When you first hear “open source software” your mind will likely go directly to developers building more tools that only developers will use. While projects like TensorFlow or React are popular for open source contributors, there are other galvanizing projects with direct societal impacts that inspire people to contribute to open source.
Specifically, one project to consider is Little Window which supports victims of intimate partner violence and provides resources to leave abusive relationships. Another is Restroom Refuge, which identifies and highlights safe restrooms for trans and gender nonconforming individuals. These are only a few of the open source (OSS) for social good projects we identified. But although they may seem to be fewer in quantity, they are definitely powerful in influence.
What we asked
To understand the scope of these unique types of open source projects, we ask the following questions…
RQ1: How do contributors define OSS for Social Good?
RQ2: What motivations do contributors have to contribute to OSS for Social Good?
RQ3: What factors do contributors consider to select an OSS for Social Good project?
RQ4: What are the current challenges to work in OSS for Social Good?
To answer these questions, we conducted a mixed-methods study of semi-structured interviews with Open Source for Social Good (OSS4SG) contributors and distributed a large scale survey to OSS and OSS4SG contributors. We identified OSS4SG projects from Ovio and the Digital Public Goods and strategically selected project topics from those platforms. Additional details on our sampling approach can be found in the paper.
In total, we interviewed 21 OSS4SG contributors and received 517 valid survey responses (8.97% response rate) from OSS and OSS4SG contributors.
What we found
For RQ1, “How do contributors define OSS for Social Good?”, we found that there are core properties that influenced participants perspective of OSS4SG:
- Whether the project targets people or communities that need help
- Whether the project aims to solve some societal issues or provide social benefits
Another priority that was mentioned but was not as prevalently featured in our survey was whether the project is for profit or not.
For RQ2, “What motivations do contributors have to contribute to OSS for Social Good?”, we found that OSS4SG contributors were slightly more motivated by the desire to solve a societal issue rather than building something for their own career or school project (more data on this can be found in the figure below). This result encourages us to think about how people decide that an OSS4SG project aligns with these motivations.
For RQ3, “What factors do contributors consider to select an OSS for Social Good project?”, we found that there are several factors that influence the selection process. One consideration that stood out was who the owners of the project are:
“If it’s even a charity organization I go and look at who their sponsors are. And if it’s a government I’m already like, no, it’s not gonna happen. A political party maybe. But government is too far for me.” (P14)
From our survey responses, we found the OSS4SG group cared significantly more about whether they trusted the owner of the project they were contributing to. Other strategies for identifying projects included having a personal connection to the projects through a friend and also resonating with the issue this project is trying to solve.
To better understand differences in the scale of impact people desired to have on these projects, we designed a set of trolley problem style questions across 3 levels of psychological distance: spatial, temporal, and social proximity. In this survey we found that contributors in our survey overall preferred to work on a project with a global need, would have long term benefits, and that would contribute to a need someone they know personally has (see Table 5 for more).
For RQ4, “What are the current challenges to work in OSS for Social Good?”, we found that there are several challenges including finding a project to contribute to and securing consistent funding:
“It is difficult to know where the projects are. Where the communities are. And getting involved in it. There are many, many, many developers that might want to contribute, but they never get, you know, an announce or publication, a post, something.” (P1)
“I honestly think the hardest thing about working on social good is very frequently they’re funded by charities, so it’s very hard to get people’s full focus on it. Like, paid full focus on it.” (P4)
Although we explicitly sought to find challenges that exist for the OSS4SG projects, these issues resonate for the broader open source community as well. However, for OSS4SG projects in particular, these challenges present higher risks for the sustainability as global citizens are more likely to be relying on these projects.
What this means
From this work, we’ve been able to capture a more confident understanding of Open Source Software for social good in hopes of being able to better support these projects. We’ve outlined a couple opportunities to help the OSS4SG community grow such as highlighting the societal impacts of the project more explicitly through semi-automated badging processes. We also suggest more safety and privacy guidelines to protect contributors working on these high impact (and sometimes polarizing) projects.
The formal research paper, Leaving My Fingerprints: Motivations and Challenges of Contributing to OSS for Social Good by Yu Huang, Denae Ford, and Thomas Zimmermann is published at ICSE‘ 21, 43rd ACM/IEEE International Conference on Software Engineering.
The presentation from my former intern Yu Huang’s talk, paper, and study materials are available locally online. This work was also featured in the 2020 GitHub Octoverse report and highlighted on GitHub’s Social Impact website.
Since this work has been published we have also conducted additional research investigating how to support developers growing their open source skills, matching developers who want to contribute to social good projects, and offering community tracking tools to the maintainers and organizers that keep these projects afloat. In fact, the follow ups from this work have also inspired how GitHub measures and sustains newcomer activity through GitHub Discussions Insights Dashboard. I am very proud of the work my interns Jenny T. Liang, Nischal Shrestha, and Mariam Guizani completed that summer.
To find the latest project updates be sure to check out the project website!
The GitHub Social Impact, Tech for Social Good Team partnered with OBI Digital to conduct a deeper investigation into open source work as digital public goods in low- and middle-income countries. This teams included Director Mala Kumar, Gina Assaf, Priyanka Patha, Yacoub Saad, and a host of others. Their in-depth study focused on the countries on India, Kenya, Egypt, and Mexico highlighting opportunities for sustainability and inclusive design. You can check out their summaries on their project website (which also includes the report available in English, Arabic, and Spanish)!