Geocoder overconfidence — why 90% coverage isn't enough, why Google's API and geocode.earth have ceilings.
View All TagsMailwoman couldn't find Tirana, or Yerevan, or 145 other capitals — not a wrong coordinate, silence. We were going to fix the parser. A ten-minute probe killed that plan, and a stale symlink nearly talked us into a GPU retrain to fix a model that was already fine. Two evals, two silent confounds, and the one dull habit that caught both.
You can point a Nominatim client at Mailwoman now and get a geocoder from a SQLite file, without PostgreSQL or a planet import. We did, and the US worked beautifully while the rest of the world resolved to its American namesake. The bug was one line: we'd told it the answer was always America. This post covers the line, the four cities that still won't move once you delete it, and the ranking database we forgot to switch on — which faked the entire scoreboard until a contradiction gave it away.
Our parser's international-order German 'collapsed' to 44%, so we retrained it three times. Then we measured the thing that can't be gamed and found out the model had been right all along. The metric was lying. Native German was sitting at 96%.
Our resolver kept dropping German addresses in New Hampshire, and our scorecard handed it a gold star every time. This post shows how a name-match metric lied to us for months, and how much a postcode hint improved placement once we measured by distance instead.