High Academic Results Make Better Programmers
skorks.com
skorks.com
I've got a pretty decent GPA and a record of academic excellence, and I make this argument.
Most of the current educational system is based on learning to repeat the teachers password (http://lesswrong.com/lw/iq/guessing_the_teachers_password/). I teach students with good grades at a top university, and they routinely resist doing anything other than repeating what I tell them.
This isn't because they are morons, it's just what they've been trained to do. Indeed, in the past, they've been punished for doing things differently.
Guessing the teaches password does establish a minimum level of intelligence; a person who is sufficiently dumb will fail at this task. But once that minimal level is established, it's hard to really care. I'll take a kid with grade inversion [1] and a cool side project over a straight A student any day.
[1] Grade inversion is the phenomenon of a student who's grades are directly proportional to the difficulty level of the class (inverting the normal relationship). Calculus, literature, CS 101: C. Topology, Operating Systems, Quantum Mechanics: A.
But why isn't it preferable to have the person who goes with: Calculus, literature, CS 101: A. Topology, Operating Systems, Quantum Mechanics: A. ?
Perhaps the C/A person was spending their time better elsewhere when they could get away with it. But even so, this doesn't really excuse them because either the C/A student:
(1) "gets it" -- in which case doing well in those easy classes would be trivial. Given that all the students know what the game is (get high grades), why would an employer trust them to take a job seriously when they refuse to perform any trivial tasks despite the high return on low investment?
(2) "didn't bother learning the easy stuff" -- in which case they probably don't actually have any deep understanding of these more advanced topics. This would suggest to me that they simply started figuring out the teachers password for those courses.
I don't see any way to interpret the C/A student as being better than the A/A student without knowing more about either the student (certainly some interaction can reveal a drone) or the classes (there are tough classes out there, which actually require understanding for high performance).
Edit: Self-awareness check -- I made the transition from "C/A" in high school to "A/A" in university. This was based on an attitude adjustment, and the realization that I'm playing a partly silly, but also partly valuable game and I might as well try to win both. I suppose this opens up a third possibility in which the C/A transition comes from a change in the student. But that should still at best put me on par with an A/A student, given no more information.
You assume that "gets it" translates to good grade. If large portions of your grade are based on busywork/class attendance/memorizing trivia, that's a bad assumption.
Given that all the students know what the game is (get high grades)...
What if the student was playing the game of learning? I prefer that student to a grade grubber.
C/A indicates a smart person who doesn't feel the need to play games. I can use a person like that.
A/A may be a person who plays games when useful, or a person who thinks producing GPA is their goal in life. The latter sort of person is useless outside of some well defined roles. Or an A/A may be a C/A person who wasn't required to attend class, or a C/C person who took the right classes.
Overall, I don't see any reason to prefer the A/A person over the C/A person. As I said, a cool side project will easily tip the balance.