From: Anders Sandberg (asa@nada.kth.se)
Date: Thu May 23 2002 - 04:53:13 MDT
On Wed, May 22, 2002 at 01:46:52AM -0700, Samantha Atkins wrote:
>
>
> Hal Finney wrote:
>
> >It's true, with a randomizer there is no loop, but there are still only a
> >finite number of states (fewer than there would have been because you're
> >using some of your resources to run your randomizer!). So now you can
> >jump around pre-explored states at random instead of running through
> >them sequentially. But you will have no memory of your trajectory.
>
> You are assuming that your own state does not change by your
> experiences, finite or not. Even if you repeated an experience,
> you are not the same on subsequent iterations. You really can't
> step into the same river twice.
Actually, when we speak about "state" here we really mean all that there
is to us - not just our current thoughts and experience, but also our
collected memories and the body state. This kind of recurrence implies
that at some point you will have to start ending up in situations you
have once been in, but you will have no memories of the earlier
occurences to change how you react to them. It is a fundamental property
of finite systems.
I think Damien has a point about this being nearly an angels on a pin
discussion; it is not an acute problem and it is rather academic. But
I'm an academic, I simply cannot let academic problems slip me by! :-)
I read the original paper on the bus this morning, and found it a nice
modernization of Dyson's old paper. It feels good to see his thoughts
dusted and cleaned a bit. Their point about quantum computing was good
(it doesn't change anything in the really long run, but can give
exponential increases in subjective lifetime) and seems to fit with my
ponderings on nearly reversible computations in my jupiter brain paper.
The wormhole speculation near the end has some interesting implications.
If we can manage to expand wormholes, then either they will be expanded
Planckscale holes that may lead anywhere in the universe, or
manufactured holes whose ends will have to be moved apart at sublight
speed. If the first possibility is true, then the amount of mass/energy
available will grow essentially arbitrarily fast and we don't need to
worry about horizons. Yay! If the second holds true (or if expanding
pre-existing quantum holes leads to local holes), then it seems that the
amount of mass-energy will grow only as the square of time - but given
the time-machine effects of relativistic holes, the subjective amount of
available mass-energy grows exponentially. Very promising, but we better
get those wormholes to work first.
Just one things bothers me: why is it called Cardassian expansion?
-- ----------------------------------------------------------------------- Anders Sandberg Towards Ascension! asa@nada.kth.se http://www.nada.kth.se/~asa/ GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:14:17 MST