Why does code reproducibility matter and how to achieve it? Presentation from 4TU.ResearchData Roadshow at OSC/e meet-up, TU Eindhoven
Abstract
4TU.ResearchData is an international data repository service for the science, engineering and design domains. Every year, we hold roadshows at our partner institutions to introduce our services and provide hands-on support. On 4th November 2025, 4TU.ResearchData together with Dutch Reproducibility Network and TU Delft Digital Competence Centre, gave a presentation about the importance of the code reproducibility and explained how it can be supported by CODECHECK. We have also discussed how 4TU.ResearchData repository has introduced the CODECHECK as service for their software publications. The presentation was completemented by a live demonstration of reproducibility check with the CODECHECK method.
Full text
Why does code reproducibility matter and how to achieve it? 4TU.ResearchData Roadshow Yasel Quintero Lares Daniela Gawehns Aleksandra Wilczynska 4 November 2025 OSC/e lunch meet-up
nice to MEET YOU ! Aleksandra Wilczynska Data Solutions Engineer, 4TU.ResearchData Daniela Gawehns Network Coordinator, NLRN Yasel Quintero Lares Research Software Engineer, TU Delft Digital Competence Center A
Before we START Do you think another team could repeat your results based on the description in methods section or your paper? Was it more difficult than you thought? Do you think someone else with access to your code and data would get the same results that you got? Have you ever tried reproducing results from a scientific paper? Do you think it is important to reproduce results from a scientific paper? A
presentation OUTLINE We’ll discuss importance of reproducibility in research and how this can be facilitated via CODECHECK initiative supported by 4TU.ResearchData repository. We will follow-up a live demo. Why Reproducibility Matters? 01 CODECHECK as service 03 LIVE DEMO 04 CODECHECK in Action 02 A
WHY REPRODUCIBILITY MATTERS? 01. D
Revisiting DEFINITIONS re·pro·du·ci·ble adjective /ˌriː.prəˈduː.sə.bəl/ A result is reproducible when the same analysis steps performed on the same dataset consistently produces the same answer. The Turing Way project illustration by Scriberia. Used under a CC-BY 4.0 licence. DOI: The Turing Way Community & Scriberia (2024). D
Why care about REPRODUCIBILITY TRACK PROJECT HISTORY WRITE, EFFICIENTLY BE TRANSPARENT GAIN VISIBILITY AND IMPACT FACILITATE COLLABORATION ENSURE CONTINUITY OF WORK COMPLY WITH REQUIREMENTS AND POLICIES Based on The Turing Way Community. (2022). The Turing Way: A handbook for reproducible, ethical and collaborative research (1.0.2). Zenodo. DOI The Turing Way Community (2025). Reproducible Research: Added advantages D
What are the BARRIERS TO REPRODUCIBILITY LIMITED INCENTIVES PUBLICATION BIAS HELD TO HIGHER STANDARDS THAN OTHERS NOT CONSIDERED FOR PROMOTION TAKES TIME REQUIRES ADDITIONAL SKILLS Based on The Turing Way Community. (2022). The Turing Way: A handbook for reproducible, ethical and collaborative research (1.0.2). Zenodo. DOI The Turing Way Community (2025). Reproducible Research: Barriers to reproducibility D
CODECHECK IN ACTION 02. D
FAIR DATA FUND 5th Edition of Find out more edu.nl/89m39 up to 5K EUR per grantee What is it Fund to make existing datasets (more) FAIR. It’s not for creating new data. Who can apply Researchers affiliated with TU Delft, UT, TU/e, or WUR. The dataset must originate from such an affiliation. 3 NOV 4 JAN A
Science Engineering Design DATA SOFTWARE 100GB data uploads per researcher per year LONG-TERM PRESERVATION guarantee of minimum 15 years preservation DATA CURATION for every dataset & software DISCIPLINE-SPECIFIC FORMATS specialised services and workflows TRUSTED REPOSITORY accredited with CoreTrustSeal badge Welcome to our community 01 DEMO 03 FAIR data in practice 04 Q & A 05 https://data.4tu.nl A
Introducing PILOT 4TU authors who published code at 4TU.ResearchData repository can request CODECHECK. At TU Delft, DCC Team performs codecheck as a service. For other TUs - volunteers. A
CODE SUBMITTED FOR PUBLICATION REQUEST FOR CODECHECK Via metadata form PRESCREENING 4TU.RD checks for necessary conditions: published paper and data RECOGNITION Once done certificate published on Zenodo; 4TU.ResearchData publication receives “CODE WORKS” badge What is CODECHECK AS SERVICE PILOT WORKFLOW At 4TU.ResearchData repository CODECHECK Codechecker tries running code, contacting authors, if needed A
RESULTS so far 6 BADGES A
Next and New Plans Expand service pilot All 4TUs involved. Invite integrated in metadata form Integrate in publishing process Collaborate with diamond OA journals Impact Monitoring Impact study on policy relevant insights Codecheckers community Grow (pro) community; Provide training
How to REQUEST A CODECHECK? SANDBOX DEMO next.data.4tu.nl/ A
WILLINGNESS & BARRIERS TO ENGAGE Welcome to our community 01 DEMO 03 FAIR data in practice 04 Q & A 05 🔸Would you be willing to have one of your own papers be checked for reproducibility by someone else? 🔸 Why? 🔸 Why not? D
LIVE DEMO 04. Y
Join LIVE NOTES edu.nl/tudnf Y