Hacker News Reader

Comment by effn | Hacker News Reader

I think this is for people that already know how to use monads effectively in Haskell and wants to learn more about the mathematical background.

View on HN

This is explicitly not a writeup for beginners ( ergo the title ), it's a side-by-side description of the /very/ general categorical description and how you can model that in Haskell. The three parameter form of morphism composition ( i.e. c y z -> c x y -> c x z ) would be quite unwieldy in day-to-day programming but it does illustrate the underlying definitions better to have categories explicit.

I don't mean to nitpick, but the three parameter form is exactly what you use in day-to-day Haskell programming. You simply let c be the infix (->), and you get (y -> z) -> (x -> y) -> (x -> z). We just tend to think of it as a two parameter form because a lot of the time, you can think of (->) as syntax, as opposed to the type that it is.

Yup. This is straight cut from Control.Category module:

  instance Category (->) where
      id = Prelude.id
      (.) = (Prelude..)

It's great for two classes of people: those who understand monads from a Haskell/software engineering perspective and want to learn the mathematical perspective, and those like me who understand them from the mathematical perspective and want to learn them from the Haskell/software engineering perspective.

Given that, I think it's great! :D

That's pretty much my point: I don't think mathematical monads are a good background for Haskell Monads. It's a bit like saying the ZF axioms are the mathematical background for Data.Set.

Reply on news.ycombinator.com