Even a lot of CS research journal papers feel more like role play — the same way startups try to pretend to be real companies with executive headshots, flashy offices, and all the other nonsense. (Instead of analytically modeling something to prove an idea, they’ll build a simple simulation and focus on its “Architecture”)
Engineering departments effectively weed out such in the first ground of engineering courses. Looks to me CS has no equivalent.