[extropy-chat] Cryonics questions and uploading

Wed May 10 08:54:56 UTC 2006

On Tue, May 09, 2006 at 06:16:48PM -0400, Anne-Marie Taylor wrote:

> So the brain has mind instances (information, remembering
> feelings, experiences, remembered sensations..... in a
> moment in space and time...)

Object instances in computer lingo typically do not include
state, and state change over time. If you instantiate two
objects of the same class in two or more location without
synchronizing their state, they're different objects. An
object could make a decision based on its internal state
(which can be changed by an input), and it would be a different
decision than an object elsewhere, with a different
state did.

If you impose the constraint that two or more spatially
separated object have to be syncronized this is no longer 
possible. Notice that this condition does not allow
differing inputs to change the inner state. Whether you
you disallow different input (including nondeterministic
noise in the object itself) or force the state to be
the same (either by replicating the state of a master
object, or constraining the evolution trajectory of
each individual object so that it may not bifurcate)
is not relevant. OpenCroquet prevents divergence by
two mechanisms: by evolving the state using a deterministic
process, which is the same on all remote computers,
and by synchronizing differences in input. It can also
use a master approach to replicate state elsewhere 
(e.g. if you use a noisy system or a system which
is speedier than any other, so everybody saves work
and you don't have to wait until the slowest member
catches up before allowing all worlds to move on).

In any case two or more objects are incapable of making
a different decision. They are all the same object,
existing in two or more places.

> And the brain is the matter that directs all these
> mind instances.

The brain makes no sharp distinction between state and
structure implementing it. You could run all kinds of objects
on an all-purpose computer. You can only run one person
on a specific brain. 

With uploading you use one all-purpose substrate for
the computation, so the hardware is the same for all
systems. What is different is the state of emulated
animal objects. Notice that the hardware has almost
no complexity, it is entirely in the object state
and the transformation function of the object state
over time.

>   Am I understanding this properly?
>    
>   Cryonic reanimation may occur in the futur by 
> "Waking up the mind" from a frozen state.

Theoretically, you could repair the damage (you were 
dead, after all) and the suspension artifacts in the one
object, remove the cryoprotectants and rewarm it so
that the CNS activity constituting a person would 
resume spontaneously.

If you make copies of frozen bodies, they start to
diverge by the moment they resume activity. They
have become two distinct, albeit very similiar 
persona. This is very similiar to identical twins:
they start pretty close (though not as close as
an exact copy), and diverge further during
life.

>   Does cryonics also consider the posibility of 
> removing the brain matter and transfering to a new
> body? (Brain transplant).

It is one theoretical possibility.

>   >I understand that technology is no where near this posibility just curious to know 
> >if it's still as feasible as the idea of "waking up the mind". 
>    
>   Uploading is taking mind instances and transfering
> them to a computer. 

Uploading is making numerical models of animals, including
body and environment. You can think of it as a video game
(with game AI controlling non-player characters) on steroids.

>   I'm assuming this is what i'm doing right now.  Typing
> and transfering into a computer. I would assume, in the

You're not transferring your state into your machine.
Well, yes, a little, by serializing some of your inner state,
which is being interpreted by a human elsewhere because
the coding is sufficiently accurate for it to build a
very primitive, abstract model of your inner state.
You're using the computer as a communication channel.

> futur, this would be done very differently.
> What is the general opinion on how this process 
> will occur?
> Won't this only just create a super-computer?

It would require a very large (humongous, by current standards)
supercomputer. But that supercomputer would contain a person,
which subjectively sees something very different (it sees whatever
the world and body model are faking).

>   Again, just want to be sure i'm grasping the basics.  

It is really not very complicated.

> Thanks for the replies

-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820            http://www.ativel.com
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 191 bytes
Desc: Digital signature
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20060510/450dde14/attachment.bin>