= 27d567bc2d48d9f40e9ccc20ea370b15)
Eliezer_Yudkowsky comments on Non-Malthusian Scenarios - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (88)
Population doubling takes place in alternative branches of a coherent quantum superposition, so that each single branch has plenty of capital. Parents don't get to talk to their children much, but everyone maintains a higher standard of living - in exchange, perhaps, for lower measure.
You could take the current splitting as showing there is already a vast rate of population increase. However, the analysis of per-capita income is robust to splitting or joining individuals across quantum branches. If people can grow kids within a branch faster than they can grow wealth within a branch, incomes fall regardless of how fast the number of branches grow.
I don't believe in MWI, but I try to understand it, and I can't get this to make sense. Reproduction is made conditional on some quantum event, so that it happens in some futures and not in others - OK. These futures are supposed to be in coherent superposition - not sure what the point of that stipulation is; is the idea that the separate futures are given a chance to recombine at some point?? But they can't recombine unless they come to resemble each other, which is a bit of a problem when one future contains lots of people who don't exist in the other one! And finally, doesn't this situation result in a Malthusian scenario in the branch which does reproduce??
I think you're misunderstanding an important part.
I think the idea is that you flip a quantum coin, and if it's heads you do nothing, otherwise you replace yourself with your child. So the population of each branch stays the same.
But the coherent superposition part doesn't make sense to me either. And the scheme also doesn't say what prevents people from reproducing the normal way and driving up the per-branch population.
This would be accomplished using quantum suicide to edit proportions of the population out of individual world lines. At each juncture 50%, say, of the population would be edited out of each world line. You aren't gambling on which world line gets to reproduce, you're putting half of the population into each one.
Correct; and the point of specifying coherent quantum superposition, as would exist in a large-scale quantum computer, is that so that both branches "go on existing" - which I think they do anyway, but I wasn't sure about the quantum suicide aspect - actually, the more I think about this, the more I start to think that quantum suicide ought to work, since could you really say someone was 50% dead if they lived on a still-coherent quantum computer and only existed in half the branches? Maybe you could. I don't understand Born probabilities or anthropics.
Your original proposal is extremely close to quantum suicide; in particular the phrase "at the expense of smaller measure" is the key assumption of quantum suicide. This is, as far as I can tell, the rejection of the Born measure. But we observe the Born measure and our normal decision theory values it as probability. To reject Born measure is to say that it does not "all add up to normality."
Making reproduction conditional on some quantum event sounds like a reproduction lottery to me. It reduces the birth rate, but the main difference from other means of population control is that the decision about who has kids is made randomly - resulting in an eventual decline in genetic quality of the population.