Most never accepted the validity of the concept.
"Just get an adequate computer. That's management's problem."
OK, another story. A friend of mine was a Physics student at
Rolla (University of MO sub-campus) and also very good at
programming/analysis. He had a roommate who was an ME student,
and just learning programming. He saw his professor fiddling
with one of those 15-tile puzzles, and said "Hey, I could
write a program to solve those". His prof said it would be a
good exercise, and gave him an account on the 370/168
at the main campus. He wrote it, and tried it, but it kept
running out of time. He assumed an infinite loop, and eventually
mentioned it to the Physics student. My friend sensed a
teachable moment and said, Hmm, you have 16 nested do loops
from 1 to 16. Do you know how many total iterations
The guy whips out his trusty HP45 and starts calculating.
My friend smiles, as he already knows that 16^^16 will not
be computable on the calculator. So, they trudge over to
the comp. Sci. building and he shows the guy the System/360
storage and timing estimates book, and they work out that
JUST the loops, themselves, will take something like
10 billion years to complete. He eventually chopped out
useless iterations and got a program that ran in a minute or
So, the moral is that there are relatively modest problems
that can become TOTALLY un-computable with a poor
algorithm. No computer ever made could have finished
his poorly thought out program.