It's not the same thing, because when you split things you abstract, which is a form of saying you "generalize the code". Generalizing the code usually makes it easier to understand, because the parameters will not be directly tied to the specific algorithm at hand, and would be easily reusable. You can easily cut much of the code by doing this, it also makes it easier to modify things as you iterate on the code, a question of comprehension. Obviously, as you mentioned, there is such a thing as "overabstraction", but this existing don't mean abstractions are EVIL, this just mean the dose makes the poison. EVERYTHING, including a false notion of simplicity, if forced upon a situation, will be poisonous.