Frédéric Saby is curator of libraries and head of the “Bibliothèques et Appui à la Science Ouverte” (DGD BAPSO) department, which serves all students and teacher-researchers at the Université Grenoble-Alpes.

Université Grenoble-Alpes implements an open publication, data, and code policy. The Open Science Blueprint is subdivided into four challenges and 21 actions, including challenge number three “Improve support throughout the life cycle of data, code and software.” For example, UGA recommends archiving research codes in Software Heritage, in conjunction with a HAL record. Frédéric Saby discusses both organizational and strategic issues.

Key takeaways

“The role of libraries is evolving, and we need to explain to librarians that Open Science isn’t just about publications: it includes data and software.”

“Software represents an emerging topic but is perhaps not as far removed from the scope of libraries as one might think.”

“Software Heritage meets needs not met by other infrastructures.”

Organizational challenges of open science

How has the library’s role in open science evolved?

Université Grenoble Alpes libraries have been involved in Open Science for many years. A particularly revealing sign of this commitment is the change in the department’s name: four years ago, the documentation department became the Direction générale déléguée chargée des bibliothèques et de l’appui à la science ouverte, (Directorate for Libraries and Open Science)or BAPSO. This change makes a great deal of sense, as it demonstrates in a tangible way that open science is fully within the scope of university libraries. Open Science is taken into account in the functional organization of the department itself.

The department is divided into several divisions, including the Direction d’appui à la recherche et à la science ouverte (Research and Open Science Support Division) or – DARSO. This department is responsible for managing Open Science. The director of DARSO, Lucie Albaret, is also in charge of the institution-wide Open Science mission, alongside Violaine Louvet, research engineer in scientific computing and administrator of research data here at the university.

Work on publications is firmly rooted in librarians’ activities, and the number of open-access publications has steadily increased. Libraries are now working on research data and source code.

How does your institution foster Open Science, particularly for data management and software curation?

Collaboration between the libraries and Gricad, an intensive computing and data infrastructure, has played a decisive role. The university presidency appointed Albaret and Louvet, former director of Gricad from 2016 to 2022, as Open Science project managers. This dual appointment is not neutral, and allows us to benefit from complementary skills: Gricad is home to computer scientists and mathematicians, who contribute their knowledge of software engineering, while BAPSO contributes from its area of documentary expertise.

Another specific feature of the Grenoble context is the existence of a university-wide Open Science unit, to which BAPSO actively contributes. BAPSO is the university’s main contact for Open Science issues.

Behind the scenes of the open science blueprint

What actions have the libraries taken?

In the field of Open Science, our actions can be divided into three components. First, the libraries are dedicated to verifying data consistency and metadata quality in HAL software repositories. This first area of work is dedicated to moderation. Although the software is a special object, the librarians’ tasks are part of a continuum: they involve checking keywords, identifying the production context of the deposited code, working on author affiliations, etc. These tasks are currently integrated into the library’s metadata management system. At present, these tasks are included in the job descriptions of seven people, and probably more in the future, as this activity is set to expand.

The second area of work is the training of researchers and doctoral students. Librarians are involved in the Collège doctoral and Inrae seminars. The librarians also produce practical information sheets on depositing source code. These resources are available on the university website.

Finally, the third area of work is legal issues, in collaboration with the university’s research department. The head of DARSO’s thesis department has been trained in intellectual property issues.

Training and support on legal issues are essential prerequisites for software registration.

Support is a major challenge

What are users’ main needs?

Training is a fundamental need. Beyond HAL deposits, we must educate users about the broader scientific information ecosystem, including Software Heritage. We can’t assume that certain tasks, such as depositing an article in HAL, are self-evident. We need to offer support to researchers.

What does library staff need most?

This is a managerial challenge that falls squarely on library leadership. Convincing staff to participate requires strategic planning and effective communication. Staff may be reluctant to embrace the idea of developing services and skills in new areas. While the subject of publications has become familiar, the same cannot be said for software. The role of libraries is changing, and librarians too need to be made aware that Open Science is not just about publications: it includes data and software.

The deposit of source codes and the legal issues associated with opening them up are part of a university’s research strategy. And the library can support this research policy, playing an active role and even becoming a pillar of this strategy.

Legitimacy and evolution of librarians’ skills

What are the advantages of libraries working on a subject that may still be perceived as far removed from their usual activities?

