Cargo Cult Science, by Richard Feynman (1974) | Hacker News Reader

Cargo Cult Science, by Richard Feynman (1974)

yost.com

31 pointsvinutheraj17 comments

I'm just finishing "Surely You're Joking Mr. Feynman" - what a damn good book.

For a few years, I'd read references and anecdotes and cool snippets and meant to read it, but I never got around to it. And now I'm kicking myself that I didn't earlier, because it's wonderful and illuminating and my worldview is greatly expanded.

You, yes YOU, if you haven't read it yet, get your copy and read it. You'll be happy you did. It reads fast, and you'll learn a lot about society and love and engineering and science and mixing with people very different than you. Just an amazing read. It's worth taking a 30 minute break from Hacker News each day for a month to read it, it's like the essence of Hacker News distilled into witty, insightful, gorgeous easy-yet-deep reading.

Amazon, no affiliate link (really, add it to your cart and get it with your next order - no excuse not to, anyone who likes this site will love this book):

http://www.amazon.com/Surely-Feynman-Adventures-Curious-Char...

I could just hope, that William H. Gates will still have enough income to buy the right to copy this book, and put it out on Silverlight, out of pure fascination for science and integrity; so that Amazon will need not meddle with the freedom of the Right to Read.

Until then, those whom dare to click through and read seem to be in breach of the law. May you have that freedom. Lucky it's not source code with all those 4 freedoms...

Why on silverlight, why not on project gutenberg ?

ask W.H.G. ...if I were to guess, the real answer might have to do something with control for power. Speaking of integrity, power and Gutenberg: "Understanding Power" is quite a book.

> The next experiment, and the one after that, never referred to Mr. Young.

I read this years ago, and it left me with the impression that most science is cargo cult science (90% of everything?). It's a delight to read people with ridiculous levels of integrity - like Chomsky or Feynman. It's a shame that integrity seems to require Socratic heroism.

Programmers owe far more to the works of Chomsky -- more than most are aware of, or care to admit. Without his rigorous stance, I am not sure if compilers(!) and programming languages would exist, the way they do now.

Research has been more openly industrialized in the past few decades; it is hard to escape pseudoscience due to the sheer mass production. Deprecated Phd.-s are just a symptom, pseudo-Phd.-s just make matter worse. Cargo-cult is quite easy to unmask for anybody, pseudoscience is tougher by design.

"Because of the success of science, there is, I think, a kind of pseudoscience. Social science is an example of a science which is not a science; they don't do [things] scientifically; they follow the forms -- you gather data, you do so-and-so and so forth but they don't get any laws, they haven't found out anything.... You see, I have the advantage of having found out how hard it is to get to really know something, how careful you have to be about checking the experiment, how easy it is to make mistakes and fool yourself. I know what it means to know something, and therefore I see how they get their information and I can't believe they know it, they haven't done the work necessary, haven't done the checks necessary, haven't done the care necessary. I have a great suspicion that they don't know, that this stuff is [wrong], and they're intimidating people." -- The Pleasure of Finding Things Out, by Richard P. Feynman

"Context-free grammars... are a reinvention of a technique first used by ancient Indian grammarians... They were reinvented by Noam Chomsky... and independently by John Backus for the analysis of Algol-58 syntax." (Russell & Norvig 2nd edition p. 827)

I generally squint at claims like "without so-and-so, X would not exist". Independent invention is so common it's the opposite that's surprising. Chomsky did make statistical techniques unfashionable in NLP for a few decades.

The way I know the context: after WWII, it was clear for policy makers in the USA, that internationalization and (seeing the success of) science (as mentioned above) were key to global dominance relying on technology. There was a slight hope for the automatic translation of documents [i.e. scientific papers] from one language to another, and funding was abundant with the '56 landmark paper. The original goal then was not reached, but as with all good research, the by-products were of great value. Speaking of statistical techniques, not unlike GOOG's case today, with the UN documents used to train translate.google.com -- somewhat achieving the "original" dream, having gone full circle. I would therefore discount the later UNESCO report from Backus as independent work, it is rather the incarnation of the results of the zeitgeist [1].

