if we cannot frankly talk about how the chance of death goes up with old age, and how those deaths change both maintainability and the zeitgeist of any given piece of load bearing software, we're in for a very troubled two decades.
this is spurred by my rant over at https://cohost.org/NireBryce/post/2539531-edit-there-s-a-cha but it's a thing elsewhere too. Sometimes mentions of programmer ages and death are about succession, bus factor, and how hard it is for anyone outside of [someone who's been doing it four decades] to hold the whole context in their heads. The death of original authors, or long term maintainers, also loses your core reference for truth for what a tool is doing, and when you lose the strong opinions of the people with the most experience on the thing, the strongest other opinion takes over even if it's not on good footing.
succession policies are things we should be thinking about before anyone dies, but even broaching the topic seems spicy.
Let alone discuss if succession is worth the cost vs rewrites or new implimentations that have learned from the last decade, when there's already so much context required to transfer.
we're approaching the age when we lose waves of people from a given "graduating class" of zeitgeist, who made and maintain critical points of Linux and open source tooling. We could at least by ready.
I haven't thought much on this in years. But I probably should go back to it.
