in reply to Re^3: Make it good
in thread Make it good
Very well phrased, if you don't mind me saying so. I think we're in danger of violently agreeing. :)
Yes, new "methodologies" and "shifts in paradigm" arise, shine brightly, and get relegated to the growing pile of Things That Work Just Like Everything Else. What I'm hoping to point out through the parent node is that all of these methods are just approaches to making good software, but the all focus on how rather than what.
I do believe strongly in the iteration style of thought. Programming, like writing, is an iterative process. You brain-dump, then you refine, which might lead to refactors in some cases, which require futher refinement, and so on. Perl has taught me that it isn't how you do something that matters as much as what you accomplish. If one has two approaches to solving a problem and one is faster, cleaner, more secure, etc. -- and it still works -- then that is the better solution. TMTOWTDI, but not all are always the best way.
The whole point of TMTOWTDI, to me, is that each approach has its strengths and weaknesses, and which is "best" to use largely depends on what you're using it for. The "Make it Good" concept is that it doesn't matter what method you choose, so long as you are reaching toward the best balance of Function, Elegance, Security, Ease-of-use (for the target audience), Reusability, and Readability.
Now, I realize that good coding practice can't be relegated to those few short bullet points, but I strongly believe that to use these -- or something like them -- as a framework make a lot more sense than standardizing on the "chic of the week" methodology.
"Users are evil. All users are evil. Do not trust them. Perl specifically offers the -T switch because it knows users are evil." - japhy