Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Geniuses of the past were aristocratically tutored

erikhoel.substack.com

Post reply on HN
Print view View original

341–350 of 505 posts

Re: Geniuses of the past were aristocratically tutored

#341

Post by ZephyrBlu »

cushychicken wrote: post #17
hans1729 wrote: post #5

This is ridiculous. We’re seeing innovation EVERYWHERE, and looking back, this will be a golden age for progress (if the species makes it far enough to look back, that is). Think of it like this: we invented personal computers. At the time, only nerds had access to computation. These days, all kinds of scientists just script up things in python. We have how many million people who are able to code? I don’t even care…

Yeah, I knew this article wasn't for me when I read the first sentence. Genius is so widespread these days that it's almost pedestrian. It's just way less concentrated and elite than it used to be - which makes its findings harder to disseminate.

What's your definition/bar for genius?

Re: Geniuses of the past were aristocratically tutored

#342

Post by cyberbanjo »

ajsnigrutin wrote: post #331
DantesKite wrote: post #322

I read the article. I disagree with the premise that there are fewer geniuses today. There are plenty of geniuses, if not more than ever. They just don't all have a deep, cultural impact on the zeitgeist. Take for example the resolution of Fermat's Last Theorem and Poincare's Conjecture. Or the work being done on the Twin Primes problem. This article is probably better situated as an examination of culture. You could…

The other thing is, that there is not much of basic math and physics left anymore for people to study and figure out something new. Archimedes' buoyancy, newtons laws, etc. are all something that we know and learn about in elementary and high schools, and we remember their names, because they were either first to figure it out or more realistically, first to put it in writing in a way, that we can now use in "math/ph…

Heres a counter example to both "math is finished" and "contributing to research needs heavy equipment", granted not sure how common that is.

Christopher Havens, a convicted murderer serving a twenty-five-year sentence, made a world-class discovery in mathematics; from his prison cell.

https://dr-younes-henni.medium.com/from-murderer-to-mathemat...

Re: Geniuses of the past were aristocratically tutored

#343

Post by hwillis »

Gollapalli wrote: post #296
hwillis wrote: post #173

Earlier quoted context omitted.

> We don't have Einsteins because the brightest people don't work on astrophysics because there's no money in it. Einstein worked in a patent office from 1902-1909. In 1905 he published four papers on the photoelectric effect (laying the way for quantum physics), Brownian motion (proving the existence of atoms), special relativity, and the equivalence of mass and energy (leading to atomic energy). From the beginning…

Maybe, but back then you just had to be an interested party of some learning whose work was correct. Now you have to be a “professional” or endorsed by one. It’s a status competition with real resources on the table rather than an aristocratic hobby for the few who were interested.

> Maybe, but back then you just had to be an interested party of some learning whose work was correct.

1905 was very different from 1705. It was not particularly different from today, aside from cars, planes and computers. Hell, it was a golden age of physics research and incredibly well respected. The men and woman of the fifth Solvay conference [1], Emmy Noether and all the incredible mathematicians of the time- most of them did not come from means and could hardly be described as aristocrats.

> Now you have to be a “professional” or endorsed by one.

No, you just pay the publishing fee.

[1]: https://en.wikipedia.org/wiki/File:Solvay_conference_1927.jp...

Re: Geniuses of the past were aristocratically tutored

#344

Post by syntheweave »

marcosdumay wrote: post #254
dboreham wrote: post #241

Earlier quoted context omitted.

Showing a use case for matrices is extremely easy : 3D graphics. At least 50% of school students will have a good understanding of that field.

Yes, matrices got everywhere in the 20th century. There is basically no field that doesn't use them nowadays. But any demonstration requires modern knowledge, and matrices are one of the very few modern thing students see. If you want to show them 3D graphics, you will need to teach programing first. Yeah, some will know it by them, but schools also can't rely on that. (As an aside, chemistry also has some weirdly mo…

You don't have to teach programming to teach computer graphics at the level of pedagogical example. The programming is bookkeeping for assets and occasionally a clever optimization(e.g. a Bresenham line rasterization, instead of one done by linear interpolation); the algorithms that are most critical to understanding graphics are often one-liners and analytic in nature, and so can directly reflect the underlying mathematics.

The kind of example you would get for matrix math would be something like: "Here is a triangle described by these points, here is a matrix that projects them in a 3D camera. In such-and-such 3D library, points are projected with the matrix by doing these steps. Compute and plot the result of applying the matrix to each point." And then in successive examples you can add some details about how you go about building up the matrix by computing the camera, translating the poly and so forth, and make comparisons to how it can be done in some scenarios using only trigonometry, and further comparisons to camera lenses and artist's 3D projection with horizon and vanishing points. If you put all those connections there in one place, you have the starting point, and then it can be elaborated on into both the pure math topics(why does this mathematical representation work) and the computing topics(how do I automate the process of plotting points).

