Over the last few decades, the publishing landscape for scholarly articles has rapidly evolved towards greater accessibility of scientific discoveries, which is a much needed and welcome change. The shift towards more transparent and openly available research has been accelerated by the digital production and dissemination of scientific papers. In theory, this transition should have reduced the costs associated with the physical printing and distribution of journals to university libraries. However, it has instead drastically increased costs related to publications (Morrison et al., 2022; Shu & Larivière, 2024, prefix=e.g., ).

Several solutions have been proposed to make scientific articles freely accessible under the umbrella of Open Access (OA). Different OA models exist, each with its own benefits and drawbacks for authors, institutions, and readers. In this opinion piece, we first present the main OA publishing models and then discuss in detail the advantages and limitations of Green OA, concluding that Diamond OA is the only OA model that is sustainable in the long term.

Open Access (budapestopenaccessinitiative.org), the most visible component of the broader Open Science movement (UNESCO, 2021), refers to the unrestricted and free access to scholarly publications. Over the past few decades, funding agencies and governments worldwide have strongly promoted OA to ensure that peer-reviewed research outputs are openly available (see for example coalition-s.org). However, the preferred routes to achieve this goal have varied depending upon national policies, institutional support, and individual choices.

At present, the main models generally used to enable the publication of scientific papers under OA are: (i) Green, (ii) Gold (pure and hybrid), and (iii) Diamond OA (1). Although all three provide readers with free access to scientific knowledge, they differ substantially in terms of costs, responsibilities, and implications for authors and their institutions.

Green OA, also known as self-archiving (n.b., here, we only consider self-archived peer-reviewed articles), was amongst the first alternatives proposed to the traditional publishing model. Green OA aimed to make studies published in for-profit journals (e.g., Elsevier, Springer Nature, Wiley, Taylor & Francis, and SAGE) accessible to all, either after an embargo period or directly after publication depending on individual journal policies. A decade ago, at least in the geosciences, it was still considered “normal” to relinquish copyright to publishers and accepting that they make profit from research largely funded by taxpayers while readers had to pay to access the resulting papers (1). Fortunately, times have changed. As researchers became increasingly frustrated (Swartz, 2011, prefix=e.g., ) by the escalating costs of accessing their colleagues’ work, and stakeholders needed to pay to access publicly funded research, the Green OA model emerged as a practical solution. In this system, authors deposit the Author Accepted Manuscript (AAM; i.e., the peer-reviewed, accepted manuscript) not the Version of Record (VoR; i.e., the final formatted, published article), in open online or institutional repositories (e.g., EarthArXiv, HAL, Archives-Ouvertes, ResearchOnline@JCU), ensuring free access to the content while maintaining compliance with publishers’ embargo and open access policies. It is important to note that Green OA relies entirely on permissions granted by publishers and is not an authors’ right. Because most journals impose embargos of several months (where Green OA is permitted), some publishers even offer a zero-embargo option for an Article Development Charge (ADC) of US$2500 (Anderson, 2023, prefix=e.g., American Chemical Society Publications; ).

Gold OA, which is also referred to as the ’author pay’ model, is a more recent model that emerged alongside government and funding-agency mandates requiring research to be openly accessible. In this system, authors retain the right to freely share their work, and the VoR is freely available to readers, allowing unrestricted sharing among researchers and the public (1). However, this model shifts the financial burden to authors or their institutions, who must pay an Article Processing Charge (APC), typically around US$2500 per paper (Haustein et al., 2024) (2). While the real cost of publishing has dropped significantly with digital production, numerous new companies have capitalised on the pay-to-publish model (e.g., MDPI and Frontiers), charging high APCs while sometimes overlooking scientific rigour and editorial quality (Frank et al., 2023). Although institutional or funding-body subsidies occasionally offset these fees, the profits of commercial publishers have soared (Butler et al., 2023; Hagve, 2020), raising concerns about equity and the long-term sustainability of pay-to-publish systems. Less subsidised research groups often cannot afford APCs and are effectively excluded from publishing in the same venues as better funded groups. Occasional publisher waivers for authors with less bargaining power are limited and do not alter the structural outcome: the model is inequitable (Asai, 2021; Klebel et al., 2025). Moreover, the paradox at the heart of the current system, where researchers conduct the work yet must pay for its dissemination, has not only persisted but deepened, further entrenching systemic imbalances.

