Preserving, Referencing, and Valuing Research Software

Preserving, Referencing, and Valuing Research Software Morane Gruenpeter, Director of Scholarly Ecosystem GTSO@Couperin | Preserving Research Software | 16/12/2025 | CC-BY 4.0 What is software? Magic! Software as a concept ●project or entity ●the community around the project ●the software idea / algorithms / solutions Not a digital artifact Software artifacts ●Executables ○For multiple environments ●Source code A very large collection of digital artifacts GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #2/21 The Knowledge is in the Source Code “Source code provides a view into the mind of the designer.” Len Shustek, 2006 Computer History Museum https://en.wikipedia.org/wiki/Binary_file GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #3/21 “Programs must be written for people to read, and only incidentally for machines to execute.” Harold Abelson, 1985 Structure and Interpretation of Computer Programs (1st ed.), Apollo 11 source code Binary Source Code Source code is fragile! 1.4 million projects GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #4/21 What’s the the Software Collapse? Konrad Hinsen. Dealing With Software Collapse. Computing in Science and Engineering, 2019, 21 (3), pp.104-108. ⟨10.1109/MCSE.2019.2900945⟩. ⟨hal-02117588⟩ Research Software is a thin layer on top of the global software stack https://xkcd.com/2347/ The software stack: More than Research Software Can be preserved in a: - Scholarly Infrastructure - Universal Source Code Archive FRAGILE GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #5/21 Collect, preserve and share all software source code Preserving our heritage, enabling better software and better science for all Snapshot:Dec. 4th 2025 https://archive.softwareheritage.org/ GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #6/21 Open Science & Open Source in France Source: Ministère de l’enseignement supérieur et de la recherche GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #7/21 Archive ➔make sure we can access to retrieve the software (reproducibility) Reference ➔make sure we can identify the software artifacts (reproducibility) Cite (for credit) ➔make it rewarding to create software by giving credit to authors (evaluation!) Describe ➔make it easy to discover the software projects (visibility) What is at stake? GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #8/21 Researchers ●archive and reference software ●find useful software ●get credit for software ●verify/reproduce/improve results Laboratories/teams ●track software contributions ●produce reports ●maintain web page Research Organization know its software assets for: ●technology transfer, ●impact metrics, ●Strategy Curators ●verify and curate software metadata ●provide documentation on software curation ●monitor research teams’ production What are the use cases? GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #9/21 Catalogs, registries, aggregators ● ASCL ● CORE ● OpenAire ● OpenAlex ●libraries.io => Ecosyste.ms ● RSD - eScience center ● swMath ● … Scholarly repositories ● Zenodo (InvenioRDM) ● HAL ● … Scholarly publishers (on record) ● IPOL ● eLife ● Dagstuhl ● Episciences ● … Software development platforms (on page) ●GitHub ●Bitbucket ●SourceForge ●… Package managers ●PyPI ●NPM ●… In the source code (as a file) ●README ●LICENSE ●AUTHORS ●SBOM ●Package manager file ●codemeta.json / CFF file ●… Intrinsic metadata Describe & Cite: where is the metadata? Citation / Mentions in Research Articles ● Softcite - extracting 3 types ○ Created ○ Used ○ Shared GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #16/21 SWH a metadata broker: BibTeX citation case BibTeX export @sofwtare @softwareversion @codefragment Citation docs Citation blog-post GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #17/21 https://annex.softwareheritage.org/public/talks/2025/2025-09-08-OCDE-RS.pdf Monitoring at scale the software production GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #18/21 Supporting SWH with the Archives and Libraries Interest Group (ALIG) Open infrastructure Contribute to a 100% Open Source Infrastructure, valuing users input and support Tailored Support Receive assistance for your software archival questions, plans and projects Community Participate in an inclusive, innovation driven community ALIG Members https://www.softwareheritage.org/support/members-alig/ GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #19/21 Call to action: Software requires coordinated efforts ❏Build on and support mutualized, shared, open, non-profit infrastructures, ➢like Software Heritage #AvoidTechnicalBalkanisation ❏Contribute to qualified metadata ❏Archive early, archive often, even small scripts! ❏Rely on identifiers built for software #SWHID ❏Advocate for formal recognition for code sharing in funding and promotion activities 🌶 ❏Set the default to Open Source 🌶 ❏Publicly funded research software should be open source, exceptions must be justified GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #20/21 Slide from