[ExI] Meat v. Machine
jonkc at bellsouth.net
Thu Dec 30 21:34:29 UTC 2010
On Dec 30, 2010, at 11:33 AM, Anders Sandberg wrote:
> letting stars shine only lose you ~1% of their mass-energy. If you are really advanced and long-term, you will burn the matter in black holes achieving ~50% efficiency in a very cold future where every Joule is worth many more bits. So if stopping stars takes a significant effort (which seems likely, you have to star lift them and then redistribute sizeable chunks of matter), then it might not be worth it.
I don't see why harvesting photons now would prevent Jupiter Brains from making black holes in the very distant future if they wanted to, and yes living a trillion years would take a significant effort but I think it would be worth it and a jupiter brain might think so too.
> it seems that if you have a huge time horizon then the current waste might not matter much,
But Jupiter brains are smart and I don't think its smart to have a HUGE time horizon because if you look too far in the future you probably don't really understand the problem you're trying to solve and you most certainly won't know what new tools you will have in your toolkit by the time the problem becomes serious. It would be like the Wright brothers figuring out that air traffic control would someday be a problem and refusing to continue development of their first airplane until, totally ignorant of the existence of Radar, they found a solution.
And I think even a Jupiter Brain would value a lollypop right now more than a lollypop in one hour, especially a Jupiter brain actually given how much slower time would pass for it due to its much faster mind.
> you get 1/kTln(2) bits of information out of a Joule of energy, so if you wait until the universe is twice as cold you get twice as much computation.
You may be able to perform more computations per unit of energy but wouldn't they be slower? It's subjective not objective time that would be the concern and the faster the mind the slower the time.
> It seems likely that Jupiter brains will have a pretty good theory of the universe after a mere million years or so.
A pretty good theory of the universe may not be good enough to accurately predict the future a trillion years from now, you'd need an ultimate theory but it might be like the layers of a infinite onion and there may not be an ultimate theory of physics; but even if there is such a theory and even if the Jupiter Brains know it they will almost certainly not be able to PROVE it is the ultimate theory, they can never be certain that something unexpected won't come up in a trillion years telling them they should wait another hundred thousand million billion trillion years before they try to do anything. It's like the man who ends up never buying a computer because he figures he can buy a better one for less money in just a few months.
I know that Freeman Dyson wrote a paper on life enduring indefinitely in the universe, but that was before it was discovered that the universe is not only expanding but accelerating; it would be rather embarrassing if a major new fact like that cropped up making hash of your careful plans and showing that a trillion years of living in poverty was pointless. Besides, having a good physical theory is only half of what you need to make predictions, you also need to know initial conditions.
> It is not clear to me that there is any reason to think that they will get a huge physics surprise if they spend another billion years on the problem.
Even if they know exactly what the universe will be like in a trillion years they won't know what they themselves will be like then. Problems that seem of cosmic importance now may have trivial solutions by then and its the problem we cannot now even conceive of much less answer that will be the cause of big trouble. Before I committed myself to a trillion years of sacrifice I'd want to be damn sure it was worth it, and with all these uncertainties I don't see how you could be.
John K Clark
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the extropy-chat