In the field of publications, the question no longer arises, and the library is seen as the natural player in these issues. Software is an emerging subject, but perhaps not as far removed from the library’s field of action as one might think. BAPSO’s three areas of action mobilize existing skills. The software is treated as a documentary object: librarians know how to process different types of documents and have regularly upgraded their skills. Librarians have been involved in training for over ten years. Although source code differs from books, it too requires documentary processing. However, it should be pointed out that in the case of software, more than for other types of documents, the target audience is more restricted. But the dynamic remains the same, as seen from the questions raised about intellectual property. From the outset of the Open Science master plan, the libraries were recognized as the primary point of contact for other university departments.

Complementing library services

If you were to talk to another library director, how would you describe Software Heritage’s contribution to his or her department?

Software Heritage meets needs not met by other infrastructures. This logic of complementarity is comparable to the partnership between BAPSO and Gricad. Some institutions choose to develop their own tools, but delegating expertise is a better option. In the early days of the library management system (LMS), some university IT departments chose to develop local solutions rather than rely on specialized service providers. The problem is similar to source code.

Librarians and their software stories: Monocle

Saby recounts that in the 1960s, Marc Chauveinc, then-Director of Libraries at Grenoble University, developed an LMS called Monocle. The library’s software archives offer a glimpse into the past, potentially illuminating the evolution of software development at the school. This historical record could shed light on the subject for new generations of IT specialists, marking the beginning of a new software narrative.

Further reading

The university’s commitment to Open Science:

How to  promote the HAL software repository

Since 2018, a collaboration between the CCSD, Inria, and Software Heritage has enabled software deposits in HAL to be transferred to Software Heritage. Numerous resources on software deposit are available:

• Resources for end users:
  • http://doc.hal.science/deposer/deposer-le-code-source/#
  • https://youtube.com/playlist?list=PLD2VqrZz2-u3bOWtoCoBIh5Flt6iYXsq3&feature=shared  
• Resources for librarians: https://inria.hal.science/hal-01876705
• Document for research team leaders: http://doc.hal.science/essentiels/#le-depot-de-logiciels-software-deposit

The post Collaborative Open Science at Université Grenoble Alpes appeared first on Software Heritage.

The stereotype of libraries as quiet, outdated places, often symbolized by a scolding action figure (a tribute to Nancy Pearl), persists in popular culture. However, librarians’ daily work focuses on shaping the future. That’s why in 2018, the Association of Research Libraries (ARL) highlighted the increased need for collaboration to ensure efficient and equitable access to software: “Ensuring equitable and efficient access to software and software-dependent materials will increasingly require collaboration and resource sharing.” This collaboration can involve library support for open science infrastructures.

Cécile Swiatek Cassafieres is a member of the Executive Board of the League of European Research Libraries (LIBER) and sits on the Board of SPARC Europe. In particular, Cécile Swiatek Cassafieres contributed to the Unesco recommendation on Open Science and its toolbox, on behalf of LIBER. Her publications supporting Open Science infrastructures are available on ORCID.

Cécile Swiatek Cassafieres.

Key takeaways
“When it comes to open science, research infrastructures, and the open dissemination of research results, universities have every interest in being represented by their libraries. As trusted third parties, libraries are widely identified internationally as ‘engaged and trusted hubs,’ as LIBER puts it.”

“Preserving and providing long-term access to source code is an extension of libraries’ traditional missions: it’s about making knowledge available for the benefit of all. We need to provide access to both research results and the methods used to produce them.”

“Librarians and specialists in scientific and technical information need to take action on software and code to raise awareness of these issues […].”

Generally speaking, why should libraries support research infrastructures?

Academic and research libraries can provide two types of support: the first is financial support, to contribute directly to the maintenance, health, and sustainability of the infrastructure. Libraries can also provide support by getting involved in governance bodies. The aim is to make the voice of higher education and public research institutions heard through the discourse of libraries, known for their reliability, reflexive feedback, and organizational stability.
Libraries’ participation in governance bodies adds value to their institutions. When it comes to Open Science, research infrastructures, and the open dissemination of research results, universities have every interest in being represented by their libraries. As trusted third parties, libraries are widely identified internationally as ‘engaged and trusted hubs,’ as LIBER puts it. One of the challenges for libraries is to be identified as key players. This means relaying their institution’s strategy and participating in its development.

What criteria should organizations consider when selecting infrastructures to support?

