In most of these cases, waiting until we’ve reached the pain point where we’re forced to confront the issue makes solving the problem a much bigger ordeal than if we’d addressed them along the way. At the same time, premature optimization is seen as a bad thing.
How can we reconcile these two facts?
The solution is perhaps to weaponize Jevons paradox and impose real constraints to optimize against, constraints that are smaller than the sum of all available resources.

Let me tell you, if you deploy and dogfood your software on a raspberry pi, then it will be fast on a Threadripper too. If you can navigate a codebase in vim in an 80x25 terminal, then you can navigate it in a powerful modern IDE as well.

Resources and capabilities are tricky. Intuition often sells the idea that they allow us to do more, but often they allow the same, but at greater cost.