In response to these developments, many higher-education institutions have adopted Read & Publish (transformative) agreements, also known as ’Big Deals’, which grant access to most research published by major commercial publishers while allowing affiliated researchers to publish articles as Gold OA rather than behind paywalls. These agreements are typically negotiated at the national level (e.g., swissuniversities) or through university-library consortia (e.g., Council of Australasian University Librarians), and they involve multi-million-dollar contracts funded largely with public money. While positioned as a solution for authors, these agreements often obscure who ultimately bears the cost of publication, making both the funding sources and the actual expenses increasingly difficult for authors and the public to discern (Rothfritz et al., 2024). Additionally, depending on the specific agreement, there is often a yearly cap on the number of OA articles that can be published by an institution while some journals are not included in the deals (e.g., Nature titles are often not covered in Read & Publish deals with Springer Nature). Once this quota is reached, authors must publish behind paywall and deposit the AAM in an open repository (Green OA) or pay an APC to make their work (the VoR) available via the Gold OA route. Unsurprisingly, recent assessments indicate that these agreements are economically unsustainable for institutions or organisations and do not meet expectations for delivering a fully OA publishing ecosystem (Rothfritz et al., 2024; Šimukovič, 2024). Furthermore, a number of universities or institutions have already decided to walk away from those agreements, largely as a response to rising prices and opacity (see for example many US universities with Elsevier, or swissuniversities with Springer Nature, bigdeal.sparcopen.org/cancellations).

Diamond OA (sometimes called Platinum OA) is a traditional model that has been recently modernised and redeveloped in reaction to the exorbitant amount of public money spent each year enriching for-profit publishers rather than supporting research infrastructure and development. Globally, commercial publishers generate an estimated US$19 billion in annual profits (Butler et al., 2023; Hagve, 2020)a figure that has prompted many in the scientific community to reclaim control over scholarly communication. Compared to the other OA roads, the Diamond OA applies no APCs, and offer full free access to the published articles (1).

In response to the growing profits generated at the expense of researchers and public funds, several scientific communities have launched community-led Diamond OA journals to reclaim control of scholarly publishing from commercial interests. While Diamond OA options had existed for some time (e.g., Rivista Italiana di Paleontologia e Stratigrafia), and were quite dominant in other scientific fields, in the geosciences, this movement was rejuvenated with Volcanica (Farquharson & Wadsworth, 2018) and was soon followed by Tektonika (Fernandez-Blanco et al., 2023), Seismica (Rowe et al., 2022), Sedimentologika (Thomas et al., 2023), Geomorphica (Lefebvre et al., 2025), Advances in Geochemistry and Cosmochemistry (Pourret et al., 2025), Open Palaeontology (Drage et al., 2025), Geodynamica, jSEDI, and Planetary Research. All of these journals are hosted by university libraries or other non-profit institutions and most use the free, open-source platform Open Journal Systems (OJS; pkp.sfu.ca/software/ojs), developed by the Public Knowledge Project (PKP; pkp.sfu.ca). Two exceptions are Geodynamica, which runs on Janeway (janeway.systems), and jSEDI, which is hosted on Episciences (episciences.org). OJS allows for managing submissions, editorial workflow and subsequent publication (Thomas et al., 2023, prefix=see ). Alternatively, some established journals have flipped their publishing model (Constantin et al., 2025), leaving commercial publishers to operate under community or university presses. For instance, Lethaia transitioned from Wiley to the Scandinavian University Press (Doyle et al., 2022). In all these cases, articles are free to read, no APCs are charged and authors are free to distribute the VoR of their article (1).

In this model, funding bodies and university libraries redirect portions of their budgets traditionally allocated to subscriptions and APCs to subsidise not-for-profit publishing venues. Within these academic publishing houses, professional staff provide both material and in-kind services, enabling scholars to operate journals at little-to-no cost. This approach results in a system where publishing and access are both free, and authors can freely share their work. Research is disseminated in its final formatted version (the VoR), with comparable quality to that of commercial publishers. Crucially, since no profit is involved, the legitimacy of these journals rests on their reputation. This can foster greater scrutiny in the editorial process and, ideally, a stronger and more motivated involvement of editors and reviewers whose labour is no longer appropriated for commercial gain.

Copy-editing and formatting can be carried out in various ways, either by volunteers running the journal (often academics themselves) or by subsidised professionals, depending on the journal’s funding model. Although this approach may require more work from academics, it stands in stark contrast to the exploitative nature of the current Gold OA system. In that context, such efforts may be considered a necessary step toward establishing a more equitable model of scholarly communication. This opinion piece advocates precisely for that shift.

