Scientific publishing is currently undergoing a fast change towards open access (OA). Many new OA journals emerged in the past decade, every major publisher has scores of OA journals and the number of journals in general keeps growing quickly, primarily due to new OA journals. One can read everywhere about advantages and fairness of OA publishing – free science available to everyone. Isn’t OA publishing a great idea? It sounds like communism in science, providing to each according to his/her needs. But will communism in publishing lead to the same disaster as communist ideas in politics? Why should scientists, who are producers of scientific information, pay thousands of dollars for making their findings available to others? Doesn’t it contradict the laws of economics (and common sense)? No one expects farmers to pay to people who acquire their crop or cooks to pay restaurant guest for eating dishes they prepared. Maybe scientific publishing is an intrinsically unprofitable business that cannot be sustained without donations from researchers? Then why do many publishers report record profits? More I observe the quickly changing scientific publishing landscape, more questions come to my mind, and I don’t necessarily have answers to all of them. If I’m confused and ask those questions after acting as editor or associate editor of big and small journals for almost two decades, serving on editorial boards of at least 20 journals and authoring/co-authoring close to 800 papers, maybe scientists with less extensive experience are confused too? Let’s discuss this issue together.
Presumed Benefits of Open Access Publishing: Many research sponsors, especially in Europe, have made publishing OA a requirement. In August 2022, the US Office of Science and Technology Policy (OSTP) announced that it would require immediate open access for its funded research, and US funding agencies’ policy plans are due this year. The key benefit of OA publishing is that the work becomes easily and immediately available to everyone, including scientists in industry (companies subscribe to fewer journals) or in poor countries, that would not have quick and easy access otherwise (even though anyone can write to the author directly or via Research Gate to request a free PDF version nowadays). Many scientists believe that OA publishing may potentially increase the impact and citations of their work, even though my personal experience does not support this statement - novelty and significance of the work are the key factors, not the mode of publication. Nowadays, many less selective journals (and all predatory journals) publish OA, while majority of high-impact and high-quality journals require subscription. Therefore, one should be careful selecting an OA journal. One should look for the one having a reputable publisher (e.g., Springer-Nature or ACS), keeping well-known scientists on its editorial board and publishing papers that have good quality and are cited by peers. Unfortunately, those are typically expensive, currently ranging from $2000 to $5000 and more. Of course, there is significant cost involved in publishing. Hiring knowledgeable editors capable of attracting qualified peer-reviewers, as well as editing/processing manuscripts before publication costs money. As someone who has been serving as editor for journals published by professional societies and a for-profit publisher, I know it firsthand. And the cost is higher for top journals, as quality does not come for free. Fewer papers are published in a journal per year, higher is the selectivity, better are the editors, but higher is the cost.
Unfortunately, the OA system may erode the publication quality substantially. Traditionally, the publisher had to select quality papers for subscription journals – otherwise, no one would subscribe or pay to access individual papers or buy journals issues. However, in the case of upfront OA payment, the quality of the work becomes not the condition for receiving profit. More papers get published, more money is generated, and right away, no need to worry about subscription or sale of the copyright. Since profit is the driver for any industry, including publishing, not surprising that OA journals want to publish more papers and are ready to sacrifice quality for quantity. As a result, predatory journals have become the modern-day plague of science. They contaminate literature with papers that are presented as “peer-reviewed work”, but have little-to-zero value or, worse, mislead due to lack of proper review. Only experts in the field can eliminate papers that report erroneous results or have no scientific value. With the growing volume of scientific information, there is no good way to provide professional editing and high-quality peer-reviewing to each manuscript. I get several requests to review manuscripts every day but can accept no more than one per week.
Why to pay? I personally feel that publishing OA should be free for authors who put months (sometimes years) of their life, long days and sleepless nights to produce data and write a paper. There are a few free OA journals (e.g., Beilstein journals) and many depositories for not-peer-reviewed work. Also, some journals (e.g., American Chemical Society journals) publish few selected best papers OA as a means of promotion. However, typical OA journals require authors to pay. My approach is to only pay when the sponsor rules (e.g., European funding agencies) require OA publication or when my co-authors want to publish in a particularly prestigious OA journal (e.g., Nature Communications) because of requirements/expectations to publish in journals with a high Impact factor (IF) and, particularly, in Nature or Science family journals in some countries. This particularly puts pressure on junior scientists looking for a position or being on tenure track at universities in those countries.
