Yes monads in general are too expressive but the answer isn't to limit the typeclass to something between applicative and monad but rather to limit what monads are allowed. The problem is that there should only be one monad: an effect monad loaded with various effects depending on the side effect needed. Instead of defining this or that monad, there should only be the capability to define the effect you need.

In that case, everything runs within the effect monad and then no one would ever really need to learn what a monad is, just that some calls are effectful (like reading a file or throwing an exception).

It'd be nice to have a process like the following:

1. I free solo a bunch of junk in vanilla javascript with state flowing hither and thither until I'm out of coffee

2. I test the exact behaviors(s) I wanted to make possible in the GUI I just wrote.

3. The framework whitelists only the event chains from my test.

4. For any blacklisted event chains, the user gets a Youtube video screencast of the whitelisted test so they can learn the correct usage of my GUI.

I tried learning Haskell for a decent chunk of time and could make some stuff, but despite trying to learn, I still could not tell you what a monad actually is. All the explanations for it seemed to make no sense.

I'd argue the exact opposite. Compared to what you can do if you can write compilers anything that involves composing functions is weak beer and most monad examples cover computational pipelines as opposed to computational graphs. It's like that Graham book On Lisp, it's a really fun book but then you realize that screwing around with functions and macros doesn't hold a candle to what you learn from the Dragon Book.

And here I thought it was a pedantic word for “data box”

Haskell looks a heck lot like F#, even more than Ocaml if you ask me

Yes. For the same reason that the Yoneda lemma and the Cayley theorem are almost meaningless tautologies once you fully understand what they're saying. "Every small thing (of a certain type) is able to be expressed as a subcase of a bigger thing that contains every single possible subcase in existence." Well no shit.

Interesting, but it seems like he kind of proved himself wrong? Monads are the only option he presented that are sufficiently powerful for normal programs.

[flagged]