Hacker News Reader

Comment by joeberon | Hacker News Reader

What I don't understand is that if you look at the performance of Chinese academics it isn't particularly exceptional compared to those who had lazy upbringings in the west. Surely they should be basically superhuman in their academic achievements? I am extremely lazy, and had a very crappy lazy school experience, even at British university, and yet I was not significantly worse than my Chinese clearly-a-genius PhD supervisor. In fact I do not see any particular difference in research quality or output when comparing people who had intense school careers to those who had easy and laid back ones. Somehow the intense academic training in those countries does not seem to translate into actual research output.

View on HN

For all it's apparent advantages, the Chinese education system also has some pretty deep flaws. A family friend in China is the head of the Computer Science department at a local university. This was a few years ago, but he told me none of their graduates ever, as part of their studies, actually compile and run any code. They are taught java, algorithms and theory but no practical exercises, it's all on paper. That's not necessarily true of every institution, I've met some great coders from China, but they're mostly self taught.

Also even if you had the best students, they are constrained by the institutions they are in and the experience of their professors. It takes generations to develop deep institutional experience and culture. They know this, and expend a lot of effort trying to get experienced Chinese academics that have studied abroad in western institutions to come back to China.

We see this among students who come to American universities from certain parts of Asia: serious knowledge of syntax, but not able to navigate the OS, use the IDE, or compile the code. I've heard this same scenario of learning programming on paper from multiple countries.

And, also, if you know the syntax of a few languages cold, but you have no creative practice at problem-solving, are lacking foundational computer science concepts and practice, then by the time you are in college or grad school, another language may have come along, while you may have to struggle to apply existing knowledge and creativity to new tools (which is the point of a CS degree).

Some students have no mental toolkit to rely on other than brute force memorizing the syntax of a new language. The difference is highly noticeable, even at a low level of college work.

Compulsory education is all about the median student, not the outliers. I suspect that performance at the top end of the curve, i.e. people who tend to get PhDs and become professors, is influenced much more by personality and graduate-level educational experiences.

The point of all the studying is to ace the entrance exams, not learn anything about research or original thinking. At least in Japan, once you're actually in the university, you can pretty much slack off for the next four year until you graduate and face the next grueling phase of your life, becoming a junior salaryman.

One popular explanation is usually the scarcity of educational resources in China. In China, you work really hard so that you can out-exam the other students so that you can get into some decent college. The gist is that China has limited higher-educational resource (so graduates on average do not come out that well) and a large population base (so people really need to work hard to fight for the little resource).

It’s more statistically significant- check how Chinese names dominated CV conferences

Different fields always have different nationalities dominating though. You could easily say "look how Italians dominate astrophysics" but that doesn't necessarily mean anything.

AI/ML is wayyy bigger

AI/ML research is not "way bigger" than astrophysics lol

Reply on news.ycombinator.com