Chad Perrin: SOB

Okay, so the title isn't exactly accurate. They aren't all stupid. Some of them are actually kind of clever. Others are obviously just designed to see how the interviewee works. Some, however, are definitely stupid.

I ran across an old article at Business Insider: 15 Google Interview Questions That Will Make You Feel Stupid. In short, they did not make me feel stupid. They did, however, make me hope that the smart people at Google are smart enough to not take someone being stumped by a single question as a black mark on their interview, since anyone can be stumped by something that might appear simple in retrospect later.

Without further ado, here are the questions:

How many golf balls can fit in a school bus?

This question is evidently asked of applicants for a product manager position.

The answer is actually really simple. Develop a mathematical expression used to estimate. If they want a more exact number, tell them "Give me a school bus and an effectively unlimited supply of golf balls, and I'll tell you." Otherwise, your expression should allow for variable input based on the internal volume of the school bus. You may need to tell the interviewer that you need to look up something about how roughly spherical objects pack together, and you may have to ask whether you're allowed to grind the golf balls into dust to fit more of them inside the school bus.

How much should you charge to wash all the windows in Seattle?

This question is also evidently asked of applicants for a product manager position.

$50 per window (then pay other people $30 per window to actually do the work). Tell them you want to double-check the market standard rate for window-washing before giving a final answer, though.

In a country in which people only want boys every family continues to have children until they have a boy. If they have a girl, they have another child. If they have a boy, they stop. What is the proportion of boys to girls in the country?

This is evidently yet another question asked of applicants for a product manager position, though I think it's probably a good one for "software engineers".

It's time for another simple mathematical expression. This one might benefit from a little recursion. No, I'm not doing your work for you. The end result will, however, be very close to 50% each, with a negligible edge for boys — assuming a 50% gender split for birth rate.

How many piano tuners are there in the entire world?

How many of these target product managers? Here's another one.

"I need more detail for project requirements. Let's start with your definition of 'piano tuner'. For instance — is every person who has ever tuned his own piano a 'piano tuner' for purposes of this? If you're only talking about professionals, tell me how much money the person has to be making to be a 'piano tuner'. Do we include out-of-work piano tuners? Details, man! I need details."

That, or: "As many as there needs to be. There may be a few extra, or slightly fewer than there 'needs' to be, depending on externalities if people are trying to 'manage' the piano tuning economy, or something like that."

One could always just start doing math based on assumptions of how long it takes to tune a generic "piano", what constitutes a "piano tuner", and what percentage of pianos actually get tuned, among other things. Such assumptions will almost certainly lead to inaccurate results, though.

Why are manhole covers round?

Hallelujah, this one is not for product managers. It's a question for software engineer applicants. Funny how the first "software engineer" question strikes me as a "reverse engineering skills test" question.

If they were square, they might fall through the hole and kill someone working in the sewers.

Design an evacuation plan for San Francisco

Google must hire a lot of product managers. This is another question for such job applicants.

The way I thought of it was this:

"I'd just start working on details and ideas, and let them watch, since I'm sure what they want to see is how I work." As a friend of mine put it when I posed the question to him, though, "Sure." Poe-tay-toe, poe-tah-toe.

In any case, a number of questions would have to be asked (or assumptions made) at the start of the planning stage. I'm sure the questions you ask or assumptions you make (if questions other than basic question clarification aren't allowed) are what they really want out of your "answer".

How many times a day does a clock's hands overlap?

Here we go again. Product managers galore.

"22, assuming a twelve hour circular analog clock with an hour hand and a minute hand. Let me know if the requirements differ, or if you want a mathematical expression that can take variable inputs for other clock designs." The minute hand doesn't cross the hour hand in the hour leading up to one o'clock, nor in the hour leading up to twelve, but it does cross at twelve. Thus, you have one overlap per hour on the clock, minus one.

Also: s/does/do/

Please make your interview questions grammatically correct.

Explain the significance of "dead beef"

Applicant type: Software engineer! There's another one that isn't a product manager.

I have two answers to this:

1. Dead beef doesn't fight you as much as live beef when you try to throw it on the grill.

2. It's a hexidecimal representation of a fairly big number. If you lop off two of those characters, depending on which pair (assuming you're counting in twos rather than just picking characters at random), you also get the hexadecimal RGB code for one of several shades of nearly-gray.