What tends to happen is that the text doesn't allow enough of a detour to make all of those connections, so you instead get an extraordinarily brief allusion to application in a single word problem, after a completely abstract introduction.

Re: Geniuses of the past were aristocratically tutored

#345

Post by aaron695 »

defgeneric wrote: post #155

The quality of this thread is really disappointing. Conservatively, about 80% of the replies, criticisms, suggestions, re-framings, here are addressed in the article by the author. About 20% of the replies here are responding to the title alone (and therefore focused on Einstein/progress in hard science). The piece is about the kind of education the children of the aristocracy received and how it has disappeared enti…

No post body was provided.

Re: Geniuses of the past were aristocratically tutored

#346

Post by hackerfromthefu »

betwixthewires wrote: post #306
lern_too_spel wrote: post #280

Earlier quoted context omitted.

He was also aristocratically tutored, learning mostly at home (first from his parents and then from a litany of professional mathematicians who tutored him in what they were interested in) and then speeding through classes all the way through undergrad.

And the next generation will be able to do the same thing at a whim even in poverty with a little device in their pocket, and not even have to pay for the classes.

Except they will fritter their time on fb and tiktok and the other countless distractions online.

Re: Geniuses of the past were aristocratically tutored

#347

Post by sdeframond »

alfor wrote: post #284

The goal of the education system is to flatten society. Both of my kid have been homeschooled up until 5th grade. At home they used to do less than 20min per day of ’school’. After a few month at school they now find the pace incredibly slow and they can’t stand the waiting they have to endure. They both score 80-95% in all classes. My son is learning programming and video editing and piano by himself, my daughter is…

I was homeschooled and did less than one hour of "study" per day until a was 14yo. 20mn seems quite likely.

I confirm that I found school very slow when I first went there, although I am happy I did since being at home with my mom all day long was becoming seriously boring (she decided to stay home when I was 1yo).

Getting used to middle school sure was a challenge but the academic part was fine. I did end up with a master in engineering.

The social side was more uncertain although, thinking back about it, I cannot decide whether I really was a weirdo or if that was just puberty kicking in. I certainly wasn't the most popular but I did have friends. I did sports and played music in a rock band (crappy but fun!), so...

Now, if I have a kid I am not sure I would homeschool her. Kids are hardwired to learn by they do need to be around an adult to learn by mimicking. It would require a serious lifestyle change.

I'd certainly be very interested in alternative schooling such as Montessori and the like though.

Don't misunderstand me: public school is great. It brings affordable education to everyone. It takes children off the streets. It makes society better.

But to be available to everyone, it is designed to be cheap and "industrial". Same content to everyone on the same rhythm, with very little variation. It is not difficult to understand that it cannot be optimum to every child.

Re: Geniuses of the past were aristocratically tutored

#348

Post by bcrosby95 »

lkrubner wrote: post #203
wcarron wrote: post #192

I think the author has got it mainly correct. There exists a perfect analog to this 'genius problem' in plain sight: Architecture. Back in the day, especially the art-deco era, buildings were adorned with beautiful facades and friezes and carvings. They had style and substance and actual design. Today, all you get are construction companies shitting out glass-paned bricks with shoddy materials, poor usability conside…

Back in 2010, I was invited to a party in a building down at the south end of Manhattan. We were 4 blocks from Wall Street. It was a new building, my friend had just moved in. The building was for the "$400,000 a year working stiffs" of Wall Street, but that line is from 1987 (the movie Wall Street) and nowadays they all make a few million a year. I was surprised by the relatively poor quality of the building. All th…

Here's a joke an Urban planner once told me: what's the difference between condos and luxury condos?

The price.

Re: Geniuses of the past were aristocratically tutored

#349

Post by morpheuskafka »

ascar wrote: post #338
defgeneric wrote: post #155

The quality of this thread is really disappointing. Conservatively, about 80% of the replies, criticisms, suggestions, re-framings, here are addressed in the article by the author. About 20% of the replies here are responding to the title alone (and therefore focused on Einstein/progress in hard science). The piece is about the kind of education the children of the aristocracy received and how it has disappeared enti…

I don't see the author addressing most obvious criticism. There is no definition of genius. He mentions a lack of ideas, but doesn't talk about the effect that every field is much more specialized. You won't make advances in physics on the level of Newton or Kelvin. "The basics are covered" (with all the caveats that apply to this statement). He also ignores the pure genius that went into developing computing and the…

Interesting, the article does mention von Neumann who as I understand it created the CPU/bus/memory computer architecture. I was a bit surprised to see a computer scientist discussed as one of the people who received this kind of education.

But yeah, my main response to the article is that "genius" and "the Western classics" are more social constructions than any sort of objectively defined term.

Post reply on HN