Search for: All records

Past work has sought to design AI ethics interventions–such as checklists or toolkits–to help practitioners design more ethical AI systems. However, other work demonstrates how these interventions may instead serve to limit critique to that addressed within the intervention, while rendering broader concerns illegitimate. In this paper, drawing on work examining how standards enact discursive closure and how power relations affect whether and how people raise critique, we recruit three corporate teams, and one activist team, each with prior context working with one another, to play a game designed to trigger broad discussion around AI ethics. We use this as a point of contrast to trigger reflection on their teams’ past discussions, examining factors which may affect their “license to critique” in AI ethics discussions. We then report on how particular affordances of this game may influence discussion, and find that the hypothetical context created in the game is unlikely to be a viable mechanism for real world change. We discuss how power dynamics within a group and notions of “scope” affect whether people may be willing to raise critique in AI ethics discussions, and discuss our finding that games are unlikely to enable direct changes to products or practice, but may be more likely to allow members to find critically-aligned allies for future collective action. 

The involvement of companies and public institutions in open-source software (OSS) has become widespread. While studies have explored the business models of for-profit organizations and their impact on software quality, little is known about their influence on OSS communities, especially in terms of diversity and inclusion. This knowledge gap is significant, considering that many organizations have the resources to enhance diversity and inclusion internally, but whether these efforts extend to OSS remains uncertain. To address this gap, we conducted interviews with maintainers of community-owned and organization-owned OSS projects, revealing tensions between organizations and their projects and identifying the impact of internal policies on OSS communities. Our findings reveal that, on the one hand, organization-owned projects often restrict external contributions due to stringent operating procedures and segmented communication, leading to limited external engagement. On the other hand, these organizations positively influence diversity and inclusion, notably in the representation and roles of women and the implementation of mentorship programs. 

Who creates the most innovative open-source software projects? And what fate do these projects tend to have? Building on a long history of research to understand innovation in business and other domains, as well as recent advances towards modeling innovation in scientific research from the science of science field, in this paper we adopt the analogy of innovation as emerging from the novel recombination of existing bits of knowledge. As such, we consider as innovative the software projects that recombine existing software libraries in novel ways, i.e., those built on top of atypical combinations of packages as extracted from import statements. We then report on a large-scale quantitative study of innovation in the Python open-source software ecosystem. Our results show that higher levels of innovativeness are statistically associated with higher GitHub star counts, i.e., novelty begets popularity. At the same time, we find that controlling for project size, the more innovative projects tend to involve smaller teams of contributors, as well as be at higher risk of becoming abandoned in the long term. We conclude that innovation and open source sustainability are closely related and, to some extent, antagonistic. 

Much of our modern digital infrastructure relies critically upon open sourced software. The communities responsible for building this cyberinfrastructure require maintenance and moderation, which is often supported by volunteer efforts. Moderation, as a non-technical form of labor, is a necessary but often overlooked task that maintainers undertake to sustain the community around an OSS project. This study examines the various structures and norms that support community moderation, describes the strategies moderators use to mitigate conflicts, and assesses how bots can play a role in assisting these processes. We interviewed 14 practitioners to uncover existing moderation practices and ways that automation can provide assistance. Our main contributions include a characterization of moderated content in OSS projects, moderation techniques, as well as perceptions of and recommendations for improving the automation of moderation tasks. We hope that these findings will inform the implementation of more effective moderation practices in open source communities. 

How do software engineers identify and act on their ethical concerns? Past work examines how software practitioners navigate specific ethical principles such as “fairness”, but this narrows the scope of concerns to implementing pre-specified principles. In contrast, we report self-identified ethical concerns of 115 survey respondents and 21 interviewees across five continents and in non-profit, contractor, and non-tech firms.We enumerate their concerns – military, privacy, advertising, surveillance, and the scope of their concerns – from simple bugs to questioning their industry’s entire existence. We illustrate howattempts to resolve concerns are limited by factors such as personal precarity and organizational incentives. We discuss how even relatively powerful software engineers often lacked the power to resolve their ethical concerns. Our results suggest that ethics interventions must expand from helping practitioners merely identify issues to instead helping them build their (collective) power to resolve them, and that tech ethics discussions may consider broadening beyond foci on AI or Big Tech. 

While the severe underrepresentation of women and non-binary people in open source is widely recognized, there is little empirical data on how the situation has changed over time and which subcommunities have been more effectively reducing the gender imbalance. To obtain a clearer image of gender representation in open source, we compiled and synthesized existing empirical data from the literature, and computed historical trends in the representation of women across 20 open source ecosystems. While inherently limited by the ability of automatic name-based gender inference to capture true gender identities at an individual level, our census still provides valuable population-level insights. Across all and in most ecosystems, we observed a promising upward trend in the percentage of women among code contributors over time, but also high variation in the percentage of women contributors across ecosystems. We also found that, in most ecosystems, women withdraw earlier from open-source participation than men. 

Open-source software projects have become an integral part of our daily life, supporting virtually every software we use today. Since open-source software forms the digital infrastructure, maintaining them is of utmost importance. We present Climate Coach, a dashboard that helps open-source project maintainers monitor the health of their community in terms of team climate and inclusion. Through a literature review and an exploratory survey (N=18), we identified important signals that can reflect a project’s health, and display them on a dashboard. We evaluated and refined our dashboard through two rounds of think-aloud studies (N=19). We then conducted a two-week longitudinal diary study (N=10) to test the usefulness of our dashboard. We found that displaying signals that are related to a project’s inclusion help improve maintainers’ management strategies. 

Open source software represents an important form of digital infrastructure as well as a pathway to technical careers for many developers, but women are drastically underrepresented in this setting. Although there is a good body of literature on open source participation, there is very little understanding of the participation trajectories and contribution experiences of women developers, and how they compare to those of men developers, in open source software projects. In order to understand their joining and participation trajectories, we conducted interviews with 23 developers (11 men and 12 women) who became core in an open source project. We identify differences in women and men's motivations for initial contributions and joining processes (e.g. women participating in projects that they have been invited to) and sustained involvement in a project. We also describe unique negative experiences faced by women contributors in this setting in each stage of participation. Our results have implications for diversifying participation in open source software and understanding open source as a pathway to technical careers. 

The rapid growth of open source software necessitates a deeper understanding of moderation and governance methods currently used within these projects. The code of conduct, a set of rules articulating standard behavior and responsibilities for participation within a community, is becoming an increasingly common policy document in open source software projects for setting project norms of behavior and discouraging negative or harassing comments and conversation. This study describes the conversations around adopting and crafting a code of conduct as well as those utilizing code of conduct for community governance. We conduct a qualitative analysis of a random sample of GitHub issues that involve the code of conduct. We find that codes of conduct are used both proactively and reactively to govern community behavior in project issues. Oftentimes, the initial addition of a code of conduct does not involve much community participation and input. However, a controversial moderation act is capable of inciting mass community feedback and backlash. Project maintainers balance the tension between disciplining potentially offensive forms of speech and encouraging broad and inclusive participation. These results have implications for the design of inclusive and effective governance practices for open source software communities.