> "If our principal architect went on parental leave tomorrow, three production systems would be operating on faith." ^pain A collective survivability pain. Critical subsystems, customers, or processes depend on one or two experts. Incidents stall when they're unavailable. The experts are too busy to document, and the fragility compounds because their absence is fearsome enough that the organisation organises around their presence — making documentation feel even more redundant. ## Discovery questions - "Name a system in your stack where one person is the only one who fully understands it." - "What happens if your most-relied-upon engineer is on vacation when something breaks?" - "Which of your customer relationships sit in one person's head?" ^discovery-questions ## Examples - In 2016, the author of left-pad unpublished his tiny npm package, breaking thousands of builds worldwide - a single maintainer disrupting large parts of the ecosystem.[^1] - Academic study on the bus factor of GitHub projects: many widely-used repositories effectively depend on one primary maintainer, posing systemic risk.[^2] - The "Ken Thompson hack" lore around early Unix compilers illustrates how security knowledge and build-chain trust was effectively held in one individual's head.[^3] [^1]: https://blog.npmjs.org/post/141577284765/kik-left-pad-and-npm [^2]: https://dl.acm.org/doi/10.1145/3180155.3180200 [^3]: https://dl.acm.org/doi/10.1145/358198.358210