Hacker News new | past | comments | ask | show | jobs | submit login
What would Feynman do? (msdn.com)
203 points by cruise02 on Feb 14, 2011 | hide | past | web | favorite | 35 comments

When the original "If Feynman Applied for a job at Microsoft" was posted (http://news.ycombinator.com/item?id=1866305), a lot of people commented that this doesn't actually sound like something Feynman would say. Sheldon Cooper, maybe, but Feynman would have loved the simple, elegant solution that relied on no extra tools.

I felt like the middle of the post wandered around, but the final tirade felt suitably Feynmanish to satisfy me. It reminded me of some comments made about processes in his Challenger analysis. For instance, the bit about claiming a 3x safety margin on ring erosion while not establishing how it actually works to know whether that was a reasonable assumption.

Also, part of the assumption of these dialogs is not that Feynman is actually proposing these solutions, but that he is using them to Socratically establish to the listeners the flaws in their layout of the problem.


Reading this reminded me of an old story of an exam question about measuring height of a building with a barometer:


Although I found the article amusing, I have to agree with you. Surely Feynman would have seen the "correct" answer as quickly as any of us mere mortals.

That doesn't mean he'd give that solution right away. This sort of run-around seems to fit his sense of humor.

Whoever this guy is should try to make a point without putting words in Feynman's mouth. It is downright disrespectful. I know that everyone knows that Feynman did not say these things, but it is still disrespectful.

Also it shows an enormous ego. Mr. Lippert thinks he is as smart as Feynman but he is not; he just makes Feynman sound like a pedantic asshole.

Mind you I agree with his point that those questions are silly, but again he can make this point without bringing in Feynman's corpse and using it as a sockpupet.

For the record: I intended no disrespect; I have immense respect and admiration for Feynman.

And for the record: I do not think that I am as smart as Feynman; that would be ridiculous.

The piece was intended as satire, and is in a long tradition of such dialogues intended to ridicule a "straw man" of a particular position. Consider, for example, Gallileo's Dialogue Concerning Two World Systems, in which the scientist, Salviati, criticizes the position of Simplicio, who believes the earth to be the center of the universe.

I couldn't think of a better modern figure to stand for science, reason, clear thinking, and a mischievious sense of fun than Richard Feynman as my Salviati, and I hope that he would appreciate the spirit in which it was presented.

I think you defended your use of satire very well. Good job.

Would it have made you feel better if he had invoked Einstein's name? It's satire, relax a little.

People did that to Einstein all the time, claiming that he followed one religion or God or another, and it actually did piss him off.

Agreed. This was a cartoonish version of Feynman I found rather pathetic. It is missing the recurring themes of learning and experimentation found in real Feynman stories. Either the old barometer story or the manhole cover story are far superior to this blog post.

The Feynman I know from the _Feynman Lectures on Computation_ would have quickly disposed of the information-theoretic problem posed, and then have gone on to discuss all the interesting variants.

Without wanting to drag the discussion down, it seems remiss not to link to this image that's been going around forever.


..And why is the solution you were clearly driving me towards one which takes advantage of an undocumented and unreliable epiphenomenon..

Since when has the thermal performance of a light bulb been undocumented and unreliable? I would think that designers of light fixtures and shades rely heavily on the documented thermal performance of light bulbs, and rate their products for compatibility with a range of bulbs accordingly.

he calls them epiphenomenon which is correct, their primary property is light production, heat is a secondary feature and not reliable, take for example LED based lighting, or even the low energy fluorescents.

But RF was just being a well deserved smartass. Besides, who would ever interview RF, it would a waste of time, Unless you have Newton, Einstein and Larry & Sergei already on your payroll just give the man an office, s stapler and let him do whatever comes to mind.

Back when I entered Caltech, someone mentioned what Feynman's salary was. (Feynman was a prof at Caltech.) I stupidly said I thought his salary was outrageously high. I was pointedly told that Feynman was worth far more than his salary simply to have his name listed as being on the staff (bringing interest, credibility, and donations to Caltech).

And, of course, Feynman being Feynman he did infinitely more than just have his name on the roster. Only his tragic demise was able to stop him.

Any company would be out of their freakin' minds to pass up any opportunity to hire Feynman.