It’s important to ensure that the funded structure meets national, European, and international quality criteria. Following the University of Paris Nanterre’s example, we can utilize the National Fund for Open Science’s criteria and the selection processes of Scoss and IOI. The challenge is to ensure that the structure to be financed is aligned with the university’s strategy: while the use of standards and other technical criteria guaranteeing good interoperability with other systems are essential points, it is also essential to take an interest in the values and principles defended by the infrastructures.
How can libraries support research software through infrastructure like Software Heritage?

Preserving and providing access to source code is a natural extension of libraries’ traditional role: to make knowledge available for the benefit of all. We need to provide access to research results and the information that went into producing them. This content must be readable, in the original sense of the word. This is a first step, but some libraries go further, linking the question to the field of data management.

A 2023 UNESCO report (Open Science Outlook 1) identifies a lack of a standardized framework and uneven understanding of open science themes as obstacles to resource development. If we focus on software in research, where do we stand in libraries in France and abroad?

In Europe, situations vary widely from one country to another. However, we can see that research support services, including those outside libraries, are becoming aware of the need to extend the Open Science policy to all components of academic knowledge. This first stage of awareness has been reached. In Europe, the EOSC is building a framework conducive to a homogeneous, standardized approach for example through the EVERSE project from 2024 to 2027.
Finally, Unesco plays a fundamental role at the international level, in particular with the Open Science Monitoring Initiative (OSMI) which is being set up, and with countries and regions of the world where open science is still an emerging issue.

What are the strengths of libraries when it comes to working on a subject that may still be perceived as far removed from their usual activities?

It’s in librarians’ best interest to promote their key missions to show their richness and diversity: a library is more than just reading rooms and printed books. More broadly, libraries aim to document, describe, and provide access to this documentation. Libraries are resource providers, service platforms, and living spaces: as far as code and software are concerned, they play above all the fundamental role of access services and information centers.
To develop this aspect of libraries and make it more visible, I think it’s important to open up discussions within institutions, between libraries, research departments, and political teams, to define a strategy and implement actions around these emerging themes. This initial step will enable the production, dissemination, and promotion of institutional open science resources and strategies.

If you were to talk to another library director, how would you describe Software Heritage’s contribution to their department?

One must assume that working on software should be seen as something obvious, if you want to avoid limiting thoughts. We have a number of success stories and, I hope, indicators of the adoption of software and code-related topics in libraries. Librarians and specialists in scientific and technical information need to take action on software and code to raise awareness of these issues, make these developments understandable to teams to facilitate their mobilization on these issues, and enhance the value of the actions taken by the documentary structure on these issues in the eyes of all. It’s an approach that can be described as inclusive.
Peer exchanges play a major role in the circulation of ideas and influence. We share the same professional references and the same culture. Having librarians as ambassadors for Software Heritage is certainly a strategy worth considering for spreading the word about Software Heritage in the world of academia, knowledge, and research.
From a personal viewpoint, I’d say that the library, which is in charge of supporting infrastructures that foster open science at the University of Paris Nanterre, values Software Heritage’s open approach through the support we’re giving it on behalf of the university from 2025 onwards. The subjects on which Software Heritage is working are in line with those of the most advanced libraries on open science issues. In France, the Ateliers de la donnée (data workshops) focus on the link between software and data, and libraries are very active and regularly lead the way in this field.
To conclude, I think it’s essential to question the practices and tools used by doctoral students. Inviting them to talk about the codes and software they create, modify, and use in the course of their work is certainly a first step towards moving forward together on this subject: young researchers and librarians are often the initiators of dialogue on emerging subjects.

Going further

Support for open science infrastructures at the University of Paris Nanterre:
https://science-ouverte.parisnanterre.fr/strategie-science-ouverte/soutien-aux-infrastructures-de-la-science-ouverte

“Supporting and taking strategic participation in open science infrastructures at the University of Paris Nanterre” https://zenodo.org/records/8398184

Open Science Infrastructure Support from the University of Paris Nanterre (UPN):  https://zenodo.org/records/8134805

University of Paris Nanterre Open science strategy: https://zenodo.org/records/7115904

The University of Paris Nanterre academic libraries website:
https://bu.parisnanterre.fr/le-service-commun-de-la-documentation-scd

Open science infrastructures: https://hal.parisnanterre.fr/hal-04142253

“Sustainable and collaborative development of research support services at the university library of Paris Nanterre (France):” https://zenodo.org/records/10559949

Unesco recommendation on open science: https://unesdoc.unesco.org/ark:/48223/pf0000379949_fre

Unesco’s Open Science tools and resources: https://zenodo.org/communities/unesco-science/

The post How libraries shape the future of research infrastructure appeared first on Software Heritage.