Echo(”you’re here”);

I want to make sure the function is triggering at the right time, then I code it, adding whatever functions I need. If I’m lucky the new functions don’t break anything. If I break something I refactor the class, search the entire project for calls to that class and flag them or fix them.

I suppose I avoid messy code by refactoring constantly. It’s very time consuming.

Maybe one of the problems is also that when we think the project is 90% finished, the other 90% still needs to be done. But most of the time we do the other 90% with quick hacks (so half of the code will smell).

Another problem is that a lot of times you already have a working code base (the first 90%), and you don’t want to refactor that because it already works, so you hack the extra features around that code without touching it too much (to keep it stable). And with every hack you make, you don’t want to break the previous hacks. But in the end I think, all your hacks will make the code more unstable than when you should have refactored it in the first place.

Maybe the best way is to first prototype it, like you say, and then decide: 1) throw it away or 2) refactor it and commit.

And to be honest: I never regretted making quality code, but lots of times I regretted building in quick hacks.