Killed by Worse is Better What we can learn from the last time FP failed (TBR)
summary
Argues a concrete cause: functional languages lost not from theory but from adoption factors — Lisp versus Unix tradeoffs, Haskell’s slow standards and Cabal treadmill, and Go’s pragmatic design with a measured study finding 171 concurrency bugs mostly from language semantics. Shows what was built and whether it worked: Haskell as a research lingua franca, Go as a practical workplace win with surprising bug classes, and historical fixes (wash hands / chlorinated lime) as a model for cheap effective interventions. Recommends tooling, undergrad curriculum changes, and measurable industry metrics.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:04Talk overview and functional programming context
- 3:36Lisp’s contributions
- 6:25Unix’s success
- 10:57Fox versus hedgehog
- 12:26Lisp Machine rivalry
- 17:19Haskell’s limits
- 22:20Go’s trade-offs
- 24:49Learning environments
- 26:32Semmelweis resistance
- 31:52Human nature in dev
- 32:54Type systems
- 35:42Industry’s role
- 37:10Practitioners’ responsibility for correctness
- 38:55Structured programming
- 40:40Haskell’s persistence
- 42:06Worse is better
- 43:56Q&A
related talks
44mThis is why we can('t) have nice things
27mClosing Keynote
56mThe Tale of Troy
40mRust’s approach of getting things right
30mBack to the Future: Rust, Zig, and the New Age of C++ Alternatives
31mThe Rust That Could Have Been
01h 09mFunctional Programming in C++
58mThe Imperatives Must Go! [Functional Programming in Modern C++]