Well, I’m definitely filled with rage about these questions.
But I did see the wink. I wasn’t angry at you or anything, merely annoyed that at the fact that somebody feels the need to make that comment every time.
This whole thing is probably my biggest pet peeve.
Then again, if somebody didn’t make that comment I’d probably miss it! So thanks!
I found this site:
http://www.sellsbrothers.com/fun/msiview
It talks about these questions.
In part it says this:
This was just one person’s opinion, but if this really is why MS asks these questions, then it seems likely that the people that are copying MS are copying the question without knowing the purpose for asking it.
Which I find funny.
So Microsoft then, is full of people who spend their time keeping it real, man?
I got asked that question in an interview and had never heard it. I did manage to get the correct answer. I think they are looking for some sort of spatial logic technical aptitude with those types of questions.
– Xaroc
There’s another reason, and the most important: it’s to see how well you handle incidental stress. You’re given a stupidly difficult problem; now, do you freeze up, do you get belligerent, or do you just try to work through it the best you can?
In most cases, no-one gets a “no hire” for not getting the question right. Asking questions in a calm voice and seeing the problem through to some sort of conclusion, even if it is incorrect and suboptimal, will usually get you good marks. Refusing to answer, freezing up, getting enraged, or freaking out, on the other hand, indicates some reasonably appropriate measure of your immediate crisis management skills.
Nonsense.
Well, OK - obviously if the candidate were to become enraged or “freak out” that could indicate difficulties in socially appropriate behavior that might have a deleterious effect on the candidate’s ability to work in a team.
But the rest of the idea is nonsense. It’s a lousy way of determining whether a candidate is a good fit for the activities involved in performing the job. The ability to deal with an abstract question asked out of the blue in an interview does not tell you anything about the candidate’s ability to handle the kinds of crises that might arise in real life. Maybe the candidate is really good at brain teasers (or has heard this particular old saw before). Maybe he or she has difficulty with spacial puzzles but is excellent at solving word association or pattern sequencing puzzles.
It would be much better to ask questions related to the actual position. If the candidate is interviewing for a project manager opportunity, pose a scenario that might actually happen - you’ve just discovered that one of your staff badly underestimated the complexity of a task, how do you deal with situation and get the team back on track? How do you address customer concerns? What if this particular customer is very sensitive to schedule, but not as sensitive to cost?
If the candidate is interviewing for a programming job, use a crisis from software development - a requirements change from the customer means that you need to add logging support to 300 domain classes in time for a patch release tomorrow night, how would you approach the problem? Possible answers include using a base class or class template to reuse the logging code, or creating a quick-and-dirty code generator using awk/sed.
In either case, the interviewer can use the opportunity to construct a dialogue with the candidate, using additional deeper questions to elicit responses that illustrate both problem solving skills and the job-specific abilities.
–milo
I’ve used puzzlish interview questions a couple of times, although mostly with code fragments and the like. Why? Partly I do it as a test of personality, but also as a test of communications skills, willingness to work with an arbitrary set of constraints, and in some cases, to see how many hints it takes before they get it.
Wrong answers are fine, since they let me segue into debugging questions (often asked in the form of: Ah, but what if such-and-such situation occurs?). You can train almost anyone to write code. I’m more concerned that someone not do negative work (i.e. work I’ll have to undo later).
The non-technical puzzle questions are entertaining, but of marginal value when trying to determine problem solving skills. Too often those questions involve some sort of gimmick, basically a memorized answer or trick that reveals the solution. The cylinder problem is one such example.
Flip the question onto the interviewer by pointing out square manholes are suprisingly common, it’s just that most people think too narrowly to notice them. They think the covers must be cast iron instead of the plate steel from which they actually are made. Since it doesn’t fit their preconception, they don’t notice them. 
Exactly Milo! You are 100% correct and Doug is 100% incorrect.
Doug, I think that you have been fooled or have fooled yourself into believing that a very very narrow and extremely artificial situation can be a good predictor of future behavior in general and non-artificial situations.
I don’t have any idea why somebody would believe this. At the very least I would want to see some serious proof that there was a correlation.
Do you realize that academics have been wrestling for years and years with whether the SAT predicts anything? And this is a multi-hour test with several different subjects!
If the experts wrestle with the predictive power of the SAT for goodness sake, what’s the likelyhood that the 2 minute phony bullshit question has any predictive power?
I believe these questions are used by people who don’t know how to properly conduct an interview. It’s a crutch.
Milo’s link has a pretty good article, but this guy is still in love with the “impossible question.”
Smart candidates will realize that you are not quizzing them on their knowledge, and they will enthusiastically leap into trying to figure out some back-of-the-envelope answer. “Well, lets see, the population of LA is about 7 million; each person in LA has about 2.5 cars…” Of course it’s OK if they are radically wrong. The important thing is that they leapt into the question enthusiastically.
I just don’t get this. It’s ARTIFICIAL. You simply have to pick your canned response and then go with it.
Some interviewers might be looking for “enthusiasm” while others might be looking for somebody who is “real” or “genuine.”
Either way, the candidate can decide ahead of time how to react to the question and then just react that way.
I really don’t get the hard-on this industry seems to have for these stupiud questions! I guess every industry has some stupid quirk. At least ours is relatively harmless.
Although the guy also says
Finally, avoid brain teaser questions like the one where you have to arrange 6 equal length matches to make exactly 4 identical perfect triangles. If it’s an “aha!” question, you don’t get any information about “smart/get things done” by figuring out if they happen to make the mental leap or not.
So at least he’s half right.
The debate is over what exactly it predicts, not whether it predicts “anything”. The debate is over its meaning.
The debate is over what exactly it predicts, not whether it predicts “anything”. The debate is over its meaning.[/quote]
Ok Semantics-man.
Well some people say that “what exactly it predicts” is “nothing.”
Update.
I recently went through a round of interviewing. Every single interview at every single company involved logic puzzles as a major part of the interview. Every single one.
So good luck with that Spoofy. I hope you don’t plan on looking for a job any time soon.
Also, did you get the job Mr. McCullough? I think you did.
Talk about your thread necromancy. There’s nothing but bones left!
This thread was from my first MS interview back in 2003, which was the logic puzzle era; I did horribly because uh, I hadn’t memorized all the logic puzzles yet. Sometime after that Microsoft teams (and official hiring guidance) started dropping them because, well, they’re comically stupid trick questions as usually implemented by interviewers, and more importantly a terrible predictor of on-the-job performance. I got a job at MS in 2004 in an interview loop that actually asked relevant questions.
God help us. Well, the good news is that if you just order all the major puzzle books and memorize them you’ll do fine.
So…Professor Layton is actually a job hunting expense then?
I’ve done a few programmer interviews in the last few years, including one at MS, and there were 0 puzzle questions. Every question felt like a relevant problem that might come up in the course of work, and was interesting as well as challenging.
One thing that law firms actually get right in my opinion (compared to investment bankers and hedge funds, for example).
Of course, that may be because of the grotesque hours worked by many big firm lawyers - there’s a reason law firm interview questions are commonly described as, “Figuring out whether you are someone I would want to kill if I were stuck working with you at 1 in the morning.” :)