The NixOS Leadership Crisis: An Analysis Through Great Founder Theory
A Case Study in Succession Failure, Institutional Capture, and the Loss of Tacit Knowledge in Open Source Governance
A Case Study in Succession Failure, Institutional Capture, and the Loss of Tacit Knowledge in Open Source Governance
Abstract
The NixOS community experienced a profound governance crisis between 2023 and 2025, culminating in the resignation of founder Eelco Dolstra, mass departures of key contributors, and the emergence of multiple community forks. This paper analyzes these events through the analytical framework of Samo Burja’s Great Founder Theory, which posits that functional institutions are rare exceptions created by exceptional founders, and that the central challenge facing all institutions is the succession problem: successfully transferring both power and skill to subsequent generations. We argue that the NixOS crisis exemplifies a classic succession failure, compounded by institutional capture, the loss of tacit knowledge, and the conflation of borrowed and owned power. We conclude with recommendations for how the NixOS Foundation might establish owned power, solve its succession problem, and create mechanisms for capturing and transmitting tacit knowledge.
1. Introduction
NixOS and the Nix package manager represent one of the most innovative approaches to system configuration and package management in the history of computing. Created by Eelco Dolstra as part of his 2003 PhD thesis, the Nix ecosystem grew over two decades into a sophisticated technical project with thousands of contributors and a complex institutional structure including the NixOS Foundation, various governance teams, and a commercial entity, Determinate Systems.
Beginning in late 2023 and accelerating through 2024, the project experienced what can only be described as an institutional crisis: the permanent banning of prominent contributors, mass resignations from the Foundation board and moderation team, the forced resignation of the founder himself, and the emergence of multiple community forks (Lix, Auxolotl). By late 2024, five of seven moderation team members had resigned following conflicts with the newly elected Steering Committee—the very governance body that had been created to resolve the crisis.
Standard accounts of this drama focus on proximate causes: disputes over military contractor sponsorships, ideological conflicts within the community, and allegations of founder overreach. While these factors are relevant, they fail to explain the deeper structural dynamics at play. This paper applies the analytical framework of Samo Burja’s Great Founder Theory to provide a more comprehensive explanation of the crisis and to derive actionable recommendations for institutional repair.
2. Theoretical Framework: Great Founder Theory
2.1 Core Propositions
Great Founder Theory rests on several key propositions relevant to our analysis:
Functional institutions are the exception, not the rule. Most institutions are non-functional—they inadequately imitate functional institutions while maintaining narratives of effectiveness. Truly functional institutions are rare and always trace their origins to a skilled founder.
The succession problem is the central challenge. Every functional institution eventually faces the problem of transferring both power (the ability to pilot the institution) and skill (the knowledge required to pilot it well) to successors. Failure to solve this problem results in institutional decay.
Live players vs. dead players. A live player is a person or coordinated group capable of doing things they have not done before. A dead player operates from a script, incapable of novel action. Institutions can transition from live to dead when their tradition of knowledge dies.
Borrowed vs. owned power. Borrowed power can be taken away by others (titles, positions); owned power cannot easily be removed (skills, relationships, knowledge). Institutions where key actors have only borrowed power are inherently unstable.
Social technology. Institutions depend on social technologies—designed mechanisms for coordinating human action. These technologies can be lost, and their loss is often invisible until catastrophic failure occurs.
Tacit knowledge and intellectual dark matter. Much of what makes institutions function is knowledge that cannot be easily documented: trade secrets, implicit expertise, personal relationships, and long-term plans. This “intellectual dark matter” is easily lost during succession.
2.2 The Succession Problem in Detail
Burja identifies two components of the succession problem:
Power succession: Ensuring the successor inherits the formal and informal authority to direct the institution.
Skill succession: Ensuring the successor possesses the tacit knowledge and capabilities required to exercise that authority effectively.
