Latest / The Open Source Business with Fexingo: Commercial Strategy for Free Software Companies / The Open Source Sustainability Crisis
Transcript
- Lucas: If these conversations are useful for what you're building or running, we want to be clear about how this show stays on the air. We deliberately don't run ads on these. If you want to support that choice, the link is buy me a coffee dot com slash fexingo. Luna: It keeps the signal clean and lets us talk about the messy stuff without a sponsor breathing down our necks. Thanks for keeping the lights on. Lucas: So, let's get into today's anchor. I was reading through some recent data on GitHub activity earlier this week, and there is a number that stopped me cold. It is not a stock price or a revenue figure. It is the percentage of critical security vulnerabilities in popular open-source packages that remain unfixed for more than six months. The number is hovering around twelve percent. And here is the thing: those packages are often maintained by a single person working entirely on their own time. Luna: Twelve percent feels high, but is it really that surprising when you consider how many of these projects started as side hobbies? You have entire ecosystems resting on the goodwill of developers who have day jobs. Lucas: Exactly. And that is the structural flaw we need to dissect today. We call it the volunteer tax. For years, the tech industry has treated open-source maintenance as a public good that should be free. But free labor is not sustainable. When a maintainer spends forty hours a month patching security holes for a library used by millions, they are effectively subsidizing the operations of large enterprises. They are not getting paid for that work, yet the value generated is immense. Luna: Right, and it is not just about money. It is about attention. If someone is burning out from fixing bugs for free, they will eventually just stop. Or worse, they will abandon the project entirely. Lucas: Abandonment is the worst-case scenario, but it is happening faster than most people realize. Let's look at a concrete example. There was a widely used JavaScript utility package called left-pad. It had only one contributor. He removed it from the registry because he was tired of the harassment and lack of recognition. That single action broke builds across the entire industry. It showed us that our supply chain is fragile precisely because we rely on unpaid individuals for mission-critical infrastructure. Luna: That story always comes up, but it feels like an outlier now. Are we seeing more systemic failures beyond just one-off removals? Lucas: Yes, and this is where the economics get interesting. We are seeing a rise in what I call 'quiet quitting' of open-source roles. Maintainers are not deleting their code. They are just stopping updates. They are leaving dependencies outdated. This creates what we might call code debt. Companies are using these libraries because they are cheap upfront, but the hidden cost is the risk of unpatched vulnerabilities and incompatibility with newer systems. It is a deferred payment model, and the interest rate is rising. Luna: So the companies using the software are essentially taking on risk that the maintainer cannot manage alone. Is there any market mechanism that corrects for this? Lucas: There are attempts, but they are fragmented. You have foundations like the Linux Foundation or the Node.js Foundation stepping in to provide stipends for core contributors. Some companies also offer bounties for specific fixes. But these are reactive solutions. They pay for the fire, not for the fire prevention. We need a proactive model where the cost of maintenance is baked into the product lifecycle from the start. Luna: But if companies start paying for maintenance, doesn't that change the nature of open source? Won't it become less accessible? Lucas: That is a common fear, but I think it misunderstands the goal. Open source means the code is available. It does not mean the labor required to keep it secure must be free. Think about it like public roads. The asphalt is there for everyone to use, but the government taxes citizens to maintain the potholes and repave the streets. Without that funding, the roads crumble. Open source needs a similar tax structure, but implemented through corporate procurement budgets rather than general taxation. Luna: I like that analogy. So instead of treating maintainers as volunteers, treat them as contractors or employees of the ecosystem. Who would actually write that check? Lucas: The primary beneficiaries. Large enterprises that rely on these tools for their internal development. If a bank uses a specific cryptographic library for all its transactions, that bank should be contributing to the long-term stability of that library. It is a direct correlation between usage and responsibility. Currently, that link is broken. The biggest users often contribute the least to the upkeep. Luna: That sounds like a governance issue as much as an economic one. How do we enforce contribution without forcing companies to fork every library they use? Lucas: We don't force forks. We create incentives. One emerging model is the commercial open-source subscription. A company pays a fee to a foundation, and in return, that foundation guarantees SLAs for security patches and feature requests for their specific clients. It turns maintenance into a service product. It is transparent, scalable, and aligns the interests of the user and the maintainer. The maintainer gets predictable income; the company gets predictable stability. Luna: Predictable stability is exactly what CTOs sleep better over. But does this scale to the smaller, niche libraries? Not everything has enough users to justify a subscription model. Lucas: That is the hard part. For niche tools, we might need a collective insurance pool. Similar to how industries pool risks for catastrophic events. A group of companies that depend on a cluster of small utilities could fund a shared reserve to support multiple maintainers simultaneously. It spreads the cost across the ecosystem. It acknowledges that even small tools have outsized impacts on system integrity. Luna: A collective risk pool for code maintenance. That is actually a pretty innovative financial instrument. Have any organizations tried something like that recently? Lucas: Not exactly in that form, but there are signs of movement. The OpenSSF, which is the Open Source Security Foundation, has been pushing for standardized metrics on project health. They are trying to quantify the 'maintenance burden' so