The principle doesn't apply everywhere, but I generally use it about memory usage, CPU usage and performance. A change that is smaller than 2 is usually inflated away easily by other changes, and improvements larger than 2 are usually possible by making the right inventions.
3 comments:
Or realistic.
While I generally agree, it does not always mean you need to be more ambitious.
Sometimes you just want/need to prepare the code for a future far more ambitious change. Legacy code...ahem
The principle doesn't apply everywhere, but I generally use it about memory usage, CPU usage and performance. A change that is smaller than 2 is usually inflated away easily by other changes, and improvements larger than 2 are usually possible by making the right inventions.
Post a Comment