> I'd like to point out that do-notation in Haskell is a syntax sugar for something nasty. [..] Which is exactly the callback hell criticized in the article. W

> I'd like to point out that do-notation in Haskell is a syntax sugar for something nasty. [..] Which is exactly the callback hell criticized in the article. What's even worse - the resulting code gets compiled to machine code. That's usually very ugly, there is a limited number of variables (registers), and you have to juggle them around. So, obviously, it's no better than writing the machine code by hand. See, that's the purpose of compilers - to take readable abstractions and to turn them into unreadable, but efficient, mess that is actually executed.