Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

J_Thomas comments on 9/26 is Petrov Day - Less Wrong

101 Post author: Eliezer_Yudkowsky 26 September 2007 04:14PM

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

Comments (66)

Sort By: Old

You are viewing a single comment's thread.

Comment author: J_Thomas 29 September 2007 09:46:33PM 3 points [-]

A full-scale thermonuclear war would have had lots of unintended consequences. We are not in a good position to decide whether everybody would be killed. To help us decide that we have a sample of size zero.

Predicting complex systems, very often we are wrong. In as simple a question as whether an airliner accident would cause WTC to collapse, everybody who tried to predict it got it wrong.

I would guess that a sheepherder in paraguay who has never seen many effects from the industrial revolution ought to be minimally affected. He shouldn't get much radiation unless somebody intentionally uses cobalt bombs or such. He should be able to live just fine without new steel knives and such. I guess there would be no subtle interactions that would kill him, and I have no basis for that guess except my lack of imagination.

We could certainly design a nuclear war to kill everybody. It's called a doomsday device, and it would be possible to make one that was fairly certain to be effective. The question of just how small a full-scale thermonuclear war would have to be to not kill everybody, is not something we can estimate with any degree of certainty without repeating the experiment enough times to get a decent sample size.

Why would we think we know the answer to this question? If we could reasonably predict whether a full-scale nuclear war would kill everybody, why can't we design a warplane from scratch and build it from untested plans and have it fly correctly the first time?

Comment author: homunq 14 November 2011 04:18:07PM 0 points [-]

It's hard to say that everyone got WTC collapse wrong. Obviously, the people who flew those airliners had some sort of expectation in their head about what would happen, and it may well have been relatively accurate.

Comment author: gwern 14 November 2011 04:48:06PM 3 points [-]

it may well have been relatively accurate.

Depends on what relatively accurate is. Bin Laden was off by about 2 or 3 orders of magnitude, depending on how you count floors and number of collapsed buildings: http://www.gwern.net/Terrorism%20is%20not%20Effective#fn4

Comment author: Desrtopa 14 November 2011 04:29:06PM 1 point [-]

I would guess that a sheepherder in paraguay who has never seen many effects from the industrial revolution ought to be minimally affected. He shouldn't get much radiation unless somebody intentionally uses cobalt bombs or such. He should be able to live just fine without new steel knives and such. I guess there would be no subtle interactions that would kill him, and I have no basis for that guess except my lack of imagination.

Unless he's part of a subsistence community (and even in underdeveloped countries most agrarian workers are not,) he's still economically tied to industrialized civilization. He's still directly or indirectly dependent on a consumer market that's now gone.