Live data from Hacker News

The Lesson to Unlearn

paulgraham.com

581–588 of 588 posts

Re: The Lesson to Unlearn

#581
post #300

Earlier quoted context omitted.

Add air resistance into the mix, and denser materials do fall faster. Coins vs. napkins. Plums vs. crumpled balls of paper I'd actually take the lesson the other way. It's hilariously tough for model-makers to acknowledge real-world noise that prevents their theoretically precise models from delivering reliable results.

I've done so many physics questions that explicitly instruct me to disregard air resistance... Reminds me of Feynman's speech: https://v.cx/2010/04/feynman-brazil-education > There are no experimental results mentioned anywhere in this book, except in one place where there is a ball, rolling down an inclined plane, in which it says how far the ball got after one second, two seconds, three seconds, and so on. > The nu…

Thanks for that Feynman link!

> “If I ask you a question during the lecture, afterwards everybody will be telling me, ‘What are you wasting our time for in the class? We’re trying to learn something. And you’re stopping him by asking a question’.”

In my experience there are almost just 2 types of motivated students: those who want to understand the content of the course and those who want a good grade and ultimately graduate.

There's often a conflict of interest between them. For example the first group wants to learn more, ask more questions, go deeper. The second group sees this as a threat because the additional content may now be on the exam as well. And they often misunderstand this behavior of the first group as just being teacher pleasers/pets.

Or from own experience: when you tell the prof that he accidentally started to cover the same topic that we already covered in the last lecture (because you're there to learn more), you will annoy a lot of people who hoped to get one fewer lectures to study in the course. Less content, less effort, same grade.

You hear the first group having passionate discussions on physics or programming over a beer, while the other group discusses all the latest info of which prof is how demanding, what the grading criteria are, what the deadlines are, which departments you should or should not take courses from, where they have mandatory presence, who does multiple choice tests and who does free-text exams.

I don't have much to conclude, just an observation.

Re: The Lesson to Unlearn

#582
post #552

Earlier quoted context omitted.

I'm intrigued. What are you saying here?

It sounds like there could be a 100x difference between a light guy on a light bike and a heavy guy on a heavy bike. Sounds hard to believe, so I'm intrigued too :).

Could be 4x if you use mini order of magnitudes, and if you include kids and adult body builders... there’s your 4x!

Re: The Lesson to Unlearn

#583
post #567
post #291

Earlier quoted context omitted.

It is amusing how many of my peers still hold to the notion that heavier things fall faster. Not a higher terminal velocity, but the intuition that they accelerate faster. Hilariously tough for folks to shake.

If your model can't be used to predict things in reality, then what good is it? If I throw a feather down from a rooftop and then I let a brick fall down, the brick will hit the earth first. The heavier thing falls "faster". You can use a bunch of words to explain what's really going on there, but it won't change that the brick hits the earth first, hence it falls faster. Unless you can predict the future correctly a…

Because science aims to discover what is really going on under the hood. And that matters. By breaking the problem into vacuum and non vacuum you get a better understanding.

Re: The Lesson to Unlearn

#584
Is the latest essays the in-text links to notes are missing.

(My favorite from this one: "Learning is the naive algorithm for getting good grades.")

Re: The Lesson to Unlearn

#585
post #240

The tech industry has unfortunately adopted the methodology of centralized hackable tests as the canonical gatekeeping method in the form of programming interviews. Most big tech companies don't care about how good you have been at delivering some value through creating software: they want to see you deliver a very specific type of performance at a whiteboard. Interviewers are given specific math puzzle questions to…

Haha here is the email i have from my facebook interview thats coming up. "Your initial Facebook interview is coming up and we want you to ace it! Here are some tips and resources so you know what to expect and can prepare adequately. Preparing is key. Our initial interview determines whether to continue with a full series of onsite interviews. This initial interview is primarily a coding interview that will take pla…

That's good, they're trying to put everyone in a level play field given the system they use.

Re: The Lesson to Unlearn

#586
post #240

The tech industry has unfortunately adopted the methodology of centralized hackable tests as the canonical gatekeeping method in the form of programming interviews. Most big tech companies don't care about how good you have been at delivering some value through creating software: they want to see you deliver a very specific type of performance at a whiteboard. Interviewers are given specific math puzzle questions to…

Sure, but what's the alternative?

I know big companies do a lot of AB testing with interviews and change their approach over time (Google stooped asking why are manholes round and looking at AST scores because data showed them those things had almost zero correlation with performance).

Re: The Lesson to Unlearn

#587
I'm too late to comment, but my 2 cents: As long as there's something important at stake people will try to hack the test. There's no way around that.

The best that can be done is to try to keep the correlation between the "test" and what it's trying to measure add high as possible and to teach everyone how to hack it, that way everyone is on a level play field. If everyone hacks it then the difference in scores will be mostly about what is being measured and not about who hacked it and who didn't.

That's why big companies actually advice interview candidates to practice on leetcode and give them example interview questions.

But I agree with the main point that being good only at "hacking tests" in life instead of being good at your profession is very bad.

Re: The Lesson to Unlearn

#588
post #22

Earlier quoted context omitted.

We thought the same way but it’s difficult to replace the social component. Still it’s hard to see how bad the school system is and force our kids to participate. It could be so much better.

Look for an unschooling school, strange as that may sound. Free democratic and Sudbury Valley model are some of the things to search for. There aren't many around but the numbers are increasing. We send our son to one. It's a leap of faith; the kids essentially "do whatever". But it seems worth it compared to the damage done by middle and high school. And the kids that I've seen come out of this school seem to do wel…

Agile Learning Centers and Liberated Learners are inspiring networks of small schools in this vein (and both less dogmatic than Sudbury - in a good way, IMO).
Post reply on HN