Investigating developers' email discussions during decision-making in Python language evolution

TitleInvestigating developers' email discussions during decision-making in Python language evolution
Publication TypeConference Paper
Year of Publication2017
AuthorsStanger, Nigel, Rainer Austen, Licorish Sherlock A., Sharma Pankajeshwara, and Savarimuthu Bastin Tony Roy
Tertiary AuthorsMendes, Emilia, Counsell Steve, and Petersen Kai
Secondary TitleProceedings of the 21st International Conference on Evaluation and Assessment in Software Engineering - EASE'17
Pagination286 - 291
PublisherACM Press
Place PublishedKarlskrona, SwedenNew York, New York, USA
ISBN Number9781450348041

Context: Open Source Software (OSS) developers use mailing lists as their main forum for discussing the evolution of a project. However, the use of mailing lists by developers for decision-making has not received much research attention. Objective: We have explored this issue by studying developers' email discussions around Python Enhancement Proposals (PEPs). Method: Our dataset comprised 42,672 emails from six different mailing lists pertaining to PEP development. We performed multiple forms of analysis on these emails, involving both quantitative measures (e.g., frequency) and deeper analysis of specific PEP discussions (i.e., outlier analysis). Results: Out of three PEP types (Informational, Process and Standard Track), Standard Track PEPs attract a large amount of discussion (both in volume and average number of messages per proposal). Our study also identified specific PEP states and topics that generated a disproportionate amount of discussion. Conclusion: Our outcomes point to several opportunities for improving the management of an OSS team based on the knowledge generated from discussions. We have also identified several interesting avenues for future work such as identifying individuals or groups that present persuasive arguments during decision-making.