He describes the basis of a messy function as he gradually improved, and are thereby the first intermediate step and the final result award. He created in the intermediate step, an absolutely understandable, clear function with 11 lines of code and descriptive function calls. Thus, the author is not satisfied, the function called absurdly as still "too long" and sets again on hand, he grabs by almost every line in a separate function. The resulting code has almost twice as many lines as the original code it was to improve! The author hopes that the Code should be read as a story, but dismembered him in such a way that one thinks one would now read a book in which each side is only one set with a maximum of 10 words. I can his enthusiasm for such a fragmented code absolutely not understand, especially since the name of the functions is always abstruse because now no longer require only 5-10 functions but rather 25 for a meaningful description. Not to mention that you now have to read along each functional transition from the entry point, by searching the called function, there finds the next function call, to no longer know very quickly on what "level" of the call stacks are located now. In the head is something only difficult to understand, and debugger friendly which is no longer well.
Summary: I liked a lot better with 11 lines of intermediate step, he was definitely at a glance understandable and comprehensible. Then I wanted to have pointed out, because here the author is dogmatic and shoots himself in the foot, in which he leads his absurd arguments, yet very clear what it insists for beautiful code he has conjured it. This dogmatism is the point I wanted to, and because I also depart a star.