- Does it work?
Your programs should compile without warnings or errors. They should give
the correct answer, both on the example inputs given in the assignment and
possibly on arbitrary new examples we may create during testing. You can
get at most 30% of the maximum score if the program does not
run.
- Is it found correct also after reading the code? (The test set may
not have revealed all bugs.)
- The quality of the solution: e.g., is it unduly complex?
- Comments and structuring.
Here are some details of the last point.