[extropy-chat] Just curious, cryonicist living life in reverse

Stathis Papaioannou stathisp at gmail.com
Fri Mar 2 23:57:01 UTC 2007


On 3/3/07, Mike Dougherty <msd001 at gmail.com> wrote:

On 3/2/07, Stathis Papaioannou <stathisp at gmail.com> wrote:
> > Suppose there are two programs: program A is my life on 1st March 2007
> and
> > program B is my life on 2nd March 2007. Granted, the programmer needs to
> > know all sorts of details about my past life before he can write the
> > programs, and in particular he has to know what is going into program A
> > before he can write program B, but we assume that he has done his job
> > properly. Now here I am, and it's the 2nd of March by my calendar, so it
> > must be program B that is running. I certainly remember yesterday as
> being
> > the 1st of March, but does this give me any information at all as to
> whether
> > program A was run yesterday, has not yet been run, is being run
> > simultaneously on a separate machine or process, or any details at all
> about
> > its implementation?
>
> Suppose your life/program is represented as a high dimension manifold,
> then it would be possible to compute a given moment by selecting a
> point this object's the surface.  The memory/state information would
> be relative to that point, but not necessarily require iterative
> calculation.
>
> I've thought about this using the frames of a film - each 1/24 of a
> second contains stateful information about it's placement in the
> sequence.  For a movie you know well enough, you could probably
> resequence a random pile of film clips from these clues (within some
> threshold of accuracy: 1/24 of a second doesn't allow a great deal of
> variation from frame to frame)


There is a crucial difference between an external observer resequencing and
an internal observer resequencing. The external observer could do it but it
would take a lot of knowledge and effort, and as you suggest the smaller the
time slices the less difference between them and the more difficult to place
them in order. The internal observer, on the other hand, does not have this
problem. No matter how small the time slices and how thoroughly shuffled,
subjectively OM2 cannot help but feel that it follows OM1 and precedes OM3.

One very thorough way of doing the shuffling is to have a computer generate
all programs via a Universal Dovetailer. If the computer runs long enough,
it will generate OM1, OM2 and OM3 and even though this is completely useless
for an external observer - they are hidden in the background randomness -
the internal observer will still experience OM1, OM2, OM3 occurring as if
arising in what we consider the usual manner.

One complication is that the UD will generate not only OM1, OM2, OM3 but
every possible variation. Thus OM1 could experience as next moment OM2.1,
OM2.2, OM2.3... each of which will have a distinct measure, or subjective
probability. The effect of this is that although the UD is perfectly
deterministic from the point of view of an external observer, from the point
of view of the internal observer his future is indeterminate. In form, this
matches the branchings in the many worlds interpretation of quantum
mechanics.

Stathis Papaioannou
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20070303/39c51128/attachment.html>


More information about the extropy-chat mailing list