Hm. Comparing paradigm shifts with model based realism is not really valid criticism of either, yet understandable. The folks who just added one more term to t

Hm. Comparing paradigm shifts with model based realism is not really valid criticism of either, yet understandable. The folks who just added one more term to the old stuff to make it predict better - let's call them the old guard necessarily did it to point out that old models can become slightly new ones too. Even if in the long run they will be seen as the evil holdouts. Yet at the same time we know that just whipping up a new fancy maths model won't solve anything in itself. New models need to make new testable predictions. And then even new models require lengthy fine tuning, which requires costly experiments. Alas textbooks are very often terrible, but not because they emphasize predictions and models over "understanding" - but usually because they omit to elaborate on how to select the better of two models, how paradigm shifts happen, how anomalies are ever present - and thus make practical model selection even harder. Plus they regularly fuck up the math explanation part, exactly because they use terrible language and models. Finally, it's always data that cleans up the mess. Either practical usefulness - engineering, applied science. Quantum experiments, q-bits, and so on. And on the high-energy end cosmology and astronomy. Anyone harking about how the crisis is about politics usually wants to allocate more money to theorists, so we will finally get breakthrough theories. Yeah, great, we already have a lot of those, but without data we don't know which one to take seriously. Furthermore pouring money into theory is a nice idea, and comparatively cheap (compared to a new collider), but that won't solve the very pragmatic employment question for the collider builders. (Who are out of luck anyway, because the era of building bigger underground circles seems to be over. But they would gladly build anything, but they won't make good theorists, even if pundits' articles seem to imply there's a simple slider between theory and experimentation.)