Keynote – Yukihiro 'Matz' Matsumoto
summary
History shows PL calcu predates FORTRAN with a 1942 design and a first compiler only in 1998, proving ideas need implementations. Ruby appears in 1993 and ships changes incrementally: Ruby 1.9 VM rewrite (2007) yields much faster VM performance (authors report 3–50× speedups), recent work delivers Ruby 3.3 with JIT and heap compaction and Ruby 3.4 planned, teams (Shopify Cookpad Square) build the prism parser gem, add fiber schedulers and improve memory and concurrency while keeping Ruby’s syntax and compatibility.
AI-written summary · about
chapters (25)
- 0:00Introduction
- 0:48Programming language as life’s work
- 2:14Earliest languages and PL/Calcul
- 6:30Idea needs implementation and community
- 7:36Simplicity versus mental models
- 12:16Featurism and complexity in languages
- 13:21Perl’s evolution and lessons
- 17:39Second system syndrome in redesign
- 18:44Ruby’s near-rebuild and split
- 21:40Compatibility in language evolution
- 23:29Ruby’s rise and fall in hype
- 25:56New blood keeps Ruby alive
- 27:25Goals for Ruby’s relevance
- 29:50Ruby’s performance and memory
- 31:17Tools matter more than design
- 34:37Inviting ideas for Ruby’s future
- 36:31Optimize for small teams
- 38:21Ruby’s release cycle and future
- 39:25Crazy ideas for Ruby’s next decade
- 42:02Training Ruby successors
- 44:13Call to action for Ruby
- 45:15Q&A on Ruby’s GIL and WebAssembly
- 49:33RBS and type system future
- 51:00Ruby’s future in ten years
- 54:15Closing remarks