The most annoying thing about this question is that they're probably not really testing your abilities so much as your exposure to certain cultural phenomena — particularly the once-traditional use of 0xDEADBEEF as a test memory address. I'd fire the fucker asking this question of interviewees if that was his reason for asking, or at least ensure that he doesn't get any say in who's hired, because letting him keep good people out based on their lack of exposure to this one cultural tradition or let people in based on their exposure to it is a great way to get false positives and negatives in the interviewing process, and thus result in a lower overall value of employee pool.

A man pushed his car to a hotel and lost his fortune. What happened?

This is a question for people applying for a software engineer position.

My first question was "What the heck do these two things have to do with each other?" It occurs to me, after a little thought, that they might be completely unrelated.

You need to check that your friend, Bob, has your correct phone number, but you cannot ask him directly. You must write the question on a card which and give it to Eve who will take the card to Bob and return the answer to you. What must you write on the card, besides the question, to ensure Bob can encode the message so that Eve cannot read your phone number?

This question is evidently asked of applicants for a software engineer position.

First, I'd offer to rewrite the question so the interviewer doesn't embarrass himself when giving it to other applicants. Then, I'd probably just send Bob a note that said something like "Call me on my cell, or have Eve tell me if you don't have my number." It's not technically a question, but I suspect the interviewer wouldn't complain. If he would, I probably don't want to work for him. This is problem solving, not Jeopardy. Anyway, out-of-band confirmation if you're trying to avoid letting Eve in on the secret seems like the most reasonable answer, so I'm pretty sure I win the prize here.

You're the captain of a pirate ship and your crew gets to vote on how the gold is divided up. If fewer than half of the pirates agree with you, you die. How do you recommend apportioning the gold in such a way that you get a good share of the booty, but still survive?

This question is evidently asked of engineering manager applicants.

"Figure out how to give everyone exactly even shares — but don't do it. Now take the money from the shares for a number of crew members equal to half minus two, and spread that money amongst the rest of the crew, so that more than half of them are happy with you. Take the share from the last guy, and put it in your own pot, just for a CEO golden parachute. Since we're apparently being asshole CEOs about this, follow that up by asking Uncle Sam for TARP funds, then retire."

The answer they actually want is probably to split the dough evenly amongst the smallest number of crew members that is more than exactly half. That's slightly safer than my answer, but I think the difference is negligible and if you're going to be an asshole about it you might as well make out like a bandit.

I wouldn't give anyone who answered this question this way as if it's how he actually does his job any kind of management position. Sociopaths need not apply. Nobody who actually believes they should conduct their personnel management duties like this should be let within a hundred miles of an engineer manager position.

You have eight balls all of the same size. 7 of them weigh the same, and one of them weighs slightly more. How can you find the ball that is heavier by using a balance and only two weighings?

This question is (again) for project manager applicants.

Ooh, sorting algorithm stuff.

• Weigh three of the eight balls on each side of the balance.

  1. If one side is heavier, weigh one of the three on each side of the balance.

     • If one of them is heavier, you win.

     • If they're the same, it's the other ball, and you still win.

  2. If neither side is heavier, weigh the other two balls, one on each side of the balance.

     • One of them will be heavier. You win.

No matter how it works out, you've only done two weighings. I notice, after writing this answer, that it's damned near pseudocode already.

You are given 2 eggs. You have access to a 100-story building. Eggs can be very hard or very fragile means it may break if dropped from the first floor or may not even break if dropped from 100th floor. Both eggs are identical. You need to figure out the highest floor of a 100-story building an egg can be dropped without breaking. The question is how many drops you need to make. You are allowed to break 2 eggs in the process.

Get ready for something different: a question for project manager applicants.

I really hope the problems with the clarity and grammar in these questions is a Business Insider problem, and not a Google problem.

This is another sorting algorithm kinda thingie. I'm less certain of this one, off the top of my head. I guess I'd start by dropping one from either the 13th or 14th floor, depending on which I figured would take less time to get to my goal (I'd have to do math for that, or guess based on my mood at the time, and I'm not doing much math today). Then, I'd do something like this:

Given N = (14 or 13) and n = N:
until N > 100,
set N = (current + (n -= 1))

Thus, if we assume N is 14, we drop first from the 14th, then from the 27th, then from the 39th, and so on, until the egg breaks or we get to 100 without it breaking. If it never breaks, our answer is 100. If it breaks, take a step back to the last floor where we dropped it and it didn't break and start going up one floor at a time until the second egg breaks. Intuitively, I'd say that with either 13 or 14 as the starting value we shouldn't end up with more than about 15 tries before we figure out which floor, and we'd end up breaking either two eggs total or no eggs. I think.

