Radical reproducibility, real constraints: An autoethnography of open and transparent research from the inside

BACKGROUND: This paper presents a three-year longitudinal autoethnographic study of the TIER2 project, an international, interdisciplinary consortium committed to "radical reproducibility" through open research practices. We investigated how epistemic diversity, disciplinary norms, and institutional cultures shape reproducibility in practice. METHODS/MATERIALS: We take an auto-ethnographic approach. Data include periodic consortium-wide surveys, quarterly reproducibility diaries by five diarists, and fieldnotes from General Assembly discussions; these were coded inductively. RESULTS: We found that initial disciplinary differences in defining reproducibility evolved to a nuanced appreciation of its complexity. Participants developed new competencies in open research practices and experienced insights regarding early planning and collaborative transparency. Enablers of reproducibility include the Open Science Framework, co-developed tools, containerization, and "slow science" approaches, strong role modeling by senior researchers and improved documentation. Challenges include fear of exposing imperfect work, unequal engagement across career stages, and substantial time and resource demands. Epistemic tensions emerged around the applicability of reproducibility to qualitative research and the risk of standardization undermining diversity. Despite these, participants reported high professional satisfaction and intellectual growth. CONCLUSION: The study demonstrates that "radical reproducibility" is not merely a technical but a cultural and epistemic project that fosters collective learning, skill development and innovation, and requires systemic and institutional support.