UC Irvine

Open source development is not only a software development paradigm that is widely adopted nowadays, but also a social movement that has profound impacts in our society. Ideology, as “heart and soul”, drives such social movements, and so does open source software ideology (OSS ideology) to open source movement. Thus, it is essential to understand OSS ideology. However, SE literature only conducted preliminary explorations towards OSS ideology, or only focused on particular elements within OSS ideology. In this dissertation, we report our efforts in developing an empirical understanding towards OSS ideology. First, we defined OSS ideology based on a general and explicit definition to encompass our research. Following grounded theory methodology, we collected data from 21 semi-structured interviews with OSS practitioners, and 41 videos of Open Source Initiative (OSI)’s board members’ public interviews/speeches; then we developed an empirical theory of OSS ideology. Such a theory consisted of 42 themes that emerged from the collected data, under six broad categories, i.e., Norms/Values, Membership, Goals, Activities, Resources, and Positions/Group Relations. Then, based on the empirical theory of OSS ideology, we developed an assessment framework to investigate the OSS ideologies within particular open source projects. The assessment only relied on public data sources, and reflected OSS ideologies at the project level. Furthermore, we conducted three case studies by applying the assessment framework to three selected open source projects, D3, Zephyr, and PyTorch. The assessment results demonstrated how open source projects uphold OSS ideology in practice.

This dissertation contributed to SE literature with a foundational understanding of OSS ideology. First, the empirical theory of OSS ideology, featuring comprehensive, contemporary, and empirically-grounded characteristics, significantly extended the SE literature with new themes identified in the study. It also provides an ideological lens to examine open source development. Second, the practical and extensive assessment framework could facilitate both OSS researchers and practitioners to investigate open source development with an ideological lens. Third, the three case studies provided insights into the practices of open source development from an ideological perspective. Moreover, the studies in this dissertation yielded rich theoretical and practical implications regarding open source movement, they also indicated various future research opportunities to further our understanding of OSS ideology and open source movement.