The Invisible Work of Open-Source Software Maintenance

• In Event: Between Technological Determinism and Social Shaping: Participation and Isolation in Governance

Abstract

In this talk, we present findings about the work of maintaining community-based open-source software (OSS) projects, with a focus on the invisible and infrastructural work performed in such projects. Many OSS projects have become foundational components for stakeholders across academia, the tech industry, government, journalism, and activism. Although OSS projects provide immense benefits for society, they are often initially created by volunteers. Their maintainers often struggle with how to sustain and support their projects, particularly as they become adopted as infrastructure in increasingly critical contexts. Some large projects transition to corporate sponsorship or non-profit foundations --- raising different issues --- but most are maintained by a handful of individuals who must balance their responsibilities to the project with their ‘day jobs’ and life.

The software engineering skills necessary to start a successful project are not the same skills required to maintain one. In fact, many skills required to maintain the project fall entirely outside of the traditional scope of software engineering. Project leaders and maintainers resolve conflicts, perform community outreach, write documentation, review others’ code, mentor newcomers, coordinate with other projects, and more. Many OSS project leaders and maintainers have publicly discussed the effects of burnout as they find themselves doing unexpected and thankless work. In examining the work and life behind OSS projects through interviews with maintainers and analyses of code repositories, we gain a nuanced view of the tensions engineers face as they manage accountabilities to various groups, contributing to our understanding of the relationship between engineering and society.

Authors

• Dorothy Roe Howard, UC San Diego Department of Communication & Design Lab

• R. Stuart Geiger, UC-Berkeley Institute for Data Science

• Lilly Irani, University of California, San Diego

• Alexandra Paxton, University of Connecticut

• Nelle Varoquaux, University of California, Berkeley

• Chris Holdgraf, University of California, Berkeley