Transition to Open‐Access Publishing and to Open Science

Scientific publications are undoubtedly the main means of disseminating and preserving scientific knowledge. Methods of scientific publishing—from handwritten notes circulated in scientific circles before the invention of the printing press to extensive digital archives now available to billions of people—reflect profound societal changes that have taken place since. The sharing of ideas, discoveries, and methodologies through scientific journals began in the 17th century in England, France, and Germany. Philosophical Transactions and Le Journal des Sçavans, both founded in 1665, were soon followed by the German journals Miscelanea Curiosa (1670) and Acta Eruditorum (1682). Behind them all were important figures of intellectual history—from Heinrich (Henry) Oldenburg, through Denis de Sallo and Otto Mencke, to Gottfried Wilhelm Leibniz. These journals became a model for recording, verifying, disseminating, and archiving scientific knowledge. Peer review was introduced and systematically used by The Edinburgh Medical Journal from 1773 onwards and was gradually adopted by other, initially, mainly medical journals. However, a peer-review-like practice was in place in a number of old-vintage journals (e.g., Proceedings of the Royal Society of London) where the published papers were “communicated” (i.e., endorsed) by a distinguished academic. From the beginning of the 19th century, the number of scientific journals increased from tens or hundreds to hundreds of thousands by the end of the 20th century. The growth was approximately exponential (several percent per year), with rates varying by field and region. At the beginning of the 20th century, a model for the structure of scientific articles, called IMRaD (introduction, methods, results, and discussion), emerged and became a global standard after World War II. In 1971, the digitization “Project Gutenberg” was proposed, leading to the creation of the first free online digital library (https://www.gutenberg.org/). Digital books—as well as journals—then proliferated rapidly with the rise of the World Wide Web in the late 20th century. Digital scientific communication soon surpassed traditional print in many ways, including rapid dissemination, easy search and storage, quick corrections and updates, as well as collaborative and interactive distribution of scientific articles. With widespread and growing access to digital content, the model of scientific publishing has shifted from exclusively print to print-on-demand, exclusively digital, or both print and digital. With the development and deployment of digital literature search engines such as Google Scholar, Scopus, and Web of Science, citations and other metrics have become easily accessible and transparent. Scientific publishing, once a noncommercial activity funded primarily by scientists themselves or their societies, has dramatically changed since World War II into a highly lucrative industry. In 2024, this industry had a global turnover of over $12 billion and a profit margin of nearly 40%. Scientific journals have become the main product of this industry, which is now dominated by the “Big Five”: Elsevier, Springer Nature, Wiley, Sage, and Taylor & Francis. One of “Britain's most notorious tycoons, Robert Maxwell” has been recently credited with creating the academic publishing industry as we know it today (https://www.theguardian.com/science/2017/jun/27/profitable-business-scientific-publishing-bad-for-science). How did OA come about? It all began at the dawn of the 21st century, when the World Conference on Science, under the auspices of UNESCO, adopted the Declaration on Science and the Use of Scientific Knowledge [1]. This declaration stated that “in the 21st century, science must become a common property from which all humanity will benefit on the basis of solidarity.” In 2002, a group of academics, librarians, and publishers issued a statement at a conference in Budapest that introduced the OA publishing paradigm. This was followed in 2003 by the Bethesda Statement from the National Institutes of Health (NIH) and the Berlin Declaration from the Max Planck Society (Max-Planck-Gesellschaft, MPG). The Bethesda Statement and the Berlin Declaration gave rise to a movement that has since transformed how scientific information is published worldwide. The launch of the Public Library of Science (PLoS) in the United States in 2003 is but one example. The goal was to create a high-ranking journal similar to Cell, which was both subscription based and author supported, but which was free to read. As to the success of this effort, see https://www.nature.com/articles/nm0203-154b. In 2018, a group of national research funding organizations, with the support of the European Commission and the European Research Council, announced the launch of the cOAlition S initiative, where S stands for Science. (We note that the “S” could as well stand for Robert-Jan Smits [special envoy to EU Commissioner Moedas on OA in 2018] and Marc Schiltz [President of Science Europe since 2017], who were the main architects of cOAlition S.). cOAlition S is an initiative that seeks full and immediate