A big step towards open science
- Cet article est une traduction de :
- Un grand pas de plus vers la science ouverte
- Autre(s) traduction(s) de cet article :
- Un gran paso hacia la ciencia abierta
Texte intégral
1It is a great privilege to be able to say thank you to a public institution and to welcome a decision that goes in the direction of what we have long defended, without always being followed. On several occasions, Cybergeo has warned about the actions of major international publishers, by denouncing their system, just as incomplete and harmful to scientists as the one that Jonathan Nossifer had deconstructed in his film Mondovino for small wine producers: Clarivate Analytics and Elsevier had succeeded in imposing a production system that brings them profit margins of more than 35% per year by not paying science evaluators, and to convince the consumers of science to use the rankings made by them after diverting from their initial function, intended for the information of scientists, the WOS and Scopus databases, to produce force rankings of journals and other scientometry parameters supposed to measure the performance of researchers. The dramatic consequences of these practices are known: inflation of the number of journals and articles, unbridled race for the number of publications, “salami slicing” in research results, impact on the direction of scientific policy, loss of time and quality of evaluations.
2The CNRS has decided to no longer associate with these practices. The breakup began in 2024 with the unsubscribing from Elsevier’s Scopus. Starting January 1, 2026, the CNRS will also free itself from its expensive subscription to the WOS (1.4 million euros). It advocates a return to qualitative peer review that takes into account the value and impact of all research results including data sets and software, and confirms an evolution towards open science initiated in July 2018 with the first national open science plan. In the same motion, the CNRS supports the publication in diamond magazines at no cost to authors and readers.
- 1 Christine Kosmopoulos, « Entre prédation et désincarnation, l'édition scientifique à l'épreuve de l (...)
- 2 Christine Kosmopoulos, Joachim Schöpfel, 2024, Publier, partager, réutiliser les données de la rech (...)
3Several elements motivate this decision, but we will mainly note the immense cost for public finances required by publishers to make the articles freely accessible because of the APC (Author Processing Charges) system. From 2017 to 2022, the cost of APCs for the CNRS doubled and projections estimate the expenditure for the organization at 168 million euros in 2030 in an all-APC model. To this unsustainable cost, the impact of artificial intelligence in the system is added. Despite millions of euros invested by commercial publishers to compensate for the generation of articles by AI, and moreover their acceptance in peer-reviewed journals, no tool can guarantee 100% of the integrity of scientific publications to date. The result is a plethora of articles partially or fully generated by machines, evaluators who collapse under requests and end up letting irrelevant or even erroneous articles1 pass through and researchers who no longer have time to read all relevant papers and rely on syntheses produced by AI. For the CNRS, it must be noted that the publication and evaluation system fed by Clarivate Analytics and Elsevier via scientific journals is out of breath. Open science and diamond publishing models are presented as future solutions to ensure integrity, quality peer review, reproducibility and equitable sharing of research results and data2.
4The decision is courageous, and it follows the way defended by OpenEdition with its Freemium economic model that we were the first to support. It marks the beginning of a resistance that will be difficult to shoulder and maintain in the long run in the face of long-established routines and the diversity of social representations regarding the appreciation of the quality of science. We can still hope that if the return on investment benefits research, the scientific community will come out stronger and more united. Cybergeo will continue to defend these principles and value the qualitative peer review and openness of its publications. Here, we thank all the people who, through their conscientious and often voluntary scientific work, contribute to maintaining this quality, while working for creativity!
5On March 13, 2026, "The 30 years of Cybergeo: 1996 - 2026" will be an opportunity to look back on our common history and share with you our new projects!
Notes
1 Christine Kosmopoulos, « Entre prédation et désincarnation, l'édition scientifique à l'épreuve de l'IA », Cybergeo: European Journal of Geography [En ligne], Éditorial du 13 février 2025. URL : http://journals.openedition.org/cybergeo/41699
2 Christine Kosmopoulos, Joachim Schöpfel, 2024, Publier, partager, réutiliser les données de la recherche : les data papers et leurs enjeux, Presse universitaire du Septentrion. A télécharger en libre accès : https://www.septentrion.com/fr/book/?gcoi=27574100316700
Haut de pagePour citer cet article
Référence électronique
Christine Kosmopoulos et Denise Pumain, « A big step towards open science », Cybergeo: European Journal of Geography [En ligne], Éditoriaux, mis en ligne le 08 décembre 2025, consulté le 12 février 2026. URL : http://journals.openedition.org/cybergeo/42521
Haut de pageDroits d’auteur
Le texte seul est utilisable sous licence CC BY 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont susceptibles d’être soumis à des autorisations d’usage spécifiques.
Haut de page

