dspeyer comments on Rationality Quotes July 2013 - Less Wrong

5 Post author: Vaniver 02 July 2013 04:21PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (425)

You are viewing a single comment's thread. Show more comments above.

Comment author: Pablo_Stafforini 01 July 2013 10:37:07PM *  5 points [-]

Einstein’s theory of relativity suggests that there is no fact of the matter as to when “now” is. Any measurement of time is relative to the perspective of an observer. In other words, if you are traveling very fast, the clocks of others are speeding up from your point of view. You will spend a few years in a spaceship but when you return to earth thousands or millions of years will have passed. Yet it seems odd, to say the least, to discount the well-being of people as their velocity increases. Should we pay less attention to the safety of our spacecraft, and thus the welfare of our astronauts, the faster those vehicles go? If, for instance, we sent off a spacecraft at near the velocity of light, the astronauts would return to earth, hardly aged, millions of years hence. Should we—because of positive discounting—not give them enough fuel to make a safe landing? And if you decline to condemn them to death, how are they different from other “residents” in the distant future?

Tyler Cowen, ‘Caring about the Distant Future: Why it Matters and What it Means’, University of Chicago Law Review, vol. 74, no. 1 (Winter, 2007), p. 10

Comment author: dspeyer 02 July 2013 03:07:18AM 16 points [-]

They are different because when we pack the spaceship with fuel, we control with reasonable certainty whether they make a safe landing or not. As for our millions-of-years descendants, it's very hard to make any statement about us effecting them with >51% confidence (except, "we shouldn't exterminate ourselves").

A lot of what looks like time discounting is really uncertainty discounting.

Comment author: Pablo_Stafforini 03 July 2013 06:24:36PM *  1 point [-]

A lot of what looks like time discounting is really uncertainty discounting.

Cowen is explicitly discussing time discounting. As he writes, "Should we—because of positive discounting—not give them enough fuel to make a safe landing?" (emphasis added) There may of course be other reasons for treating these people differently, including uncertainty about the long-term future, but Cowen is not focusing on these reasons here.

Comment author: fractalman 07 July 2013 07:32:39PM *  0 points [-]

It feels like a terrible example for examining the effects of relativity on utility functions regarding time-discounting; the typical human utility function is going to result in something that approximates Utility(fuel)=stepfunction(fuelpurchased-“100% fuel”) at around 99-100% fuel, regardless of time-discounting. It’s a case of [lands succesffully] versus [runs out of fuel 10 seconds too soon and crashes, killing everyone in the rocket.]

If you’re time discounting heavily enough to not notice that spike, and fuel is somehow the most expensive part of the whole operation, then you’re probably discounting heavily that you’re better off launching two rockets on one-way trips with about 25-50% fuel each, depending on specifics of the rocket.

-In other words, the example fails to probe to the real heart of the mater because it doesn't matter if i use an Einsteinian reference frame or a Newtonian one, my answer is the same: either 100% fuel or very little fuel.