Nudging Scientists into adopting Open Science Practices
Abstract
While funding agencies typically require data management plans for project proposals, this does not guarantee the eventual availability of Open Research Data (ORD) that adheres to the FAIR principles. In reality, only a small fraction of funded projects fulfills these commitments, primarily because providing ORD demands additional effort. Often, this effort occurs only after the publication of research results. Large international collaborations, such as the CMS Collaboration at CERN's Large Hadron Collider, face similar challenges. There is concern that requiring the sharing of analysis code and ensuring the reproducibility of research products could delay publication. However, an increasing number of scientists within CMS are adopting Open Science (OS) practices as part of their daily work. In my presentation, I will illustrate how a small group of OS advocates has successfully made these practices appealing to the community, transforming bystanders into OS practitioners. I will also review the vital role of infrastructure providers — specifically CERN as the host laboratory — in facilitating these efforts. Additionally, I will link my argument to CERN's OS policy and the recently developed best practice recommendations for OS and data preservation in high-energy physics. Ultimately, I will propose how these policies and recommendations can be adapted to benefit OS in other laboratories and research disciplines.
Full text
Nudging Scientists into adopting Open Science Practices Clemens Lange (Paul Scherrer Institute PSI) Open Science Fair, 16th September 2025 © 2025. This work is openly licensed via CC BY 4.0
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 2 The experiments at the Large Hadron Collider at CERN are already very good at Open Research Data (ORD).
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Data Management Plans Findable Accessible Interoperable Re-usable 3 e.g.
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Data Management Plans Findable Accessible Interoperable Re-usable 3 ✅ ✅ ✅ ✅ e.g.
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Data Management Plans Findable Accessible Interoperable Re-usable 3 ✅ ✅ ✅ ✅ “We follow the CERN and experiment data management plans and open data policies” e.g.
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 From Data to Publication 4 Event Generation Reconstruction Simulation Theory model Detector Conditions Database Experiment Data Reproducible Internal Documentation Preserved Open Access Open Access MachineReadable Results
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 5 This is actually bad for the field.
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 6 Why?
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 From Data to Publication 7 Event Generation Reconstruction Simulation Theory model Detector Conditions Database Experiment Data Reproducible Internal Documentation Preserved Open Access Open Access MachineReadable Results
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Closing the Reproducibility Gap 12 Event Generation Reconstruction Simulation Theory model Detector Conditions Database Experiment Data Reproducible Internal Documentation Preserved Open Access Open Access MachineReadable Results
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Closing the Reproducibility Gap 12 Event Generation Reconstruction Simulation Theory model Detector Conditions Database Experiment Data Reproducible Internal Documentation Preserved Open Access Open Access MachineReadable Results Policies Incentives Tooling/Software Skills Training Infrastructure Competition
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Policies Reality vs. aspirational goals An existing policy avoids unnecessary and exhausting discussions (e.g. CMS+CERN Open Data policies → clear agreement) 👍 It still needs someone to act… 13
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Incentives (for the Researcher) Intrinsic: >Scientific integrity, ethical responsibility >Confidence in results, personal efficiency1 Extrinsic: >Prizes/Recognition (Swiss National ORD Prize) >Career advancement, citations2, visibility >Avoiding negative consequences 14 1 see Training 2 see Competition
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Training Should be at the centre of each Open Science initiative! ORD skills = Industry skills >Software containers >Expertise in Continuous Integration/Deployment >Workflows → Active field, work can have impact beyond academia 15 https://hsf-training.org/training-center/ Do: >Train researchers early on directly applicable skills Don’t: >Create yet another Git course
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Training 16 Analysis Preservation Bootcamp
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Training 16 Analysis Preservation Bootcamp I can run my code anywhere! My code is doing what I think it should do I can reproduce my results
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Training 16 Analysis Preservation Bootcamp I can run my code anywhere! My code is doing what I think it should do I can reproduce my results Well-trained (and successful) researchers become ambassadors and trainers
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Tooling/Infrastructure 17 What are incentives for researchers to use your platform? For quite a while, they will just lose time… (Mind also: engineers typically 100% on the project while scientists help out with a small fraction since they need to advance their careers → perform actual research)
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Tooling/Infrastructure >Use existing tooling (also from other research domains), only if needed develop own tooling >Train scientists in using tools and infrastructure Provide support → Potential to make them your advocates 18
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 Conclusions 23 A handful of motivated people can make a big difference Facilitate collaboration between platform providers and users Create actionable policies Train researchers early on Make open science interesting and rewarding Use competition to enforce policies
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 SNSF DMP Requirements […] grantees are obliged to make available to the public in an appropriate manner the research results obtained with the help of SNSF funding, […]” (see https://www.snf.ch/media/ en/lCCrvpOHZ38Hbg5Y/allg_reglement_16_e.pdf, Article 47) The SNSF favours a bottom-up approach. It provides best practice guidelines and gives each scientific community sufficient flexibility in defining and applying its own standards. In particular, the best way of managing and sharing data depends on the research field. 25
Clemens Lange — Nudging Scientists into Adopting Open Science Practices 16.09.2025 ERC DMP Requirements How to make “the project data” FAIR >SUMMARY: dataset reference and name; origin and expected size of the data generated/ collected; data types and formats >FINDABLE: dataset description: metadata, persistent and unique identifiers e.g., DOI >ACCESSIBLE: which data will be made openly available and if some datasets remain closed, the reasons for not giving access; where the data and associated metadata, documentation and code are deposited (repository?); how the data can be accessed (are relevant software tools/methods provided? >INTEROPERABLE: which standard or field-specific data and metadata vocabularies and methods will be used >INCREASE RE-USE: what data will remain re-usable and for how long, is embargo foreseen; how the data is licensed; data quality assurance procedures >RESOURCES: estimated costs for making the project data open access and potential value of long-term data preservation; procedures for data backup and recovery; transfer of sensitive data and secure storage in repositories for long term preservation and curation 26