I’m sympathetic to this concern (it’s why I don’t like the QM sequence and think thinking about many-worlds is mostly a waste of time), but I also think math has the potential to be a useful toy environment in which to practice good epistemic habits (as suggested by shev’s recent litmustest posts), especially around confusing paradoxes and the like. Many of the complications of reasoning about the real world, like disagreement about complicated empirical facts, are gone, but a few, like the difficulty of telling whether you’ve made an unjustified assumption, remain.
I’m sympathetic to this concern (it’s why I don’t like the QM sequence and think thinking about many-worlds is mostly a waste of time), but I also think math has the potential to be a useful toy environment in which to practice good epistemic habits (as suggested by shev’s recent litmus test posts), especially around confusing paradoxes and the like. Many of the complications of reasoning about the real world, like disagreement about complicated empirical facts, are gone, but a few, like the difficulty of telling whether you’ve made an unjustified assumption, remain.