Hacker News new | past | comments | ask | show | jobs | submit login
The Feynman Lectures on Physics Audio Collection (caltech.edu)
443 points by sohkamyung on May 29, 2021 | hide | past | favorite | 98 comments

Oh I wish I could tell my past Caltech student self to tape record the lectures.

Feynman gave a guest lecture to our freshman Physics class about potato chip worlds. I wish someone recorded it.

It simply never occurred to me, or apparently anyone else, even though those portable Radio Shack cassette recorders were around.

My last boss had Feynman as his physics professor at Caltech. I had read most of Feynman non-physics books and some physics books. When he found out my interest in Feynman, we used to have interesting talk about him over lunch, mostly him talking about Feynman lectures.

That makes me wonder what could we be doing now that we're not? The time now is just like the time when you were in the lecture at Caltech. I kind of wish we had recorded more of Steve Jobs and wish he was more open for university lectures/presentations. This is their legacy and although we document their story, nothing quite beats audio/video of their own self - as they express grand ideas.

We are still throwing away massive amounts of historical data every day simply because it isn't in any corporation's financial interests to keep it. Efforts like the internet archive have very recently gained steam and are nowhere near enough.

Libraries (the putative repositories of this data) have also done more than their share in destroying it:


I spied similar patterns as a student and started collecting books I valued. As a consequence I have some not very valuable but rather rare items that I cherish and couldn’t read anywhere else, even digitally (at least at the time). One example, and probably highly controversial, is “The New Romans”—a collection of essays, poems, aphorisms, and other short writings by every major Canadian writer of the 50s-70s about America including bpNichol, Margaret Atwood, Farley Mowat, Michael Ondaatje, Irving Layton, Margaret Laurence, Dennis Lee (the list goes on and on) edited by Al Purdy (and personally signed by him and Bill Bissett each with a note to someone named Arlene. A monumental undertaking gathering from all of these writers for one edition at the height of their careers. You can still find copies, and affordable, but there was one printing in paperback in 1968 and never again. https://www.goodreads.com/book/show/23362827-the-new-romans *

I’ve invested a lot of money, and lugged them and some vinyl records back and forth across the country with me.

Still, I’m wary to pin that on libraries as a whole or as an entity as if the library itself was the problem. The problem is “move fast and break things” and digital maximalism (as the article itself ends up pointing to as a culprit). I’d point to the pressure to install “disruptive” management and those kinds of trend-chasing moves in organizational structure (it’s invaded publishing, too)

It’s funny, my day job is working with computers and I love programming and putting computers to good use, but the more I learn and the more I grow the more I prefer the physical world to our new digital one. (As I curse my phone for autocorrecting the wrong words, but still persist in typing this comment)


* Bonus, here’s a favourite poem from the book by John Robert Columbo https://rpo.library.utoronto.ca/poems/oh-canada

Although I don't really consider a lot of what we (programmers) do engineering, it does make me happy that we do is documented reasonably thoroughly if it's open source.

There's so much brilliant work of the past only accessible in either archives or spoken word, whereas code can at least be preserved as-is.

What I love about Github is not only having archival storage in the cloud, but all the forks are copies on individuals' computers.

git may turn out to be a far more valuable invention than Linus ever imagined.

The distributed aspect was a goal from day 1

Potato chip worlds?

Go on...

Worlds that appeared flat to the inhabitants but had weird geometry because they were curved in strange ways.

The point was how your perspective could hide your world's true nature.

Ah, yes. Following in the delightful tradition of Flatland [1]. Rudy Rucker's Geometry, Relativity and the Fourth Dimension (1977) may be a fun read {perhaps less fanciful than Potato Chip World} for those interested.

[1] https://en.wikipedia.org/wiki/Flatland

Or the Planiverse by A.K.Dewdney

Reminds me of Greg Egan's Dichronauts

The first thing that came to mind was a hyperbolic universe.

There's a potential subset of people who don't know who Richard Feynman is, but might recognise a segment of one his lectures from the movie theatre puzzle in The Witness (the 2016 video game by Jonathan Blow).


(The section clipped in The Witness begins at 38 minutes, 34 seconds.)

