Tests are greater than types
summary
Claim: when deadlines are tight, writing tests yields more immediate return than adopting TypeScript because tests catch runtime regressions that types miss. Examples: TypeScript introduces a build step, hundreds or thousands of upstream bugs, and wrong runtime behavior when types disappear from published JS (5 + "4" becoming "54"). Workarounds include adding runtime checks, linter changes, and comment hacks. Recommendation centers on write real tests and linters and avoid adding a transpiled type layer unless the team can pay the ongoing cost.
AI-written summary · about
chapters (9)
- 0:00Intro
- 1:06Types are great
- 2:50Why not both
- 3:51Drawbacks of types
- 4:56Typescript
- 6:10HTTP2 sessions
- 7:07Public API
- 8:04Hidden Costs
- 9:25Conclusion
related talks
26mThe Impact of Testing on a DevOps Pipeline
45mBetter types = fewer tests
56mIf You Won't Test Your Code, Who Will?
17mWrite tests. Not too many. Mostly integration.
44mTesting: Testing
24mTesting Less, Achieving More: How we mastered The Art of Strategic Testing
33mWhat people get wrong about testing and how can you leverage tests to develop a lot faster!
21mTesting Javascript: Building JavaScript applications you won't hate