically costs about half as much.
Macking an impact
Schimmer says he is not looking to drive traditional publishers out of business. For example, he notes that physicists continue to subscribe to peer-reviewed journals even though they upload their papers to the freely accessible arXiv server (prior to review). However, argues Schimmer, what must change is publishers’ business model. He believes that, rather than relying on what he sees as opaquely priced reader subscriptions, they should charge authors or their institutions for each paper published.
Unsurprisingly perhaps, traditional journal publishers see things differently. The Nature Publishing Group, for example, produces a number of open-access journals and also lets authors archive their papers online starting six months after publication. But, as the company outlined in 2011, it does not believe “one size fits all”. It argues that while open-access publication is most suited to cheaper, lower-circulation journals, subscriptions are best for the leading titles. That way, it says, the higher cost per paper – due to lower acceptance rates – can be spread among the greater number of readers.
Indeed, some researchers worry that requiring authors to foot the bill risks a drop in standards. They argue that open-access publishers can reject fewer papers in order to make more money. In fact, Frontiers has been criticized for making it very difficult for reviewers to reject research, no matter its quality. But Markram defends Frontiers’ peer-review process, saying that as they have published more papers their journals’ impact factors – which record the average number of citations per paper – have actually gone up.
Schimmer believes that even the most prestigious journals, such as Nature or Science, could be made open access – both because the high costs per paper would be diluted by the lower costs of other, more numerous journals, and because many scientists would still be prepared to fork out for the prestige associated with publication in such journals. But time is running out. “If publishers refuse to change then change will be forced on them,” says Schimmer. “People who are in their twenties will not tolerate such a ridiculous and antiquated system. They will simply pull the plug.”