The more we test, the worse?
summary
Three concrete failure modes: missing prevention where a medical device lacks safety checks and a high‑fidelity music player spends months testing but misses market timing; misplaced automation that produces flaky end‑to‑end tests and slow feedback while visual regressions need dedicated tools not generic UI scripts; and misleading metrics where bug quotas and 100% code coverage produce combative KPIs and false confidence (example XCOM code). Recommended fixes include build‑quality practices, shift‑left collaboration, and outcome metrics like DORA rather than vanity numbers.
AI-written summary · about
chapters (5)
- 0:00Introduction
- 3:28Quality is more than just catching bugs
- 9:47Automated tests and law of instruments
- 15:25Harmful quality metrics
- 21:45Summary
related talks
56mThe End Game
31mMeasure it or it didnt happen
24mWHY ARE WE FAILING WITH TEST AUTOMATION?
46mThe state of the Art in Tackling Flaky Tests
24mHow to solve intermittent issues, aka Random Failures
14m3 Automation Testing Mistakes & How to Correct Them?
50mWhy Everything Is Spoiled
41mExits on the (QA) Highway to Hell