About the influence of Pāṇini it is true. Thanks for drawing my attention to it: I've learnt something[1]. But I am inclined to think, that the pieces of today's compilers fell to place as a byproduct of his work at MIT (right time, right place).

[1] "Speaking of linguistic law in general is like trying to pin down a ghost" -- Ferdinand de Saussure

[2] http://www.hinduonnet.com/fline/fl1825/18250150.htm

When asked what suggestions he would give to individuals who are trying to raise questions and challenge conventional doctrines, Chomsky said: "It's the same advice you'd give to a young scientist. Be honest, be thoughtful, be creative."

Those refs to Panini in Russell & Norvig blew me away a little -- context-free grammars and knowledge representation in 350 B.C.!

On Backus, here's what Google found at http://archive.computerhistory.org/resources/text/Oral_Histo...:

Booch: And giving examples, and stuff like that. I had read you were influenced by some of the work of Noam Chomsky, that led you to that.

Backus: Yeah, well, that’s a funny story. That’s what I said and what I believed, and yet… Who was it? Somebody sort of proved that I was wrong about it, that I hadn’t got it from Noam Chomsky, because the dates were all wrong somehow. But-- God, who was that?

Unrelated quote on the next page:

Booch: Really! Interesting. So as a programmer, what did you program in most of the time?

Backus: I never wrote many programs. I was not good at doing that. What kind of programs was I going to write anyway?

...an avid TiVo user, or rather Linux that is ;)

Chomsky espouses some pretty strange political views.

Chomsky's an odd one. He gets a lot of respect from academics and laypeople alike but a) no one in academia has ever gotten his theories to model anything in the real world and b) the laypeople who like him tend to already share his politics.

He's a very, very talented public speaker and writer - talented enough that no-one questions what he's saying.

a) Chomsky, N.: Three models for the description of language, IRE Transactions on Information Theory (2): 113–124. 1956.

I would like to bet, that you are using formal languages. In case you don't, consider the following: http://www.jstor.org/pss/411934 [GOOG feels as if it is what RAND used to be, as Daniel Ellsberg noted in an @ GOOG talk]

b) He is still at MIT, an Institute with strong DoD/WhatNot links (follow the money). Judging by the daily news, I am in doubt if they really do share politics...

I don't know about Chomsky, either. There has been some harsh criticism about integrity on e.g. Cambodia.

It seems to me that Chomsky notes something obvious -- USA has an egoistic foreign policy and lies about it, like most every country (see "realpolitik"). When Chomsky then gives his explaination for the lies, he seems to border on conspiracy theories].

For instance, has this been answered? http://jim.com/chomsdis.htm

Here is a cute list: http://www.paulbogdanor.com/chomsky/200chomskylies.pdf

I read those a while ago, but don't really know (or care) enough to build a certain opinion. If I would respect Chomsky as an intellectual, I'd first want to see serious answers to those.

<quote> One time I sat down in a bath where there was a beautiful girl sitting with a guy who didn’t seem to know her. Right away I began thinking, “Gee! How am I gonna get started talking to this beautiful nude babe?”

I’m trying to figure out what to say, when the guy says to her, I’m, uh, studying massage. Could I practice on you? </quote> :)

"Another example is the ESP experiments of Mr. Rhine, and other people. As various people have made criticisms--and they themselves have made criticisms of their own experiments--they improve the techniques so that the effects are smaller, and smaller, and smaller until they gradually disappear. All the parapsychologists are looking for some experiment that can be repeated--that you can do again and get the same effect--statistically, even. They run a million rats no, it’s people this time they do a lot of things and get a certain statistical effect. Next time they try it they don’t get it any more. And now you find a man saying that it is an irrelevant demand to expect a repeatable experiment. This is science?"

Feynman noticed the scientific procedure problems in parapsychology "research," and later investigators responded to his suggestion to be better research by carefully debunking claims about ESP research.

http://en.wikipedia.org/wiki/Project_Alpha

Reply on news.ycombinator.com