I bought The Witness (the PC version) a couple of years ago, even though it's not a game that I'm inclined to try to play (and I might not even be able to play it). I bought it partly because I was inordinately curious about superficial packaging details like whether or to what extent dynamic linking would be used, how assets would be bundled, and whether they would be obfuscated so they couldn't be viewed or played outside the game. But I also bought it because I knew it had some interesting audio and video recordings, like a recording of Brian Moriarty's lecture "The Secret of Psalm 46". Anyway, I found the audio of that Feynman clip you referenced, and it was indeed playable outside the game, in an ordinary audio player. Now I'm curious about this movie theatre puzzle and the significance of the Feynman clip in the context of the game.

It's not a game for everyone, but I'd suggest giving it a reasonably serious try if you're curious. It's by far my favorite puzzle game, and I generally do not like puzzle games.

I played through it with a local retro gaming group over a series of sessions. That made it a lot more relaxed and fun, as we could share the mental load when getting a big fatigued. Most of it we could figure out through simple intuition, persistence, observation, etc, but there is one puzzle at the end that we hand to resort to pen and paper, coming up with a formalism for it.

All in all it was a really cool experience.

I thought it would be nice to listen to these with my podcast player on iphone. Here's my too-complicated process that worked:

- in Chrome inspector, look at the network tab when you click play on a Feynman lecture. Right click the mp4 and do "copy as cURL".

- Go to command line (unix style) and paste. Then append to that command line something like "--output flp1.mp4". That will download the file locally with that file name.

- Put the file on Dropbox or something that will get it to your phone.

- From dropbox on iphone, share and export the file, then choose your podcast app. The podcast app that worked for me is "Pocket Casts".

- Now in Pocket Casts -> Profile -> Files, you should be able to play the mp4s with nice podcast-style controls and learn physics and be happy!

There's a wget one-liner here to download everything: https://news.ycombinator.com/item?id=27323235


Required 'brew install wget' first for me. ;-)

Nice, I tried doing the same, but by writing the curl command myself. It failed, it seems it requires ~the cookies~ because I was getting 403s. Thanks for the better idea!

Edit: Looks like all it needed was the Referrer header.

Feynman started fixing radios and antennas as a kid.

During the Great Depression, people did not have enough money to hire a radio repairman, so they hired him instead. He started taking incrementally more difficult jobs, until he got very skilled at it.

Another cool fact about Feynman is that he had some built-in syntax highlighting for equations in his head. A very specific type of a condition known as grapheme-color synesthesia.

Feynman used to think about the problem before doing anything, so his neighbor asked him why he wasn't working on the radio but standing there staring off into space. After Feynman fixed his radio, his neighbor told everyone in amazement "He fixes radios by thinking!" You could take this to mean he thinks about the problem first, but I think the neighbor (this was in late 1920's, early 1930's) meant that he was was amazed because Feynman appeared to reach into the radio with his mind to fix it.

Physicists joke about the Feynman Problem Solving Method:

  1) Write down the problem.
  2) Think really hard.
  3) Write down the solution.

It was after that that he enrolled into every physics class he could.

I think being exposed to early electronics (before transistors) must have been an incredible learning experience for a kid.

So much dislike for these!

I dug through Apple’s audio book section at some length and purchased Feynman’s overall survey of math, a sort of toss-off in the middle of his physics lecture series, and I love it. It’s just the most profoundly enthusiastic holistic summary of basic mathematics I can imagine packed into a one hour talk. I’m going to listen to it again today.

The pleasure of these lectures is getting to spend a little time in Feynman’s head, appreciating the world the way he does, unpacking it the way he does. I don’t think they deserve the side eye - they’re a really unique product of a very unique person, and a culturally valuable thing to have around.

I don't understand why everyone so much adore feynman lectures there are plenty of other prof. who explain much much better for example suskind

My theory is something I call "The Feynman effect". Feynman has a talent of making the listener believe that they (the listener) understand everything at a very deep level. So it gives the feeling he's an amazing teacher -- and if you don't actually try to apply that knowledge, you might never notice that you're wrong.

I realized that after reading his lecture on "the principle of least action" coming out with the feeling that I deeply understand (among other things) calculus of variations - a field I didn't even know existed until I read that. So I tried to use it -- and realized that, other than recreating Feynman's example, I can't really use it for anything.

