Building the thing, one stubborn evening at a time

I am not a product developer by training. I know enough to build something, and apparently also enough to make it take much longer than it should. Building Double Stop from scratch meant making a lot of small decisions I had never thought about before — how posts should expire, what to ask users for, how carefully to handle someone's location. I spent an embarrassing amount of time on the map picker. This is not something I am proud of.

One rule I set early: this would cost nothing to run. Not because I am especially frugal, but because I wanted it to keep existing without needing to grow or charge anyone. Leaflet instead of Google Maps. Nominatim for geocoding. A free Postgres tier. Every choice was basically the same choice: keep it alive without it needing to justify itself.

The first time it worked end-to-end — post created, post visible, post unlocked — I sat there for a moment. It was a small thing. It was also the thing I had wanted to exist since that Tuesday evening in Tokyo.