Bias reduces noise—if you know <em>roughly</em> what to expect, then being biased by those expectations means you won’t get distracted by less relevant data points.

We usually complain about systems ‘getting in our way’, with arbitrary criteria that determine success. But this goes the other way too. Much of my success and that of those around me is similarly mechanical. Not luck, effort, or nepotism.

The utility of violence isn’t in the violence itself, but only in the threat of it. It creates immediate behaviour change, but only for so long as the threat is active.

Meditation generally involves either acknowledging or excluding thoughts, but can be problematic. Movement-based meditations (e.g. yoga, running) are better for people who can’t sit with themselves. More generally, many everyday activities meet the broad criteria.

Many aspects of my PhD were surprising to me, but in hindsight, they didn’t have to be. Here’s my reflections on how I’d go about it if I’d known.