This year, we are celebrating open access week by talking to a number of researchers who have chosen to publish their work openly. We are discussing the reasons behind their choice, and the benefits they have seen for their research and career, and the role that institutions as well as Transformative Agreements can have in assisting with funding. In this interview, Dr. Thilo Hagendorff, expert in applied ethics and lecturer at the Hasso Plattner Institute at the University of Potsdam in Germany, shares his thoughts on publishing open access.
Justified reasons for limiting open access to research results can hold true in fields like IT security, synthetic biology, or nuclear physics research. Here, publication restrictions can serve as precautionary efforts to prevent the malicious use of scientific discoveries. Other than that, however, open access ensures public availability of in many cases publicly funded research, raises its visibility and citability so that science can influence other social systems, fosters the discovery of (interdisciplinary) cooperation opportunities, makes research controllable and replicable, helps with preventing experiments from unnecessarily being carried out twice, makes code transparent and auditable, and much more.
I am an expert in applied ethics, especially technology and AI ethics. My research concerns methodological questions of AI ethics as well as specific subfields like industry involvement, publications norms, quality data, or privacy in AI research. Luckily, my university supports open access publishing, hence many of my papers are freely available to the wider public.
My papers that are openly accessible receive considerably more citations than papers that appear in closed access journals where only researchers at institutions which have a licence can view them. I assume that the success of my most impactful paper titled “The Ethics of AI Ethics”, which appeared in the journal “Minds & Machines” under open access conditions, is in part due to its free availability.
My discipline, which is machine learning and AI research, is not much different from many other scientific disciplines, embracing the idea of open access. New journals that enter the field and that stand under the umbrella of closed access are facing lots of criticism and even calls for a boycott.
Projekt DEAL helps a lot with OA publishing. By selecting one of the various journals that has a contract with my university about OA under Projekt DEAL, I just have to talk the accepted publication over with someone from the OP team of my library, and they are covering the costs. A short notification suffices, and OA publishing is successfully underway.
Go for it! Open science raises the level of transparency and reproducibility. This, then, enhances the efficiency of the scientific process as well as the ease with which new collaborations can be built upon, which ultimately benefits society.
Open access is not a binary variable. There are gradations in the way research results can be shared, for instance in case dual-use applications are investigated, where a staged publication release can be adopted during which various (negative) effects of the released scientific insights can be monitored.
Dr. Thilo Hagendorff’s article was published OA under the Projekt DEAL agreement with Springer Nature. This agreement means authors affiliated with participating institutions can publish OA with fees covered, in more than 2,000 Springer hybrid journals and more than 500 fully OA journals across the Springer Nature portfolio.
Dr. Thilo Hagendorff is an expert in applied ethics, especially technology and AI ethics. He is the author of several non-fiction books. He works for the Cluster of Excellence "Machine Learning: New Perspectives for Science" and at the International Center for Ethics in the Sciences and Humanities at the University of Tuebingen (Germany). He is also a lecturer at the Hasso Plattner Institute at the University of Potsdam (Germany) and a member of various working groups on AI governance.