OA to the results of publicly funded scientific research. The latest significant international development in open science took place at the UNESCO General Conference in 2021. On this occasion, UNESCO issued the “Recommendation on open science,” which aimed to “provide an international framework for open science policy and practice.” In this 36-page document, UNESCO highlights “the urgency of addressing complex and interconnected social and economic challenges for humanity and the planet,” recognizes “the vital importance of science, technology, and innovation (STI) to respond to these challenges,” and emphasizes “the transformative potential of open science to reduce the existing inequalities in STI.” The UNESCO “Recommendation” defines open science “as an inclusive construct that combines various movements and practices aiming to make multilingual scientific knowledge openly available, accessible and reusable for everyone, to increase scientific collaborations and sharing of information for the benefits of science and society, and to open the processes of scientific knowledge creation, evaluation and communication to societal actors beyond the traditional scientific community. It comprises all scientific disciplines and aspects of scholarly practices, including basic and applied sciences, natural and social sciences and the humanities, and it builds on the following key pillars: open scientific knowledge, open science infrastructures, science communication, open engagement of societal actors and open dialogue with other knowledge systems.” Open science is based on values such as quality and integrity, and on principles such as transparency, control, and reproducibility—all of which are enforced through rigorous peer review. This is illustrated in figure 2 in [2]. OA publishing is currently the most widespread component of open science. How far along is this publishing model? In figure 1 in [2], global production of research and review articles by OA category is shown. Currently, almost half of all research and review articles are freely accessible. OA publishing benefits science itself. Science, as an institution of organized criticism, can function properly only if scientists are able to consider the work of others and use it in their own research. The advantages of OA publishing include a higher citation rate for OA articles than for articles behind paywalls. An example: OA articles in Wiley journals receive over 90% of all citations, whereby nearly 50% of Wiley articles are published OA. Only the remainder goes to articles with closed access. Among open questions regarding OA publishing is covering its costs (including peer review, copy editing, formatting, and archiving). This is addressed in some detail below. In addition, there is Diamond OA, where readers do not pay for reading and authors do not pay for publishing. The costs are borne by institutions or companies, which must themselves secure all the services—otherwise provided by publishers—from their own resources. Diamond OA has the support of the EU, which: “ENCOURAGES Member States and the Commission to step up support to the development of aligned institutional and funding policies and strategies regarding not-for-profit OA multi-format scholarly publishing models in Europe with no costs for authors or readers, and to set and implement roadmaps or action plans for a significant expansion of such publishing models” (https://data.consilium.europa.eu/doc/document/ST-9616-2023-INIT/en/pdf). However, according to a recent analysis (https://doi.org/10.1007/s11024-023-09519-7), the Diamond OA model works only in well-coordinated niche fields for journals with a low total number of articles, and generally appears less sustainable. In the past, the libraries of research institutions paid publishers an annual subscription fee for new journal issues. This subscription was a reading fee that had to be paid for the current fiscal year before the end of the previous year. Among the factors that led to the emergence of the OA movement were the excesses of the subscription business model: authors were forced to transfer copyright for their publications to publishers. Publishers described the confiscation of copyright as the “bedrock” of the subscription system. In addition, subscription contracts were not transparent, often remaining hidden from the public eye due to nondisclosure agreements between libraries and publishers. This allowed for arbitrary pricing (see also the OA white paper, https://oa2020.org/wp-content/uploads/pdfs/MPDL_OA-Transition_White_Paper.pdf). The term “publishers” includes both commercial publishers (Wiley, Elsevier, Springer, etc.) and scientific societies (American Chemical Society, ACS; American Institute of Physics, AIP; American Physical Society, APS; etc.). In 2015, the Max Planck Digital Library (MPDL) found that, under the subscription system, libraries worldwide paid approximately €7.6 billion for 2 million research and review articles published annually, or €3800 per article. Subscriptions were typically increased by 3%–4% per year. For large publishers, these fees were bundled (in so-called big deals) as part of negotiations with libraries