Ruby’s Core Gem
summary
Proposes extracting Ruby’s core into Ruby core in Ruby (distributed as a gem or bundled) and leaving a tiny set of C/primitive hooks; cites Rubinius origin and TruffleRuby’s existing implementation and shows actual wins: TruffleRuby runs ~2,386 core methods in Ruby versus MRI’s ~2,194 C methods with ~64 primitives, enabling inlining and JIT optimizations that sometimes make Ruby code faster than C. Demonstrates tooling (yjit style inlining, Sulong for C ext) and notes real tradeoffs: startup and memory versus debuggability and shared, optimizable core.
AI-written summary · about
chapters (7)
- 0:00Start
- 7:48Best of both worlds
- 13:31Advantage - understandability
- 14:09Advantage-shared code
- 14:50Advantage - optimisation
- 17:56Disadvantage - memory - mitigate it
- 22:41A potential way forward
related talks
57mRuby: 2020
35mRuby’s C Extension Problem and How We're Solving It
49mDeconstructing the hype: What people think is wrong with Ruby?
52mRubinius
44mRubinius in One Act OR Rubinius: A War on Two Fronts
01h 23mThe Ruby Implementors panel
30mFast Metaprogramming with Truffle
01h 08mStrengthening the Ruby Ecosystem Part I: Rubinius