[bob1029]

You can get pretty damn close if you establish some reasonable constraints.

For me, a perfect system is one that is stable over long periods of time. Achieving these outcomes first requires careful selection of tools, frameworks and platforms. Building anything on unstable foundations is never going to end well over time.

We have large chunks of code that haven't been modified since ~2014. How many shiny frameworks and languages have come and gone between then and now? You can be certain I have basically forgotten about this code. Not out of disdain or neglect, but because it just works 100% of the time now.

We have a lot of problems to deal with in front of us, so we find the courage to properly settle things and move on.

[tluyben2]

Yep, that's my idea as well. We have code that has been updated for security but nothing else from the early 00's. It makes building very large systems with tiny teams much easier if you don't go after the latest & greatest.