Four outcomes are possible:
Power SuccessionSkill SuccessionOutcomeSuccessSuccessInstitution remains functional and liveSuccessFailureInstitution becomes piloted but dead (unskilled leadership)FailureSuccessSkilled individuals exist but lack authority to actFailureFailureInstitution becomes unpiloted and dead
The NixOS crisis, we will argue, represents a complex combination of outcomes three and four: the founder’s resignation transferred formal power to new structures, but these structures lacked the skill to pilot the institution, while those with skill were progressively excluded.
3. Analysis: The NixOS Crisis Through Great Founder Theory
3.1 Eelco Dolstra as Great Founder
Eelco Dolstra fits Burja’s definition of a great founder: he created something genuinely novel (the Nix approach to package management), built a functional institution around it, and served as its pilot for over two decades. His 2003 PhD thesis represented not merely an academic contribution but the foundation of what became a live tradition of knowledge.
The Nix project under Dolstra’s leadership exhibited the characteristics Burja identifies with functional institutions:
Production of notable effects: Nix and NixOS demonstrably outperformed alternatives in reproducibility and declarative system configuration.
Shared methodology: The project developed distinctive approaches (derivations, the Nix expression language) that constituted genuine social technology.
Master/apprentice relationships: Core contributors learned the Nix approach through close interaction with Dolstra and early contributors.
Living tradition of knowledge: The project continued to innovate and adapt, indicating a live rather than dead tradition.
Critically, however, Dolstra’s position exhibited characteristics that would prove problematic for succession:
Informal authority: Despite having no formal BDFL title, Dolstra functioned as one, exercising veto power over significant decisions.
Tacit knowledge concentration: Much of the project’s direction and decision-making rationale existed only in Dolstra’s understanding.
Owned power confusion: Dolstra’s authority derived partly from owned power (his technical expertise and foundational role) and partly from borrowed power (his Foundation position), but these were never clearly distinguished.
3.2 The Failure of RFC 98 and Early Governance Attempts
The 2021 proposal RFC 98 (Community Team) represents an early failed attempt to address governance gaps. Authored by Irene Knapp, a non-Nix contributor with a background in labor organizing, it proposed creating a Community Team with broad powers to “model and enforce social norms” and combat “ideas rooted in fascism or bigotry.”
From a Great Founder Theory perspective, RFC 98 failed for predictable reasons:
Borrowed power without skill: The proposal would have granted significant borrowed power to individuals who had not demonstrated the tacit knowledge necessary to pilot such authority effectively.
Counterfeit understanding: The proposal’s authors appeared to understand the form of governance mechanisms (moderation, codes of conduct) without understanding their function within the specific context of the Nix ecosystem.
Institutional capture risk: By explicitly politicizing the moderation function (”fascism,” “bigotry”), the proposal created vectors for capture by those whose primary allegiance was to ideological goals rather than the project’s technical mission.
Jon Ringer’s contemporary critique proved prescient: “It creates a situation where there’s a moving target in what is considered acceptable behavior, only for the benefit of the moderation team.”
3.3 The Moderation Team as Dead Player
The moderation team that emerged after RFC 102 (2022) operated with borrowed power from the Foundation but increasingly as a dead player—capable only of executing scripts (ban procedures, CoC enforcement) rather than adapting to novel situations.
Evidence of dead player characteristics:
Script-bound behavior: The srid ban (November 2023) followed a rigid escalation procedure despite the unusual nature of the case (demands to remove a steak photo from a personal profile).
Inability to adapt: When challenged on their decisions, moderators responded with “We are not going to revisit the decision” rather than engaging substantively.
Self-selection for ideological conformity: New moderators were added through unanimous consent of existing members, creating an echo chamber rather than a tradition of knowledge.
The key insight from Great Founder Theory is that this was not merely bad moderation—it was moderation by individuals who had acquired the form of moderation authority without the tacit knowledge of how to exercise it wisely. They possessed counterfeit understanding.
3.4 The Sponsorship Crisis as Trigger
The Anduril sponsorship controversies (NixCon EU 2023, NixCon NA 2024) served as the proximate trigger for the crisis, but through the lens of Great Founder Theory, we can understand why they proved so destabilizing.
