Open Source Is Not One Thing: A Typology of Open-Source Software Sub-Genres

Technical Summary: Open Source Is Not One Thing

Problem Statement
Empirical research in open-source software (OSS) frequently treats the field as a homogeneous entity. Studies often sample broad sets of popular projects (e.g., from GitHub) to model contributor retention, onboarding, or code quality, implicitly assuming that findings generalize across all OSS. However, the paper argues that OSS is not a single phenomenon; projects differ fundamentally in their purpose, governance structures, funding models, and community behaviors. These differences are not merely matters of scale but of distinct sub-genres. Consequently, models or interventions validated on one type of OSS (e.g., a company-backed product) may not hold for another (e.g., a volunteer-driven distribution or a humanitarian project). Coarse metrics like star counts or commit volumes obscure these structural differences, leading to findings that may not generalize and practical decisions that fail to account for specific project contexts.

Methodology
The authors conducted a "light, multi-source review" rather than an exhaustive systematic review. The process involved:

• Data Collection: Queries were run against two scholarly indexes, OpenAlex and arXiv, using 46 queries grouped into 15 facets (one per candidate sub-genre plus a discovery facet). This yielded 5,771 records.
• Normalization and Deduplication: Records were normalized to a common schema, and duplicates were removed based on DOI or title, resulting in 3,925 unique papers.
• Screening: A rule-based screening process retained papers where the title or abstract explicitly referred to open source and signaled that the work defined, classified, or studied a specific kind of OSS project or community. This resulted in 399 papers.
• Synthesis: The screened papers were analyzed to identify patterns in drivers, governance, and funding, leading to the construction of a typology. The authors explicitly note that the resulting typology is a "working hypothesis" intended to be confirmed, split, merged, or extended, rather than a settled ontology.

Key Contributions
The paper makes three primary contributions:

1. A Typology of 14 OSS Sub-Genres: The authors organize OSS into fourteen distinguishable sub-genres based on the primary driver (who starts and sustains the project and to what end). These include:
   • Established genres: Community-driven, Company-backed, Foundation-governed, and Research/Scientific Software (RSE).
   • Growing genres: Multi-company co-opetition, InnerSource, OSS for Social Good (OSS4SG), Government/Civic OSS, Educational OSS, and Open Hardware/Open Data.
   • Emerging genres (thin empirical support): Open-source appropriate technology (OSAT), Hobbyist/Solo OSS, Protestware, and Critical Digital Infrastructure.
2. A Comparative Framework: The sub-genres are placed in a framework (Table 1) that records each genre's primary driver, governance model, funding sources, maturity in the literature, and representative projects. This framework highlights that the primary driver does not strictly dictate governance or funding (e.g., foundation-governed and multi-company co-opetition both involve competing firms but differ in authority structures).
3. A Research Agenda: The paper proposes a research agenda centered on the question of transferability: "Do findings established on one sub-genre transfer to others?" It calls for empirical studies to treat sub-genre as a variable to be reported and controlled for, similar to how studies currently control for programming language or project size.

Results and Findings

• Divergence in Behavior: The review confirms that sub-genres exhibit distinct behaviors. For instance, OSS4SG communities are described as "sticky" (retaining contributors), whereas conventional OSS projects are often "magnetic" (attracting many who do not stay). The pathway from newcomer to core contributor also varies by sub-genre.
• Literature Gaps: The analysis reveals significant disparities in empirical support. While community-driven, company-backed, and foundation-governed OSS have substantial literature, other critical areas are under-studied. Specifically, multi-company co-opetition, protestware, and open-source appropriate technology are identified as having only a handful of dedicated studies despite their practical importance.
• Sustainability Risks: The typology highlights that "Critical Digital Infrastructure" (e.g., OpenSSL, curl) often relies on a few unpaid maintainers and under-funded ad-hoc sponsorship, creating a sustainability risk that differs from firm-employed teams or grant-funded research software.

Significance
The paper argues that recognizing OSS as "plural" is essential for both theory and practice.

• For Research: It prevents the over-generalization of findings. A retention model or onboarding program validated on a single-vendor product may fail when applied to a volunteer distribution or a research library. The typology provides a mechanism to stratify analysis and control for sub-genre.
• For Practice: It aids developers, maintainers, and funders. Contributors can better understand the governance and community behavior of a project before joining. Maintainers and funders can identify which sub-genres lack support (e.g., critical infrastructure or protestware) and direct resources accordingly.
• Modesty of Claims: The authors explicitly state that the contribution is the typology and the research agenda, not a complete census. They acknowledge limitations, including the exclusion of non-English literature and grey literature, the non-mutually exclusive nature of the sub-genres (many projects span multiple categories), and the fact that the "maturity" labels reflect the retrieved literature rather than a citation count. The work is presented as a provisional framework to guide future empirical work.