[ChuckMcM]

This is a fun math problem to use as an interview question, it can expose how people think.

The interesting trick is to assume 75% of the earth is covered in ocean, And sea levels are expected to rise 20 meters (or more). Since you can meet "half way" between the rise in the sea level vs the lowering of the sea floor, you remove 10M of sea floor, and re-distribute it over the 25% of the "land", that means that all of the land goes up by 10M/.25 (the ratio of land to sea) or 40 meters.

So burying the continents under 40M of sand would counteract a 20M rise in sea level, sort of :-)

[Retric]

Moving that much of the crust around would likely to create fairly massive earthquakes and the surface of the earth is effectively floating so you are limited on how much you can change things. Still if you’re willing to move a lot of rock you could more or less get it to work.

But, your solution would raise the land and lower the sea levels at the same time. You want the difference (added land) + (lower ocean) = 20M. (.25 land : .75 ocean = 1:3) Removing 5M of sea floor would add 5 * 3 = 15M of land 5 + 15 = 20.

Of course there is little need to raise all land just the low lowing areas would be enough. Or you could also make new land vs messing with any of the existing land.

[ChuckMcM]

Oh, so you want to go for bonus rounds!

Bonus question #1: If sea level rise is an exponential with respect to time, at what rate would we have to extract sea floor material to keep the level at a current coastline constant?

Bonus question #2: What is the minimum sea wall you would need along the coast of all continents in order to hold back sea level rise?

The point being such questions are always useful for eliciting the interviewee's response who questions they might consider to be "unknowable" when they start the interview.

[Retric]

I have no idea why people like to use HS math problems for interviews.

#1 If the ratio is A:B, then you need to remove A/(A+B) * f(x). So, using your numbers it's 1/(1+3) * f(x) = 1/4 f(x). #2 = rise + safety factor. But, safety factor is going to be some BS number. Area wise it's got to deal with rivers not just coast lines, but for a minimum wall you could in theory just use pumps and ridiculous amounts of energy.

Honestly, Google actually tracked this stuff and these types of problems are very poor indicators of workplace performance.

PS: For more fun with exponential functions grub starts a reaction that produces 1 atom the first day 2 atoms the second day... 2 ^n. Now your blast away from the area in a space ship at .9c, approximately how long do you get to live? Hint the earth has around 10^50 atoms it's radios is ~4,000 miles and the speed of light is 1.6 * 10^10 miles per day.

[ChuckMcM]

   > I have no idea why people like to use HS math problems
   > for interviews.

Did you go through Google's hiring class? I did. There is a lot going on in an interview, and much of it has to do with how people respond to stressful situations. That is especially true of people interviewing at a place like Google because they often feel like it is their "dream job" and if they blow it they will end up homeless or something.

The important thing about interview success is getting an understanding of three things, can the person listen, can they reason under stress, and can they get stuff done. The candidates always over think the questions, it is the nature of things. Locking someone up mentally because you've over stressed them a question on tensor physics really doesn't help determine if they would be a good candidate or not. If you start with the HS math and they blow through it, you can work your way up to multi-variable calculus :-)