Far away from home

A race of very tiny creatures live on a perfectly spherical planet of radius 2 m 2 \text{ m} . Due to an unknown aberration in space-time, all the creatures are teleported to a random point on the sphere. What is the distance the average creature has to walk to get back home?

Inspired by erratic systematic wormhole mathematics .


The answer is 3.14.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

2 solutions

Julian Poon
Jul 26, 2015

There has been something bugging me about this question, is it right to assume a 2 D 2D circle for this problem?

Here is an illustration of the cases considered when assuming a 2D circle, as used in the other solutions:

As you can see, the cases are not evenly spaced. By right, there should be a higher possibility of teleporting a distance of π \pi as compared to a distance of 2 π 2\pi .

However, there is no doubt the answer is π \pi due to symmetry, I even verified it with a little calculus to make sure:

0 2 π 8 π x sin 2 ( x 2 ) 0 2 π 8 π sin 2 ( x 2 ) = π \frac { \int _{ 0 }^{ 2\pi }{ 8\pi x\sin ^{ 2 }{ \left( \frac { x }{ 2 } \right) } } }{ \int _{ 0 }^{ 2\pi }{ 8\pi \sin ^{ 2 }{ \left( \frac { x }{ 2 } \right) } } } =\pi

Moderator note:

Right, the symmetry argument is that "It is equally likely for the creature to be teleported by π + k \pi + k , as it is to be teleported by π k \pi - k . Hence, the expected distance would be π \pi ."

You are right in saying that the cases are not evenly spaced. There is a higher possibility of teleporting a distance of π \pi , than any other distance.

Calvin Lin Staff - 5 years, 10 months ago

Monte-carlo also yields π \pi .

Thaddeus Abiy - 5 years, 10 months ago

Log in to reply

Yeah, I agree the answer is π \pi , but the method of visualising it as 2D is incorrect. Its a great question though.

Julian Poon - 5 years, 10 months ago

Hmm... my argument is this:

Every great circle that can be drawn on the sphere has an equal chance of being chosen. For each of these great circles, the arc lengths between 0 0 and 2 π 2\pi along half of the great circle have an equal chance of being chosen. This can lead us to believe that the average distance along the arc on the half circle (and thus every circle that can be drawn) would be π \pi .

You're probably right that it's a bad explanation, I just don't get why is all. Maybe it has to do with the initial assumption that each great circle has an equal chance of being chosen?

Brock Brown - 5 years, 10 months ago

Log in to reply

I'm not sure about this, but I think that assumption depends on how the Great Circles are arranged, and from my interpretation of the solutions here (Which might be wrong), the great circles are arranged like the picture above.

Julian Poon - 5 years, 10 months ago
Brock Brown
Jul 24, 2015

EDIT: This solution is dumb and bad, please ignore it.

Let d d be the expected distance someone will have to walk home.

Since our creatures will be walking straight home, this can be reduced to a two-dimensional problem.

The shortest possible distance is if the creature teleports right back into its house ( 0 m 0\ m ). The longest possible distance is if the creature teleports to the other side of the planet ( 2 π m 2\pi\ m ). Since the possible outcomes are equally spaced apart (by ϵ \epsilon ) and the chances are uniform, we can find the answer with the average of the maximum distance and the minimum distance:

d = m a x + m i n 2 = 2 π + 0 2 = π d = \frac{max + min}{2} = \frac{2\pi+ 0}{2} = \boxed{\pi}

Moderator note:

This solution has an incorrect interpretation of restricted probability.

I disagree with "the possible outcomes are equally spaced apart so the chances are uniform". It is much more likely that the creature will be displaced by π ± ϵ \pi \pm \epsilon , then by 0 ± ϵ 0 \pm \epsilon .

Calvin Lin Staff - 5 years, 10 months ago

Log in to reply

Oh, never mind, I get it, because the surface area on the sphere for all points that are just barely far away from the initial warping point is much smaller than the surface area on the sphere for all points that are barely away from the equator. I feel kinda dumb.

Brock Brown - 5 years, 10 months ago

Log in to reply

That's right. I elaborated further in Garett's note.

Don't feel dumb. This is similar to Bertrand's Paradox , where confusion over uniform probability is the root cause.

Calvin Lin Staff - 5 years, 10 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...