I shared the sentiment with others over lunch the next day (a couple of other undergrads and two graduate students), and they were all familiar with that feeling....

I interviewed a Caltech grad for a job once. She was completely qualified (overqualified?) and I figured that out in all of about 3 minutes. So I asked her if she'd taken Feynman. She smiled and said she'd sat in on a seminar where he lectured and, similar to your Feynman effect, she said that he had the ability to take the most complicated idea, crystalize it, explain it and that you would understand it. This effect lasted for about 5 minutes after which you would confuse yourself.

BTW, she didn't get the job which was not my doing. My boss was a woman who felt threatened by having another woman who was massively smarter than she was.

As for Feynman, I'm more than ok with his lectures. Clarity is no substitute for application but it damn well helps application.

I wish I could frame this comment for every first-year student. The profound irony of the Feynman Lectures is that although they are revered as master works, the whole endeavor was a miserable failure.

Feynman was teaching an intro sequence, and he delivered a lovely set of lectures that grad students and professors enjoyed. We still give newly minted physics majors a lovely bound copy of the Lectures, because some of us are in on the joke.

It depends on a person. For me the lectures deserve all the praise they've got and more.

I've read Feynman lectures twice. First time in the secondary school -- it was above my head but it gave me an understanding, the right models to work with later. The second time I read it in the university after I'd already studied the topics by other means (through complex math using e.g., Landau and Lifshitz textbooks). Now, I could appropriate the full depth of the lectures.

It is often useful to have several perspectives on a subject , to know it better in particular to solve complex problems (the more tools you can apply, the better).

Oh, I definitely agree. I appreciate the perspective. The Feynman lectures are great at giving an intuition and perhaps a general framework for thought.

What they don’t give you are tools; like you say, you had to learn them in other places before really appreciating the lecture.

But those lectures do seem to leave many people with the feeling they also got the tools and not just the gist.

They are enjoyable. They are informative. But they are not at a textbook level, even though they leave the impression that they are.

The reason why they are so popular is because school/textbook leaves the opposite impression to most people, i.e. we're given tools without the "gist".

I'd been in a prep school for 3 years, and people were just regurgitating what they had been shown, applying formulas with a very vague idea of what they're doing or what their results means.

School is just about remembering things and finding patterns.

There's a Feynman story in one of his books where he shows a university class a french curve (a very curvy plastic shape used in technical drawing) and explains to them it has the amazing property that no matter how you turn it, the lowest point is tangent to the horizon. No one in the class realized he was just messing with them.

The point of lecture is not to teach you the subject throughly. That’s what practice is for. The point of lecture is to get you so profoundly interested in the concepts you’re learning that you’ll go and do the practice without feeling like you’re doing any work.

I think this is a very easy trap for the layperson to fall into with physics.

Interestingly, "false understanding" by lay people seems more common in physics than perhaps any other field I am familiar with.

Happens in other fields too. When I worked at Tera, I (and plenty of others) would talk to Burton Smith in the halls. Whatever the topic, he always made us feel included and smart. Then, after the conversation, as we moved apart, our IQ would drop and our understanding would fail.

Except maybe some of it stuck. Hope so.

That is becase feynman lectures, or any physics lecture, will not give you a solid math framework.

I think you could get a relatively solid math framework just by studying physics and learning the math concept when you encounter it in your physics education.

Yes, I find Leon N Cooper was also very good at explaining physics.

His 'Meaning as Structure of Modern Physics' is as good an resource for freshman physics as Feynman lectures.

Also there are topics Feyman himself thought was covered hastily, if I recollect correctly, and that included thermodynamics.

Feynman is an engineer's physicist, and we're on an engineering board afaict

That doesn’t make sense. Every physicist is an engineer and vice versa

Ever heard of string theory?

I would say every physicist believes they are an engineer...

Susskind is great, and he's very much inspired by Feynman.

Of course there are plenty of good professors that will help you understand physics, but I haven't heard anyone lecture in such an interesting way, have you?

In particular, Feynman himself thought his lectures were a failed experiment.

