What Pulls the Strings? Understanding the Characteristics and Role of Argumentation in Open-Source Software Usability Discussions
Arghavan Sanei, Chaima Amiri, Atefeh Shokrizadeh, Jinghui Cheng
TL;DR
This study analyzes how argumentation shapes OSS usability discussions on GitHub Issues across five popular projects. By combining qualitative coding with a formal argument-quality taxonomy, it reveals that issue posts are more argumentative and of higher quality than comments, with visual content and premises strongly boosting cogency and reasonableness. The findings show that well-structured arguments influence participant engagement and discussion dynamics, highlighting a need to improve visual communication and broaden participation to achieve collective intelligence around usability. The work offers practical implications for enhancing OSS usability discussions and provides a framework applicable to other distributed, asynchronous collaboration contexts.
Abstract
The usability of open-source software (OSS) is important but frequently overlooked in favor of technical and functional complexity. Argumentation can be a pivotal device for diverse stakeholders in OSS usability discussions to express opinions and persuade others. However, the characteristics of argument discourse in those discussions remain unknown, resulting in difficulties in providing effective support for discussion participants. We address this through a comprehensive analysis of argument discourse and quality in five OSS projects. Our results indicated that usability discussions are predominantly argument-driven, although their qualities vary. Issue comments exhibit lower-quality arguments than the issue posts, suggesting a shortage of collective intelligence about usability in OSS communities. Moreover, argument discourse and quality have various impacts on the subsequent behavior of participants. Overall, this research offers insights to help OSS stakeholders build more effective arguments and eventually improve OSS usability. These insights can also inform studies about other distributed collaborative communities.