R. Clayton (rclayton@monmouth.edu)
(no date)
Is it better to be late than wrong?
If your assignment's a lot wrong after you've been working on it for two weeks,
chances are good it'll still be a lot wrong when you turn it in, which means
not only will you get a low score, but you'll also get whacked for the late
points. This argues for tuning in your assignment on time and taking your
chances with just a low score, rather than going with a guaranteed loss of 5 to
25 points on top of whatever low score you'll probably get.
On the other hand, it depends on what's wrong. Assignments that don't even
compile get a grade of no more than 50 (out of 100). Assignments that compile
but don't work usually get grades in the range 65-75 (out of 100). In this
case it makes sense to take the two or three extra hours it would take to fix
your assignment so it compiles; you lose 5 points for being late (assuming you
can fix your assignment before midnight) but gain around 15 points for
submitting something that compiles for a net gain of 10 points.
The decision is trickier if your assignment works, but you just want to "fix a
few, simple things." I have committed some of my worst programming atrocities
when I tried to make a few, simple fixes to my code at the last minute. I have
learned never to make changes at the last minute, no matter how simple. My
recommendation is that you submit you assignment as it is. And if you cannot
resist making changes at the last minute, make a copy and change the copy.
(Of course, I have absolutely no expectation that you'll pay any attention to
my recommendations. Plenty of smart people who's opinions I respected warned
me about everything I'm telling you here, and I ignored them all.)
This archive was generated by hypermail 2.0b3 on Mon Dec 15 2003 - 19:45:06 EST