Re: Haskell, anyone?

[Hadrien Grasland <hadrien.grasland@gmail.com>]

Le mardi 17 novembre 2015 13:01:24 UTC+1, G.B. a écrit :
> > the horrors of recursion
> 
> Recursion, I speculate, is one of the two things that
> a programmer new to FP style needs to learn, the other
> being combinatorial control of the evaluator's mechanics.
> It is like learning how to write assembly programming,
> with one difference: the "paper" is turned by 90°
> and FP is a little less explicit about instructing the
> machine.(*)

I would say that recursion is the goto statement of functional programming languages : necessary in order to achieve Turing-completeness, and occasionally useful, but makes control flow messy and programs very difficult to reason about, and should thus only be used in situations where no higher-level abstraction will fit.

The recursive hoops that one has to jump through in order to "modify" a complex data structure are especially gruesome. Just implemented a radix tree in OCaml today, as an exercise, and the insertion function gave me a glimpse of what FP hell looks like. Sometimes, immutability can really make things much worse.

I enthusiastically approve of the creative uses that FP languages find to discriminated types, though. It gives me plenty of ideas to experiment with in my future Ada programs.