Most of CS also revolves around making things easier (correctly, of course) for yourself.
If there's a method that will get things done with less work, you are usually encouraged to take it, provided it gives the right output - nobody cares too much unless its completely slow or bloated.
That was the biggest reason for me over going into engineering - CS actually rewards you for thinking outside the box, and it thinking like that really helps most of the time.
... where on earth do you work? Dont get me wrong, I love CS; however, I dislike when people romanticize it. Read thedailywtf.com over and keep in mind that there is no shortage of stories ever to get posted on that site. and as for:
- "What invariants must this process satisfy? What's the cleanest way to organize this structure? How should these subsystems work together?"
I know a lot of people who graduated with me that I would not trust to ever answer any of those questions.. I have seen their programs and do not believe they know a good way to cleanly organize themselves out of a cardboard box.