Someone needs to insure quality in education, and IMHO that job should be more in the hands of professional societies and journal peer reviewers than in college admission boards for one reason. It doesn’t matter how smart people are, Harvard can only admit 30,000 students. It doesn’t matter how smart people are, the stupid weed out classes in UT Austin science and engineering insures that only a small fraction survive. (And the reason that I want to use MIT as the model is that MIT doesn’t have physics weed out classes. It is tough, but its like the Marines, most people that make the program, make it through. The nasty stuff I saw at MIT undergraduate education is minor compared to the really nasty stuff I’ve seen at the UT Physics department.)
In each case the limits are artificial, and as people become smarter and better trained, the result of these artificial limits are to insure that what determines if someone survives *ISN’T* quality but rather luck, the ability to game the system, and personality traits such as obsessiveness. And it is just going to get worse. People are getting smarter, and the social systems designed to weed out people just can’t cope with this influx of smart people.
Professional societies and journals don’t have these hard artificial limits.