A bit off-topic, but I remember a college class in which I was completely lost. I wasn't even clear on the topic of the class. Nothing ever made sense. I got D's on most of the tests. I was enormously frustrated. Then, at the end, I got my final grade: B+.

My sense of relief was comical and fleeting. It was replaced by anger. How was it possible that most of the students in the class did worse than I? What an absolute waste to subject us all to such nonsense.

I remember speaking to an engineering statistics (IIRC) professor once, who said that no one had ever gotten an 'A' in his class. (Presumably, the final grades were curved like yours.) He said it proudly. I considered asking him if he was teaching basket-weaving or underwater archaeology to his statistics students.

I heard a story from an academic coordinator once, of an instructor who had been hired for a required CS computer architecture class because he was a friend of the department chairman. He was an electrical engineer, which made some sense, but then students started showing up in the coordinator's office crying and trying to drop the class well after the last drop date. It seemed he thought CS undergraduates were supposed to be the same as electrical engineering grad students, and wanted to fail the entire class. (He did not, nor did he get hired for further classes. After many years, though, his friend was the department chairman again and hired him as a tenured professor and the department's external relations coordinator. This is part of the reason I did not go into academia.)

Many instructors are just stinking bad. Many aren't, and manage to tie together both interesting lectures and more active assignments. But the bad ones do leave marks.

Are you just talking about grading on a curve? This happened to me in a graduate math class and I agree with you that it felt unnecessarily demoralizing. But it also seemed like a natural outcome of grading a hard class on a curve.

I had some upper division math classes that were offered for both undergrad and grad credits. I enjoyed the classes, but one thing I noticed was that the teacher seemed to be under some pressure to ensure the grad students passed. They didn't seem to care about the classes and performed horribly, I would do OK, and at the end of term all the undergrads like me would exit with an almost-guaranteed A.

Really helped to shape my perception that grades are meaningless and ultimately political.

In most graduate programs, grades are meaningless---what matters is that dissertation at the end.

It's not just that the grading was on a curve, it's that they learned nothing at all but still got a better grade than most of the class (implying that no one learned anything). Almost better to just have the whole class fail, then at least the department will notice that the professor is useless.

That wouldn't really be fair to the students who care about their GPAs, though.

> Almost better to just have the whole class fail, then at least the department will notice that the professor is useless.

You know it's the professor themselves that adjusts the grades? Of course the professor wouldn't fail the class if, as you suggest, it would make the professor look "useless".

I get the sense that history teachers talk about how history is the most important, physics teachers talk about how physics is the most important, gender politics the same, arts the same etc...

There's incentive to keep yourself employed however useless, bloated or out of time what you're teaching is.

I'm not sure how the school plan is evaluated in different places, but i feel like for example religion in a country like Sweden where most people don't believe in it[1] should be brought down to make space for something Swedes think is important.

[1]: https://en.wikipedia.org/wiki/Irreligion_in_Sweden

I had to learn driving a car this spring, with manual transmission. Knowing these things helped a lot. I made notes of all the traffic regulations document, and made tables of all long and intricate rules (e.g. lists of places where u-turn is forbidden, or backpedal, or what are speed limits for different kinds of vehicles/roads).

This helped avoid learning all those 1000 test questions, what most students did.

And using computer simulator with 3 pedals helped to automate the movements, and think of subtleties. Reportedly, those who had no practice before driving school, under the pressure of the exam, didn't stall the engine, but failed exactly in higher-order matters, like they'd cross continuous lines or not notice speed limit signs.

Do you live somewhere with an unusually hard driving test? I don't remember studying much at all for the written test (Canada), let alone memorizing a bank of 1000 questions. I never heard of anyone failing the written test.

Plenty of people failed the driving test, though.

I live in Russia. Our country still hasn't adapted vision zero policy in street planning, the ministry of interior love cars and high speed. So the technocratic govt is trying to fight with traffic accidents (though it did reduce by x2 in the last decade) by ever harsher surveliance and exams.

If you're curious whether road police is corrupted, as many believe, the industrial approach to fighting this actually works and gradually decreases corruption. In 2000-2008 you could literally pay for a licence and get it without any effort. In 2010 more rules were introduced, and the risk for the bureaucrats for doing this increased drammatically, so you had to go to driving school, but could bribe the exam inspector. Since then, inspectors would try extorting bribes by making you fail. The federal govt is fighting this as well: couple of years ago, the exams started to be videotaped. The video is kept for some months. Any case of incidents or if bribery is reported, a team of investigators starts analyzing videos, and they may take away the issued licenses as well as make criminal charges against the exam inspectors. (When I was driving, the instructor was talking on the phone discussing that our local road police was being under such inspection right in those days, and everyone there were scared.) Stakes for bribery increased dramatically.

I personally would prefer less industrial approach and more social capital, trust and honesty among people, but that's another story.

You have compulsory 50 hours of lectures (though it's up to each driving school whether to enforce it or not), 27 compulsory driving lessons 1.5 hours each. Theory test is standardized for the entire country, to be exact it's 800 same questions for the whole country issued every year. The exam is done on computers, over internet, so that local police can't help the examinees. The exam is also videotaped to catch those who cheat.

Our driving school made preliminary test of 40 questions, and I think the success rate was 80%.

At driving test, roughly half of people failed.

In the UK, the pass rate for the theory test was 54% for the last year. Lots of people fail it. You need to correctly answer 43 or more questions from 50 they ask, and there are over 700 possible questions. In addition to which, there's a simulated hazard perception test, which lots of people fail too.

Also, under 50% of learner drivers pass their first practical test, which involves a 40 minute drive on public roads with an independent driving section and manouvres like parallel parking and bay parking, plus a bunch of questions you have to get right. You'll fail for stuff like not checking your mirror often enough, not checking blind spots, going too fast, going too slow, incorrect positioning at junctions, and so on.

If you can, teach your kid to drive at 13. On private property obviously, no traffic rules, just the mechanical part. Everything after that will be much easier.

I also had a similar experience in college, but I'm having trouble understanding what the lesson you learned was. Did you eventually improve?

I think this is a problem solvable by doing practice problems.

For me with Math this is something I learned the hard way many times. In more advanced math tests you will be asked to prove something novel (to you) using skills/material from the covered subject. So it’s not enough to know the theorems, you need to understand them at a deeper level so you can apply them in new situations or use a similar practice from the theorem to prove something similar.

This is pretty straightforward for most math/physics/other stem courses since they’re usually accompanied by problem sets in their textbooks, and it’s rare to have to do absolutely all of them as HW.

For me, the act of programming has taught me epistemic humility much more forcefully and consistently than exams.

The only thing that taught me is imposter syndrom and how to wake up in sweat after dreaming of still being at university.