3 ms·

"There is nothing to be gained." Not completely true, you could argue that it would be more maintainable in the future but I don't think that outweigh the risk

• joosters 11y ago
  Yes, maintainability / extensibility is good. But I'd be wary of refactoring code purely for the sake of potential maintainability / extensibility, as it's work for no gain. You're trying to predict the future for the code, and that's a losing game. If someone has a reason to extend the code, then you have the choice of hacking in the new feature, or refactoring the code to make it nicer. Each case really needs to be considered on its merits. But IMO it's almost always better to prefer to keep the code as-is rather than making big rewrites. Tiny refactoring is the way to go.
  • nkassis 11y ago
    I'm OK with that stance, I try to advocate the same myself but you have to be careful of the boiling frog issue of making small incremental refactoring that lead to dead ends. Stepping back and thinking about the long term is valuable. Doesn't mean you have to do the work now for that potential future but thinking of the potential drawbacks of the current design and it's limitation may guide the decision as to when it's appropriate to do a major refactoring. Also this particular example seems like something that will probably rarely need modifications. I'd be interested in seeing how often this code changes. My guess is the effort to refactor this will probably be equivalent cost to the future changes.