I got an offer from Bloomberg right out of college before going to my masters. The last interview was fairly grilling, in a social way, not technical questions.
The hardest question after I mentioned my interests in business was why I wouldn't start a company right after getting a job. At the time I answered truthfully: I thought I needed experience at a real job before knowing what it takes to start a company.
Since then, I'm convinced I could have had a fair shot at success by starting a company on the spot.
So the biggest question I would be asking if I were Google: why don't you want to work for yourself right away?
Google is like Microsoft 20 years ago. Everybody should read "How Would You Move Mount Fuji" -- a detailed book on the topic of IQ, puzzles, and job interviews.
I heard a funny one about someone who created a popular programming language and now works for a large company (not Guido/Google, sorry), from the horse's mouth:
After a few of these questions, he stops the interviewer and says - listen, you guys are interested in Flub, and use Flub a lot - do you want to hire me, or not?
The parting line confirms my suspicion that marketing and SEO types secretly worship/unconsciously patronize Googlers and their ilk. Frankly, I don't care whether or not I have what it takes to work at Google.
oh, man. i HATE questions like this. i know they're checking to see if you're game and up for a hard challenge, but personally i'm just thinking that they want to make me dance, and my brain rebels at such shenanigans. most of the smart people i know love to dig into challenges like this, and i'm sad that i don't. as much as i'd like to work at google, i think questions like this would cause me to get up and walk out of the interview.
I don't get either one. The announcement just seems redundant. Everybody already knows that there is at least 1 so-and-so. In fact, everybody already knows that there is more than one. I can't see how the announcement adds knowledge.
It took me a long time to articulate what new information was being given; much longer than answering the question itself. The trick is to imagine a smaller scale. Lets say there were three each of brown- and blue-eyed people. And let's call the blue-eyed people 1, 2, and 3. Everyone knows that everyone sees at least one blue-eyed person, so no new information, right?
Look at this fact: before the announcement, 1 doesn't know if 2 knows if 3 sees a blue-eyed person. (2 knows if 3 sees a blue-eyed person, because 2 knows that 1 has blue eyes, but 1 doesn't know any of that.)
Then, after the announcement, the first night passes and 3 doesn't go home -- so now 1 knows that 2 knows that 3 sees a blue-eyed person. That's the new knowledge.
To finish the riddle, now 1 can reason like this: if my eyes are brown, then 2 knows her eyes are blue, because 2 knows 3 sees a blue-eyed person. (That's the new knowledge.) So if 2 goes home, my eyes are brown. And if my eyes are blue, then 2 will know that 3 sees a blue-eyed person that isn't me, and 2 won't go home. Either way, 1 goes home the 3rd night. And 2 and 3 are actually in the same position, and they go home too.
(It seems like the brown-eyed people go home the next night, but I forget and am too lazy to think about it rigorously again.)
If you want to analyze this for n=4, start with this fact: 1 doesn't know if 2 knows if 3 knows if 4 sees a blue-eyed person. Don't try this if you're sleepy.
Another way to think about it: forget about how many people there are in total, I can observe x blue eyes. Let m be the minimum globally known number of people with blue eyes. After the announcement m = 1.
On that day (and any subsequent) if your x < m you know you have blue eyes. If at the end of the day noone leaves, m must be at least m + 1.
(N=1 blue) 1B goes home 1st night.
(N=2 blue) 2B go home 2nd night.
(N=2 blue) 3B go home 3rd night.
To see why this works at N=4
B1, sees 3 blues so if his eyes are not blue everyone goes home on the 3rd night. But they don't go home on 3rd night so he and all the other blues go home on the 4th night.
etc So at N=100 all blue go home on the 100th night and everyone else stays.
Note: brown and green don't get to go home because having 2, 1, or 0 green eyes presents blue with the same information. AKA Blue and brown eyed people know their eyes are not blue but they could be brown or green. So having 10 or 10,000 people with brown eyes does not change what happens.
Yeah, I was wondering the same thing. My best guess is that the question is mistranscribed, and the orginal version was something inspired by the birthday paradox (http://en.wikipedia.org/wiki/Birthday_paradox).
Some of these questions are simply stupid. I mean, if I'd be shrunk to the size of a nickel, I would have bigger problems than imminent blending. Assuming constant cell size, which cells would I lose? Or do they shrink the atoms too? Wouldn't the required pressure needed to pump blood through my veins go up? Terrible, terrible question (for computer scientists at least, might be great for biologists).
But I think 9, 13 and 16 are predictive of certain logic skills that you'd need for e.g. debugging. I hate to say that I failed 13 - the answer is so simple.
"How many mini shampoo bottles should you order annually if you owned every hotel in California."
The reason for questions like these is not to see what answer you come up with, but how you reach it. It's also a cool hack because you can get a good idea of how creative someone is in just 15 minutes.
It's also a bit of a personality test. My interviewer said that about 20% of people won't even try or get frustrated and angry after just a few minutes.
It lets them eliminate everyone except people who are really good at answering these types of questions. I've heard that Larry and Sergey have proudly stated that they wouldn't make it through Google's current hiring process.
I'm curious as to how long you'd have to answer all these. Many are pretty fun to think about, but I'd be pissed if they only gave me 60 seconds to answer some of them.
The hardest question after I mentioned my interests in business was why I wouldn't start a company right after getting a job. At the time I answered truthfully: I thought I needed experience at a real job before knowing what it takes to start a company.
Since then, I'm convinced I could have had a fair shot at success by starting a company on the spot.
So the biggest question I would be asking if I were Google: why don't you want to work for yourself right away?