The sponsorship decision involved a classic conflict between:
The founder’s tacit knowledge: Dolstra understood, implicitly, that broad sponsorship acceptance served the project’s long-term interests and that politicizing sponsorship decisions would create dangerous precedents.
Activists’ explicit ideology: A faction within the community held strong explicit beliefs about military-industrial complex involvement that they sought to impose on the project.
What made resolution impossible was that:
Dolstra’s owned power was illegible: His authority to make such decisions derived from tacit understanding of the project’s needs, but this understanding could not be easily articulated or defended in explicit terms.
The activists possessed concentrated borrowed power: Through positions on the moderation team and as Foundation observers, they could apply sustained pressure that Dolstra’s diffuse owned power could not easily counter.
No succession mechanism existed: There was no way to transfer Dolstra’s tacit understanding of “what the project needs” to a successor or governance body.
3.5 The Save-Nix-Together Letter as Institutional Capture
The April 2024 “save-nix-together” letter represents a textbook example of institutional capture. Burja warns:
“If an institution built to transfer a tradition of knowledge gains power or prestige, it will attract people who want to use the institution for other purposes than the preservation and development of the tradition.”
The letter’s authors:
Demanded the founder’s resignation
Sought to restructure governance to privilege specific identity groups
Characterized technical disagreement as evidence of moral failure
Used an ultimatum structure designed to maximize pressure rather than facilitate compromise
The anonymous authorship is particularly revealing. Burja notes that “live players frequently conceal themselves”—but here the concealment served not to preserve strategic advantage but to avoid accountability for what was essentially a power grab.
The letter succeeded in forcing Dolstra’s resignation, but this “success” merely accelerated the succession crisis. Power was transferred to individuals and structures that lacked the skill to exercise it effectively.
3.6 Jon Ringer’s Ban as Symptom
The treatment of Jon Ringer—one of the project’s most prolific contributors (9,000+ PR reviews, three terms as Release Manager)—illustrates the pathology of counterfeit understanding in governance.
Ringer was suspended and eventually permanently banned not for technical failures but for “derailing sensitive discussions and willfully furthering the division in the community.” His actual offense was articulating a perspective that conflicted with the moderation team’s implicit ideology.
From Great Founder Theory’s perspective, this represents a catastrophic error:
Destruction of tacit knowledge: Ringer possessed vast tacit knowledge about the Nix ecosystem—how packages actually worked, where technical debt lay, how to coordinate releases. This knowledge was irreplaceable and was lost to the project.
Prioritization of borrowed power over owned power: The moderation team’s borrowed power (to ban) was used against an individual whose owned power (technical expertise, relationships, knowledge) constituted a core asset of the project.
Failure of verification mechanisms: A living tradition includes mechanisms for correcting errors. The Ringer ban revealed that no such mechanism existed—there was no way to appeal, no way to demonstrate that the moderation decision was mistaken.
3.7 The Constitutional Assembly and Steering Committee
The Constitutional Assembly and subsequent Steering Committee elections represented an attempt to solve the succession problem through formal mechanisms. 450 contributors voted; seven members were elected.
However, Great Founder Theory suggests this approach was doomed to produce suboptimal results:
Committees cannot receive tacit knowledge: Burja observes that “contrarian ideas—as all new technologies are by definition—almost never survive committees.” A committee cannot possess the tacit understanding that resided in Dolstra’s mind.
Democratic legitimacy is not the same as skill: The Steering Committee possessed borrowed power (electoral mandate) but this conferred no guarantee of the skill necessary to pilot the institution.
The wrong problem was solved: The community diagnosed the problem as “concentration of power in the founder” and prescribed “distribution of power through democracy.” But the actual problem was “failure to transfer tacit knowledge,” which democracy cannot solve.
The post-election conflict between the Steering Committee and the moderation team—resulting in five of seven moderators resigning—demonstrates the instability of borrowed power structures without underlying traditions of knowledge.