Explain a database in three sentences to your eight-year-old nephew.

Why does a product manager need to explain a database to his eight year old nephew?

"Say you have a bunch of checkerboards and a crapton of checkers, each of which has a different piece of information on it, and you store your checkers with their information in the squares of the checkerboards. Each board is sort of a categorical topic area, with each column and row corresponding to a value type so that when you combine the value types you get unique values that correspond to each piece of information. There, I did it in two."

Yes, any eight year old nephew of mine that would get my explanation of a database would be pretty smart. If he was dumb, I'd say "It's like a well-organized bucket that you fill up with things you know."

Note that nobody said it had to be a relational database, and I went beyond the call of duty assuming it was a technical implementation of a database as one uses with a DBMS anyway so that there'd be something to explain (otherwise it really is just a bucket of information).

I'm sorry. This is really just an incredibly stupid question. Maybe I should have used one sentence: "Here's the URL for the Wikipedia page for 'database'." Yes, a question that tests one's ability to explain technical things in simple terms can be very important, but there's not nearly enough context here to actually give an answer that will suit the needs of whatever scenario the interviewer has in his head.

You are shrunk to the height of a nickel and your mass is proportionally reduced so as to maintain your original density. You are then thrown into an empty glass blender. The blades will start moving in 60 seconds. What do you do?

What do questions like this have to do with being a product manager?

Considering the dimensions I'd have at that point, I might just try lying down and letting the blades whir over my head. I'd hold on tight for good measure. This is assuming I couldn't disable the blender, of course. This is a terrible question. Maybe they'd be happy with the answer "fire the guy who shrank me and put me in the blender, and get his replacement to get me out," since this question is so stupid.

The Rest of the Story

I looked at the provided answers after I wrote the rest of this SOB entry. The answer to the question about the guy pushing the car is an even stupider, more culturally dependent answer than the thing about dead beef, and if getting hired depends on the answer to that, there's something deeply fucking wrong in Google's hiring process. Seriously. What if I pushed my wookie, instead of my car? Maybe I'm playing the Star Wars Edition, numbnuts.

Before the list of fifteen questions from Google interviews that are supposed to make you feel stupid, the Business Insider article offered three criteria for getting hired at Google that have nothing to do with these questions:

• Google prefers Ivy Leaguers.
• It cares about your GPA, even if you're in your 30s.
• It wants people who want to change the world.

I'm okay with the last item in that list, but the first two are ridiculous. Seriously? This is another case of that tendency hiring managers and HR people have to believe that something that narrows the field is good even if it doesn't actually test for something particularly relevant to job performance. Sure, a higher GPA may correspond to a slightly higher tendency toward good job performance, but by throwing out anyone whose GPA sucked (or may not have attended college at all), you may be ignoring a completely different, more directly relevant metric that could have a much more important and reliable impact on your ability to hire the very best. Don't even get me started on the Ivy League school preference. Remember that correlation does not imply causation, and you'll avoid these asinine mistakes that major corporations all seem dead-set on making.

I'm also not sure the last item on the list has led to good things. How exactly do these people want to change the world? Judging by Google's performance, this requirement has resulted in better-than-average outcomes for the "don't be evil" motto, and for the corporation's success, but "better-than-average" for a major tech corporation isn't saying much when you're comparing it against examples like Microsoft and Oracle. If average is somewhere south of -85% on the Goodness Scale, I don't think a "better-than-average" result of -80% is a "good" outcome. Maybe Google should change that motto to "Don't be as evil as Microsoft." At least then it'll probably correlate more strongly with results.

A few weeks ago (three? four?), I put an end to a seven-year famine in my life. It has been seven years since I moved across the country and, as part of the move, sold my motorcycle to a friend (since I had no reasonable way to either transport or store it, and it wasn't terribly valuable). This month, I got a new motorcycle.

The new bike is a 2000 Buell Blast. That's the first year Buell made the Blast. Buell has been a subsidiary of Harley Davidson ever since HD bought the company in the '90s, until HD closed down its Buell division last year "to focus on the Harley Davidson brand." As you should be able to see from the pictures, a Buell is not a gigantic cruiser. Buell started out making racing bikes and their street-legal equivalents. The Blast was created for the express purpose of having a Harley Davidson built "starter bike" that was suitable to use in Harley Davidson motorcycle classes, so HD wouldn't have to buy little Honda 250s for its own bike classes.

