Comments on: A Monad Tutorial for Ocaml http://enfranchisedmind.com/blog/posts/a-monad-tutorial-for-ocaml/ programming, politics, & other religious issues Thu, 11 Mar 2010 20:50:13 +0000 hourly 1 By: Elias http://enfranchisedmind.com/blog/posts/a-monad-tutorial-for-ocaml/#comment-37086 Elias Sun, 13 Dec 2009 08:02:52 +0000 http://enfranchisedmind.com/blog/archive/2007/08/06/307#comment-37086 i liked it :) i'm bookmarking, thank you. i just missed an explanation on doing functional I/O with monads -- this is something that looks very arcane and i never got it very well.. i liked it :) i’m bookmarking, thank you.

i just missed an explanation on doing functional I/O with monads — this is something that looks very arcane and i never got it very well..

]]> By: Louis http://enfranchisedmind.com/blog/posts/a-monad-tutorial-for-ocaml/#comment-36843 Louis Sat, 05 Sep 2009 15:30:48 +0000 http://enfranchisedmind.com/blog/archive/2007/08/06/307#comment-36843 I have a question about the int_val example. I'd have expected that read_int_val should have been something like let read_int_val = SerialMonad.bind int_val (fun r -> (fun () -> !r)) just to make the types work out. What's wrong with how I am thinking about it? Also, do you have sample code for the thread example? I have a question about the int_val example. I’d have expected that read_int_val should have been something like

let read_int_val = SerialMonad.bind int_val (fun r -> (fun () -> !r))

just to make the types work out. What’s wrong with how I am thinking about it?

Also, do you have sample code for the thread example?

]]> By: Paulo Villela http://enfranchisedmind.com/blog/posts/a-monad-tutorial-for-ocaml/#comment-36035 Paulo Villela Tue, 09 Jun 2009 06:35:35 +0000 http://enfranchisedmind.com/blog/archive/2007/08/06/307#comment-36035 Your great tutorial demystifies monads and makes them accessible to OCaml programmers. Thanks! Your great tutorial demystifies monads and makes them accessible to OCaml programmers. Thanks!

]]> By: Benjamin L. Russell’s Adventures in Programming Language Theory Wonderland http://enfranchisedmind.com/blog/posts/a-monad-tutorial-for-ocaml/#comment-34013 Benjamin L. Russell’s Adventures in Programming Language Theory Wonderland Wed, 07 Jan 2009 09:10:08 +0000 http://enfranchisedmind.com/blog/archive/2007/08/06/307#comment-34013 [...] Scheme — Benjamin L. Russell @ 6:09 pm Recently, there was an interesting reference to a monad tutorial for the O’Caml programming language that provided an alternative perspective in helping to [...] [...] Scheme — Benjamin L. Russell @ 6:09 pm Recently, there was an interesting reference to a monad tutorial for the O’Caml programming language that provided an alternative perspective in helping to [...]

]]> By: Nicolas Pouillard http://enfranchisedmind.com/blog/posts/a-monad-tutorial-for-ocaml/#comment-34001 Nicolas Pouillard Mon, 05 Jan 2009 20:38:05 +0000 http://enfranchisedmind.com/blog/archive/2007/08/06/307#comment-34001 @Norry Ogata Your last mapD definition is not syntactically correct (fun (h x) -> ...), maybe you want to write (fun h x -> ...) let mapD (h:’a->’b) (m:’a->’a list) = fun h x -> (List.flatten(List.map h (m x)));; @Brian Thanks for this post! That's always a pleasure to see this concept being more-and-more popularized to other functional languages. I have a small remark about your monads, in particular the return function. One have to remember that OCaml has a call-by-value semantics such that : let return x = fun () -> x OR let return x = lazy x Don't defer any effects, they are just there to fulfill the type, then expanding the definitions will not lead to the same result, "return (print_int 3)" is different from "lazy (print_int 3)", however this is maybe not that problematic. Moreover, there is a typo in your bind definition (ill-scoped/ill-typed): let bind m f = lazy (Lazy.force g (Lazy.force m));; => let bind m f = lazy (Lazy.force (f (Lazy.force m)));; @Norry Ogata
Your last mapD definition is not syntactically correct (fun (h x) -> …), maybe you want to write (fun h x -> …)
let mapD (h:’a->’b) (m:’a->’a list) = fun h x -> (List.flatten(List.map h (m x)));;

@Brian
Thanks for this post! That’s always a pleasure to see this concept being more-and-more popularized to other functional languages.

I have a small remark about your monads, in particular the return function. One have to remember that OCaml has a call-by-value semantics such that :

let return x = fun () -> x
OR
let return x = lazy x

Don’t defer any effects, they are just there to fulfill the type, then expanding the definitions will not lead to the same result, “return (print_int 3)” is different from
“lazy (print_int 3)”, however this is maybe not that problematic.

Moreover, there is a typo in your bind definition (ill-scoped/ill-typed):
let bind m f = lazy (Lazy.force g (Lazy.force m));;
=>
let bind m f = lazy (Lazy.force (f (Lazy.force m)));;

]]>