3.8 The Forks as Creative Destruction
The emergence of Lix and Auxolotl represents what Burja calls “creative destruction”—the replacement of sclerotic institutions through competition rather than reform.
Burja notes:
“Disruption should be the backup rather than the first choice for innovation. That disruption is often the first choice instead results from poor institutional health.”
The forks indicate that:
Succession failed: Live players with skill (fork founders) could not obtain power within the existing institution.
The institution became unpiloted: Creative destruction became necessary because the main institution could no longer adapt.
Knowledge is fragmenting: Each fork will develop its own tacit knowledge tradition, leading to divergence that may prove irrecoverable.
4. Diagnosis: Why the NixOS Succession Failed
Synthesizing the above analysis, we can identify several structural factors that caused the NixOS succession to fail:
4.1 Tacit Knowledge Was Never Externalized
Dolstra possessed vast tacit knowledge about:
Why certain technical decisions were made
How to evaluate contributor readiness for increased responsibility
What the project’s implicit values and priorities were
How to balance competing stakeholder interests
This knowledge was never systematically documented or transferred. When Dolstra departed, it departed with him.
4.2 Owned Power Was Never Established
The Foundation and governance structures operated entirely on borrowed power. No individual or body possessed the kind of owned power—skills, relationships, resources that cannot be taken away—necessary to pilot the institution through crisis.
Dolstra himself confused his owned power (technical expertise, founder status) with his borrowed power (Foundation position), leading him to believe that transferring the latter would solve the succession problem.
4.3 No Mechanism for Identifying Successors
The project never developed means for:
Identifying individuals with the tacit knowledge necessary for leadership
Testing whether candidates possessed genuine vs. counterfeit understanding
Gradually transferring authority as skill was demonstrated
The moderation team’s self-selection mechanism actively worked against this, producing ideological conformity rather than skill development.
4.4 Institutional Capture Was Not Prevented
The project’s social technology (RFCs, governance structures, moderation policies) included no defenses against capture by those whose primary loyalty was to external ideological goals rather than the project’s technical mission.
5. Recommendations
Based on the foregoing analysis, we propose the following recommendations for the NixOS Foundation and similar open source projects facing succession challenges.
5.1 Establish Owned Power
Problem: Current governance structures operate entirely on borrowed power, making them inherently unstable and susceptible to capture.
Recommendation: The Foundation should establish owned power through:
Financial reserves: Build an endowment sufficient to sustain core operations independent of any single sponsor or funding source. This provides material independence that cannot be easily captured.
Infrastructure ownership: Ensure critical infrastructure (build farms, package caches, domain names) is held by entities with strong legal protections and clear succession provisions.
Skill development programs: Create structured apprenticeship programs that develop owned power (skills, knowledge) in emerging leaders, rather than merely conferring borrowed power (titles, positions).
Reputation capital: Develop mechanisms for recognizing and rewarding demonstrated technical contribution, creating a form of owned power that is visible and defensible.
5.2 Solve the Succession Problem
Problem: The project has no mechanism for transferring both power and skill to successors.
Recommendation: Implement a structured succession process:
Identify potential successors early: Rather than waiting for crisis, continuously identify individuals who demonstrate both technical skill and sound judgment.
Gradual authority transfer: Implement mechanisms for gradually increasing authority as skill is demonstrated. The Release Manager role historically served this function but was insufficiently integrated into broader governance.
Explicit skill verification: Develop verification mechanisms that test for genuine rather than counterfeit understanding. This might include:
Requiring candidates to articulate the reasoning behind historical decisions
Testing ability to handle novel situations rather than merely follow procedures
Evaluating judgment through simulated scenarios
Multiple succession paths: Avoid single points of failure by developing multiple potential successors in different domains (technical, governance, community).
Founder documentation: Require founders and key leaders to document their tacit knowledge in accessible form. This should include not just “what” decisions were made but “why”—the reasoning and values that informed them.
