I'm sure that you can get away with a less thorough "pre-op", but how many times have we all made changes to code and then realized our changes didn't work afte

I'm sure that you can get away with a less thorough "pre-op", but how many times have we all made changes to code and then realized our changes didn't work after-the-fact? Even in very complex systems I've worked with, I actually find that not spending enough time understanding the code tends to result in small errors rather than big ones: by the time I feel confident enough to write code, I can usually write something that's basically (but not entirely) correct. As I read it, the author is imploring us to practice putting in the time upfront to avoid these kinds of mistakes.