3 ms·
In the end, it's all about pragmatism and experience. Over the years I've been bitten for over abstracting and under abstracting. Over the years you develop a w
by mindcrime 16y ago
In the end, it's all about pragmatism and experience. Over the years I've been bitten for over abstracting and under abstracting. Over the years you develop a weather sense for how extensible a particular piece of code needs to be.
Bingo. It's easy to sit back and say "never guess about extensibility that might be needed in the future." But I posit that there are degrees of guesswork; that is to say, there's a difference between a wild assed guess with no basis at all, and a guess based on years of experience solving similar problems in a related domain. Experience counts for something, and part of that is noticing familiar patterns (not "design patterns" necessarily, just patterns in the general) that tend to crop up in certain situations. I think making some speculative decisions based on the latter form of guesswork can be a net positive more often than not.
Of course I don't have scientific evidence to back that up, but I believe it very strongly based on my own experience.