In scholarly publishing, we are increasingly aware of issues of diversity from a societal point of view. It has become a part of our efforts to promote, facilitate, and ensure Diversity, Equity, Inclusion & Accessibility (DEIA) in different areas of the publishing ecosystem. But other forms of diversity can also be found in academic publishing.

Researchers, for example, use a wide range of options to access research — from freely accessible to subscription to open access (OA) journals. Many researchers without institutional access get in touch with the authors or colleagues to get hold of journal articles. Over the past few decades, global programs and coalitions have been making thousands of journals accessible to researchers in the Low- and Middle-Income Countries (LMICs) or the Global South. Preprints have been around since the early 1990s, but recently have expanded to many major disciplines. Many journals are now working with preprint servers. On rare occasions, all published articles on a particular topic can be made free to read and use, as we saw in March 2020, at the beginning of the COVID-19 pandemic.

If we look into OA a bit deeper, we will find diversity in OA models: Diamond or Platinum OA, Green OA, Gold OA, and Bronze OA. Many reputable publishers have launched new online OA journals in recent years, along with making their established journals OA. Many long-running journals also follow a hybrid model by offering both subscription and OA options. OA megajournals showcase a new (disruptive) dimension in scholarly publishing. Diversity can also be seen in the Article Processing Charge (APC) model: flat fee agreements between a publisher and institutions, equity models that take into account an institution’s country’s economy, and community models distributing publication costs more equitably among the participating institutions. Transformative agreements between libraries and publishers allow a shift from subscription-based reading to OA publishing. Regional OA journals are being launched to improve equity in open access models.

We also see diversity in the peer-review process. In addition to, single-anonymized and double-anonymized peer-review, triple-anonymized peer-review has been advocated for. On the other hand, some journals are not only sharing peer-reviewers’ names with the authors, but also publicly disclosing those names along with their reviews. Open, community-based peer review can be seen on preprint servers and other research platforms. To cut down peer-review time, options like transferable peer-review, AI-assisted peer-review, and collaborative peer-reviewer pools are in practice. While in almost all cases, peer-review is a voluntary service, some have argued that a paid peer-review model would expedite the review process, while others have argued to the contrary. It however may be more common in Southern journals, which traditionally pay a small amount to their in-country reviewers. While post-publication peer-review is practiced by some journals, peer-review can now be done on experimental design and methodology before starting data collection. Octopus represents new platforms where researchers would publish all steps of their research cycle and receive feedback from other registered users on a regular basis.

If we explore other aspects of scholarly publishing — publication format, workflow, data sharing mechanisms, copyright, or licensing — we will find diverse options in practice. We may explain such diversity as a manifestation of the vibrant innovation culture of this industry driven by the needs from its stakeholders. To understand what value such diversity brings about, let’s compare it with the biodiversity we see around us.

First, in a natural ecosystem, be it a forest or a lake or a salt-marsh, we often find many similar species performing similar activities. Let’s take for example the world’s largest mangrove forest, the Sundarbans shared between my country Bangladesh (60%) and India (40%). Home of about one thousand plant and animal species, including the world-famous Bengal Tiger, Bangladesh Sundarbans, for example, has 37 species of butterfly, 13 species of orchid, and 7 species of woodpecker. This type of redundancy makes an ecosystem resilient to any shocks or stresses. It means that if due to some pest invasion one species is wiped out, other similar species will fill in the gap and the whole ecosystem will continue functioning without any significant changes. Similar redundancy in the scholarly publishing industry in the form of digital publishing, preprints, megajournals, diverse peer-review systems, for example, has already helped us to survive the initial surge of the COVID-19 pandemic, which I pointed out in a Learned Publishing article.

Second, each species plays a specific role within an ecosystem, such as a rainforest in Indonesia, a peatland in Russia, or a coral reef off the coast of Mexico — it’s called the “ecological niche” of that species. In scholarly publishing, as context changes, our understanding evolves and new demands are raised, and we respond to them by creating new options (like species) and by putting them in our existing system (like finding a niche for the option within the scholarly ecosystem). These new options or innovations are in addition to what we already have, without replacing anything. That’s why, in 2022, we can see all the versions of research communication — from printed journals (as was published in the 17th Century) to the Octopus platform and every other form in between. Nothing seems get lost in the scholarly publishing ecosystem. This coexistence of similar options, being neutral to each other, highlights the less competitive, soft nature of scholarly publishing around the globe.

Third, a list of all the species found in an ecosystem is a basic representation of its diversity. But all species don’t exist in equal number in an ecosystem. When the abundance of each species is put into the equation, we get a weighted expression of all species. Such calculation is important because with the increasing number of a particular species, the overall functional diversity of the ecosystem decreases, and the dominant species start dominating the ecosystem. Thus, how diversity is measured and presented matter. Although several OA models exist, for example, their collective proportion varies: In 2019, OA articles represented more than 30% of total published articles, but when expressed in journal publishing market value, OA occupied a little above 7% (US$ 763m). So, when we see diverse options are available or are recommended, we need to ask ourselves what proportion of the population (or journals in this case) have adopted them. Dominant species which influence an ecosystem by their sheer number or size are crucial to transform an ecosystem. It is also true for the scholarly ecosystem where a handful of publishers guide the overall directions of the industry.

Fourth, a species’ survival and continuity depend on sufficient resources in the ecosystem — for plants, the nutrients in the soil; for predators, the prey; for fungi, dead matter. But, given the exponential growth of the scholarly publishing industry, it seems, at least theoretically, there are no limits to the sustenance of journals. Unlike many other businesses, scholarly publishing remained largely unaffected during the COVID-19 pandemic, not only because it showed amazing resilience — defined by robustness, resourcefulness, redundancy, and rapidity — but probably also due to journal articles’ intrinsic relationship with academia’s research funding, scholarship, recruitment, tenure, and promotion. Further, species in nature go extinct over time; in recent decades, the rampant over-exploitation by humans has accelerated the extinction rate 100 times faster. But, the “death of a journal” seems to be a myth — my recent Google Scholar search came up with only 23 documents; just three articles bear that phrase in their titles.

All of these make me wonder, is academic publishing thriving by offering new ideas and innovations in policy and practice, by embracing and capitalizing on digital advancements, and by acquisitions and mergers? Or is the diversity discussed above is just a way to brand ourselves as critical, unique, relevant? We indeed define our brand on our own terms, otherwise how could we ask authors anything between US$ 100 and US$ 10,000 to process one manuscript?

Finally, we use diverse ways to measure the impact of our research and publications. With so many indicators, indices, factors, matrices, rankings, and scores, every researcher and their institution, every journal and its articles, have a number to talk about. It seems we have turned ‘research impact’ into ‘impact of numbers.’ As journal publishers, we follow the number of times authors cited us, social media mentioned us, appropriate agencies ranked us, but not what is happening on the ground based upon the articles we publish. We know that not all research outputs have immediate, direct, practical implications or applications. But have we ever searched how many of our articles were translated into reader-friendly briefs to reach out to policymakers? Or were cited in public strategies benefitting vulnerable people? Or quoted in donors’ working papers to channel funds to deprived regions? What is stopping scholarly publishers from innovating in measuring research impact? I wonder.