I'd "interview" him for the opportunity to spend an hour with him, not that there's any chance in hell I'd say "no hire".

As for his position and job at the company, who cares. Whatever he wants.

Just in case anybody didn't know already of this gem: http://longnow.org/essays/richard-feynman-connection-machine...

I love their response to Feynmen's desire to "do something real".... "So we sent him out to buy some office supplies."

their primary property is light production, heat is a secondary feature

Hmm, not necessarily. Back in the days when this problem was originally offered incandescent bulbs were by far the most popular light source for facilities, and if you look at a graph[1] of such a bulb's output, you'll find the fast majority of it is below the threshold of visible light -- the real primary property of incandescent bulbs is IR (heat) production. As a physicist Feynman would be well aware of this, and hence I think he would approved of the low-tech intended solution over the suggested alternatives.

If you posit fluorescent lights, the traditional solution would still work. LED lights weren't stipulated as a special condition and even they are stated, they're not 100% efficient and hence might well work with the traditional solution.


It doesnt matter whether they are phenomenon or epiphenomenon. The assertion was that they are undocumented and unreliable. Do you believe that there is no documentation on the thermal properties (epi or not) of a light bulb? If there is no such documentation then I imagine that the designers of light bulb fixtures and shades cant sleep easy..

"Since when has the thermal performance of a light bulb been undocumented and unreliable?"

By making assumptions as to what type of a light bulb it is (which in this setting one could not know beforehand). F.ex. LEDs give off very little heat.

Right. When this question first became popular — I remember hearing it from my grandpa sometime in the 1990s — CFLs and LEDs were not common at all (though fluorescents were, in stores/offices; but I think most people assume this is a room in a house.)

But LEDs are not light bulbs. :-)

Furthermore, he's applying for a position at Microsoft.

... who mandate darkness as an industry standard. Why are they even asking about light bulbs?

"Microsoft technical support, how may I help you?"

"Yeah, one of our light bulbs just went out."

"Have you tried selling your house, and moving into a new one?"

Before turning it off and back on again? What tech support manual are you reading from? ;)

Can't read - The light's out.

I don't think Dr. Feynman talks in such a cryptic language, if you've read his books you would know that he actually talks/writes in a very lucid language. But the person mentioned in the interview does sound like a certain "Sheldon" from the "The Big Bang Theory".

Brain teasers and real life just don't work the same way. The trick is to distinguish them.

If a bridge can only support 2 people, maybe it is better to not cross it at all. If you do have to cross it, maybe you can trust your tamed tiger with the goat. Pirates aren't rational agents that use silly rules for sharing the treasure. An egg that doesn't take a scratch when falling from the 13'th floor belongs to wonderland. Few women would kill their cheating husband right away. Fewer still would rely on the perfect rationality of others, and the mayor's to do so.

And so on. Brain teasers are fun, but many people (not just Richard Feynman) don't accept their weird assumptions right away, and instead assume a real-life setting.

I find this story unlikely, but probably for a pretty uncommon reason. Here's the thing: I heard this question a number of times before (never in an interview; always on the interwebs), but never attempted to solve it and never read a solution. As I was reading this paragraph:

  Can I assume that the lights and the switches are correctly
  wired according to the National Electric Code of the United
  States? That is, that the switches interrupt the hots, not the 
  neutrals, that the switches are standard-duty switches rated 
  to interrupt 15 amps of 120 volt alternating current, and so on?
I instantly realized the intended answer. I don't doubt for a second that if that just happened to me, it would also have happened to Feynman (because I blame being a physicist for what just happened), immediately after first thinking of that important signalling word that he was supposedly about to speak.

Of course Feynman knows the intended answer. He's way ahead of the interviewer, and is having some fun with him by deliberately avoiding the "correct" answer.

Reiterating how brilliant Feynman was completely misses my point, which depends on the assumption that Feynman would start out as he did.

Feynman would have a better approximation for the resistance of a human. 1K is just way too low.

"Does your team usually write code whose correctness relies upon undocumented and unreliable correlations, correlations whose magnitudes can vary widely as a result of implementation details?"

I loved this implication on a Microsoft website.

Registration is open for Startup School 2019. Classes start July 22nd.

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