Open source software, in its variety of forms (including operating systems such as Linux and Android, systems software like SQL and Apache server, internet browsers like Mozilla Firefox, and applications such as LibreOffice), provides the foundations for global digital infrastructure. It is quite startling then to learn that given the scale of public benefit, it is built on and sustained by the unpaid labour of software developers around the world. The developers, who contribute code, fix bugs and write guidance for open-source repositories often do so without direct remuneration for their work. Of course, the reality is rather more complicated; some contribute in the context of a paid role and others have found ways to monetise their contributions via sponsorship. Nonetheless, this voluntary aspect of OS raises questions about the sustainability of open source labour, the cross-subsidising of unpaid work for commercial interests and the longer-term implications for the development and maintenance of open source projects and digital infrastructure.
In January 2021, as three researchers at the University of Southampton with backgrounds in Sociology and Computer Science and supported by a grant from the Web Science Institute, we set out to explore the experiences, motivations, and priorities of a diverse group of contributors to open source projects. We sought to understand not only what contributors did, the nature of their OS labour (type, frequency etc.) and why they did it (their motivations) but importantly, how they did it? Our questions included:
- How they managed to fit contributions into their working week[end]?
- How they juggled their paid work with unpaid work and potentially with caring responsibilities or public roles?
- How their unpaid work was supported financially by various forms of income and resources?
In essence, what makes open source labour possible, or, at times, less possible?
We conducted online interviews with twenty individuals. Potentially suitable participants were identified in GitHub, an online development platform used by millions of developers to build and maintain software. Purposive sampling was undertaken based on recent contribution to open source and we sought to achieve diversity in terms of gender and country of residence. We extracted data from a range of GitHub listings, for instance: “Top Trending Developers (in February 2021)”, the “Women Who Code” and “Made in Nigeria” communities amongst others. From those lists we identified the online profiles of contributors. We sent research invitations via email or direct message on Twitter; and encountered approximately a 20% positive response rate from the 100 or so contributors whom we contacted.
Our 20 interviewees, aged between 22 and 57, came from ten countries across five continents and included seven women and seven participants from the global south. They were full-time software developers in large corporations, freelance developers, academics, early-career developers and those in OS community support roles. These participants coded in a range of software languages, contributed to a variety of software repositories and supported a range of OS communities. However, there was an element of selection bias to this sample given that many of those who responded had expertise in open source or were stakeholders in the field. It was a challenge to recruit those for whom OS was less central in their work.
Our semi-structured online interviews explored the working lives of participants, their education and training, their careers including both formal employment, freelance work, internships, voluntary roles and other elements of work identity. We asked about their relationship with open source, when and where they contributed, and factors related to this such as paid employment and domestic life. We probed how OSS contributions were supported financially and how participants earned a living. The perspectives that we obtained were as varied and diverse as our participants and offered intriguing insights into how contributions to OSS play out in different careers and at various life stages. The interview data that we collected has been professionally transcribed, pseudonymised, and coded using the data analysis software NVivo.
Early findings and future plans
We are in the early stages of the analysis process and analysing broad themes that include careers, internships, organisations, motivations and rewards, diversity and barriers, and financial resources. In the coming months, as part of our analysis we will describe the broad range of activities that contributors undertook, their diverse profiles, the geographical and cultural factors that have shaped contributions to open source. We will explore the role of OS in individual career paths and how it is embedded within organisations and professional practice. We are currently focused on understanding the patchworking that employed, freelance and student developers do to piece together their working weeks through combinations of paid and unpaid developer work, writing, consultancy and mentorship. We are also interested in understanding the barriers to diversity and exploring how women and developers in the global south have accessed the industry; especially the role of mentorship for guiding junior developers within global OS communities. Watch this space…
We would like to thank all of our research participants for taking part in this study and for their enthusiasm to learn about the findings of this research.
If you have any questions about this research or you would like to comment on this piece, then please contact us via email: firstname.lastname@example.org or tweet us: @os_sproject