Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Guidelines for Developing Bots for GitHub

Mairieli Wessel, Andy Zaidman, Marco Gerosa, Igor Steinmacher

TL;DR

This article provides guidelines for developing and maintaining software bots, backed by multiple studies with practitioners, that can be annoying and disruptive to the community.

Abstract

Projects on GitHub rely on the automation provided by software development bots to uphold quality and alleviate developers' workload. Nevertheless, the presence of bots can be annoying and disruptive to the community. Backed by multiple studies with practitioners, this paper provides guidelines for developing and maintaining software bots. These guidelines aim to support the development of future and current bots and social coding platforms.

Guidelines for Developing Bots for GitHub

TL;DR

This article provides guidelines for developing and maintaining software bots, backed by multiple studies with practitioners, that can be annoying and disruptive to the community.

Abstract

Projects on GitHub rely on the automation provided by software development bots to uphold quality and alleviate developers' workload. Nevertheless, the presence of bots can be annoying and disruptive to the community. Backed by multiple studies with practitioners, this paper provides guidelines for developing and maintaining software bots. These guidelines aim to support the development of future and current bots and social coding platforms.
Paper Structure (7 sections, 2 figures)

This paper contains 7 sections, 2 figures.

Table of Contents

  1. Research overview
  2. Guidelines for developing bots
  3. Designing bot interaction
  4. Facilitating bot adoption
  5. Recommendations to platform builders
  6. The mediator bot
  7. Conclusion

Figures (2)

  • Figure 1: Methodology employed to identify challenges, build a prototype, and create guidelines. The result from Phase I was published at CSCW (Wessel2018Wessel2021CSCW), and Phase II at ICSE 2022 Wessel2022ICSE. We added a graphical mark ( ) to identify the challenges related to noisiness, which crosscut the three categories of challenges.
  • Figure 2: Prototype of the interventions in a real-world scenario on GitHub. It shows the relationship between the design strategies for the mediator bot derived from Phase II (S#1-6) and our proposed guidelines (G#1, G#2, G#4)/platform recommendations (R#1, R#2). The interactive version of the prototype is publicly available on Zenodo mairieli_wessel_2021_5675702.