An Empirical Study of Multi-discussing Pattern in Open-Source Software Development

Abstract

GitHub enables developers to expediently contribute their comments on multiple issues and switch their discussion between issues, i.e., multi-discussing. Discussing multiple issues simultaneously is able to enhance work efficiency. However, multi-discussing also relies on developers’ rationally allocating their focus, which may result in the different influence on the resolution of issues. Therefore, investigating how multi-discussing affects the issue resolution is a meaningful research question that can help developers understand the benefits and limitations of multi-discussing. Using quantitative and qualitative methods, this paper proposes a groundbreaking study of the impact of multi-discussing on issue resolution in GitHub. First, we collect and analyze data from 624 GitHub projects to explore how multi-discussing affects the overall issue resolution of the project. Further, we investigate how multi-discussing affects the resolution of a single issue. We find that multi-discussing is a common behavior in GitHub. Also, multi-discussing is connected to a shorter average issue resolution latency of the project. However, during a single issue resolution, more multi-discussing behaviors tend to bring longer issue resolution latency. We also conduct the qualitative analysis to explore the developers’ experiences and expectations of multi-discussing.