Functional Programming is Terrible

on YouTube ↗35m

" summary

Argues that Scala forces workarounds for pure FP: no general tail-call elimination in Scala, so composing many functions needs trampolines with memory and CPU overhead that hurt performance; higher-kinded types require manual kind annotations or ugly Lambda/unapply hacks (State[A,S] can’t be partially applied); lack of kind/type inference, boxed function objects, and missing existential/universal types force boilerplate. Recommends tail-call bytecode, kind inference and better type inference; nevertheless modularity, compositionality and reduced cognitive load make FP worth it.

AI-written summary · about

" chapters (18)

1. 0:00Intro
2. 2:53Tail recursion
3. 3:45Inter-method tail calls
4. 4:23Tail calls in FP
5. 5:34Trampolines
6. 10:03Kinds
7. 12:34Applicative functors
8. 13:15Code reuse?
9. 14:29Currying?
10. 15:07Unapply
11. 17:27FP made less terrible
12. 19:44is FP just a hair shirt?
13. 20:59Modularity
14. 22:07Functions are modular
15. 23:10Compositionality
16. 24:11Functions are compositional
17. 26:10FP = Happiness
18. 30:02Summary

" related talks

• 49mFunctional Programming EssentialsKelley Robinson · Scala Days Copenhagen 2017
• 48mObject Functional ProgrammingRahul Goma Phulore · Functional Conf 2014
• 19mScala for the Rest of UsPaul Cleary · Scale By the Bay 2017
• 59mIntro to FP in Scala: Exploring a Toolbox for Solving Problems - Part 3 of 6Pawel Szulc · LambdaConf 2017
• 37mIntro to FP in Scala: Exploring a Toolbox for Solving Problems - Part 6 of 6Pawel Szulc · LambdaConf 2017
• 01h 00mIntro to FP in Scala: Exploring a Toolbox for Solving Problems - Part 1 of 6Pawel Szulc · LambdaConf 2017
• 38mIntro to FP in Scala: Exploring a Toolbox for Solving Problems - Part 5 of 6Pawel Szulc · LambdaConf 2017
• 42mPolymorphic Function Types in Scala 3Guillaume Martres · Scala Days 2023 Seattle