3 ms·
I agree with all of that. My first reaction is that you’re probably in startup land where you’re mainly building new things. People do bat an eye at estimates t
by dahart 8y ago
I agree with all of that. My first reaction is that you’re probably in startup land where you’re mainly building new things. People do bat an eye at estimates to update old and large existing systems. It’s the new and small ones that are easy to update. If you’re even talking about MVPs, you’re probably not seeing much over-engineering yet. Or, you might have some and not know it yet, it might take a couple of years before you realize you paid large opportunity costs solving issues that seemed important. Having to decide whether to solve an issue before vs after MVP can be hard, but if you know you need to solve it either way, you’re way ahead of solving something that you aren’t absolutely sure you’ll need.
A common but not-common-enough (IMO) programming philosophy is to only write the minimum code necessary to solve problems you actually tangibly have, not for any problems you think you’ll have, only for problems that absolutely require solving now and people are screaming for them. Extreme programming and Agile and lots of other software development frameworks all have variations on this tenet.
The biggest over-engineering mistakes I’ve seen are when teams decide to rewrite from scratch a code base that they deem too crufty and too difficult to maintain and update anymore. Devs make this call far too easily, and fail to take into account how many things still work and how much effort it took to get there. I’ve been witness to two very large such rewrites that, many years later, both companies admitted were colossal mistakes.