It's a bit odd that people often call these side-effects. They're just "effects". Side effects are things you don't want to happen, i.e. accidental/incidental effects... which is exactly what a pure language like Haskell aims to prevent.
That's not what "side effect" means in this context.
Side effects are the opposite of pure functions. A function with side effects will mutate global state, a function without side effects will not (it will simply return a value).
This is true and correct, but I would add that sometimes in FP books and papers you do see authors merely writing "effects" as well. I think either would be acceptable.
"Effects" in an FP context generally refer to effects explicitly modeled as values in the program, whereas "side effects" are implicit. `Either` is an effect, throwing an exception is a side effect.
