We talk a lot about bad science, so I would guess that the opposite of that is what the author may be talking about. Which is probably a good idea, since bad sc

We talk a lot about bad science, so I would guess that the opposite of that is what the author may be talking about. Which is probably a good idea, since bad science sucks and it's getting more air time than it used to. "Better" could probably even be quantified in a helpful way if desired. Politics are just another human factor. Stimulus and response. Can also be quantified if addressed situationally. As a catch-all it's a low-leverage descriptor though. I like that the author is at least talking about this. It could lead to a system-level improvement. > models that reflect reality Or models that elegantly allow for multiple perspectives while contributing to a desired outcome. Reality when held up as an accountability target opens up a bit of a can of worms, starting with the subjectivity question: Whose reality? Or which reality? Humans are remarkable in their ability to create, entertain, use, and appreciate even conflicting models of reality.