Solve the problem before you code the solution. That may be obvious to this crowd, but I learned it the hard way! That's likely far from obvious to this crowd

Solve the problem before you code the solution. That may be obvious to this crowd, but I learned it the hard way! That's likely far from obvious to this crowd (if it has been paying attention). Paul Graham advocates the exact opposite: "exploratory programming", where you start writing the program without even knowing what the problem is, let alone the solution. I personally agree more with you than with Graham, though.