They are really more Feynman memorabilia and nerd status objects than used as educational resources, similar to TAOCP

The recording linked span nearly 3 years. Why would he consider them to be "a failed experiment"?

he said it himself literally in the preface to the lectures

"The question, of course, is how well this experiment has succeeded. My own point of view—which, however, does not seem to be shared by most of the people who worked with the students—is pessimistic. I don’t think I did very well by the students. When I look at the way the majority of the students handled the problems on the examinations, I think that the system is a failure. Of course, my friends point out to me that there were one or two dozen students who—very surprisingly—understood almost everything in all of the lectures, and who were quite active in working with the material and worrying about the many points in an excited and interested way. These people have now, I believe, a first-rate background in physics—and they are, after all, the ones I was trying to get at. But then, "The power of instruction is seldom of much efficacy except in those happy dispositions where it is almost superfluous.” (Gibbon)"


I think his remarks need to be put in perspective. Compared to any other college lecturers, how do his lectures measure up?

Sure, a couple dozen students who attended out of hundreds perfectly understood everything in the lectures and demonstrated it in exams to Feynman's unreasonably high standards. Does that mean that all of the rest failed to gain any substantial physics knowledge which was valuable in their academic or professional careers? I think not.

Looking at the wikipedia entry for the lectures (https://en.wikipedia.org/wiki/The_Feynman_Lectures_on_Physic...), it says that Feynman taught "the course" once. It's clearly a beefy 3-course sequence that was taught over the span of 3 years. Sounds like it could have been 1 semester a year for each of the three courses in the physics intro sequence. How many students is that? Somewhere in the low hundreds, I guess. 5-10% of the students with "perfect" scores could be "two dozen". That's not unusual and not a failure if larger fractions got a good-enough grounding in physics at Caltech to proceed. I think they did.

Feynman's remarks, I think, are an elliptical self-deprecating way of saying the lectures didn't meet his own standards. That doesn't mean they were "a failure".

i'll leave it to the guy who designed and ran the course to give me an assessment of it over some random that wishfully compiles circumstantial evidence (like how long the courses were taught for).

Feynmans last lectures have only been published last year: https://arxiv.org/abs/2006.08594

This would be a great project for ML-enhanced audio.

I've listened to a few things lately that could benefit from that. You wouldn't happen to know any good models available for that would you?

I really wanted to follow this class taught by Steven Pinker [1] recently, but the audio becomes so bad after the first lecture that it's almost impossible to follow. It made me also think that there could probably be some interesting post-processing applied, to improve the quality.

[1] https://stevenpinker.com/classes/psychological-science-scien...

Or izotope RX8.

Wow this is awesome. Like other has asked, is there a way to download these? Would love to listen to these when I’m driving or at work.

Is it just me, or is Feynman's voice clipping a lot even in the first couple of minutes?

I think it's just the first lecture, the tape of which was damaged somehow - https://twitter.com/preskill/status/1398371618372526080.

Are there accompanying slides or pictures of the blackboard for these lectures?

It would be good if some audio engineer could somehow remaster these.

Is there a way to download these?

For MP4 (2.6GB)

    wget -qO- "https://hastebin.com/raw/amekevecum" | wget -nv --referer="https://www.feynmanlectures.caltech.edu/flptapes.html" --user-agent="Mozilla/5.0" --input-file -
For OGG:

    wget -qO- "https://hastebin.com/raw/esukabunew" | wget -nv --referer="https://www.feynmanlectures.caltech.edu/flptapes.html" --user-agent="Mozilla/5.0" --input-file -
Titles are here: https://www.feynmanlectures.caltech.edu/flpplaylist.js

    wget -qO- "https://hastebin.com/raw/amekevecum" | wget -nv --referer="https://www.feynmanlectures.caltech.edu/flptapes.html" --user-agent="Mozilla/5.0" --input-file -

thanks so much

The URL to the media file is easy to see in inspector, in firefox or chrome. it downloads happily with curl as long as you preserve the cookies.

The filenames follow a (mostly) predictable pattern, amenable to a simple for loop. After downloaded, I extracted the audio with ffmpeg.

Just renaming to M4A should also work - the MP4 files only have a audio stream.

Thank you for this! Can't wait to try to start listening to these during workouts.

Thank you for the share @sohkamyung.

Anyone else noticed that its missing lecture #21?

You're welcome. :-)

As always, it is a great pleasure to listen toFeynman talk ...

