Harnessing other languages to make Ruby better
summary
The talk describes using Ruby for data analytics at a startup, the limitations encountered, and a solution that borrows ideas from LISP and ActiveRecord scopes to add data-science-level performance and expressive power to Ruby. It also discusses lessons Ruby can learn from other languages while explaining why Ruby remained the team's preferred language.
AI-written summary · about
chapters (24)
- 0:00Start
- 2:40The Challenge
- 3:43The Requirement
- 6:29Plain Ruby
- 7:28Vectorisation
- 8:59Bitmap C Extension
- 10:05GSL
- 11:01NArray and Parallel
- 12:22Pushing Ruby Analytics
- 12:56Performance, Abstraction & Complexity
- 16:13The JVM: Scala
- 17:32The JVM: Clojure
- 19:16Why Python?
- 22:46Bonus extras
- 24:41Pandas issues by label
- 26:20Problem statement
- 26:54API Inspiration
- 27:21API implementation problem
- 28:54Lessons from LISP
- 30:08S-Expressions
- 30:58Simple s-expression example
- 31:29Example with nesting
- 32:59Closures (Blocks) are surprising
- 34:45Added benefits
