Globally, state funders and philanthropic organisations have joined forces and pushed through the creation of platforms where the research they fund must be published open access (OA). But the form of that OA varies by region; Europe is mostly gold, while North America and Asia Pacific is generally green. Rapid advances in artificial intelligence (AI) and technology mean these platforms are flourishing – they are interoperable, and content is easy to access and showcase.
As a result, there are fewer subscription-based journals. A number of broad science, gold OA mega journals with low article publishing charges exist to publish content not captured by open platforms. Major society journals remain active, many operating a gold OA model, but struggle for manuscript submissions, so revenue is low. Pre-prints thrive in this world and are linked to the final article versions, which are still recognised as the authoritative version. Researchers benefit from access to data in a variety of ways, for example, via bite-sized publications and dynamic notebook-style articles.
The advances in AI and technology have also provided new methods of generating and communicating results. While research quality is still an important measure of performance, journal publication plays a diminishing role in determining a researcher’s career progress. Increasingly, research is assessed against agreed societal impact standards.
Overall, global research and development (R&D) investment is holding stable. There have been regional shifts – intensity (R&D investment as a proportion of gross domestic product (GDP)) has reduced slightly in North America. And although increases in R&D intensity in China have plateaued, overall R&D investment continues to rise, as China’s GDP grows steadily.
Funders in China, the West and the developing nations have come together to establish shared goals for both basic research and some major applied challenges (for example, climate change, energy and food), which are now the key focuses of national funding agencies and philanthropic organisations. Funding for exploratory blue-sky research has reduced; the emphasis is on rapid development of practical solutions.
Funders have also collaborated to create guiding principles for open science and scholarly publication, as well as metrics of assessment (such as societal impact, data dissemination, peer review and the success of collaborative processes).
Thanks to this joined-up approach, global and interdisciplinary collaboration has increased, aided by virtual reality and augmented reality tools. Researchers are now rewarded more for collaboration and the usefulness of their research, and less for novelty or being first to publish. The EU has focused on strengthening its internal approach to research and initiatives like the European Open Science Cloud, an environment for hosting and processing research data to support EU science, have gained good traction. This has prompted China to adopt a similar approach, with other emerging research nations in Asia following in their footsteps. Researchers demonstrating interdisciplinary skills are the most successful. Collaboration via social platforms is common and post-publication evaluation and comment is the norm.
Funders are driving interdisciplinary, cross–institution, global collaborations and reward the sharing of data as it enables research to be more open. To support this, high-technology content management, collaboration and dissemination products are vital. Tech companies are partnering with information solution providers, major research institutions and state funders to provide them. These solutions tend to be globally interoperable and can be personalised to meet most needs. Importantly, they promote accuracy in data, contributing to improvements in reproducibility, which are further aided by the availability of data sets in large-scale data repositories. Funders and publishers have also partnered to create a global web of open citations – most article references are now freely available.
The research article is still valued as a channel for communicating the stories behind discoveries, but has become atomised with the growth in popularity of electronic lab notebooks and other tools that facilitate fragmentation of the research and publication process. This means the article has evolved into a notebook-style paper containing (as applicable) experimental methods, data and observations, source code, and claims and insights. There are funder requirements around engaging with the public; each grant proposal must be accompanied by a public engagement plan and researchers are mandated to communicate their research findings – and their benefits to society – in an easy-to–understand way. This has helped to increase public trust in science, supported by increased access to raw elements of research (e.g. raw environmental and ecological monitoring data).
The interoperable open repositories include both pre-prints and peer-reviewed manuscript versions. Open access (OA) publication in journals is the norm: a number offer green OA; however, adoption is uneven across geographies and disciplines. With pressure to release information as widely, and as close to real time as possible, green OA embargo periods are approaching – and in numerous cases have reached – zero months. Many journals have transitioned to gold, others have folded; consequently, there has been a resurgence in authors choosing to publish in gold OA, broad-discipline mega–journals after a lull in the early part of the decade. However, the appetite for OA involving article publishing charges (APCs) is not universal, primarily due to funding priority challenges, and this has helped to force down the cost of APCs. Prestigious journals play a role, but their influence has waned. Across a range of subject areas, researchers increasingly post pre-prints.
of their work to communicate research outcomes. As a result, new research metrics supplement the existing indicators, which typically measure citation activity.
Revolutionary developments in artificial intelligence (AI) mean hypotheses can now be data-driven – although take-up varies across the sciences – and the speed and volume of research has accelerated. AI also supports peer review by checking manuscripts are logical, consistent and comply with editorial standards. Easy-interface, off-the-shelf products have made coding relatively simple. Researchers are broadly comfortable with accessing large data sets and interrogating them (using programming skills) and working alongside data scientists; however, there are still skills gaps.
On the education front, universities have resisted pressure to commercialise, but have diversified; they now offer more online courses and lifelong learning. More cross-disciplinary degrees are available and modules on data science and writing for the public are common. Although competition between universities remains, there is more collaboration (e.g. on shared research priorities).
This is an extract from “The Research Futures report”
