Popperian-style hypothesis testing is an important part of science, but it’s not all of science. Observational science (e.g. most pre-20th C botany) and probabilistic sciences (e.g. meteorology, medicine) have been a part of science for just as long as physics.
Despite the jokes about “stamp-collecting”, observational science is just as vital and “scientific” as experimental science. And it’s the probabilistic sciences that make the most obvious contributions to human wellbeing.
This is related to my standard whinge about early Philosophy of Science; Popper was a physicist by background, and he based his theories on the idea that all science is or should be done in the same way as physics. It isn’t, and it shouldn’t be.
It’s ultimately a fight over the demarcation problem, which is still an unsolved issue. So far, all attempts to come up with a rigorous definition of “science” and the “scientific method” have failed. It’s easy to invent definitions, but it’s also easy to find exceptions to those definitions in both directions (i.e. non-scientific things inside the definition, and clearly scientific things outside).
Some (including myself) argue that the reason for this is that there is no such thing as “the” scientific method. Science isn’t a single thing; it’s a collection of things, thrown together haphazardly over the course of history.
The only thing that the sciences really have in common is that they have proven themselves to be relatively reliable means of producing increasingly accurate approximations of true answers to interesting questions. Any method that can do that eventually gets incorporated into science; fields of inquiry that fail or do not attempt the cumulative-accuracy trick stay outside the bounds of science.
OTOH…
