concept rule in category haskell

This is an excerpt from Manning's book Get Programming with Haskell.
This code clearly isn’t Haskell, because both tick and reset violate the rules we established. Neither function takes any arguments nor returns any value. The question is, then, what are these functions doing, and how is this different from functions in Haskell? It’s not a long shot to suppose that tick is incrementing a counter and that reset restores that counter to its starting value. Even if we’re not exactly right, this reasoning gives us insight into our question. If you aren’t passing an argument to a function, you must be accessing a value in your environment, and if you aren’t returning a value, you must also be changing a value in your environment. When you change a value in your programming environment, you’re changing the program’s state. Changing state creates side effects in your code, and these side effects can make code hard to reason about and therefore unsafe.
Learn the rules for writing recursive functions
For each goal you establish in rule 1, you need to figure out what the result will be. In the case of washing dishes, the result is that you’re finished washing the dishes. With functions, you need to return a value, so you have to determine what value should be returned at the end state. A typical problem programmers face is trying to think of the goal state in terms of being the end of a long recursive process. This is usually unnecessary and overly complicated. Often the answer is obvious when you ask the question, “What happens if I call my function on the goal state value?” For example, the end state of the Fibonacci sequence is to arrive at 1; by definition, fib 1 = 1. A more mundane example is determining the number of books you have by counting the number on each shelf. The goal state is to have no more shelves to count; the number of books on no shelves is 0.