Beyond the journal: The future of scientific publishing

The scientific publishing system is buckling under increased pressure from soaring article output, affordability issues, and threats from AI and paper mills. Amid widespread calls for reform, new models are emerging to reshape scholarly communication.
Beyond the journal: The future of scientific publishing
Like

The global system of scientific publishing is under unprecedented strain. With growth in article output outstripping growth in the number of researchers around the world, the publication workload per scientist (writing, reviewing and editing) has increased dramatically and unsustainably (Hanson et al, 2023). Longstanding concerns over affordability, access and equity, which the open access movement has struggled to solve, have been compounded by threats to research integrity from generative artificial intelligence and paper mills.

Small wonder, then, that bodies including the International Science Council, cOAlition S and the European Council have issued calls for reform of the scientific publishing system in recent years. Yet what might a reformed system of scientific publishing look like?

Replacing the journal

Some proponents of reform have argued that the solution is to replace traditional journals with a decentralized network under the governance of the scholarly community (Brembs et al, 2023). Such a network, which would be based on open-source technology, appears both technically possible and socially desirable given the outsized profits and limited public access that commercial publishing models have delivered to date. Yet, as the authors of the proposal acknowledge, the entrenched nature of the academic reward system, the scale of funds that would need to be redirected, and the level of international coordination required make fundamental reforms of this nature unlikely in the near term. In the meantime, three emerging models offer pointers to the future shape of scholarly communication.

  1. Funder platforms

Over the past few years, funding agencies have taken an increasingly active interest in how the research they support is shared. Wellcome, the Gates Foundation and the European Commission, among others, have created dedicated platforms on which beneficiaries of their funding can publish their work in open access at no charge. The coming years will see the EC, in conjunction with European national funders, take steps to extend its Open Research Europe platform into a not-for-profit European publishing venue for all. Responding to the European Council’s call for transparent, equitable and open access to scholarly publications, the goal is to ensure all researchers have a high-quality, non-profit alternative to commercial and society-owned journals, irrespective of their access to funding (full disclosure: I have supported the EC in developing plans to operationalise the platform).

  1. Publish, review, curate

cOAlition S, a group of mostly European funding agencies, has already demonstrated its ability to reform the publishing world with its Plan S initiative. The plan undoubtedly served to accelerate the transition to immediate open access following its launch in 2018, but many argue that it has created a host of other problems in its wake. The group’s 2023 proposal, ‘Towards Responsible Publishing’, aims to tackle the inequity of current publishing models based on article publication charges, eliminate delays in the sharing of research outputs and reduce inefficiencies in peer review. The proposal, which encourages the early sharing of preprints and opening up of the peer review process, favours wider adoption of the ‘Publish, review, curate’ model implemented by eLife in 2021. This would see growing connectivity between preprint repositories like bioRxiv/medRxiv and preprint review services, with the results aggregated into curated collections of high-quality or impactful papers.

  1. Deconstructed publishing

Other emergent models seek to deconstruct the publishing process and challenge the dominance of the journal article as the sole or primary research output. FEBS Open Bio’s Research Protocols is one of many examples of this trend, enabling researchers to share an experimental protocol as a standalone output, providing more detail than that typically included in the methods section of a research article. Other initiatives go further by enabling multiple research components to be shared on a single platform. For example, Octopus, funded by UK Research & Innovation, offers eight publication types that are aligned with the research process and aim to make primary research open to all. Meanwhile, DeSci Labs leverages decentralised web technologies to break down the traditional publication into modular components that can be linked and reused. This allows researchers to publish individual figures, tables, data sets, code and other elements of their work, with each module assigned a digital object identifier (DOI) and citable on its own.

A step too far

As a long-term observer of scientific publishing, I believe each of these models can enrich the publishing ecosystem. Nevertheless, I am sceptical that any of them can replace the academic journal as the dominant mode of scholarly communication, for two main reasons.

Firstly, in emerging economies, even more so than in the West, the journal remains the currency of scientific evaluation and career progression. It is these countries - most notably China, but also Brazil, India and Indonesia, among others - which will determine the future shape of global scholarly communication, and all the signs are that they remain committed to the journal. For example, China’s China Journal Excellence Action Plan (CJEAP), aims to create a portfolio of 400 world-class journals which can attract a global author base. The most influential publishing developments of the next decade are more likely to be Chinese-owned journals than curated preprint collections or funder platforms.

Secondly, past attempts to transcend the journal as the venue of publication have shown a tendency to founder beyond a certain level of uptake. In the early 2010s, PLOS ONE’s seemingly inexorable rise led some to conclude open access ‘megajournals’ would revolutionise the publishing world. Yet after publishing in excess of 30,000 articles in 2013, PLOS ONE’s popularity slumped, and it has stabilised at around half that level in recent years. More recently, we have seen the rise of ‘disciplinary mega-journals’, exemplified by MDPI’s International Journal of Environmental Research and Public Health (IJERPH), which used guest editorships and special issues to drive submissions. Yet after publishing almost 20,000 articles in 2022, IJERPH and similar journals have also suffered a steep decline in submissions after many of them were delisted from the Web of Science due to a lack of ‘content relevance’.

The scientific community’s continuing reluctance to let go of the journal, even when ostensibly superior technical solutions are available, suggests it continues to play an important role in the eyes of researchers. Whether it is understood as a community, a club or a brand, the journal holds an enduring value for authors, readers, users and evaluators of research. There is a pressing need to reform the access, editorial, governance and business models that underpin today’s scholarly journals. Yet efforts to replace the journal itself may prove a step too far.


Photo by Maarten van den Heuvel on Unsplash

Join the FEBS Network today

Joining the FEBS Network’s molecular life sciences community enables you to access special content on the site, present your profile, 'follow' contributors, 'comment' on and 'like' content, post your own content, and set up a tailored email digest for updates.

Go to the profile of Gavin McStay
about 1 month ago

Very interesting Rob. I direct Biomed News (hhtps://biomed.news) which is a machine learning based platform to discover new biomedical research literature. It is made of up of Reports curated by an Editor on a particular topic. These are released weekly and almost act as a personal journal for each report editor. This is freely shareable to anyone who wants to subscribe. 

Go to the profile of Rob Johnson
about 1 month ago

Hi Gavin, thanks for sharing this, it's really interesting to see. Can you share a little more information on how these reports are assembled? I'm interested to understand how you are using machine learning in combination with editorial curation, and specifically what you feel is the optimal balance between the two.

Go to the profile of Gavin McStay
about 1 month ago

The reports are started when an editor provides the seed paper(s). Terms and phrases are then used for subsequent weekly issues to identify abstracts which contain those terms. This is provided in a ranked list. The editor then select papers relevant to the topic and these are added to the machine learning for the following week. This continues as each issue is released. Everything is available on our website - all previous issues and can be shared freely.