Write It Down

A recurring theme in Vivek’s essay How to Be Good at Research is unglamorous and easy to skip: write everything down. Not for posterity — for yourself, right now.

The case has three strands, each borrowed from someone who knew.

Writing finds the gaps. Paul Graham’s observation is that a thought can feel complete in your head and fall apart the moment you try to put it in a sentence. The head papers over holes; the page does not. Writing is how you discover you don’t actually understand the thing you were sure you understood.

It defends against self-deception. Richard Feynman’s line is the one to keep: “the first person you must avoid fooling is yourself.” Charles Darwin ran a procedural version of this — he made a rule to write down, immediately, any observation that cut against his theory, because those were exactly the ones his mind would quietly discard. A log is a commitment device against your own motivated forgetting.

So keep an actual log. The essay’s suggested format is plain: before each experiment, record the hypothesis, the setup, and what you expect to happen; afterward, the result and how your belief changed. The prediction is the important part. A forecast plus a correction, repeated a few hundred times, is how judgment gets built.

There’s a bonus at the end of this habit. Chris Olah and Shan Carter argue that clearly explaining what you’ve learned — paying down the field’s “research debt” — is itself a genuine contribution, and an unfakeable credential besides. The writing you do to understand your own work is the raw material for the writing that helps everyone else.

You will not remember what you didn’t record. Write it down.

References

• Vivek (@itsreallyvivek). How to Be Good at Research. X — for the Graham, Feynman, Darwin, and Olah & Carter examples and the logging format.