On Wasted Contributions: Understanding the Dynamics of Contributor-Abandoned Pull Requests

TL;DR

The paper tackles the problem of contributor-abandoned PRs in open-source software by leveraging a large, multi-project dataset and a mixed-methods design. It combines quantitative analyses with machine learning to identify features that predict PR abandonment and employs qualitative coding to uncover underlying abandonment reasons, complemented by maintainer perspectives. Key findings show that abandonment is driven more by review process, contributor experience, and project dynamics than by PR content alone, with long discussions and novice contributors at higher risk. The study provides actionable guidance for maintainers, including triage practices, clearer contribution signals, and better support for newcomers, aiming to reduce wasted effort and improve contributor retention. Overall, the work offers a nuanced view of how PR abandonment emerges across different projects and times, with implications for governance and tooling in OSS communities.

Abstract

Pull-based development has enabled numerous volunteers to contribute to open-source projects with fewer barriers. Nevertheless, a considerable amount of pull requests (PRs) with valid contributions are abandoned by their contributors, wasting the effort and time put in by both the contributors and maintainers. To better understand the underlying dynamics of contributor-abandoned PRs, we conduct a mixed-methods study using both quantitative and qualitative methods. We curate a dataset consisting of 265,325 PRs including 4,450 abandoned ones from ten popular and mature GitHub projects and measure 16 features characterizing PRs, contributors, review processes, and projects. Using statistical and machine learning techniques, we find that complex PRs, novice contributors, and lengthy reviews have a higher probability of abandonment and the rate of PR abandonment fluctuates alongside the projects' maturity or workload. To identify why contributors abandon their PRs, we also manually examine a random sample of 354 abandoned PRs. We observe that the most frequent abandonment reasons are related to the obstacles faced by contributors, followed by the hurdles imposed by maintainers during the review process. Finally, we survey the top core maintainers of the studied projects to understand their perspectives on dealing with PR abandonment and on our findings.

Figures (28)

• Figure 1: Spearman's $\rho$ correlation among different pairs of features across the combined data of the studied projects.
• Figure 2: ALE plots showing how review_participants_responses varies the abandonment probability of PRs across the studied projects.
• Figure 3: ALE plots showing how review_contributor_responses varies the abandonment probability of PRs across the studied projects.
• Figure 4: ALE plots showing how contributor_acceptance_rate varies the abandonment probability of PRs across the studied projects.
• Figure 5: ALE plots showing how contributor_pulls varies the abandonment probability of PRs across the studied projects.