The Best Way to Learn Anything: The Feynman Technique 96 points by peterkshultz on June 26, 2016 | hide | past | favorite | 16 comments

 I'm not being funny, but I thought the Feynman technique was: - Write down the problem. - Think very hard. - Write down the answer.Still, this is a useful update to my understanding of his techniques :)
 https://en.wikipedia.org/wiki/Richard_Feynman#Popular_legacy: 'In a 1992 New York Times article on Feynman and his legacy, James Gleick recounts the story of how Murray Gell-Mann described what has become known as "The Feynman Algorithm" or "The Feynman Problem-Solving Algorithm" to a student: "The student asks Gell-Mann about Feynman's notes. Gell-Mann says no, Dick's methods are not the same as the methods used here. The student asks, well, what are Feynman's methods? Gell-Mann leans coyly against the blackboard and says: Dick's method is this. You write down the problem. You think very hard. (He shuts his eyes and presses his knuckles parodically to his forehead.) Then you write down the answer."'In other words, Feynman himself never promoted that. I believe Feynman did extol the virtues of learning things until you can explain them to others.I have seen many people use the real Feynman technique to good success. While I have to admit I've never quite used it directly, I can also attest to the virtues of trying to teach others to make sure you understand something yourself.I have seen many people attempt to use the "write problem - think - write solution" method. Universal abject failure. It's a joke and it always was.
 Ah I never caught it was a joke.. thanks for clarifying. The 'real' Feynman technique sounds much more interesting and beneficial ;)
 > I'm not being funny, but I thought the Feynman technique was: - Write down the problem. - Think very hard. - Write down the answer.It's funny, because he was the very antithesis of that.
 Wouldn't surprise me, the guy was a legend.
 That is for solving a problem, not learning.
 That's the one I've seen.
 I think the issue with this is that it doesn't allow for the existence of black boxes. I think an improvement might be to recurse until you hit something you don't think is worth understanding fully, and then move back upward.
 In this context what is a black box? And can you explain further "move back upward"?
 Well, it's like if learning to write a computer program you might not feel the need to understand assembly language or machine code. So you stop at "then the machine translates it into machine code" (handwave, handwave), stop the recursion, and go back upwards.
 The focus on simple language is important for comprehension, and politicians have been trending their level of complexity simpler over the years:
 This is a great technique. I wish everyone, and especially teachers, would adopt it.I think the fellow who writes waitbutwhy.com is using something like this.Also, I once read an interview with the wife of Jonas Salk, the inventor of the polio vaccine. She said she met a great many people working in the sciences and technology, and the really smart ones could explain the heart of what they were doing in a clear and simple manner.
 "Without using the new word which you have just learned, try to rephrase what you have just learned in your own language. "
 It only applies to procedural knowledge, of course, but there’s a similarity here between “teaching it to a toddler” and writing a program that does it. I often find huge gaps in my domain knowledge and understanding, not while I’m composing a readme up front but while I’m writing code.
 Explaining something in a way a toddler can understand sounds a lot like writing code to me. Computers have only as much context as they have libraries/instruction sets and you have to "explain" the rest in those terms.
 working my way through Herb Grossman's Calculus Revisited on MITs OCW. He is constantly approaching the same result from a different angle in much the same way. Best calc teacher i ever had and he passed away years ago...

Search: