The space of Jupiter-brains is indescribably immensely larger than that of human brains. A human brain almost certainly has fewer than 10^10^20 distinguishable configurations, while a Jupiter-brain probably has more than 10^10^40 configurations. Don’t let the fact that the number at the top is only twice the other fool you, these numbers are incredibly different in magnitude.
The fraction of Jupiter-brain configurations delayed due to exploring the space of all possible human brains first is less than 0.000000...(10^40 zeroes)...0001%. So don’t worry about the waste of computing power.
Of course, both are utterly impractical to explore. With a computer the size of the observable universe that explores 10^100 configurations per second, it will take 10^10^20 seconds to explore just the human brain configurations. It will also take 10^10^20 years, because on these scales the change in the exponent between seconds and years is so small that it would require more decimal places than is worth typing. For that matter, the ratio in computing power between a universe-sized supercomputer and an abacus is likewise irrelevant.
People saying things like “just simulate all possible brains” are making implicit assumptions that can’t be supported in anything we’ve observed of the universe.
Because the number are so ridiculously huge that there is essentially no difference between seconds and years.
More precisely, I could have said “it will take 10^10^19.999999999999999999 seconds”, and “it will take 10^10^19.99999999999999999991″ years, respectively. Both would be wrong, because that communicates a degree of precision that isn’t present.
It’s really more like “it would take between 10^10^15 and 10^10^20 seconds”, because I don’t know the exact size of the reasonable state space for human brains. If I convert to years, it’s still “it would take between 10^10^15 and 10^10^20 years”.
People saying things like “just simulate all possible brains” are making implicit assumptions that can’t be supported in anything we’ve observed of the universe.
Sure, from our current (very limited) understanding of physics, simulating all possible brains seems to be impossible. But should we assume it will forever remain impossible?
It would take more than 10^10^20 seconds to find prime factors of some large integers. But after we’ve discovered one weird trick, we can (theoretically) do it many-orders-of-magnitude faster. Maybe there are similar tricks for searching in the space of all possible minds.
Judging by the history, the Clarke’s first law is more fundamental than any law of physics:
When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong.
Fromthe linked article:
On 29 December 1934, Albert Einstein was quoted in the Pittsburgh Post-Gazette as saying, “There is not the slightest indication that [nuclear energy] will ever be obtainable. It would mean that the atom would have to be shattered at will.” This followed the discovery that year by Enrico Fermi that if you bombard uranium with neutrons, the uranium atoms split up into lighter elements, releasing energy.
I should’ve been clearer, by “exploring the space of jupiter brains” I meant “making a single large jupiter brain”. In general I think computational ressources are better spent to build a few centralised very large brains than to simulate a very large number of small ones. I agree with everything you’ve said.
The space of Jupiter-brains is
indescribablyimmensely larger than that of human brains. A human brain almost certainly has fewer than 10^10^20 distinguishable configurations, while a Jupiter-brain probably has more than 10^10^40 configurations. Don’t let the fact that the number at the top is only twice the other fool you, these numbers are incredibly different in magnitude.The fraction of Jupiter-brain configurations delayed due to exploring the space of all possible human brains first is less than 0.000000...(10^40 zeroes)...0001%. So don’t worry about the waste of computing power.
Of course, both are utterly impractical to explore. With a computer the size of the observable universe that explores 10^100 configurations per second, it will take 10^10^20 seconds to explore just the human brain configurations. It will also take 10^10^20 years, because on these scales the change in the exponent between seconds and years is so small that it would require more decimal places than is worth typing. For that matter, the ratio in computing power between a universe-sized supercomputer and an abacus is likewise irrelevant.
People saying things like “just simulate all possible brains” are making implicit assumptions that can’t be supported in anything we’ve observed of the universe.
How can it take both seconds, and years, at the same time?
Because the number are so ridiculously huge that there is essentially no difference between seconds and years.
More precisely, I could have said “it will take 10^10^19.999999999999999999 seconds”, and “it will take 10^10^19.99999999999999999991″ years, respectively. Both would be wrong, because that communicates a degree of precision that isn’t present.
It’s really more like “it would take between 10^10^15 and 10^10^20 seconds”, because I don’t know the exact size of the reasonable state space for human brains. If I convert to years, it’s still “it would take between 10^10^15 and 10^10^20 years”.
Sure, from our current (very limited) understanding of physics, simulating all possible brains seems to be impossible. But should we assume it will forever remain impossible?
It would take more than 10^10^20 seconds to find prime factors of some large integers. But after we’ve discovered one weird trick, we can (theoretically) do it many-orders-of-magnitude faster. Maybe there are similar tricks for searching in the space of all possible minds.
Judging by the history, the Clarke’s first law is more fundamental than any law of physics:
When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong.
From the linked article:
I should’ve been clearer, by “exploring the space of jupiter brains” I meant “making a single large jupiter brain”. In general I think computational ressources are better spent to build a few centralised very large brains than to simulate a very large number of small ones. I agree with everything you’ve said.