Do Programming Language Features Deliver on their Promises

on YouTube ↗52m

" summary

A full compiler is implemented in a points‑free style and turns out harder to read and prone to duplicated logic so the points‑free compiler experiment works but increases cognitive load; a comparison shows a 17‑line compiler versus a nanopass thousand‑line version and a larger project that becomes 1500‑line instead of an expected 50–100k lines; a static‑analysis effort targets real codebases (about 5 million lines of APL) and proves feasible because APL is “statically dynamic.”

AI-written summary · about

" chapters (11)

1. 0:00Programming language promises
2. 3:11Simple vs complex design
3. 6:02Points-free programming pitfalls
4. 13:09Hidden costs of DSLs
5. 20:10Reflection in dynamic languages
6. 25:09Garbage collection examined
7. 30:04Control flow complexity
8. 36:51Static typing trade-offs
9. 42:52Formal methods future
10. 45:39Scaling code complexity
11. 46:44Q&A

" related talks

• 55mHow bad is meta-programming still today?Peter Gottschling · Meeting C++ 2016
• 01h 03mto pass and return, the story of functions, values and compilersDawid Zalewski · Meeting C++ 2021
• 56mPractical C++17Jason Turner · Meeting C++ 2017
• 01h 00mJust-in-Time CompilationJF Bastien · CppCon 2020
• 01h 50mUnderstanding Compiler OptimizationChandler Carruth · Meeting C++ 2015
• 01h 01mThree Unlikely Successful Features of DAndrei Alexandrescu · NDC London 2017
• 37mA (Not So Gentle) Introduction To Systems Programming In ATSAditya Siram · Strange Loop 2017
• 01h 04mWhat Can Compiler Benchmarks Reveal About Metaprogramming Implementation StrategiesVincent Reverdy · CppCon 2022