Evals
Saying a system is careful is easy. This is how I check. A list of questions with what should happen for each — some it has to answer confidently, some it has to refuse, some it has to hold its ground on when pushed. Then it gets asked all of them and graded.
Red isn't hidden here. A suite that has never failed isn't testing anything, so failures stay on the page rather than getting quietly reworded until they pass. If you see red below, that's the point working — the three groups underneath say which reds I'd stop a release for.
reading latest run…
Three layers, because a failure means three different things
Regression — it used to do this right, and it must keep doing it. Launch — it has to do this before the site is worth showing anyone. Those two turn the board red and stop a release. Future — I want it to do this eventually. A red one there is an ambition, not a fault. It's on the page so the gap stays visible instead of being quietly dropped.
Cases
Loading…