Hackers & Painters · Luis Larota

(spoiler: the chapters don't talk to each other i believe, so neither will my notes)

tribes are the default setting (by history) - the only move is picking one. pg's read on why school is savage isn't that kids are cruel, it's that they're sealed off: "the inhabitants of all those worlds are trapped in little bubbles where nothing they do can have more than a local effect." the history i wanted: adolescence is a ~120-year-old invention. G. Stanley Hall named it as a life stage in 1904; the word "teenager" barely shows up in print before 1941. before that you went kid > apprentice, doing real work next to adults who needed you. so the bubble isn't nature, it's a policy we (humans) wrote. kids aren't left behind by their peers, they're left behind by us (adult now) who stopped having a use for them.
the craft split is real but porous: "architects decide what to do, and engineers [figure out] how to do it" — and immediately, "what and how should not be kept too separate." you don't receive a spec, you should sketch one: "you should figure out programs as you're writing them." everyone should fear premature optimization; pg says fear premature design just as much - the plan you commit to before you've built enough to know. and the hiring tell that stuck: at viaweb "the main thing we cared about was what kind of software they wrote in their spare time." the side project is the interview!!
(i disagree here) "when a piece of code is being hacked by three or four different people, no one of whom really owns it, it will end up being like a common-room." that's tragedy of the commons, fair - pg on this side. but the failure on the other side is worse: sole ownership is a bus factor of 1 and exactly one point of view on the design. you building it first doesn't mean i can't improve it — my change goes through review anyway, and the reviewer is you. treat a codebase like a marriage: both contribute, both decide. a commons rots when nobody is accountable, not when more than one person is.
the two directions of thinking + pg's best line is the outward one: "looking at things from other people's point of view is practically the secret of success." promotion - what does your manager need to walk into their review with? sport - what's the guy across from you planning? and the thing i couldn't figure (how is something associated but not correlated?) he answers himself ... one sentence later: "lack of empathy is associated with intelligence, to the point that there is even something of a fashion for it in some places. but i don't think there's any correlation." also, "you need a brain that's in the habit of going where it's not supposed to", paired with "draw a sharp line between your thoughts and your speech." my test for it: take something people said with a straight face in 1850, then ask what we say now that lands the same way in 2150. outward is how you win; inward is how you get something worth winning with.
what is a plan if it changes? pg: "plans are just another word for ideas on the shelf." viaweb shipped "three to five releases a day" while desktop shops shipped twice a year — once shipping gets cheap, a roadmap stops being a promise and becomes a guess you keep around. same for the friday-night release: the question isn't whether it's on the plan, it's whether you can undo it by saturday. and the feedback loop only works from both ends — "the least sophisticated users show you what you need to simplify and clarify, and the most sophisticated tell you what features you need to add." but who tells you what to delete tho?

mental model: the customer is always right, but different customers are right about different things — the question is never "who's right", it's "right about what".
"if you want to make a million dollars, you have to endure a million dollars' worth of pain." a conservation law — though i'd say it holds to the first million and after that it's compounding, not suffering. (papi musk crossed a trillion on june 12, 2026; no way that's a trillion dollars of pain. or... maybe.) underneath it are two dials: measurement (can your output be told apart from the team's? ceo yes, entry-level mostly no) and leverage (do your decisions move anything?). the tell that leverage is real is that you can lose — "upside must be balanced by downside." and aim at wealth, not money: "what most businesses really do is make wealth. they do something people want." money is the receipt. edison's ratio is the tax on all of it — one percent inspiration, ninety-nine percent perspiration. sudar frío!!
who you stand next to sets your ceiling, and the sharp version isn't the twain line about arguing with idiots (he never said it — it traces back to cocteau). it's norbert wiener, 1950: "any labor which competes with slave labor must accept the economic conditions of slave labor." he was writing about automation. seventy-six years later that stopped being a metaphor. so don't compete on the sort of axis where the machine is free — go find the room where the axis is something else. not just silicon valley. the marina too, lol!
you don't show up with taste, you accumulate it: "as you continue to design things, you'll get better at it. your tastes will change." brian armstrong gives the mechanism in three words — "action produces information" — do the wrong thing on purpose, find out what right was. and pg's list tells me what to actually look for: "good design uses symmetry... there are two kinds of symmetry, repetition and recursion." symmetry bro! which also repairs shaw's line — "he who can, does; he who cannot, teaches" only holds when teaching came instead of doing. i've seen plenty of that in south america. the version worth respecting: those who have already done it, teach.