This was apparently pretty typical for English law hundreds of years ago. As I understand it, wealthy people had a lot of influence over the government and were able to lobby for extreme punishments to protect their fortunes. Many of the death sentences were not even carried out by the executioners, because they were so common.
I believe that Thiel is saying that the current structure of higher education, in tiers of increasing eliteness, bundled with a living and partying arrangement over an essentially-fixed time period, is bad, not that the course material or content is bad. I'm a CS student at a very prestigious university in the UK, and I'm not going to claim that I would have learned all of the same things that I have learned if I had self-taught, but I totally didn't need the whole living arrangement thing, and I'm not really sure why I had to fight with thousands of other students for my place, when most of the teaching easily parallelises if you just record the lectures and open up the course notes.
I'm not sure having better predictors of what businesses are good is necessarily a good idea. Part of the attraction of silicon valley is that it takes some of the risk out of trying new things, even if they might be bad ideas. This culture of trying things leads us to find the occasional really good idea. If we sit around all day plugging our ideas into models to see if, statistically speaking, the will succeed, we won't find the really novel ideas that look bad but are actually good.
Mathematica has a few of these features. If you paste in an image from a previous operation, then the image is shown inline in the piece of code you're typing.
The median-of-medians approach is also somewhat tricky to implement (avoiding array copies and so on), although it admittedly looks neat on the slides of the algorithms lectures.
This talks about our ability to reach or describe the worst case, so if we say it's Ɵ(n^2), we're saying that we can tell that the worst case is at least quadratic, and it's also no worse than quadratic. We might not be so good at the math and have only discovered that the worst case was at least linear and no worse than exponential (which is true of quicksort).