Say More
summary
Acceptance tests become assertions about accessibility, tab order, keyboard interaction and network churn rather than vague behavior descriptions. The talk shows how to build wrapping helpers like fillIn by label and keyboard click, a naive calculateTabs check, a track requests with Mirage/Pretender utility that surfaces extra HTTP calls, and a small Babel plugin to auto-generate keyboard variants. Demos are concrete: code sketches, failures that pinpoint missing labels or extra requests, and simple implementations that actually work.
AI-written summary · about
