What open source means to me

Open source is not an abstraction for me. It is how I have learned, how I have built, and how I want to work.

Three things make it matter. The first is reliability: software you can fork and depend on, certain that no feature you rely on will one day disappear behind a paywall. The second is flexibility: because you can fork and adapt it, you can bend a tool to your organization’s needs instead of the other way around. The third, and most important, is the people: open-source tools tend to be built by people who genuinely care, because the work is voluntary. That care shows in the result, in a way it rarely does in features shipped only to satisfy some lenient acceptance criterion.

Open source is also how I learned. It was the scikit-learn documentation, as much as any course, that made me proficient with machine-learning algorithms during my master’s, documentation written by people with a cause, who wanted to equip newcomers to do real things with data. I have never forgotten that a well-documented open-source project is a teacher.

Which is why I care about the thing that keeps people out. There is an implicit gatekeeping around contributing to open source (the podcasts, the conferences, the quiet sense that it is all for insiders), and the people it shuts out are exactly the younger, less experienced ones, like me, who have no obvious way to start. I believe documentation is the way in. If we first show newcomers that serious companies run entirely on open-source stacks, so the skills make them employable, and then guide them to read a tool’s documentation and improve it, something happens: by the time they have made the docs better, they already understand the tool. They have the confidence to use it, to contribute further, and to bring other people in after them.

That progression, from reader, to contributor, to advocate, is the one I want to help other people make. It is, after all, the one I am making myself.