Does reaching the very top tier of any domain require contrarian, first-principles thinking rather than following consensus behavior?

Paul Graham · published Jul 2023

Knowledge has a fractal edge full of gaps that are hard to see because the brain simplifies; many discoveries come from questioning what everyone takes for granted.

Your brain wants to ignore gaps in order to make a simpler model of the world.

source How to Do Great Work

Paul Graham · published Oct 2023 · hedged

Learning may be the most important case combining exponential growth and thresholds, because knowledge is fractal: pushing at the boundary of one area can open a whole new field whose discoveries you get first crack at.

Some thresholds act like machine tools (reading speeds all later learning), and the most important thresholds are new discoveries — as with Newton, Durer and Darwin.

source Superlinear Returns

Dominik Richter · published Dec 1, 2023

You have to be contrarian and think from first principles to outperform the 99.9%; following the crowd leaves you merely above average

If you follow crowds and don't ask the tough questions or form your own view, you end up above average but never in the top 0.1% of any domain

49:51 Contrarian first principles thinking is required to outperform the crowd

Steve Jurvetson · published Aug 3, 2020

Reasoning from first principles with physics training and working backward from a five-hundred-year vision is what lets founders like Elon Musk overcome the activation energy required for bold moves such as reusable rocket boosters

The physical object becomes a vessel for code — a car is an AI on wheels, a rocket is an autonomous vehicle — so minimizing the physical thing is how you get there

35:00 Physics first principles training plus long horizon vision overcomes activation energy for bold technical bets