IMO the main reason this overcomplication tends to happen is that many of the Monads for Idiots guides are written by people just out of the period where they struggled with the concept, and they mistakenly think that the entire thought process that led them to get it is the best way to present the topic.
But teaching doesn't work that way, and more often than not presenting the 'long way' through a topic obscures the essential simplicity that actually helps people figure it out for themselves.
That burrito one is a good example; maybe interesting once you already know what's going on, but if you don't it's most likely just going to frighten you away.
That is the problem with the book Real World Haskell. The authors will spend most of the chapter showing you the wrong way to do it, then a paragraph at the end saying BTW this is the right way.
Kinda sorta the point of having any sort of pedagogical aid, like a book, is that it gives you at least a prod in the right direction. Learning something just to need to unlearn it a few pages later wastes everyone's time.
But teaching doesn't work that way, and more often than not presenting the 'long way' through a topic obscures the essential simplicity that actually helps people figure it out for themselves.
That burrito one is a good example; maybe interesting once you already know what's going on, but if you don't it's most likely just going to frighten you away.