paperbot · PL 论文追踪

全部论文 待读 收藏

P2 Almost there

差一点:开源项目中准贡献者为何失败
ICSE Proceedings of the 40th International Conference on Software Engineering2018匹配度 1/10 · 该论文研究开源项目中准贡献者PR被拒原因及其对继续参与的影响,属于经验软件工程与开发者社区研究,与用户关注的PL×AI、静态分析、浮点/贝叶斯程序分析及LLM/Agent交叉方向基本无关。引用 93 inbox
Igor Steinmacher, Gustavo Pinto, Igor Scaliante Wiese, Marco A. Gerosa
crossref: 2026-09-12

AI 速览

一句话:通过分析21个GitHub项目的准贡献者,揭示其PR被拒原因及对继续参与的抑制影响。

开源项目面临贡献人力不足,外部开发者常因首次贡献障碍而失败。本文研究“准贡献者”(提交贡献但未被接受的开发者)为何失败。作者从21个流行GitHub项目识别准贡献者,分析其PR,并对准贡献者和集成者问卷调查,还人工分析263个代表性PR。结果发现10,099名准贡献者,约占实际贡献者70%,提交12,367个未接受PR;五个项目中准贡献者多于实际贡献者。约三分之一受访者不认同拒绝,约30%称被拒使其失去动力或不再提交PR。主因包括PR被取代/重复、开发者与团队愿景不匹配,以及社区关系、经验或投入不足。该实证研究有助于促进OSS社区参与和留存。

问题
开源项目难以找到持续贡献者,外部开发者常因首次贡献障碍而失败;本文关注准贡献者为何无法让贡献被接受。
方法
收集21个流行的非平凡GitHub项目数据,识别准贡献者并分析其pull-request,结合对准贡献者与项目集成者的问卷调查及263个PR的人工分析。
评估
基于10,099名准贡献者和12,367个未被接受的PR进行大规模实证分析,并通过问卷与人工标注交叉验证拒绝原因。
原文摘要(Abstract)

Recent studies suggest that well-known OSS projects struggle to find the needed workforce to continue evolving---in part because external developers fail to overcome their first contribution barriers. In this paper, we investigate how and why quasi-contributors (external developers who did not succeed in getting their contributions accepted to an OSS project) fail. To achieve our goal, we collected data from 21 popular, non-trivial GitHub projects, identified quasi-contributors, and analyzed their pull-requests. In addition, we conducted surveys with quasi-contributors, and projects' integrators, to understand their perceptions about nonacceptance. We found 10,099 quasi-contributors --- about 70% of the total actual contributors --- that submitted 12,367 nonaccepted pull-requests. In five projects, we found more quasi-contributors than actual contributors. About one-third of the developers who took our survey disagreed with the nonacceptance, and around 30% declared the nonacceptance demotivated or prevented them from placing another pull-request. The main reasons for pull-request nonacceptance from the quasi-contributors' perspective were "superseded/duplicated pull-request" and "mismatch between developer's and team's vision/opinion." A manual analysis of a representative sample of 263 pull-requests corroborated with this finding. We also found reasons related to the relationship with the community and lack of experience or commitment from the quasi-contributors. This empirical study is particularly relevant to those interested in fostering developers' participation and retention in OSS communities.

链接与引用

DOI 原文 ·

BibTeX
@inproceedings{paperbot4428,
  title = {Almost there},
  author = {Igor Steinmacher and Gustavo Pinto and Igor Scaliante Wiese and Marco A. Gerosa},
  booktitle = {ICSE},
  year = {2018},
  doi = {10.1145/3180155.3180208}
}