March 2019: functional

calendar Mar 25, 2019 · 2 min read · functional  ·
Share on: linkedin copy

Meetup #58. Please RSVP at Meetup.

We get started at 6:30 pm.

Sponsors

Venue provided by Startup Edmonton.

Food sponsored by Bellstone Engineering.

Talks

Functional Go: How well does Go Support Functional Programming?

Axel von Bertoldi (Kira Systems) took a practical look at how well Go supports functional programming, scoring each concept as supported (A), neutral (B/C), or prevented (F). He walked through the FP pillars — immutable variables, referential transparency, and first-class functions — and related ideas like tail recursion, higher-order functions, monads, pattern matching, currying, and algebraic types. Go scored well on first-class functions (generators, closures, and the functional options pattern) but only B- on immutability and C on referential transparency, since package-level state and in-place operations like sort.Strings are allowed. Tail-call optimization, higher-order functions, and monads all scored F: the first by design, the latter two because they need parametric polymorphism (generics). His conclusion was to use immutability and referential transparency by convention, lean into first-class functions, avoid deep recursion, and wait for generics before adopting HOFs and monads.

Key points:

  • Scored Go against core FP concepts: A supported, F prevented.
  • First-class functions: A — closures, generators, functional options.
  • Immutability: B- — const and immutable strings, but mutable slices/structs.
  • Referential transparency: C — global state and in-place stdlib calls are allowed.
  • Tail recursion: F — Go has no tail-call optimization.
  • Higher-order functions and monads: F — they need generics to be practical.
  • Practical advice: apply FP by convention where it doesn't hurt performance.

Axel von Bertoldi (slides)

News

Events