[greenyoda]

"1. Solve the problem before you write the code."

Sometimes the problem is difficult or huge (or both) and you don't know how to solve it. By starting to write some exploratory code, like a solution to a small part of the problem that you do understand, you can frequently get ideas about how to solve the bigger problem. Even if you end up ultimately throwing this code away, it could still be a useful learning experience and help you make progress toward understanding the real solution.

[dllthomas]

"Write it; throw it away; rewrite it." Not always the best solution, but a good voice to have in the mix.