Fellowship for Maintainers

The Sovereign Tech Fund is piloting a fellowship program to pay open source maintainers, aiming to address structural issues and support open digital infrastructure in the public interest.

Over the past two years, STF has successfully contracted over 40 FOSS projects, enhancing their technical sustainability through targeted milestones. While some contracts are with individual maintainers, most involve software development companies or foundations. Despite this success, a new and innovative program is needed to acknowledge the lived reality of how many maintainers work: stretched across multiple technologies, multi-faceted, and often behind the scenes.

Most maintainers are unpaid, working in their spare time, which both impacts projects’ stability and can lead to stress and burnout. The Tidelift Open Source Maintainer Study found that 59% of maintainers have quit or considered quitting, posing a risk to the digital infrastructure we all rely on. To even begin to mitigate this risk, it's crucial to understand the role of maintainers, who typically lead and oversee project development, review changes, manage community interactions, release updates, and fix security issues.

These leadership activities are hard to quantify for funding applications, as the demands and challenges vary and can change quickly. This is where the fellowship for maintainers comes into play.

With this new fellowship program, we’re directly investing in the people behind the code by paying maintainers of critical open source components for their work. This covers a range of activities rather than specific deliverables, broadening the impact since maintainers often contribute to and play key roles in multiple projects.

We’re exploring options to structure the fellowship as service contracts or employment agreements, always taking into account the particular circumstances of the people doing the work. The fellowship will cover typical maintainer responsibilities like technical reviews, community management, release engineering, and security triage, but can also be tailored to specific technologies or communities.

We’re using an iterative approach and starting with a pilot program in 2024. This will help us to learn from the maintainers we support in this pilot phase, and to refine the program as we go.

To inform the design of the pilot, we conducted a survey of maintainers with a series of questions to understand the situations and preferences of the respondents. This survey was answered by 536 people, to whom we owe a big shout-out: Thanks to everyone taking the time earlier this year to help us collecting valuable insights!

The initial findings from that survey and how they’re informing the pilot can be summarized as follows:

1. There is no doubt that there is a substantial demand for direct support of maintainers and maintenance work.
   61% of the participants are interested in the program, another 33% are not sure yet. A significant majority of respondents (75%) want to change something in the way they get compensated (either they don’t get paid at all currently, but want to, or they get paid but want to increase the amount of paid maintainer work).
2. The specific situations of and demands on maintainers are diverse.
   To account for this, the fellowship pilot ideally should accommodate the broad range of situations and needs by offering different types of work agreements, numbers of hours, and lengths of engagement. This is especially important in order to make participation possible for maintainers who don’t have the financial security; professional or social capital; or personal flexibility to participate in a more standardized program. This will help increase the diversity among the maintainers: in fact, 20% of the participants identified as part of a marginalized group.
3. Mentoring and coaching need to be part of the program from the beginning.
   Maintainers also want support to become mentors and coaches themselves for up-and-coming maintainers. The pilot ideally will be accompanied by an offering from STF and/or partners that support this desire for coaching and development.

We’re looking forward to sharing more findings from the maintainer survey as we continue to evaluate and analyze the results.

As we’ve learned from both the survey and our conversations with maintainers and other core contributors to FOSS projects, their situations and needs are varied — and this needs to be reflected in the structure of any initiative to strengthen the open source ecosystem. In the pilot program of the fellowship for maintainers, we currently plan to pay up to five maintainers. Depending on their situations, needs, and tailored to the technologies and communities, there will be different options for structuring the fellowship, including but not necessarily limited to contract or employment types, working hours, and coaching or mentoring.

This type of flexible composition allows for the best diversification of approaches and enables us to support many technologies.

For example, 28% of respondents said their preferred weekly time commitment would be below 20 hours. From the comments submitted, there is bigger group of maintainers who would not be able to leave their current employment or customers. For them, it still would make a significant difference to get compensated for their part-time open source maintenance work.

We’re limiting the fellowship to 12 months because we want to learn from and iterate on this first pilot. This is also based on the fact that the largest group (30%) selected 12 months as the ideal fellowship length, among the options we provided.

The survey also revealed a clear demand for both mentorship support and trainings in open source software engineering. To address this need, we’re exploring how to create a mentoring program for open source maintainers, either directly or in collaboration with partners.

The application phase will start by the end of the third quarter of 2024, and with the goal that selected maintainers can begin the fellowship in the fourth quarter. The first fellowship pilot will run throughout 2025, and we will evaluate it on an ongoing basis. Based on these evaluations, our experiences running the fellowship, and feedback from participants, we’ll determine how to expand and grow the program for a stronger and healthier open source ecosystem.

In particular, we’re aware that the different fellowship options — necessary to accommodate maintainers’ particular situations — will need to be evaluated differently. With the support of an external evaluator and researcher, we’ll be able to better understand the impact and success of the different approaches.

We look forward to sharing the results and hope they’ll inform future endeavors in this area.