Information Practices and Knowledge in Health. Группа авторов

it, and there is a risk that it does not sufficiently take into account the ethical risks that are their corollary.

1.7. Scientific information in Health: a world of no journals?

      Preprint servers are now an integral part of the scientific communication ecosystem in Health [PEN 20]. They have appeared over the last 10 years and illustrate the dynamism and importance of this method of dissemination [TEN 18; PEN 20]. Supported and set up by researcher communities, particularly early-career researchers [NIC 19], these servers represent a new generation of thematic open archives that offer an advanced digital service around preprint.

      The most famous are those grouped around the extension “Xiv” of the eponymous archive arXiv, which founded the Green Road of Open Access. Using this extension in their names as a distinguishing mark conveying values, the new preprint servers claim both a filiation and a specific identity [BOU 19b]. In addition to openness, a discursive marker of Open Access, their rhetoric also includes “acceleration” as a strong justification, which indirectly points to the unsuitability of journals to ensure publication within a sufficiently short period of time.

A representative example is bioRxiv.org, a preprint server launched in 2013 in the biomedical sciences. It now lists over 77,800 preprints and has grown exponentially since its inception¹⁶. More recently, in 2019, we saw the birth of medRxiv.org in the medical field. A recent study lists up to 44 preprint servers within the Health domain [KIR 20], attached to funding agencies, public research actors or even scientific communities. These servers have attracted the attention of funding agencies that take into account the articles posted to these platforms. This started with the NIH, which promotes the appropriation of bioRxiv and medRxiv by researchers [ABD 19]. The researchers themselves are on board and have shown unwavering enthusiasm since 2018 [ELS 19].

The interest of these platforms is that they have become spaces for exploration and experimentation. In a “Bibliodiversity” approach¹⁷ aiming at diversifying the models of Open Access publication, preprint servers, equipped with proofreading functionalities (comments, open peer review, etc.), alternative metrics or Altmetrics (mentions in social networks, downloads, etc.) make it possible to consider new ways to shake up the “old” model of the scientific journal [BOU 17b]. Thus, it is not insignificant to see that one of the preprints on the treatment of Covid-19 by Professor Raoult, available through biorXiv, has been downloaded more than 600,000 times, giving its author the opportunity to talk about it in TV interviews, presenting this extensive consultation as a guarantee of relevance and scientific credibility.

In the same vein, preprint repositories are becoming observation sites for scientific publishers and for research funding agencies. Publishers identify articles that are “buzzing” and offer to submit them to their journals [PEN 20]. Or, conversely, articles submitted to journals are automatically uploaded and shared on preprint servers like PLoS journals that allow authors to directly deposit the submitted article on bioRxiv¹⁸. On the other hand, bibliometric databases, such as Dimensions, or the academic search engine Google Scholar, are starting to take these servers into account and are adapting their algorithms for counting article citations. Authors have in fact integrated the issues of visibility and impact into their strategies for valorizing their research through, among other things, preprint repositories such as bioRxiv [SEV 19]. Fu and Hughey [FU 19] have thus demonstrated that:

      Articles with a preprint [on bioRxiv] had, on average, a 49% higher Altmetric Attention Score and 36% more citations than articles without a preprint.

      These preprint platforms are now attracting the attention of private funding agencies, which are showing increasing interest in new Open Access publishing models. In 2016, the Wellcome Trust decided to launch its own publishing platform: Wellcome Open Research [BUT 16].

The idea behind Wellcome Open Research is to allow Wellcome grant recipients to publish their findings more quickly and to create a model that, according to the charity, other funders might adopt in future.

In addition, as of 2019, the NIH has established an Open Access publishing pilot in partnership with the publisher PLoS to cover the cost of publishing in PLoS¹⁹ journals. The Bill & Melinda Gates Foundation has also set up a publishing platform, Gates Open Research, which, like that of the Wellcome Trust, aims “to accelerate the publication of articles and data from research funded by the charity”.

      These initiatives are part of the processes of platformization [MIR 18] currently operating in scientific communication and are based on the following elements: rapid dissemination and publication of research results (concerns and aspirations found in both the open archives movement and mega-journals), upstream funding of articles by research programs, open science methods (release and publication of research data; open peer review) and, finally, new forms of mediatization of scientific and technical information as previously specified. This platformization questions the place of traditional health journals insofar as they are no longer the unique way to disseminate research results.

      Large national funding agencies and private research funding organizations are thus helping to (re)shape the scientific communication landscape with Open Access publishing platforms [ROS 18].

1.8. Conclusion

      The field of Health is currently in a period in which the issues of publishing and disseminating validated scientific information have perhaps never been so crucial. The Covid-19 health crisis has demonstrated its importance as well as its complexity.

Open Access is largely part of this complexity, through a political injunction, which is carried by not only public research organizations but also public and private funding agencies. The latter now intervene directly in the determination of dissemination models and in the legitimization of new models of scientific communication. What we are witnessing in Health is the redefinition of a model of scientific publication that is more directly linked to the organization and financing of research.

      The Covid-19 pandemic has suggested possible channels of scientific communication that are becoming independent of the journal model in favor of near-real-time dissemination of articles on preprint servers. Research “in the making” requires an acceleration for which journals are not ready; it therefore turns to platforms that play the role of intermediary device, facilitating and accelerating dissemination.

The question of scientific validation, and thus of peer review, remains the ultimate lock allowing journals to keep their central place in the scientific communication system, and giving information its guarantees of integrity and credibility. But here again, recent events, such as the LancetGate affair, illustrate^20,21 to what extent the walls of scientific publishing are becoming increasingly cracked in the field of Health. The rise of predatory journals and the phenomenon of Fake Science in recent years question more broadly the status of scientific soundness, integrity and credibility of scientific and medical information that circulates on the Web and that affects both scientific communities and the general public.

      Open Access to scientific information in Health has not only improved the dissemination and circulation of scientific information but also reshuffled the cards of the game between stakeholders, allowing some of them, notably the research communities and funding agencies, to take a more important part in the definition of models that are better able to take care of the imperatives of contemporary scientific practice.

1.9.