Scholarly publishing is changing rapidly as a result of digitization, the open access (OA) movement, and a third, counteractive force—the global pandemic. While there has been a marked rise in transformative agreements between academic institutions, libraries, and various consortia along with publishers recently, the pandemic has put stakeholders at a crossroads. For instance, many academic institutions have had to suspend journal subscriptions for financial reasons in the last two years. They also have to limit access to researchers and the general public by putting articles behind paywalls. With all that said, however, the recent emergence of OA alternatives cannot be dismissed.
One can draw a parallel between the transformation of publishing and the evolution of the music industry. The music business witnessed a major disruption in the 80s and 90s when albums were sold for high premiums as CDs—think back to the days when our laptops still had CD drives and our headphones were wired. But with the advent of the MP3 format, iTunes, and streaming platforms that are breaking into the scene, the way we consume our favorite tunes has completely changed. A similar transformation is taking place in publishing, and the industry is changing as we know it.
The future of scholarly publishing is being driven by advancements in technology. Let’s take a look at the top publishing trends for 2022.
Artificial Intelligence for publishing
Artificial intelligence (AI) is everywhere. Amazon and other large businesses use AI to understand individual purchase history and promote similar products; Netflix tells you what movies and shows you’ll like based on your past viewing behavior. We’ve been consuming AI technologies for many years; now, we just need to figure out how to best apply these use cases to publishing.
AI, Machine Learning (ML), and Natural Language Processing (NLP) are all terms that are sometimes used interchangeably, and it’s important to grasp the differences between them in order to develop a good solution that will provide the best outcomes.
For example, using an AI algorithm to run a copyright test or conduct a reference check can reduce the strain of teams to complete an otherwise time-consuming manual task. In another scenario, to address the problem of consumers’ attention spans being very short (less than 8 seconds!) while browsing content, an AI toolset can develop summaries of popular scholarly literature to engage potential readers.
Another popular application of AI is voice assistants. The New York Times, The Economist, and The Washington Post, to name a few, have adopted AI voice assistants to provide audio briefings through voice search. Similar use cases for making intellectual content discoverable with voice search can be found in digital libraries and repositories.
But how far will AI push the boundaries? Will it replace content writers and reviewers? It’s crucial to remember that in the past, any new disruptive technology has had a favorable influence on all stakeholders, most notably end users. Similarly, AI will not replace content writers or reviewers; rather, it has massive potential to enhance the overall user experience and keep users engaged.
Innovations in peer review systems
Peer review exists to ensure that the research is technically sound and that any flaws are found and fixed. But this method has been fraught with challenges like non-participative researchers, unrealistic timelines, plagiarism, lack of trust in open access publications, and institutional pressure on authors to publish papers.
Over the last decade, technology has helped peer reviewers overcome some of their obstacles. The peer review process has been getting easier, faster, and more efficient over time. Now, manuscripts can be reviewed online. In recent times, a number of online reading and annotation services have emerged with user-friendliness as their primary goal. The key to a successful platform is a healthy combination of user-friendliness and a connection with the author and editor workflow.
The use of AI in peer review can simplify the process, guide reviewers’ decisions, improve the quality of published papers, and save reviewers’ time. Already, a few academic publishers are experimenting with AI technologies to complete pilot projects like selecting reviewers, validating statistics, and summarizing the conclusions of a research article.
Tools like Penelope.ai are using AI to enhance the plagiarism checking process and to verify whether a manuscript’s references and structure fulfill the journal’s standards. Publishing platforms are also exploring NLP and ML to examine manuscripts.
Peer review still has a ways to go in adopting this technology, and the possibilities are endless. However, most tools powered by AI and machine learning have an Achilles’ heel; there is a risk that biases from previously published papers could be reinforced. This is a risk that developers of peer review systems need to address with care.
Preprints for faster dissemination of research
Preprint servers are online archives that hold works or data related to scholarly articles that have not yet been evaluated or accepted by academic journals. Posting research papers or other work on preprint servers has a few advantages. Publishing is a time-consuming process in general, but research can now be made available before the final version is published, allowing readers to see the work much sooner. And because the preprint is a public record, authorship is preserved. Typically, a digital object identification (DOI) is assigned to an author’s work. And in terms of reviews, publishing a preprint allows other scholars to provide constructive criticism that can help improve the work before it goes through a formal peer review process. What’s more, with preprints, research will be made publicly available so other academics can cite and expand on it quicker.
