> Since it wasn't done, it's likely to not actually be worth it. That doesn't follow. If one company needs a widget, it might make sense for them to build it t

> Since it wasn't done, it's likely to not actually be worth it. That doesn't follow. If one company needs a widget, it might make sense for them to build it themself. If 100 companies need a widget, forming a widget company that supplies those hundred companies is probably more efficient. So if we collectively decide "let's scale up nuclear power" then it makes sense that the components that go into the plants might be modularized and built on a (cost-saving) larger scale.