IMHO the big problem in experimental sciences is what the article kindly describes as power imbalances. I'd call it perhaps neofeudalism. I've seen many cases where principal investigators / professors are completely disconnected from research, including an absolute lack of any basic knowledge related to the field. They are just middlemen. For someone coming from pure mathematics, where professors routinely come up w…
Right now, individual fellowships to grad students and postdocs are scored, in part, on the “training environment” but weirdly, research grants to a professor (to hire grad students and postdocs) are not. This would be an easy thing to flip. Faculty applicants could submit a letter from a trainee as part of their package too.