[smitty1e]

Scope context matters in hitting the Goldilocks spot in the engineering.

But given the choice of over- vs. under-engineering, overshooting a modest amount absorbs the inevitable scope creep more readily.

[rqtwteye]

Totally agree. Looking ahead a little during design can save a lot of time later. You will not guess always right but usually you can make a reasonable guess where things will be going.

I have had this fight with Agile absolutists. They claim you should only think about the next sprint and not design any further.