Week 12: What a Fair One-Shot Machine Would Look Like

Week 1 asked you to notice the exact moment you weren’t sure what to do next, at a parking meter. Eleven machines later the question hasn’t changed — only how confidently, and with what evidence, you can now answer it.

Outline#

• twelve machines, one recurring question: who benefits when the design is bad? — reviewed across parking meters, kiosks, ATMs, tolls and turnstiles
• what a redesign has to keep, not just what it gets to remove: fraud prevention, throughput, and whatever legitimate constraint made the original design defensible in the first place
• the difference between a fairer machine and a machine with no operator left to keep it running

The argument#

Eleven machines have each asked the same question of a different target: who benefits when this fails? None of the answers were “nobody” — every one of these machines works, reliably, for whoever wrote its spec. That’s the uncomfortable throughline: bad design here isn’t a mistake anyone is trying to fix, it’s a solution someone is trying to keep.

A fair redesign has to do more than remove friction. Fraud prevention, throughput, and whatever other legitimate constraint shaped the original machine don’t stop being legitimate just because the design around them was unfair — a redesign that removes all of them hasn’t produced a fairer machine, it’s produced one no operator would keep running, which helps nobody either. The harder work is keeping the legitimate constraint while removing the one that was never legitimate to begin with.

Question to carry: of everything your own redesign removes, what’s the one thing you deliberately chose to leave in — and could you defend it to the person it protects?