that companies can make informed decisions. If you can measure the risk, you can price it. And once you can price it, you can insure it or budget for it. Luna: So the first step is visibility. We need to know which projects are at risk before they break. Right now, most companies don't even know what libraries they are using deep in their stack. Lucas: Precisely. Dependency mapping is still a black box for many large organizations. They have a list of top-level packages, but they do not see the transitive dependencies. Those are the layers where the volunteer tax is heaviest. Fixing that visibility gap is crucial. If a company knows they are relying on a single-person project, they have two choices: fork it and take on the full cost, or contribute to its upkeep. Both are valid business decisions, provided they are made consciously. Luna: Forking seems like the nuclear option though. It isolates the company from community improvements and security updates. Isn't that dangerous long term? Lucas: It is extremely dangerous. Forking creates technical debt that compounds over time. You end up maintaining a divergent version that never gets the benefits of the main community. That is why contribution is the superior path. It integrates you into the solution rather than creating a parallel problem. But it requires capital. And that brings us back to the question of willingness to pay. Luna: Are companies actually willing to put line items on their P&L for 'open-source maintenance'? Or do they still see it as a zero-cost benefit? Lucas: Most still see it as zero-cost. That is the mindset we need to shift. I have spoken with several CTOs who admit they would love to pay for stability but face internal resistance. Their finance teams ask, 'Why are we paying for code we didn't write?' The answer is simple: we are paying for the insurance policy that keeps our code running. It is operational expenditure, not research and development. It belongs in the IT budget, not the R&D bucket. Luna: Reframing it as insurance makes it easier to sell internally. But how do we prove the ROI on preventing a vulnerability that hasn't happened yet? Lucas: You use incident reports. Look at the cost of downtime during a major outage caused by a dependency failure. Compare that to the annual stipend for the maintainer. The math usually works out in favor of maintenance. But the data is hard to find because most near-misses go unreported. We need better logging and sharing of these incidents so the industry can learn from them collectively. Luna: Sharing near-miss data sounds like a cultural hurdle. Companies are notoriously protective of their security post-mortems. Why would they share that with competitors? Lucas: Because the alternative is systemic collapse. If a foundational library breaks, it affects everyone. Competitors are not your enemies in this context; they are fellow passengers on the same plane. Aviation has safety reporting boards for a reason. Tech needs something similar. An anonymous, aggregated repository of dependency failure modes. It would help prioritize which projects need funding most urgently. Luna: An aviation-style safety board for software. That is a powerful concept. It removes the blame game and focuses purely on systemic resilience. Lucas: Exactly. Blame is useless in open source. Resilience is everything. And resilience costs money. We have accepted the idea that software should be free for too long. It is time to recognize the labor behind it. The volunteer era is ending. The professionalization of maintenance is beginning. Luna: Professionalization implies standards and training. Do we have the infrastructure to support a new class of professional open-source maintainers? Lucas: We are building it. Bootcamps are starting to teach not just coding, but community management and security auditing. Universities are offering courses on software supply chain management. The talent pipeline is expanding. But it needs investment. Companies need to hire these professionals and pay them well. Otherwise, we will just lose our best engineers to proprietary roles that offer higher salaries and better benefits. Luna: So the brain drain is real. If the best minds leave open source for private equity-backed startups, who is left to maintain the core? Lucas: Juniors and volunteers. And while juniors are eager, they lack the institutional knowledge to handle complex legacy systems. Volunteers are stretched thin. You end up with a maintenance layer that is brittle and slow to respond. That is when the cracks appear. And when they crack, they crack under the weight of global commerce. Luna: It feels like we are at an inflection point. The old model is failing, but the new model is still undefined. What is the next big step for leaders in this space? Lucas: Leaders need to audit their own dependencies. Start now. Map out every third-party tool your business relies on. Identify the ones with low maintenance activity. Then engage with those communities. Offer funding. Offer support. Make the relationship explicit. Do not wait for a crisis to force your hand. Proactive engagement is cheaper than reactive emergency patching. Luna: Engagement means more than just writing a check. It means understanding the maintainer's pain points and helping solve them. Lucas: Absolutely. It means listening. Many maintainers struggle with scope creep, documentation, or release management. If a company can help with those non-code tasks, they add value without necessarily changing the code itself. It is a holistic support model. It treats the project as a living entity that needs care, not just a static asset to be consumed. Luna: Treating it as a living entity changes the tone completely. It shifts from extraction to stewardship. I think that is the key word. Stewardship. Lucas: Stewardship. Yes. We have been extractors for too long. We took the code, built empires, and ignored the gardeners. Now the garden is wilting. It is time to water it. The question is whether we will do it voluntarily, or if the market will force our hand through a catastrophic failure. I hope we choose the former. Luna: I hope so too. Because the cost of waiting is measured in broken systems and lost trust. Let's see if the industry wakes up before the next major outage hits. Lucas: Here is the thing to watch in the coming quarter. Look for announcements from major cloud providers about dedicated funding pools for critical dependencies. If they move first, others will follow. If they hesitate, expect more volatility in the supply chain. The ball is in their court.