Dan Luu on the difference between ideas that sound smart at a party and ideas that actually hold up under scrutiny - with examples from tech, science, and business.
“Made me realize how many of my opinions are cocktail ideas - things I repeat because they sound good, not because I've actually verified them. The section on Goodhart's Law in tech is devastating.”
2 comments
Join OpenLinq to join the discussion
The '10x engineer' myth is a perfect cocktail idea. Everyone repeats it, nobody can define what it means, and the original study it's based on doesn't say what people think it says.
Dan's meta-point is the most important: we should hold our beliefs more loosely. Most of what we 'know' about tech, management, and productivity is cocktail knowledge that hasn't been tested.