about multiple journals published by the same publisher. In the United States and elsewhere, part of these subscription costs was covered by indirect costs, also known as “facilities and administration” (F&A) costs. Publishers charged subscription fees for their journals to cover production costs. At the turn of the 20th century, these began to include digitization and digital archiving. When OA publishing emerged at the beginning of the 21st century, publishers offered individual authors the option of making their articles available to all readers free of charge for an additional fee called an article processing charge (APC). Some funding providers (grant agencies), such as the NIH in the United States, even mandated OA publishing of research results obtained under their auspices, which became an important incentive for publishers to create the option of paying APCs in subscription journals. The costs of APCs were not visible to libraries (or research institutions) because they were usually covered by the direct costs of individual researchers–authors. For authors working in institutions that subscribed to the very journal in which they published OA for an APC, this meant double payment for the same publishing service. This practice was called “double-dipping.” To preclude double-dipping, so-called off-setting agreements were concluded between libraries of research institutions and publishers. These exempted academic institutions that had subscription contracts with publishers from paying APCs for a certain fraction of their publications, which would then be published OA. An alternative to off-setting agreements emerged in mid-2010, when the first phase of so-called transformative agreements (TA) was negotiated and implemented. “A transformative [or transformational] agreement arises when an institution or group of institutions enter into partnership with a publisher to enable a large-scale transition towards open access. A transformational agreement allows the migration of funding from reading towards publishing—whilst recognizing that the subscription element remains an important part of the agreement” (https://www.wiley.com/en-us/network/publishing/research-publishing/open-access/transformational-agreements-at-wiley-how-far-have-we-come). We note that the first transformative agreements were, solely out of pragmatism, constructed so as to realize a budget-neutral transition from the subscription era to the OA era. Only then could the academic institutions and the publishers reach an agreement on the transition. The annual amount paid to the publisher of a given journal by an academic institution—or a consortium of academic institutions, or even a nationwide consortium such as DEAL (Deutsche Allianz Lizenzen, https://deal-konsortium.de/en/) in Germany—was divided by the (average) number of articles published annually by that institution in that journal. This was then taken as the basis for a mutually agreed fee per published article, with the acronym APC taking on a new meaning, namely, article publishing charge. For this fee, each article from the contracting institution is published open access, with that institution (or consortium) having access to the journal's electronic portfolio, or even to the entire portfolio of the publishing house. In the second, ongoing phase of transformative agreements, it is being considered that only the costs of publishing should be covered, and no longer the costs of access to reading. At the same time, publication costs should be realistic and transparent and should no longer be determined on the basis of historical subscription rates. Since the payments for publishing are due only after the article is published, institutions usually have to pay the publisher 6 months later. This essentially saves half of the annual subscription budget. Ádám Dér from MPDL commented: “[W]ith post-payment models, libraries can guarantee that their financial commitments align directly with the publishing preferences of their authors, allowing them to allocate resources in line with their mission. […] Institutions and research communities negotiating TAs are making unprecedented levels of new research publicly available, eliminating author-facing APCs and generating substantial savings (Brayman et al., 2024), curtailing the potential of commercial publishers to monetize the works of authors, increasing transparency around the financial streams of scholarly publishing, and optimizing OA publishing workflows with community-driven standards” (A. Dér, “What Gets Missed in the Discourse on Transformative Agreements,” https://katinamagazine.org/content/article/open-knowledge/2025/transformative-agreements-discourse-what-gets-missed). Although at the national level (as evidenced by the DEAL agreements in Germany, see also fig. 3 in [3]) or at the consortium level (as is the case, e.g., with the University of California system or the Big Ten universities in the United States, https://oa2020.org/wp-content/uploads/B16_Session_1_Willmott.pdf), the costs of publishing have fallen as a result of the transition to OA, this has not necessarily been the case at the level of individual institutions. Institutions that are