Heuristics That Almost Always Work An Article by Scott Alexander astralcodexten.substack.com Sometimes there’s a Heuristic That Almost Always Works, like “this technology won’t change everything” or “there won’t be a hurricane tomorrow”. And sometimes the rare exceptions are so important to spot that we charge experts with the task. But the heuristics are so hard to beat that the experts themselves might be tempted to secretly rely on them, while publicly pretending to use more subtle forms of expertise. …Maybe this is because the experts are stupid and lazy. Or maybe it’s social pressure: failure because you didn’t follow a well-known heuristic that even a rock can get right is more humiliating than failure because you didn’t predict a subtle phenomenon that nobody else predicted either. Or maybe it’s because false positives are more common (albeit less important) than false negatives, and so over any “reasonable” timescale the people who never give false positives look more accurate and get selected for. expertiseheuristicsprediction
Beauty and compression An Article by Scott Alexander astralcodexten.substack.com The Buddha discusses states of extreme bliss attainable through meditation: Secluded from sensual pleasures, secluded from unwholesome states, a bhikkhu enters and dwells in the first jhāna, which is accompanied by thought and examination, with rapture and happiness born of seclusion. ...If you could really concentrate on a metronome, it would be more blissful than a symphony. The jhāna is also a strong contender as a theory of beauty: beauty is that which is compressible but has not already been compressed. The Abode of the Unsymmetrical beautysilencesensesattention
Negative Creativity An Article by Scott Alexander slatestarcodex.com Coming up with entirely novel ideas is really, really hard. Misinterpretation as inspirationSit Down And Think About It For Five Minutes ideascreativitymetaphor
Deadlines are bullshit An Article contrariantruth.substack.com In software development deadlines are a necessary evil. It is important to understand when they are necessary, and it is important to understand why they are evil. External vs. internal deadlinesWhy are internal deadlines evil?Engineers who love their work Hofstadter's LawThe Thing-deadline calculusNever enough timeDriving engineers to an arbitrary date is a value destroying mistake bureaucracysoftwareprocesswork
External vs. internal deadlines When are deadlines necessary? Contractual obligations Technical liabilities (e.g., dependency EOL) Compliance, government, investors, and other external stakeholders What do all of these deadlines have in common? They are all important. They are all deadlines that cannot be missed. They are all external. When are deadlines evil? Your manager says you have a deadline Your software development methodology says you have deadlines What do all of these deadlines have in common? None of them are important. They are arbitrary. They are all internal. They are all bullshit.
Why are internal deadlines evil? Estimation: When estimating engineering work a substantial time investment is required by an engineer in order to get an accurate estimate. Misaligned Incentives: There is an incentive to lie and give estimates much longer than the feature is truly expected to take. Low Morale: Deadlines are likely to be missed often. Repeated failure has a cost to the morale of the team. Micromanagement: Deadlines are wielded by middle managers as a whip to harass and annoy engineers working on features. High Stress: When engineers feel the pressure of other stakeholders holding deadlines over their heads it creates an environment of high stress. High Turnover: On teams with high turnover rates the best engineers have an easy time finding new work and leave quickly, the worst engineers have a difficult time finding work and remain. This selects for a lower quality team over time.
Engineers who love their work The resolution is simple. Never have internal deadlines. Operate on a prioritized and ordered list of features. Estimate only when necessary to prioritize and do so in a t-shirt sizing way. Trust your engineers and they will begin to love their work. Engineers who love their work are happy and productive. Building is never a straight line