Pascal apologised for writing a long letter on the grounds that he had not had time to write a short one, and the line survives because everyone who has written anything recognises it immediately. What is less often noticed is how completely it generalises. The long letter, the complicated architecture, the paper with six contributions, the strategy with eleven priorities: in nearly every case, the complicated version is not the ambitious one. It is the one that was shipped before the thinking was finished.

This is worth stating plainly because we tend to read complexity as a signal of sophistication. A dense design suggests the designer was grappling with hard constraints. Sometimes that is true. Far more often the complexity is simply undigested material: every consideration that came up has been accommodated somewhere in the structure, because working out which ones actually mattered would have required another week of thought that nobody had. Simple is what remains after you have understood the problem well enough to know what can be thrown away, and that understanding is the expensive part. The complicated version is cheaper precisely because it lets you skip it.

The practical difficulty is that the cost is paid by the author and the benefit is received by everyone else, which is a poor incentive structure. But the compounding returns are enormous, because every future reader, maintainer and user pays a tax on complexity for as long as the thing exists. So, as a reflection for today: what have you shipped recently that was complicated mainly because simplifying it would have cost another week?