Why OSS funding cannot be treated like public infrastructure
I read Dries’ article on funding open source like public infrastructure and I get the instinct. I’m a big proponent of OSS and have been a heavy user for most of my life, but I’m skeptical of the public infrastructure analogy. When governments fund infrastructure - say water systems or public transit - they improve assets they own (or via concessions with strong step-in rights) and can hold to strict service levels. Donating to an OSS foundation typically buys neither ownership nor enforceable SLAs, nor a binding say on priorities. That’s not a critique of OSS; it’s, arguably, the point of it. It does mean public money isn’t turning into public assets with clear accountability the way infrastructure spend does.
That dulls the incentive for governments or citizens to donate purely for the good of it: there is no asset on the books, no control rights, and no guaranteed service levels to point to. There is also the human element: when you donate to something you use, part of the motivation is knowing you also help others forward. That sense of shared benefit is not always straightforward in the current geopolitical climate, which can make collective investment even harder to frame.
What is the answer? I'm not sure. I do not think there is a single, clean answer; it depends on the type of OSS and its strategic relevance. You can think along the lines of paying for LTS policies and release engineering, or providing capacity by actually funding maintainers' wages. Each path has pros and cons.
In the end, it comes down to a simple tension. OSS runs on permissionless collaboration, shared stewardship, and the option to fork. Public spending runs on ownership, enforceable obligations, and long-term strategy input. Those pull against each other.