Universities just need to do a better job of separating teaching and research in my opinion. Someone with a PhD in algebraic graph theory isn't going to want to

Universities just need to do a better job of separating teaching and research in my opinion. Someone with a PhD in algebraic graph theory isn't going to want to teach 20 year olds what a group is, or grade their tests, and they shouldn't have to. For every undergraduate class, there are probably more than enough 25 year olds with degrees in that subject capable of teaching that class better than that professor might. But, whoever teaches the class should first and foremost be a good teacher who actually wants to teach, and not whichever professor/grad student could be drafted into duty. The main problem with universities these days is that they are given the duty of both teaching and research, but they force people to teach that don't want to (some professors, grad students). In addition many people view universities more as job training than learning about a specific subject. If you go to college to get a degree in CS, but all you want to do is get a job at Google, you don't need CS experts to teach you intro to data structures or how to write SQL. I think the long term solution is to entirely separate the realms of research and teaching, or to make teaching as a research professor entirely "opt-in" outside of discipline specific classes (the algebraic graph theory professor should probably be the one teaching algebraic graph theory). My university has been relying more and more on teaching-only professors to teach CS classes and I think it's a wonderful solution. The teaching-only professors are much more personable and engaged in the actual teaching, and the research professors get to focus on what they're good at.