Learning

Knowledge doesn't spread on its own — it concentrates in whoever happened to build something, unless it's actively reproduced in other people. This is social reproduction: not writing knowledge down once, but continuously passing it on through shared practice, as deliberately organised as the infrastructure itself. That means creating moments to work on the stack side by side, not just documentation to read alone: pairing on a deployment, debugging something together, a working group with a dedicated channel, and occasional meetups, webinars, or hackathons that turn maintenance into something social rather than a private chore. The goal isn't more content, it's fewer bottlenecks — every person who learns by doing is one less person the system depends on exclusively.