Aaron N. Tubbs bio photo

Aaron N. Tubbs

Dragon chaser.

Twitter Facebook Google+ LinkedIn Github

I know the first two parts of the series were a little dull, so it’s time to look at a different species of interview question. Back when I was graduating from college, the industry was still sold on the idea that interviewing wasn’t about testing trivia or evaluating whether or not a candidate could write code. Who cares if they understand data structures and algorithms. What really matters is whether somebody can think outside the box!

That leads us to today’s interview question, carefully rendered by myself using a Uni-Ball Vision Elite 0.8mm Bold pen. It won’t leak in flight. And it helps prevent check fraud. This entry isn’t actually sponsored by this pen or anything, but … this pen is great. It’s bold. And it doesn’t leak on a plane. I’ve worked with less-than-bold pens, and don’t care to repeat the experience. Pens that leak in flight are a much bigger deal, and I’ve lost shirts to this problem. You want this pen. As to the check fraud thing, I’m going to call bullshit, because the nature of the pen leads to not pushing particularly hard, so there’s no transfer to the carbon sheets in your checkbook.

I don’t have any clue why it costs a few bucks to buy one, and why I go through them like candy, but I’m hooked and you should be too, because these pens are great.

Anyway, the red details are via a Sharpie, but that’s much less interesting, so let’s get to the point. Here’s the problem:

Stated in words, you’re on an island that’s thirteen miles long, and there’s an easterly wind at two miles per hour. The island is narrow and long, running east to west. You’re a mile west of the west end of the island, and the entire western mile of the island is on fire. The remaining 12 miles east of you is not yet on fire, but it’s going to be. The island is covered in light brush that will burn and kill you, but the brush is passable. You’re in great shape, and can run at five miles an hour for as long as you want. Unfortunately, you’re allergic to burning to death, so you’d rather not end up on fire.

Your Task

Live.

For the astute, this problem description is not identical to the illustration. That’s because the problem gets made up as one describes the problem, because the whole point of the problem is for it to explore outside-the-box-thinking. This means that the interviewer has to be able to bullshit for a while without breaking a sweat (a good skill, anyway). If you’re not, just resort to orbital lasers and tell the candidate to shut up. It works like this “No, you can’t do that, you’ll be shot by lasers from space.”

A Massive Quantity of Suck Pulled Through Your Brain Spigot Into Your Mind Hole

So, this is all about the candidate asking creative questions and providing means of breaking the problem, and the interviewer clarifying the world to mitigate the solution thus far. For example, if the candidate says “whatever, I’d jump in the water, next problem, softball interviewer dude,” then it’s time to resort to something like this:

That is, the water is really hydrofluoric acid, with huge monsters swimming in it (they’re made of Teflon, and impervious to the acid). Also, the island is a mile high, and there are razor-sharp spikes just off the coast.

So, the candidate wants to clear some of the brush away, and sit safely. This won’t work, because the brush is a mile deep, and the island is not an island, it’s a giant slab of concrete.

“Aha!” says the candidate, then it won’t really burn that fast!

Just kidding, it’s a concrete island with a mile of magnesium filament mixed with all sorts of other nasty stuff that makes it burn very hot and bright and fast.

What about the fumes, then?

Orbital death ray, kid. Orbital death ray.

There are thousands of random ideas, some of them clever, some of them not, and it’s an entertaining little exercise. All sorts of thinking outside the box is possible. It’s actually fun, but it’s completely pointless.

Whatever, you get the idea, here’s the spoiler

The point you’re supposed to get to is that you take some brush, run east, light the island on fire, let it burn out, sit in the burned out zone, and let the fire blow over. Of course, the fumes, downwind smoke, and newfound lack of any vegetation (even if it’s toxic fast-burning razor-sharp hedgehog-infested vegetation) means starving to death is the only option, but at least it’s not death by burning or via dissolving in an acid bath.

What’s the point?

The point of this is that there is no point. A younger me believed this question had merit. That there was some magic way to measure aptitude by seeing how somebody thinks about a puzzle on the fly. There are two take-home lessons here. The first is that youth is a motherfucker. Remember that, if you forget everything else. Everything you know is wrong. Continually. The second is that there is virtually no way to measure aptitude, beyond giving somebody some tools, and seeing if they apply them iteratively when faced with a new problem that can leverage them, and even then it’s a complete crapshoot.

This question and all questions like it are crap. Don’t use them. Programming questions should be challenging and require an understanding of computer science and applied programming skill. This covers neither of those areas. Here’s a heuristic: If your question makes sense to your non-programmer friends, and they can solve it, it’s a bad question. The general populace is not cut out to do what you need your newest engineer to do.

Lateral thinking puzzles and thinking outside the box is a nice exercise that tells the interviewer nothing. Fun at the bar on a napkin. Pointless when you’re trying to ship version 1.0.

Pros

  • Possibly a great way to waste time, if you’ve got nothing better to do.

Cons

  • Tells the interviewer nothing.
  • No code to write for this one. If you can’t write code, it’s not a good interview question.

In Conclusion

Orbital death rays.