I am a software engineer at Google, and have done (and still do) a massive amount of interviewing, and echo +Colin McMillen's sentiments. Even outside of software engineering interviews, where these kinds of questions are more common, the goal is as much to see/hear how you think about these problems, not just what your solution to a wacky hypothetical is.
Even for the coding questions, while geting a good/right answer is important, seeing how you solve a problem and think about it really important.
Even for the coding questions, while geting a good/right answer is important, seeing how you solve a problem and think about it really important.
Are you a software engineer? Do you want to work at Google? If so, ignore this WSJ article, and ignore the book it comes from, titled "Are You Smart Enough to Work at Google?" The article (and I'm presuming the book) is full of assertions that Just Ain't True about the sort of questions you get asked at a Google engineering interview (I don't know anything about hiring for other positions, such as Sales or Product Management, so it might be more true there -- but I wouldn't know.)
Don't worry about figuring how to escape from a blender, or dividing treasure among infinitely-rational pirates, or why manhole covers are round. Instead, study the things that might actually be useful in your job: algorithms, data structures, medium-to-large-scale systems design, testing, debugging, parallel programming, operating systems, and most importantly how to code. (Oh, and practice writing code on a whiteboard, especially if you've not done it before.)
Another pro tip: if you find a supposed Google Engineering Interview Question in the pages of a major newspaper or book, it's probably not one you'll actually be asked, because we pay attention to that sort of thing. :) Most of the questions you might see in books are either 1) apocryphal in the first place, 2) questions misattributed to Google that were actually asked at some other company, or 3) really old questions that noone has asked in years, precisely because they became too well-known to be useful. ("What's the most efficient way to sort a million 32-bit integers?" isn't a bad question, except that it's been printed and reprinted a thousand times.)
For more detailed useful advice on doing a Google software interview, you should read +Steve Yegge's blog article, "Get that job at Google":
http://steve-yegge.blogspot.com/2008/03/get-that-job-at-google.html
Don't worry about figuring how to escape from a blender, or dividing treasure among infinitely-rational pirates, or why manhole covers are round. Instead, study the things that might actually be useful in your job: algorithms, data structures, medium-to-large-scale systems design, testing, debugging, parallel programming, operating systems, and most importantly how to code. (Oh, and practice writing code on a whiteboard, especially if you've not done it before.)
Another pro tip: if you find a supposed Google Engineering Interview Question in the pages of a major newspaper or book, it's probably not one you'll actually be asked, because we pay attention to that sort of thing. :) Most of the questions you might see in books are either 1) apocryphal in the first place, 2) questions misattributed to Google that were actually asked at some other company, or 3) really old questions that noone has asked in years, precisely because they became too well-known to be useful. ("What's the most efficient way to sort a million 32-bit integers?" isn't a bad question, except that it's been printed and reprinted a thousand times.)
For more detailed useful advice on doing a Google software interview, you should read +Steve Yegge's blog article, "Get that job at Google":
http://steve-yegge.blogspot.com/2008/03/get-that-job-at-google.html
How to Ace a Google Interview