Multi-method analysis shows stable productivity for core developers amid shifting OSS contributions during COVID-19.
The social isolation measures resulting from the COVID‐19 outbreak changed work practices in various sectors, especially with the shift to working from home. However, the implications of the pandemic on the maintenance and evolution of open‐source software (OSS) still deserve further studies. In this paper, we analyze the effects of COVID‐19 on the development activity of OSS and how social isolation changed the productivity of OSS contributors. We conducted a mixed‐method study composed of (i) a mining software repositories analysis of 155 popular and active OSS projects on GitHub, selected from an initial dataset of 1500 repositories based on activity thresholds (commits, pull requests, and size), and (ii) a survey with 57 core developers identified using an established literature‐based heuristic. The mining study analyzed commits, code churn, pull requests, and pull request latency to assess changes before and after the pandemic, applying statistical tests and a mixed‐effects Regression Discontinuity Design. The survey collected self‐reported perceptions of productivity and engagement during the pandemic, enabling triangulation with repository activity trends. Our results show that while core developers' productivity remained stable, there was a sustained decline in newcomer participation and a temporary increase in core developer turnover. In the early days of the outbreak, we observed an increase in accepted pull requests, followed by a stabilization of most activity metrics. Some findings are supported by our survey study, whose results indicate that most of our survey respondents consider that COVID‐19 did not change their productivity substantially. These findings offer insights into OSS resilience and sustainability in the face of large‐scale disruptions, contributing to a broader understanding of the outbreak's impact and providing actionable lessons for managing distributed development in crisis scenarios.
No takes yet. Share an insight, caveat, or question.
Okimoto et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: