The CNRS is breaking free from the Web of Science

Since the publication of the CNRS's first Roadmap For Open Science in 2019, the organisation has been actively implementing a policy of opening up research results and changing the way scientists are evaluated. Along with many other research organisations, the CNRS opposes quantitative evaluation based on bibliometric indicators, instead of favouring a qualitative approach. This, then, is the context underpinning the CNRS's decision to cut access to Clarivate Analytics' Web of Science (WoS) which the organisation considers an essential step for its open science policy. The aim of this decision is to be coherent as regards the organisation's stated principles for research evaluation which firstly call for a shift from the use of quantitative bibliometric indicators and secondly for the accelerated development of alternative solutions focusing on open and transparent data. 

In favour of qualitative evaluation

The use of impact factors in the evaluation of researchers has contributed to the distortion of scientific publishing practices and, in turn, of research practices. Alain Schuhl, the CNRS's Deputy CEO for Science (DGDS), explains that "for too long now, research has been trapped by indicators that actually have nothing to do with the intrinsic quality of the scientific advances set out in a publication". The DGDS considers that dictating the rules of the game for the assessment of researchers by defining the prestige and attractiveness of journals is "a reductive view of science that we should no longer be condoning. This means we have to take action now to guarantee the quality and the ethics of the research assessment system".

Promoting the development of open databases

The CNRS began this process in 2024 by unsubscribing from Elsevier's Scopus database and is pursuing its policy today by cutting access to Clarivate Analytics' bibliometric database. Alain Schuhl reminds us that "other prestigious institutions have led the way" with this approach. This decision by the CNRS will help the organisation save €1.4 million annually in subscription fees, funds that will be redirected towards initiatives promoting open science, particularly the development of open databases.

Thus, CNRS researchers are now encouraged to turn to open databases like OpenAlex. This offers a greater level of visibility for non-English-language journals and a larger number of journals than the WoS which was considered to be insufficiently representative for several disciplines. The humanities and social sciences were poorly covered, for example, as were computer science and mathematics. 

Reversing the Web of Science's historically constructed advantage 

Up until now, a major reason for the highly prominent position of the WoS was the collective contribution of public sector staff members who have been enhancing the quality of WoS affiliation data for years. Alain Schuhl explains that "we have worked for free to lock ourselves collectively into a paid system when we are all aware of its biases and incompleteness. Institutions have been trapped in the WoS system for all this time because having the majority of the world's higher education and research institutions listed there means comparisons can be made between them". Leaving the WoS means temporarily losing the ability to compare ourselves with others and position ourselves in relation to our counterparts through such comparisons founded on a common scope for evaluation. Alain Schuhl adds that "this decision means the CNRS therefore accepts that inter-institutional or inter-personal comparisons will now need to be made on the basis of other principles that still need to be reinvented". 

Alain Schuhl is well aware that a lot of work remains to be done in terms of metadata curation before OpenAlex can be used to its full potential. However the DGDS believes "it is now high time to consecrate our energy to improving the quality of open bibliometric databases" and stresses that "the CNRS is actively working to improve OpenAlex. We're confident this database will soon be capable of giving full satisfaction to all scientists in their bibliographic research work."

Emancipation can never be achieved alone

In an era of increasingly fragile international scientific cooperation, the objective is more than ever to regain sovereignty over our scientific choices and decision-making so these are guided by enhanced transparency achieved through using open tools with more transparent methods. "Working collectively on the development of open databases has become a priority and we invite all our partners to also draw a line under the Web of Science", concludes Alain Schuhl.