> College partially helped dictate the direction but why should I be paying 100,000 grands for just directions ? In general, I agree with your sentiment: the

> College partially helped dictate the direction but why should I be paying 100,000 grands for just directions ? In general, I agree with your sentiment: the best learning is done when you aren't being pressured by anyone other than yourself to do it, and at this point the sticker price for college is well out of wack of any value it could provide to students who aren't able to use the credential it provides to line up a high paying job or who aren't coming from very well off families. I do think, however, that there are things that it is hard to teach yourself, and that a college or college-like situation (i.e. where you have a few months to study a small set of subjects without regards to your immediate output and in the presence of other people who are studying those subjects) is probably going to be the best or only environment for providing. The first thing that I think it's very hard to teach yourself is how to approach problems in a field at a high level. For instance: I doubt I would have been able to get myself to my current proficiency (roughly 'beginning phd student') in mathematical proofs without a lot of work with a professor and graders. The material gets to a point where it becomes too hard to approach with your current mental toolbox, and I found that I usually needed some guidance if I was going to get myself to the next 'level' persay. The other thing that I found valuable and think is hard to find outside of a college was learning what problems fields other than your own find interesting, and how they approach those problems. I found that some of the most valuable experiences for me in college were taking upper level courses in subjects I didn't have a lot of experience in, particularly when those courses largely consisted of reading papers/books that were aimed at other scholars in that field, with a professor giving us context to the conversation that those papers and books were part of and a classwide discussion of the ideas presented. What I got out of those courses was an understanding of how ideas developed and what tools fields would draw on and apply to supporting or refuting those ideas. That was incredibly valuable to me because each course like that which I took gave me an entirely new toolbox for looking at the world and what's in it. (Sometimes I also wasn't super fond of that toolbox, and that was also enlightening.) This approach to learning wasn't confined to the soft sciences/humanities, either, although it seems to kick in sooner in those fields: graduate computer science and math seminars use an almost identical model. I think both of these can largely be categorized by "it's hard to teach yourself how to think." I think the first example is something where I at least needed the help of an instructor and grader to really grok it. It's things like that which I think you need some sort of college setting to really truly learn. The second could maybe be moved to online courses, and I know some OCW materials have that kind of feel to them (particularly for graduate courses) but I think it would require a different approach to the courses. Online courses often seem to focus on imparting the material, rather than about teaching about the process of how the material came to be. The courses involved a certain amount of discussion that was (sometimes) just as valuable as the actual reading and surrounding clarification that I don't think you'd get at all with most of the current iterations online courses.