From: Lee Corbin (email@example.com)
Date: Thu Mar 13 2008 - 08:43:33 MDT
> I think we may be talking about slightly different things. To restate
> the problem, we have two separate computers which are going to run
> Monday and Tuesday, respectively, of a person's life. If the final
> state of Monday is transferred to the second machine, by any reliable
> means, so that Tuesday can be computed, then my understanding is that
> you would agree that there is no break in the stream of consciousness
> of the person being simulated (assuming he is staying up late).
Correct---indeed I would so agree.
> On the other hand, if the run on the second computer starts with initial
> state not causally connected to the first computer, even if by chance
> it turns out that this starting state is in fact the one that would
> have occurred had the causal connection been present, then your claim
> is that there will be a discontinuity in consciousness between Monday
> and Tuesday.
Infinitesimal, but yes.
Now let's be *very* careful here. Your Sj is the last state of Monday
and Sm is the first state of Tuesday, where Sj -> Sm is the usual
calculation depending locally and reductionistically on the information
flow between Sj and Sm.
On the other hand, to just carry Sj over to Tuesday and *start*
Tuesday with Sj defeats the most important part of our TE.
Clearly, if Tuesday starts with Sj---no matter how obtained,
no consciousness not even for a pico second is interrupted.
> My problem with this latter position is that, apart from being an
> unnecessary complication, there is no way to specify what counts as a
> causal link and no obvious mechanism whereby the absence or presence
> of this link could affect the consciousness of the computation. For
> example, what happens if the transfer of information occurs by a
> friend passing you a number of files of which only one is the correct
> one. If you guess right does that count as information transfer?
What is on those files? If it is Sm on those files, well, where did
Sm come from? By chance? In that case, there was an infinitesimal
absence of consciousness in the person being emulated. But as I
said, if it is Sj, then that's no different than benign teleportation.
So am I right in thinking that your next question is either irrelevant
or not properly specified?
> What if your friend tells you which is the right file but you're not
> completely sure you believe him? What if you implement a series of
> computations using all the files, knowing that one of them has to be
> the right one?
> Stathis Papaioannou
This archive was generated by hypermail 2.1.5 : Mon May 20 2013 - 04:01:20 MDT