[extropy-chat] The Simulation Argument (was: Atheists launchinquisition...)

[Hal Finney]

Kevin Freels writes:
> Eugen said: "Jeez. People, this isn't philosophy. Technology has
> constraints, and in
> evolutionary scenarios, costs. This Pearly Gates here are *expensive*."
>
> That was my point. Which is why I think that the simulation argument is
> wrong.

No, you are not disagreeing with the SA if you make this point.  You are
agreeing with it.  You are agreeing that one of the three clauses is true,
namely, the one which says that posthumans will not run many simulations.
This is exactly what the SA says, that one of the three clauses should
be true.

I suspect that I am being pedantic here, and that what you really mean to
disagree with is an extended version of the SA.  This version takes the SA
and adds the assumptions that we will become posthuman, and that posthumans
will run many simulations, from which it follows that we probably live
in a simulation.  You are disagreeing with one of the assumptions of
this extended SA, namely that posthumans will run many simulations.

Maybe we could call this SA+ to distinguish it from the SA.  I think
most people who complain about the SA or say they disagree with it are
actually disagreeing with the SA+.

As far as Eugen's point about costs, this is discussed in Nick's paper
in some detail.  He looks at various estimates of how much it takes to
simulate a brain, and compares it with people's guesses about how much
compute power a future posthuman civilization will have.  Add a bit of
hand-waving and it looks like simulations would be cheap compared to the
overall capabilities of such a civilization.  On the other hand there
might well be more useful things they could do with that computing power,
so it's hard to say what they will decide.

Hal