https://doi.org/10.5281/zenodo.8138126 Join the effort [email protected] https://www.softwareheritage.org/ Browse the source code archive https://archive.softwareheritage.org/ Become a member https://www.softwareheritage.org/support/members-a lig Volunteer as an ambassador https://www.softwareheritage.org/ambassadors/ Subscribe to the newsletter https://www.softwareheritage.org/newsletter/ GTSO@Couperin | M.Gruenpeter | Preserving Research Software | 16/12/2025 | CC-BY 4.0 | #21/21 Links to be shared with your users: 🔑 Guidelines for researchers: https://www.softwareheritage.org/howto-archive-and-reference-your-code/ 🏷 Describe a software (intrinsic metadata made easy) : https://codemeta.github.io/codemeta-generator/ ❓ The FAQ: https://www.softwareheritage.org/faq/ 🗺 The SWH guided tour: https://archive.softwareheritage.org/?guided_tour=0&guided_tour_next=https://a rchive.softwareheritage.org/ 🎯 "Open Science tutorial: source code deposit" (how to build services upon SWH): https://youtube.com/playlist?list=PLD2VqrZz2-u3bOWtoCoBIh5Flt6iYXsq3 👋 “Book” a SWH ambassador: https://www.softwareheritage.org/ambassadors/ The Software Metadata Curation Roadmap Short-term (0-2 years) 1. Support and training: Academic institutions should invest in curation and training activities for researchers and support staff, acknowledging the significant effort required. 2. Infrastructure curation capabilities: Infrastructures encompass a wide range of platforms, including aggregators, publishers and scholarly repositories. 3. Adopting and adapting metadata guidelines: Institutions should require infrastructures to provide metadata capabilities that align with community-based guidelines, such as the FAIR-IMPACT Research Software MetaData (RSMD) guidelines, CodeMeta, and/or CFF metadata standards. 4. Community effort: Infrastructures should actively engage in community-driven efforts (e.g., the SciCodes consortium) to develop and implement standards, guidelines, and best practices. 5. Recognition & acknowledgement: ○Career evaluation: Institutions should integrate curated metadata records into activity reports to highlight software as a recognized and valued research output, using this information in the career evaluation of researchers as an incentive. ○Citation standard: Strengthening the connection between researchers and their software outputs by using the BibTeX @software type in articles and ensuring that software is properly cited and credited in scholarly work. 10.5281/zenodo.14509418 Read the full report The Software Metadata Curation Roadmap Medium-term (2-5 years) 1. Adoption: Institutions and funders should support the adoption of infrastructures that propose curation capabilities to ensure the maintenance and sustainability of these infrastructures. 2. Robust curation processes: Stakeholders should implement robust curation processes utilizing the capabilities of infrastructures. 3. Automation: Automation should be employed wherever possible to streamline the curation process and reduce the manual burden on researchers and curators. 4. Interoperability: ○Identifiers: Infrastructures should support both intrinsic (e.g., SWHID) and extrinsic (e.g., DOI) identifiers for software identification and ensure that Software Hash Identifiers (SWHID) are exposed for archived resources whenever possible. ○Exposing metadata: The adoption of standard APIs is crucial for making metadata harvestable across different platforms, thereby enhancing interoperability. 5. Feedback mechanisms: The research community should establish regular feedback channels, similar to those used in Open Source communities, to enable users to report issues and suggest improvements of functionalities and workflows to the infrastructures. The Software Metadata Curation Roadmap Long-term (5-10 years and beyond) 1. Monitoring: Institutions should actively monitor the software production within their laboratories to ensure the quality, visibility, and impact of research outputs. Effective monitoring relies on the curation capabilities of the infrastructure in use, which must be equipped to track, manage, and preserve software throughout its lifecycle. 2. Maintenance of community standards: Funders and institutions should provide monetary and human resources to continue maintaining community efforts, such as CodeMeta, CFF and SWHID. 3. Sustainability: ○Define and communicate measures: Infrastructures should define and communicate their sustainability measures, including governance, retention, and end-of-life policies, as suggested by the SciCodes best practices (Task Force on Best Practices for Software Registries et al., 2020) and POSI principles (Bilder G, Lin J, Neylon C, 2020). ○Establish funding models: Funders should establish funding models to ensure sustainable support for projects, infrastructures, and training initiatives. ○Foster institutional collaboration: Institutions should actively collaborate across national and international levels to share resources, knowledge, and expertise, enabling sustainable practices and reducing duplication of effort.