~~

Working through a few tutorials and books on Haskell I think I’m starting to feel that it’s all very good until it comes to Stateful Fucking Computations.

Well, I know that I do not know Haskell. And I know that I cannot wrap may head around all of its concepts right now (Monad Transformers? Arrows for god’s sake?). But on the other hand I see how it becomes, well, almost explosionary more compilcated as you start adding Stateful Fucking Stuff. Monad + Monad + Monad Transformer + Another one… And if you just have missed somehow that you will need, e.g. ST + IO + Error, and have just implemented smth without Error… Well welcome to ErrorT rewriting of great bunch of you functions. And it doesn’t seem to be always easy.

And of course I still have troubles with types. X :: A b c d WTF? Don’t yet clearly understand the difference between type, newtype & data. Well data is for algebraic datatypes. But type and newtype do somewhat confuse me.

And actually, writing an interpreter seems to be much harder (Parsec, State, IO through most of the program, Error throwing anywhere) than writing a compiler (Parsec -> Tree, work on the tree in a purely functional manner, generate code in a purely functional matter, you will only have to handle errors somehow in a good way. But I think it’s not that hard to do this w/o using monads here, actually. And IO only for Input once & Output once, all in main :: IO ()).

Frankly, Scheme seems to be MUCH easier to grasp. But after all, that’s one of the points why I actually did start to grok Haskell.

Good luck me.

Tags: code, hs, Thoughts

This entry was posted on Thursday, May 8th, 2008 at 1506 and is filed under Haskell, Programming. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.