Skip to content

Archive

30 issues · 83 links · Updated weekly

  1. Issue 30
    • 01My Approach to Building Large Technical ProjectsMitchell HashimotoLarge projects should be organized around how quickly they produce a real result, with frequent demos that expose mistaken assumptions early.
    • 02Unbundling Tools for ThoughtFernando BorrettiA single knowledge graph is not automatically better than focused tools; links should emerge from use rather than become a burden imposed in advance.
    • 03Seeing like a software companySean GoedeckeOrganizations make work legible through processes and metrics even when the untracked work of judgment and relationships is what keeps the system working.
  2. Issue 29
    • 01Why Story Points Don't WorkkqrAn estimate that cannot be falsified cannot become accurate, so story points create the appearance of measurement without a learning loop.
    • 02Betting on Things That Never ChangeMorgan HouselStrategy is often stronger when it is built around durable human constants rather than predictions about which changes will win.
    • 03The Three Sides of RiskMorgan HouselRisk has a probability, an average consequence, and a tail consequence, and the tail is often what determines the historical outcome.
  3. Issue 28
    • 01The Decline of DevianceAdam MastroianniBoth good and bad forms of deviance can decline when life becomes more expensive and cautious, making unusual behavior a deliberate choice.
    • 02Reading the news is the new smokingAdam MastroianniNews consumption can become entertainment that makes people feel informed without improving their models of the world.
    • 03Build vs BuykqrBuild deeply integrated, hard-to-evaluate components when becoming an expert is the only way to buy them safely; buy the rest.
  4. Issue 27
    • 01Every Man a MicroserviceGrant SlattonA service should belong to someone who can hold its whole context in mind; shared ownership without context creates accumulation rather than stewardship.
    • 02The Bus Ticket Theory of GeniusPaul GrahamGenius often grows from an apparently useless obsession that keeps someone engaged through the long, boring stretch where practical minds stop.
    • 03Superlinear ReturnsPaul GrahamMany important rewards are superlinear because they come from compounding or thresholds, while people instinctively expect linear returns.
  5. Issue 26
    • 01Why So Few Matt Levines?GwernExceptional educational newsletters are rare because they require a rich domain, a writer who could act instead, and patience for repeating the explanation.
    • 02Doing nothing at workSean GoedeckeHigh-impact engineering is often driven by rare opportunities, so leaving some capacity unused can matter more than continuously finishing low-value tasks.
    • 03Human BottlenecksFernando BorrettiAI tools cannot multiply an outcome when the real bottleneck is motivation, knowledge, executive function, or a concrete reason to act.
  6. Issue 25
    • 01Beware The Man Of One StudyScott AlexanderA large pile of studies and expert signatures can still be misleading because research results naturally vary and selective evidence is easy to find.
    • 02On PriesthoodsScott AlexanderExpert communities create useful boundaries but can also isolate themselves and become vulnerable to biases that appeal to their own intelligence.
    • 03On Seeing Through and Unseeing: The Hacker MindsetGwernThe hacker mindset deliberately looks beneath abstractions because useful abstractions leak and hidden assumptions become vulnerabilities.
  7. Issue 24
    • 01Ads Don't Work That WayKevin SimlerAdvertising often works by changing what a product means in public culture, not by secretly implanting a feeling in each consumer.
    • 02Better Than FreeKevin KellyWhen copies are free, durable value comes from qualities that cannot be copied, such as trust, immediacy, interpretation, and access.
    • 03Trapped Priors As A Basic Problem Of RationalityScott AlexanderStrong priors can twist every new observation into support, so escaping them requires gradual exposure and explicit mechanism-based reasoning.
  8. Issue 23
    • 01Everything that turned out well in my life followed the same design processHenrik KarlssonGood work and good lives often emerge by letting the situation teach you their form instead of forcing a vision onto them.
    • 02Looking for AliceHenrik KarlssonImportant relationships are discovered through particular people, not categories, and social advice can steer us toward safer but flatter choices.
    • 03Crony BeliefsKevin SimlerSome cherished beliefs serve social identity more than truth; the strongest emotional reactions can reveal when that is happening.
  9. Issue 22
  10. Issue 21
    • 0195%-ile isn't that goodDan LuuBasic deliberate practice is rare enough that modest improvement among practitioners can place someone near the top of the whole population.
    • 02Why is it so hard to buy things that work well?Dan LuuMarkets often reward claims of quality more reliably than quality itself because customers cannot easily evaluate what they buy.
    • 03Don't Read History for LessonsCedric ChinHistory is better for learning concepts and calibrating what is possible than for extracting simple lessons that supposedly apply everywhere.
  11. Issue 20
    • 01Staring into the abyss as a core life skillBen KuhnDecisions improve when you face uncomfortable possibilities early instead of protecting yourself from them with optimistic stories.
    • 02Searching for outliersBen KuhnWhen important outcomes have heavy-tailed distributions, searching for exceptional possibilities beats optimizing a single merely acceptable choice.
    • 03You don't need to work on hard problemsBen KuhnThe world rewards important problems more than difficult-looking ones, and smart people often mistake difficulty for leverage.
  12. Issue 19
    • 01Face it: you're a crazy personAdam MastroianniTreating your unusual traits as information about the life that fits you can be more productive than trying to appear normal.
    • 02Science is a strong-link problemAdam MastroianniScience is often governed to prevent its weakest work, even though its value may come from a small number of exceptional contributions.
    • 03Surely you can be seriousAdam MastroianniSome games reward you with more of the game; the way out is often to stop, not to win harder.
  13. Issue 18
  14. Issue 17
    • 01Reality is Very Weird and You Need to be Prepared for ThatSlime Mold Time MoldWhen a reliable theory meets a stubborn counterexample, the right response may be to accept a more complicated reality rather than discard the evidence.
    • 02Littlewood's Law and the Global MediaGwernIn a world of billions, rare events happen constantly, so anecdotes become nearly worthless evidence at scale.
    • 03Notes on PuzzlesNabeel QureshiExperts spend far more time trying to falsify an idea than amateurs do, but excessive skepticism can kill projects that must be tested in the world.
  15. Issue 16
    • 01Heuristics That Almost Always WorkScott AlexanderExperts can look correct almost all the time while adding little value; judgment matters most in the rare cases when the heuristic fails.
    • 02Epistemic Learned HelplessnessScott AlexanderPersuasive reasoning is not evidence by itself, because false and true arguments can be equally persuasive.
  16. Issue 15
  17. Issue 14
  18. Issue 13
  19. Issue 12
  20. Issue 11
    • 01High AgencyGeorge Mack · 30 min readIf your problem doesn't defy the laws of physics, it's not an unsolvable problem -- regardless of what other people say.
    • 02Taste for MakersPaul Graham · 11 min readWhen you're forced to be simple, you're forced to face the real problem.
  21. Issue 10
  22. Issue 09
  23. Issue 08
  24. Issue 07
  25. Issue 06
    • 01Being GlueTanya ReillyGlue work is the difference between a project that succeeds and one that fails.
    • 02How Will You Measure Your Life?Clayton M. ChristensenYour decisions about allocating your personal time, energy, and talent ultimately shape your life’s strategy.
    • 03Inventing on PrincipleBret VictorCreators need an immediate connection to what they’re creating.
  26. Issue 05
  27. Issue 04
  28. Issue 03
  29. Issue 02
  30. Issue 01