Yes, prediction, even with lots of data points and the ability to extrapolate, is very difficult. Ex post facto analysis is facile. If we consider an alternativ

Yes, prediction, even with lots of data points and the ability to extrapolate, is very difficult. Ex post facto analysis is facile. If we consider an alternative definition of strong vs weak. Where technologies that stand the test of time are considered strong. And those that fall into obsolescence are considered weak. Then every technology eventually degrades into obsolescence. Unless based upon first principles: such as fire, the wheel, or books. As another example, we see this play out in real time with SaaS. I certainly believe web delivery to be a strong tech. And most users prefer its advantages over the alternative. Which is usually some frankenstein hybrid of legacy app servers delivered via virtual terminals on remote windows desktops. Yet the economic activation gap is sufficient, despite rapid development frameworks such as React / Flutter, that investment in a new interface and training and data migration is deemed unnecessary. User pain is subservient to short term cost savings. And so the weak avoid obsolescence much longer than their organic shelf life would warrant.