Of course, Harley Davidson's idea of a "small starter bike" is somewhat different from what you'd expect out of a Japanese motorcycle company. The 2000 Blast is a 492cc single-cylinder engine. It's a zippy thing, too. Best of all for my purposes, it's about the closest thing you can get to a UJM these days.

Back in the '80s, the major Japanese motorcycle companies all made a class of street bike that came to be known colloquially as the UJM, or Universal Japanese Motorcycle. A UJM was smaller than a touring bike. It did not recline like a cruiser, with the foot pegs out in front. It did not require the rider to fold his legs under himself to reach foot pegs that were high up the sides and back somewhat. Instead, the rider would just sit upright on a UJM with the foot pegs a comfortable distance pretty much straight down below his hips. My first motorcycle was a UJM, and I found it quite comfortable and well-suited to riding around town.

The Blast is built kind of like an old UJM, in that it's not huge, it doesn't recline, and it doesn't "encourage" the rider to hunch forward over the tank all the time. Unlike the general run of UJMs (including my old XJ650 Midnight Maxim), the Blast doesn't have a transverse mounted four cylinder engine — it has that single-cylinder pounder instead. It also has a somewhat smaller engine than my 650 had (about 150cc smaller), but it feels a lot more zippy and powerful, probably because my first motorcycle was in terrible shape. It was, in fact, in such bad shape that it rapidly became apparent it could prove downright dangerous, and as quickly as I fixed stuff, other stuff broke. My favorite was when I was pulling up to an intersection, squeezed the hand brake on the right handlebar, and got brake fluid squirted out of the reservoir across my hand.

I found out that day that it's really difficult to stop a motorcycle using nothing but the engine and the soles of my boots. Thank goodness I had a really solid pair of sturdy motorcycle boots, and nobody drove through the space in the middle of the intersection where I managed to get the thing to stop.

The Blast, by contrast, is in excellent shape. It has been well-cared-for by owners who have loved and ridden motorcycles for decades, and used as an occasional fun ride, secondary to their primary motorcycles — cruisers (they were Harley guys). The man who sold it to me said he was getting too old and slow to handle this bike, and needed to sell it basically so he wouldn't be tempted to ride it sometimes. He's sticking to his touring-cruiser Harley now. He's an awesome old life-long biker with a very impressive white beard and an infectious, laid-back, happy attitude. I should give him a call and let him know how much I'm enjoying the bike; I'm sure he'd be happy to know it's in the hands of someone who appreciates it.

Life has been busy lately, which is basically why I haven't been doing much (any) writing here at SOB lately. The motorcycle is actually one of the things that makes it more busy, though definitely among the least busy-making of those things, and among the most fun. Soon, I'll be getting new boots for motorcycle riding (amazingly, I don't have much in the way of good biking boots right now), and a new motorcycle jacket (not so amazingly, my old leather jacket is falling the hell apart — a zipper actually came apart in my hand recently). The SigO is going to get a helmet, both for times when she might be a passenger and as part of her plan to get a motorcycle endorsement on her driver's license again (she lost hers moving to Colorado years ago).

Hopefully, within a year or so, she'll have a motorcycle too. I'll probably post pictures when the time comes.

note: Those pictures were from before I got the Blast. I haven't changed much, but the license plate has certainly changed.

More Details Here

In addition to being the primary name on TechRepublic's IT Security column, I have another (semi-)regular writing gig now. I'm not on an actual schedule at Troll in the Corner, but the idea is for me to write as close to once a week there as I can reasonably manage amidst the rest of the doings in my already overburdened schedule.

As of today, my first article at Troll in the Corner has been published: Finding Inspiration for NPC Concepts. Hopefully you'll like it as much as the enthusiastic first commenter did.

I have vague plans in mind about topics for other articles soon, including a stat block and description for a new critter, some stats and background for NPCs in support of a module, and maybe an overview of what niche Pathfinder RPG fills.

I've been kicking around the idea of starting up an RPG-focused Weblog of my own, but I don't think I'd get around to it any time soon, and might not maintain any kind of writing schedule at all, so a gig writing for someone else will probably prompt me to write more often and has already prompted me to start writing much sooner than I probably otherwise would have started writing. So . . . there.