I think you have a fundamentally wrong conception of what it takes to learn advanced concepts. In most cases you can't autograde or award points for everything, it also isn't what University education should be about. It is sad enough that during the Bologna reforms in Europe University education was "industrialised" as much as possible (ECTS points, core modules, etc.).
In the best case a University education in a STEM field will lead you pretty quickly (within 3-5 years) to do research in a lab that does cutting edge work. Most of that lab work can't be distributed or done online, nor can any of the preparatory lab courses in small groups with constant supervision (which are a large part of Physics, Chemistry, Biology, Medicine degrees). Even in subjects such as math and computer science a huge part is exercises that are graded and then discussed in small groups of 10-20 people. This wouldn't easily translate to an online experience either.
Universities like Cambridge even offer 1:1 supervision in Math and Physics, this can be totally awesome and is non-replaceable by video chat or online forums.
My biggest challenges was to find material which explained to me certain concepts. Stuff like 'why can you transform this mathematical notion to this short form? How do you know that? Oh i learned that in the gym'
YouTube helped, other students helped but it was time consuming.
It would have helped me much more if i would have had a proper central source of high quality material. That would have allowed me to understand it easier and better and with less effort for others around me.
THIS then should free up those people to do more 1:1 supervisions (which might be common in Cambridge but is not everywhere the same). It should also allow those people to optimize how material is explained.
And this allow more people to get education which is key.
And i don't think this should just start at university level. Why not start with when you are baby for your parents?
Well these centralised resources are books. There are high quality (almost standard) books for most subjects. Want to learn Analysis at a Graduate level? Pick one of the popular books Rudin (for example) and work through it. Same goes for Quantum Field Theory (Peskin & Schroeder, Zee, Srednicki,...) and tons of other subjects.
The lectures are often only a guided tour through one of those books (at least at the undergraduate and early graduate level). In many ways lectures are still la huge advantage of course because most of these books are too long to be useful (I still have some guilt when I think about how little I absorbed of the 500+ pages experimental condensed matter book we used). Often the exercises then cover things that were not explicitly explained in the lecture but can be worked out with the help of the book.
In the best case a University education in a STEM field will lead you pretty quickly (within 3-5 years) to do research in a lab that does cutting edge work. Most of that lab work can't be distributed or done online, nor can any of the preparatory lab courses in small groups with constant supervision (which are a large part of Physics, Chemistry, Biology, Medicine degrees). Even in subjects such as math and computer science a huge part is exercises that are graded and then discussed in small groups of 10-20 people. This wouldn't easily translate to an online experience either.
Universities like Cambridge even offer 1:1 supervision in Math and Physics, this can be totally awesome and is non-replaceable by video chat or online forums.