Practical applications

From my class notes (given by the professor, not the ones I take):

"There are several fundamental limitations of [this approach] including accuracy of [the figures used for the entire set of calculations] and the length and time scale limitations that need to be considered in the context of specific simulations."

So basically, your underlying assumptions aren't accurate, and the calculations you do with them don't take into account practical reality. Way to go, method. You are truly a winner, and I hope we use you a lot in this class. Oh wait! We will.

From the same professor in class the other day, addressing the concerns of a chemist-turned-engineer (not me) about an approximation he made: "You're an engineer now. One percent error, this is okay. Ten percent is maybe a problem."

This is a rather different mindset for me. One of my other professors takes more points off for sign errors than for method errors, if your method error still got you to an answer similar to the right one. He told a story about a student a long time ago who got the whole process right but switched one sign and got the opposite of what he should have gotten. He marked the student almost completely off on the test question, and when the student complained, "But I understand the process! I just made this one careless error!" The professor's response was, "But you couldn't have made a bigger error!"

It's just a whole new world, this engineering thing.