One of the main obstacles to adopting these new models lies in academic culture itself. Many researchers, particularly senior academics, still respond to new Diamond OA journals with one question: “What is the impact factor?”. Because most of these journals are recent, they either do not yet have an impact factor or deliberately choose not to participate in this system, given that journal impact factors are widely recognised as flawed and biased metrics (Vanclay, 2009; Dimitrov et al., 2010, prefix=e.g., ). Yet the impact factor continues to dominate perceptions of journal prestige and is misused in researcher evaluation. Originally, the impact factor was developed to help librarians decide which journals to purchase, but it has unfortunately become a proxy for evaluating researchers. To counter this misuse, 26774 individuals and organisations across 171 countries have now signed the San Francisco Declaration on Research Assessment (DORA) (sfdora.org/read; consulted on January 19, 2026), advocating for an evaluation system that eliminates journal-based metrics. Adopting DORA implies recognising that peer-reviewed scholarly publications are valid scientific contributions regardless of the journal’s impact factor. In theory, this principle should put Diamond OA journals on an equal footing with their commercial counterparts.

Currently, many countries career evaluation systems still use impact factors as necessary metrics, preventing widespread use of journals that do not provide such metrics. Additionally, many (particularly senior) researchers, who are often the ones responsible for evaluating others, continue to rely on numerical metrics rather than assessing the scientific merit and impact of research, especially when it falls outside their field. These persistent dependences on outdated measures remain one of the greatest barriers to achieving a fair and sustainable publishing system. Pushing reform in researcher assessment, the Stockholm Declaration (Sabel & Larhammar, 2025), which endorses DORA, was recently published and emphasises that (i) academia should take back control of publishing; (ii) quality should be valued over quantity, and metrics should not be used as proxies for excellence; (iii) tools should be developed and used to detect and track fraudulent publications; and (iv) policies should be implemented to increase the quality and integrity of scientific publications. These four pillars of the Stockholm Declaration are readily implementable through community-led Diamond OA journals, which are run by scientists for scientists and operate on a not-for-profit basis.

Another common concern about Diamond OA journals is their long-term sustainability and the preservation of published papers. A frequent question that is fielded to these journals is: “What happens if the journal stops?”. The usual response is: “and what happens if Elsevier shuts down?”. In reality, nothing lasts forever—except diamonds. While large, well-established commercial publishers may seem too big to fail, academia and society takes their permanence for granted, but the future is never certain. It has happened in the past that journals and even publishers were obliged to shut down, or be abandoned (Kincaid, 2023, prefix=see for example the case of Hindawi; ). Fortunately, Diamond OA journals using platforms like OJS, have robust digital preservation strategies. For example, the PKP Preservation Network (PKP PN) plugin leverages the LOCKSS program (Lots of Copies Keep Stuff Safe; lockss.org), which ensures multiple copies of content are securely stored across a geographically distributed, decentralised network of servers that are maintained by libraries. In this sense, an institutionally backed, internationally replicated and standardised open system offers one of the most sustainable approaches to long-term preservation.

Because community-driven Diamond OA journals require only modest (and transparent) funding to operate, they offer a genuinely sustainable, accessible, and equitable alternative for scientific publishing. For example, Sedimentologika operates at approximately US$2500 per year (including Crossref indexing, hosting, and in-kind support from the Open Access Publications team at the University of Geneva, Switzerland), roughly equivalent to a single average APC.

Redirecting the substantial financial support currently granted to commercial publishing houses—whose credibility has been steadily eroding for years (Dwianto, 2025; Heen & Vogt, 2024; Walter & Mullins, 2019)—towards a coordinated effort to re-empower academics in their own publication ecosystem is both feasible and necessary. In an equitable system, publicly funded research should be accessible to all at every stage of the research cycle, from submission to dissemination. While the OA movement has undoubtedly transformed scholarly communication, its ultimate goal remains unfulfilled: under the current oligopolistic landscape dominated by Gold OA, neither fair costs nor universal dissemination have been fully realised (Dulong de Rosnay, 2021; Ma et al., 2023). Within this context, Green OA plays a complex role. While its purpose is to make knowledge widely available, in disciplines and countries where OA has become mandatory, it often serves primarily to uphold legacy structures, thereby delaying any meaningful transition toward a more sustainable publishing model.

Therefore, we urge researchers to invest in their Diamond OA ecosystems, as authors, reviewers, editors and supporters, while limiting support to for profit publishing venues. The wider adoption of the Diamond OA system is firmly in our hands (Andringa et al., 2024; Dulong de Rosnay, 2021) and public stakeholders in European countries (Arasteh-Roodsary et al., 2025; Hahn et al., 2023) or Canada (van Bellen & Céspedes, 2025), for example, have prepared infrastructures and regulations for such transformation. We also urge academic societies to seize this opportunity to anchor themselves as supporters of this transition. While they may lose some direct income by moving away from subscription-based and APC-driven models, they can become key actors of this transformation, by providing support, expertise and guidance through their experience and networks. Similarly, scholars should support their network by subscribing to their professional societies, wherever possible.

Additional abstract translations