more research-oriented and therefore publish more will pay more in the OA era than institutions that are less research oriented. The latter institutions include national libraries and other institutions that are more focused on reading than on publishing. Their costs will decrease compared with those in the subscription era. However, the average fee per “article” or “review article” has fallen; for example, under the DEAL consortium agreement with Wiley, the fee has fallen to €2750 under the OA model. Figure 1 compares the OA article share and the revenue share, attesting to the cost-effectiveness of OA publishing. Moreover, the figure illustrates that the subscription model imposes a greater financial burden on the research community than OA publishing. Despite all the evidence, the myth that OA has increased the cost of scientific publishing prevails—in some communities more than in others. In some disciplines, publishing on preprint servers gains more and more attention which complicates discussions at the universities (https://www.journalmetrics.org/blog/preprint-servers-guide?utm_source=chatgpt.com). A general deduction in advance for OA costs across all faculties would be a general model, some universities OA costs to individual researchers–authors. This to discussions in the scientific system at which that publishing, in should not be a business with a profit The transition phase academic institutions with an to make the of a library institutions must now create an information to cover all publishing costs by their a large information at first to be a burden for the of academic institutions. In the a information to be a that academic institutions can and will use to the In Germany, the German Research has provided open access publication funding to the transition to OA, including to and publication at universities and other research institutions. In to the costs of the OA business model and its fiscal transparency, the to published articles are not to the publishers (as was the case with the subscription business but with the author a where stands for with the which is the for the and reach of scientific A with transformative agreements is that they have not been across the scientific community. This a and institutions, and in the of the of scientific knowledge. The annual by the on APCs and is under which includes the of the MPDL who contracts for all Max Planck Institutes with publishers. In addition, the individual approximately million per this amount essentially the of one per a additional budget. This total amount of million should be compared to the total annual of approximately This means that for the which is a the share of the entire on publication and subscription costs is only more than from this the the publication costs of all its of the journal in which they The of the transition from the subscription to the OA business model at the is illustrated in Figure Currently, approximately of all publications authors are with are published open access, and all APC costs are covered by this budget. also has a with Springer for which includes of APCs of for publications in these many articles in these journals. The has not OA is that some authors out of OA publishing, because they are not and they will have to pay APCs from their own research that the publishing fees would have been covered It should be that there are very journals that are from this namely, the American for the of Science that Science, because it not OA publishing, and the Journal of the Society of because its fees are We note that the MPDL not consider the APCs of journals to be given the and importance of these journals. all have OA publishing. in these on conference with an annual APC payments are not allowed or there is a on their total annual costs. open science and OA as means of the costs of publishing scientific on publishing through agreements between institutions or their and publishers amount to savings for who a large part of the for scientific research. the of the transition to open science and OA publishing action by the research community and the publishers. However, even in the case of Germany, where the is nationwide by the German Science and and the DEAL consortium it is up to individual institutions and their to the principles of open science and to see to it that the transition to OA publishing is taken the OA business model, it is in to increase both APCs and the number of articles APCs under is primarily the of research it is up to the of individual journals to that only articles are for However, should also that the APCs charged by their journals are transparent. It that the of journals on the principles of open science be of their publishers and on the of to publishers. Only then can open science journals the scientific community with the it We note that open science in an additional challenges that have to be with we note that there has been a of online journals that to be scientific and OA, but in are It is not to them from the as it is not to from the Their should not for OA or to its advantages and social or journals have to do with OA. the of the which online publishing is not by the and of all that the World Wide Web