5.3 Capture and Record Tacit Knowledge
Problem: Critical knowledge exists only in individuals’ minds and is lost when they depart.
Recommendation: Create systematic mechanisms for knowledge capture:
Decision logs with reasoning: Require all significant decisions to be documented with explicit reasoning, not just outcomes. This creates a record that future leaders can learn from.
Oral history program: Conduct recorded interviews with founders and long-term contributors about the project’s history, values, and unwritten norms.
Architecture decision records: Adopt formal ADR practices that document not just what was decided but what alternatives were considered and why they were rejected.
Mentorship requirements: Make mentorship of junior contributors an explicit expectation for senior roles, creating ongoing knowledge transfer.
Exit interviews: Conduct systematic exit interviews with departing contributors, particularly those who leave under difficult circumstances, to capture their perspective and knowledge.
Living documentation: Maintain documentation that evolves with the project rather than becoming stale. Assign ownership of documentation to individuals responsible for keeping it current.
5.4 Defend Against Institutional Capture
Problem: The project’s governance structures were vulnerable to capture by those with external ideological agendas.
Recommendation: Implement capture-resistant governance:
Mission primacy: Establish and enforce a clear hierarchy: the project’s technical mission takes precedence over all other considerations. Governance bodies should explicitly affirm this.
Contribution requirements: Require meaningful technical contribution as a prerequisite for governance positions, not merely ideological alignment or community participation.
Separation of concerns: Keep moderation, technical governance, and strategic governance separate, with different accountability structures for each.
Appeal mechanisms: Establish robust appeal mechanisms for governance decisions, including external review where appropriate.
Transparency requirements: Require transparency in governance deliberations, making capture attempts visible before they succeed.
Term limits and rotation: Prevent entrenchment through term limits and mandatory rotation, while ensuring knowledge transfer during transitions.
6. Conclusion
The NixOS crisis of 2023-2025 represents a paradigmatic example of succession failure in an open source project. The founder created a genuinely functional institution—a live player with a living tradition of knowledge. But the mechanisms for transferring that institution to successors were never developed. When crisis came, borrowed power structures without underlying tacit knowledge proved incapable of piloting the institution effectively.
The result was predictable: institutional capture by those with explicit ideologies but counterfeit understanding; destruction of tacit knowledge through bans and resignations; fragmentation through forking; and a community that, even after “solving” its governance crisis through elections, found itself with new leaders who lacked the skill to exercise their borrowed power wisely.
Great Founder Theory suggests this outcome was not inevitable. Succession can be solved, tacit knowledge can be preserved, and institutions can defend themselves against capture. But doing so requires explicit attention to these challenges—attention that the NixOS community did not provide until it was too late.
For the Nix ecosystem, the path forward requires acknowledging the magnitude of what was lost, rebuilding traditions of knowledge where possible, and implementing the structural changes necessary to prevent recurrence. The forks may or may not succeed in building their own functional institutions. The main project may or may not recover its former vitality.
What is certain is that the NixOS crisis offers valuable lessons for the broader open source community. Technical excellence is not sufficient for institutional health. Governance structures without underlying tacit knowledge are houses built on sand. And the succession problem—the challenge of ensuring that what one generation built can survive to the next—is the central challenge facing every functional institution, including those that build software.
References
Burja, S. (2020). Great Founder Theory. Manuscript. Retrieved from www.SamoBurja.com/GFT
Dolstra, E. (2006). The Purely Functional Software Deployment Model. PhD Thesis, Utrecht University.
save-nix-together.org. (2024). Open Letter to the NixOS Foundation.
Ringer, J. (2024). NixOS Drama Timeline. GitHub Gist.
NixOS Foundation. (2024). Board Announcement: Giving Power to the Community. NixOS Discourse.
NixOS Steering Committee. (2024). Election Results. nixos.org.
Various. (2023-2024). NixOS Discourse threads, Matrix logs, and GitHub discussions as cited in text.
This paper represents an independent analysis and does not claim to present the views of any party involved in the events described.