Is it fair to ask authors to pay? While I certainly support sharing published scientific information as widely as possible, I find it unfair that people who produce knowledge (research data) pay for publishing it. Why should not consumers of the knowledge pay? I hear arguments that if taxpayer money paid for research, the research results should be freely available. I may agree, but then those papers should be freely available to readers in a particular country, or maybe a particular community. Communities can get access to those publications through libraries (also, remotely). Why should taxpayers (and scientists) in one country pay for providing knowledge to researchers in other countries, including unfriendly ones? This is an especially sensitive issue in the fields of science and engineering, where new knowledge gives a strategic technological advantage. No one suggests stopping the flow of scientific information, but wouldn’t it be fair to expect that if one country paid for doing research, the consumers in the other country that want to benefit from the results should pay for access? If we want to drink Scotch, we pay Scotland for it, we don’t expect Scotts to pay for the privilege of making this magic drink and distributing it all over the world. Of course, countries can agree to share the cost of research and share the produced knowledge, as it is done within the European Union. It’s also important to provide access to scientific knowledge to developing countries, which may not be able to pay the same price for journal subscriptions. However, this kind of support should probably come from deep discounts provided by publishers (they all want to do business in developing countries) or the same sources that support other humanitarian causes, rather than from research budgets of funding agencies or universities. If the government decides that research products should be freely shared, it should at least provide funding for publishing open access, like the European Commission does. However, keep in mind that money for publications likely come from the same pot as research funds, so less money will be available for research. US funding agencies don’t currently offer much money for publications – usually we get a symbolic amount, which won’t be enough to publish a single paper in an expensive journal like Nature Communications, and publication costs are among the first to be cut when sponsors ask to decrease the budget. Also, enforcing OA rules hurts the most productive authors, the ones who produce multiple good papers as they may not have enough money to pay for publication. The ones who publish a paper or less per year, may be less affected. Finally, when it comes to supporting an additional student or publishing several papers OA, what choice should a professor make?
How to change the system. One option would be a wider use of free depositories, such as arXiv, where everyone can deposit preprints and this is how most (80-90%) results of research, especially evolutionary and repetitive studies, or results of unsuccessful experiments, should be shared. Other researchers can read, comment (community peer-review) and cite the work they find useful. Compare to YouTube use by artists to promote their music. Corrections can be made to adjust the deposited work by posting updated versions. Publishing an erratum nowadays is a cumbersome process that may even hurt someone’s reputation, and there is no good mechanism for updates and minor changes in archival publications. Comments and citation of researchers working in the field will determine which preprints are valuable and useful, and which are not, just like “likes” and the number of views on YouTube. If the research community, including university administrators and tenure/promotion committees, starts evaluating publications based on the number of reads, citations, Altmetrics scores and reputation in the community (use of the publication by others), instead of requiring publications in “peer-reviewed” journals, predatory journals will disappear and much of the money spent on publications will be reinvested into research. A major expansion of archives can be expected in the future, as online-only mode of publication becomes increasingly common. The cost of maintenance of electronic archives is much lower compared to journals, so funding agencies, private foundations and individual research institutions should be able to cover the cost, especially if volunteers help with moderation of publications, like it’s done with social media groups. The best and most important results that were posted as preprints can be then submitted to archival peer-reviewed journals for publication, with a preprint satisfying the requirement for OA publication. Moreover, if someone wants to give OA to a paper in a journal, subscription journals gladly take author’s money to allow free access to papers. Thus, the necessary level of OA can be achieved by using a combination of archives and subscription journals. Whether this or other model is implemented, a change is needed to enable the research community to disseminate knowledge, while keeping the quality of research publications. Since the current OA system stimulates quantity over quality of publications, punishes the most productive researchers, and goes against fundamental laws of economics (all of which are logical), it should be revamped, and this may prevent the looming collapse of the scientific publishing system.