>Well, when I look at the code and find something that may be written better, I usually change it while I am looking at the code, not at some late point. That's just how I usually approach code, even at the risk of breaking working stuff.

What is "better"?

Is it better to write new code that will force you or someone else to spend hours/days/weeks/months so you can say you saved x amount of time worth of CPU in y number of years? I'll bet whoever pays for it will beg to differ :)

I can think of many more comparisons for "better" but this one is probably the most controversial :)

Is it worth to make a change in a program that will have an arguable small impact (and this can be positive or negative) in any metric you choose, but will greatly annoy your co-workers?