Monkeywrench
The kernel panic hit at 2:47 AM. Production server. Six hundred users scheduled to hit it by 6:00.
I’m not going to name the company. Doesn’t matter. What matters is that the thing keeping that box alive wasn’t a vendor contract or an enterprise support agreement. It was a mailing list post from 2019 written by a guy in the Netherlands whose name I never learned, describing the exact same race condition I was staring at. His fix worked. The system came back up. Nobody sent him a thank-you note.
That’s the part nobody talks about.
The Foundation Was Never the Product
There’s a story the tech industry tells about itself. Brilliant teams at well-funded companies built the modern internet. Smart people, hard problems, breakthrough moments. The logo gets the credit. The press release goes out. The stock price moves.
What the story leaves out is the 40 years of unpaid labor underneath it.
Linux runs roughly 90% of the public cloud. Apache and Nginx serve most of the web. OpenSSL encrypts the traffic. PostgreSQL and MySQL hold the data. cURL fetches half the world’s HTTP requests. Git manages nearly all production code. Every single one of those tools was built, maintained, and debugged by people who, in most cases, were not compensated by the companies getting rich on top of them.
This isn’t ancient history. It’s Tuesday.
What “Giving Back” Usually Looks Like
The standard corporate response to this observation is the word “contribute.” We contribute to open source. We’re good citizens of the ecosystem.
Sometimes that’s even true. Google has done real work on Kubernetes and Android. Red Hat built a business model around actually paying people to work on Linux. A handful of companies genuinely pull weight.
But most of them don’t. Most of them consume. They fork the repo, strip out the branding, wrap it in a subscription, and call it a platform. Then they hire a developer relations team to go to conferences and talk about community.
I have a mental list of tools that solved nothing but got a conference talk anyway. It’s longer than it should be. The talks always have the same structure: here’s a problem we had, here’s the open source thing we used to solve it, here’s what we named our internal wrapper, here’s the QR code for our Slack. Nobody in the room asks who maintains the thing underneath the wrapper.
The Maintenance Problem Nobody Funds
Here’s the actual crisis, and it’s not theoretical.
Core open source infrastructure is maintained by a terrifyingly small number of people. Some of them are doing it nights and weekends. Some of them burned out years ago and just haven’t officially quit yet. Some of the most critical libraries in the world are sustained by one person who has a day job and a family and a finite amount of hours.
The Log4Shell vulnerability in 2021 exposed this. A logging library used by billions of devices, embedded in everything from enterprise software to Minecraft servers, maintained by a handful of volunteers. When the critical zero-day dropped, the response from the companies whose products depended on it was to open emergency Jira tickets and demand patches. From volunteers. Who didn’t sign anything.
The entitlement was so thick you could taste it.
I trust logs more than I trust people. Always have. And the logs on this one are clear: companies extracted billions in value from free labor and contributed legal threats, feature demands, and the occasional branded hoodie in return.
The Gratitude Is Performative, The Extraction Is Real
Open source maintainers get appreciation the same way infrastructure workers get it during a crisis. Enthusiastically, briefly, and never in a way that changes the underlying conditions.
There are exceptions. The Linux Foundation does real work. The Apache Software Foundation has kept the lights on for decades. Organizations like Sovereign Tech Fund are actually trying to fund maintenance rather than features. These things matter.
But they’re fighting upstream against an industry that has decided that “free as in freedom” also means “free as in we don’t have to pay anyone.”
The real architecture of the internet is a weird patchwork of passion projects, academic research, corporate forks, and underfunded foundations, all duct-taped together by people who mostly just wanted to build something that worked. It held. It keeps holding.
The corporations built the roof. They’ll be the first to tell you about it.
They just don’t mention who poured the concrete, laid the pipes, and wired the walls. Or that those people are still doing it, mostly alone, mostly for free, mostly because no one else will.
