Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
The Feynman Technique: The Best Way to Learn Anything (fs.blog)
59 points by chamoda on Jan 27, 2018 | hide | past | favorite | 13 comments


They frame it as a learning technique, but it's also a great problem-solving technique.

When I was a kid I made money by helping my dad with his programming. When he got stuck on problems he'd come out and find me, and offer me $20 to solve it. Then he'd explain what he was trying to do in the simplest way that he could, and explain what he'd already tried. About half the time he'd figure it out before he finished explaining it (and I'd get nothing).


What you describe is most often called something like "Rubber duck debugging", https://en.wikipedia.org/wiki/Rubber_duck_debugging . I learned it as "talking to the wall" debugging.


This is similar, but I meant figuring out how you're going to solve a problem using code, not figuring out how you're going to fix your code.

One example I can remember is my dad coming out and holding up a pen, and asking me how I would view all sides of the pen without moving my head. I said, "I'd spin the pen around," and my dad yelled, "of course!", threw a twenty at me and ran back to his office.


I still full immersion is the best.

Basically, pick any subject and consume every form of media - podcasts, books, articles, videos, talk to people, Quora questions - and you’ll eventually get it on a very fundamental level.

Constantly ask questions and seek answers and rinse and repeat.

No tests too.


I think you left out "practice".

What you describe is hard to apply to, say, tango dancing, which requires a kinesthetic learning essentially impossible to get from anything other than practice.

There are also fields where that immersion approach doesn't work well. Think about the topic of "how to lose weight" or "how to raise a child."

There is a huge amount of media on the topic, and everyone has an opinion, but I don't think the immersion will result in understanding those topics on a very fundamental level.


4 step technique:

1) study some topic and write down everything you know / learned about it

2) pretend to teach to somebody or explain to somebody or write down in simple terms

3) when you are explaining use that process to expose the gaps that you don’t think you fully understand if you struggle to explain simply

4) simplify the topic further and keep iterating on your explanation and filling gaps you don’t understand until you can explain the topic in a concise way or simpler language


I am sorry but you cannot teach everything to a toddler. It needs years of learning to understand something. Feynman knew this, he knew science was hard, he would never say those things.


And he didn't. The closest I know is the account described at https://en.wikiquote.org/wiki/Talk:Richard_Feynman :

> Feynman was a truly great teacher. He prided himself on being able to devise ways to explain even the most profound ideas to beginning students. Once, I said to him, "Dick, explain to me, so that I can understand it, why spin one-half particles obey Fermi-Dirac statistics." Sizing up his audience perfectly, Feynman said, "I'll prepare a freshman lecture on it." But he came back a few days later to say, "I couldn't do it. I couldn't reduce it to the freshman level. That means we don't really understand it."

"Beginning students" here means beginning college students. Not toddler.


I do this, it's useful. It helps during meetings as well. I figured on what's being said and can retain the key ideas long enough to jot down afterwards.


But where am I going to find a toddler at this hour? Perhaps I could fill the void by training a neural network instead.


> In fact, he created a formula for learning

Citation needed. It's a good technique for your toolbox, but people like to stick Feynman's name on their own things: quotes, the "Feynman algorithm", a book titled The Feynman Processor...


> Citation needed.

There aren't any, I searched the last time the "Feynman Technique" was posted on HN and found that the term originated on Youtube [0] in 2011.

When challenged by a commenter for a source the Youtuber stated:

> The technique is inspired by Feynman, not used directly by him.

> I was inspired by the story I shared in the beginning to formulate

> his more informal approach into a method. You can see him talk

> about understanding the math ideas in his book "Surely You Must Be

> Joking, Mr. Feynman"

From there it appears that "inspired by" has transformed into "created by"

[0] https://www.youtube.com/watch?v=FrNqSLPaZLc


Thanks, that's what I thought. The article could be clearer about this.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: