Comment by sacnoradhq

It sounds like you're arguing from a hypothetically perfect view with 20/20 hindsight. Scaling problems are good problems, solvable at the right time without the evils of premature optimization. Doing work without a need is wasted effort.