Background: The emergence of the COVID-19 pandemic has impacted all human activity, including software development. Early reports seem to indicate that the pandemic may have had a negative effect on software developers, socially and personally, but that their software development productivity may not have been negatively impacted.

Aims: Early reports about the effects of the pandemic on software development focused on software developers’ well-being and on their productivity as employees. We are interested in a different aspect of software development: the developers’ public contributions, as seen in GitHub and Stack Overflow activities. Did the pandemic affect the developers’ public contributions and, of so, in what way?

Method: Considering the data from between 2017 and till 2020, we study the trends within GitHub’s push, create, pull request, and release events, and within Stack Overflow’s new users, posts, votes, and comments. We performed linear regressions, correlation analyses, outlier analyses, hypothesis testing, and we also contacted individual developers in order to gather qualitative insights about their unusual public contributions.

Results: Our study shows that within GitHub and Stack Overflow, the onset of the pandemic (March/April 2020) is reflected in a set of outliers in developers’ contributions that point to an increase in activity. The distributions of contributions during the entire year of 2020 were, in some aspects, different, but, in other aspects, similar from the recent past. Additionally, we found one noticeably disrupted pattern of contribution in Stack Overflow, namely the ratio Questions/Answers, which was much higher in 2020 than before. Testimonials from the developers we contacted were mixed: while some developers reported that their increase in activity was due to the pandemic, others reported that it was not.

Conclusion: In Github, there was a noticeable increase in public software development activity in 2020, as well as more abrupt changes in daily activities; in Stack Overflow, there was a noticeable increase in new users and new questions at the onset of the pandemic, and in the ratio of Questions/Answers during 2020. The results may be attributed to the pandemic, but other factors could have come into play.