This made my day! Thank you so much :)

Remarkable. Thank you!

Wait what?? There were audio recordings of these???

Thy Myth. The Man. The Legend

Feynman lectures are the SICP of physics. Out of date, not all that relevant but fetishized to death.

I think people just like to make this kind of comment to sound deliberately contrarian. And smarter than they really are, also.

Can you elaborate on what is "out of date" or "not all that relevant" about the books?

What a lot of people miss about these books is the joy and wonder Feynman brings to the subject. There are a dime a dozen more appropriate books to study for in preparation for university exams for example. And while they contain fascinating material, they tend to be quite dry & drab in their presentation.

Feynman's infectious love of physics and exploration really shines in these volumes and for me that gives them value in a different way to most textbooks. There is more to life than passing university exams and if someone can make a subject like this both accessible & enjoyable that is quite a remarkable achievement in my opinion

Out of date? Sure.

But it still adds very valuable insights, and new perspectives. Feynman Lectures are great. Especially Vols I and II.

Just because something has a community fetishizing it, it does not lose its value because of that.

I read most of Feynman Lectures in my undergrad, and the experience was profoundly rewarding.

It is not about gaining up-to-date knowledge of the topic. It is about learning the core and crux of the subject, and how to think about the subject.

As a layperson to physics, I really appreciate his approachable style, both in writing and in speech. Do you have any suggestions that are as approachable and fun for technologists interested in physics?

Khan Academy hands down.

Mainly because they give you a map and a road map if you will to explore the subject. Long way to go to improve the map and ways to explore it based on individual personality and skill levels but that will happen with time.

Khan academy is great, but nothing is quite like listening to a certified genius.

I think that is part of it. You can catch glimmers of how Feynman thought process worked. How he sees the world. And that is really cool.

Physics from symmetry is good.

You don't listen to Feynman because you want the most efficient way to prepare for your Physics exam, or because you want to hear about the latest developments.

You listen to him because he was one of the greatest minds of the century, as well as a very entertaining and charismatic speaker.

It's not hard to see why that combination of qualities attracts a lot of people.

Some of the accompanying problems are not bad preparation for quals. They are like the ones on the old UChicago quals

In what sense are they out of date? Like vol 2 is electrodynamics, the basics didn't change much. Do you think there is a newer textbook which does not actually teach the same content but newer things? Could you name that book?

SICP is more relevant than ever depending on where in the stack you are working on.

883 pages. What is the book for dummies, like myself? I can put up websites, but definitely need more education on Programming.

Or, what are the must know chapters in Structure and Interpretation of Computer Programming should I study?

How to Design Programs (the 2nd edition)—<https://htdp.org>—is more recent and more approachable, but in some ways follows in the footsteps of SICP. It's a great place to start before digging in to SICP or one of the other "classics," but it's also a good choice even if you just want to read one book.

Start with The Little Schemer [1] as a prequel to SICP. If you're going to read SICP, look at the beautiful (and unofficial) typeset version [2].

[1] https://www.amazon.com/Little-Schemer-Daniel-P-Friedman/dp/0...

[2] http://sarabander.github.io/sicp/

I’d suggest reading from the foreword and working through chapters 1-3. You’ll know if you want to finish by then.

It’s a really fun read; it’s an exercise in thinking differently. If you’re looking for a new/different perspective on programming this is it.

People either love or hate this book, but it’s popular for a reason. It’s not uncommon to struggle a bit through the exercises. Some people dislike it for this reason.

The intro CS course I took waded based on this book and I never even saw this book. It wasn’t until after the course was over that I found, read, and worked through (most) of it. I took my time with it—a little each day.

It’ll definitely grant you some “aha!” moments.

And it is totally worth it to do the exercises.

You can also take the course programming languages from the University of Washington on Coursera. The 3 courses use standard ML, racket and ruby to teach concepts a learned programmer should know.

It’s really not.

Well that solves that.

Personally I would not learn physics for the first time from the Feynman lectures. The Feynman lectures are aimed at senior undergrads who already took advanced physics, in my opinion. I read the Feynman lectures after I took advanced physics courses, and I got a lot more out of it than I would have at first I think.

Applications are open for YC Winter 2024

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