As scientists around the world researched to fight the COVID-19 pandemic, the status quo of publishing changed quite dramatically. bioRxiv and medRxiv have several thousands of studies related to the pandemic. The advantages of the servers are obvious—results can be transmitted fast, potentially informing policy and speeding up research that can lead to vaccines and therapies. But the success of preprints draws attention to the level of scrutiny that these studies are put through. It’s difficult to assess the quality of work without peer review, and publicizing bad science can be damaging, especially when the research has urgent implications for medical practice. As a result, platforms like bioRxiv and medRxiv have improved their standard screening processes. The vetting method for BioRxiv and medRxiv is two-tiered. In the first stage, articles are reviewed by in-house staff who look for plagiarism and other concerns. Then, the submissions are reviewed by volunteer academics or subject experts, who look for non-scientific content as well as health or safety concerns. Principal investigators are mostly used by bioRxiv, and this work is finished within 48 hours. Because medRxiv employs health specialists and the topics covered are directly related to human health, the overall review time is 4-5 days. Because of the increased checks and the sheer volume of submissions lately, the preprint servers have had to hire extra staff.
Admittedly, preprints aren’t a panacea for the problems in scientific research, but they have proven to be efficient, especially in these uncertain times.
Audio/video formats for scholarly content
Since Gutenberg developed mass printing technology in the 15th century, the industry has undoubtedly moved beyond the consumption of print. Authors are no longer just writers in the digital age; instead, they now create their compositions digitally. Similarly, end-users are no longer limited to reading content in print; they can now consume it in a variety of digital formats.
The most recent trend is to listen and watch instead of read. There is a growing demand for audio and video content from the younger generations that prefer bite-sized content like audiobooks, podcasts, and vodcasts. Creating content in these formats is no longer overly expensive; tools, equipment, and learning resources are available for anyone to venture into audio and video content. This type of content can make research a lot more accessible and engaging, and every stakeholder in publishing stands to gain from wider audiences consuming research.
Blockchain for Shared Governance
The amount of data we now have at our disposal is unparalleled. We’re no longer just consumers of content but also the creators. Knowledge is now just a click away. Do you want to learn about a particular subject area? Take an online course and/or subscribe to a scholarly publication. Do you want to try your hand at writing? Find tips in a podcast or a video. Are you looking for a new hobby? Find it on YouTube.
This abundance of data (which comes from the 5Vs of big data – volume, variety, velocity, veracity, and value) can be both a blessing and a curse. Scientific research is expanding at a remarkable pace, posing new obstacles like predatory journals and wider issues like academic integrity.
Blockchain is a distributed ledger technology that can be tailored to address some of the difficulties that scholars face. Consensus Protocols on the blockchain can be used to focus on a single purpose, such as Shared Governance. If each original research paper is a “record” in the blockchain, Consensus Protocols’ voting process can be used to validate and arrange these records. In this setup, a peer reviewer would evaluate the manuscript and provide feedback using the voting process.
By mapping the publication workflow to blockchain architecture while remaining agnostic to a particular publisher, the Shared Governance model can improve the openness and integrity of publishing amongst publishers, scientists, researchers, reviewers, students, and the general public.
All the trends we discussed above have one thing in common—technology. None of the developments in publishing would’ve been conceivable without the development and application of technology. This technology is changing and growing by the day, and it will continue to do so through 2022. Every industry has taken on this continuous development as a challenge and is fighting to thrive, grow, and stay relevant. As the American writer, futurist, and businessman, Alvin Toffler said, technology is “the great growling engine of change.” And this holds true for every industry in the world, including publishing.
Fraser N, Brierley L, Dey G, Polka JK, Pálfy M, et al. (2021) The evolving role of preprints in the dissemination of COVID-19 research and their impact on the science communication landscape. PLOS Biology 19(4): e3000959. https://doi.org/10.1371/journal.pbio.3000959