The eleventh webinar in the CCSD’s “Let’s Talk Open Science” series offered a fascinating exploration of the PathOS and COMMONS research projects. These two pioneering projects examine one of the most important aspects of open science: what happens to research results once they are made openly available?
Beyond merely making publications and data available, these two projects focus on concrete uses, reuses, and impacts on society.
These innovative approaches rely on the analysis of digital traces to observe practices and identify different types of uses and users, including non-academic ones.
PathOS : mapping paths to open science
“To understand the impact, it is necessary to understand the usage.” — Simon Apartis
Simon Apartis, research engineer at the CNRS and head of open science at the Centre for Internet and Society, presented the ongoing work of the PathOS (Pathways to Open Science) project. This European project, funded by Horizon Europe, involves eight organisations in six countries for a period of three years.
The French team, which includes Simon Apartis’ work, is focusing on the case study “Research data and knowledge use in non-academia” which examines French open science infrastructures and their impact on non-academic communities.
The project employs a counterfactual reasoning methodology to address the following question: “In the absence of infrastructures such as HAL and OpenEdition, what might the consequences have been?”
The approach developed is based on the analysis of connection logs from HAL and OpenEdition. These digital traces enable us to observe the 437 million access attempts recorded between January 2023 and September 2024, confirming the global use of these platforms.
The researchers developed a methodology to map “who accesses what.” They classified IP addresses by type of organisation, divided into 22 socio-economic sectors, then categorized resources by scientific field (26 disciplinary fields according to the OpenAlex classification), and analysed the correlations between open access availability and consultation volume.
Preliminary results suggest that opening up resources on HAL generates a significant consultation advantage over “closed” resources. The so-called “hard” sciences (chemistry, physics, medicine) benefit most from open access in terms of number of consultations, while the public, academic, and administrative sectors are most sensitive to the opening up of resources. Future results will supplement and adjust these preliminary findings.
This initial work aims to guide future qualitative research to better understand the transition from access to use, and then to impact.
COMMONS : observing practices from platforms
The COMMONS project, coordinated by OpenEdition, adopts a complementary approach by examining open science practices, uses, and policies through the lens of dissemination platforms. The objective is ambitious: to construct an observatory of uses that will provide a more in-depth understanding of the open science ecosystem by combining quantitative and qualitative analyses.
Mohsine Aabid, a doctoral student in computer science at the OpenEdition Lab, presented the work carried out within the fourth element of the COMMONS project. The aim of the project is to streamline access to OpenEdition services.
The methodology adopted is based on the analysis of connection logs from OpenEdition platforms (Journal, Books, Calenda, Hypothèses) and its search engine.
Once the data is usable, the team can process it using two types of analysis. Transition analysis examines the probabilities of moving from one resource or platform to another, enabling the construction of transition matrices that reveal distinct browsing patterns. Preference analysis is a tool that identifies common interests shared by groups of users. This is achieved by analysing clicks and grouping similar sessions. This process enables the detection of user communities and their characteristics.
Initial analyses have revealed some very interesting browsing patterns. Users tend to remain on the same platform: once on OpenEdition, for example, users generally do not browse between platforms. Some sites act as “hubs,” attracting particular traffic, such as Presses universitaires de Rennes, CNRS éditions, and éditions de l’EHESS.
Analysis of sessions has identified four distinct types of navigation:
- Single-site sessions (navigation concentrated on a single site),
- Linear sessions (direct progression without backtracking),
- Semi-linear sessions (linear start followed by more erratic exploration),
- and complex sessions (multidirectional navigation with numerous back-and-forths).
The data also reveals marked temporal patterns, with peak activity after 12 noon, a decline at night, and a sharp drop in traffic on weekends, confirming the dominant professional and academic use of these platforms.
While the initial findings of this exploratory work are fascinating, they cannot yet be considered conclusive at this stage. We will have to wait a little longer to discover the final results.
A complementary ethnographic approach
Beyond quantitative analysis, the COMMONS project incorporates an ethnographic dimension led by Ioanna Faïta, a doctoral student in information and communication science at OpenEdition Lab at the Elico laboratory, who studies information practices within these platforms. This qualitative approach aims to understand the negotiations between users and technical devices, focusing on prescribed uses as well as misuses and appropriations.
The present exploratory study focuses on the readerships of three publishing entities: the journal Monde des migrations, the journal of the Bibliothèque de National Universitaire de Strasbourg, and three collections published by Éditions de l’Université Grenoble Alpes. The methodological approach combines questionnaires, semi-structured interviews with readers, publishing staff, and technical intermediaries, as well as browsing simulations.
The initial results from the interviews reveal a variety of discovery practices, with Google, Google Scholar and Isidore dominating (consistent with previous studies conducted on OpenEdition’s readership in 2009 and 2012), despite the existence of a search engine specific to OpenEdition. The study differentiates between individual and collective practices of resource mobilisation, demonstrating how a resource becomes information that circulates and transforms, serving personal objectives as well as those of disciplinary or practice communities.
There are three main perceptions that audiences have of resources:
- a fragmented view at the article level (the resource is perceived in isolation, at the article level, without any connection to the rest of the site or journal),
- a perception of the site as a whole (the user perceives the site as a coherent whole, beyond the individual content items),
- and a perception of continuity with the traditional journal (the online journal is perceived as a continuation of the printed journal, with its traditional editorial and periodic logic).
Different perceptions that may echo the navigation paths identified by Mohsine Aabid.
However, the study reveals persistent “confusion” among the readerships studied regarding the infrastructures themselves. Ioanna Faïta identifies confusion between Cairn and OpenEdition, and uncertainty about the roles of the various players in the open science ecosystem.
The testimonies collected demonstrate that the academic and para-academic audiences surveyed understand openness in a synchronic way, without considering the long-term developments of open science. They express a certain uncertainty about the practical modalities of implementing openness, with a tendency to confuse open science with open access, publications, and data.
In pursuit of a more comprehensive understanding of the impact.
By going beyond simple quantitative metrics, these projects provide a better understanding of who actually uses open access resources, how these resources circulate in different socio-economic ecosystems, and what concrete impacts can be measured beyond the academic world.
The innovation lies in the ability to detect unexpected readers and identify peaks in access that reveal new appropriations of open science by non-academic actors.
This approach, which is focused on actual uses, is an important step towards a more refined and accurate assessment of the impact of open science on society.