Le 20-juin-06, à 08:47, Hal Finney a écrit :
> I'll offer my thoughts on first-person indeterminacy. This is based
> on Wei Dai's framework which I have called UD+ASSA.
I guess you mean your UDist here.
> I am working on
> some web pages to summarize the various conclusions I have drawn from
> this framework. (Actually, here I am going to in effect use the SSA
> rather than the ASSA, i.e. I will work not with observer-moments but
> with entire observer lifetimes. But the same principles apply.)
Why? How? This is not obvious, and in some context even not consistent.
> Let us consider Bruno's example where you are annihilated in Brussels
> and then copies of your final state are materialized in Washington and
> Moscow, and allowed to continue to run. What can we say about your
> subjective first-person expectations in this experiment?
> Here is how I would approach the problem. It is a very straightforward
> computational procedure (in principle).
I don't think so. See below.
> Consider any hypothetical
> subjective, first person stream of consciousness. This would basically
> be a record of the thoughts and experiences of a hypothetical observer.
> Let us assume that this can be written and recorded in some form.
OK. I would add that such a record is annihilated in personal use of
teleportation. If not you can no more distinguish it from a third
> Perhaps it is a record of neural firing patterns over the course of the
> observer's lifetime,
This hides some ambiguity. " neural firing pattern" does not typically
belongs to first person experience (unlike pain pleasure ...).
> or perhaps a more compressed description based on
> such information.
? (such a compression has nothing to do with the experience, so I don't
> The question I would aim to answer is this: for any proposed,
> first-person lifetime stream of consciousness, how much measure does
> this hypothetical subjective lifetime acquire from the third-person
> events in the universe?
Which universe? And how do you link first person experience and third
person description. You are talking like if the mind body problem was
> The answer is very simple: it is the conditional Kolmogorov measure of
> the subjective lifetime record, given the universe as input.
(Conditional) Kolmogorov measure is not computable (even in principle)
unless you have an oracle for the halting problem. It is well defined,
and does not depend on the chosen Universal machine (except for a
constant), but it is not computable. Still it is computable "in the
limit", but for using that feature, you need to take into account the
invariance for delays in "reconstitutions", but this changes the whole
frame of reasoning.
I am not sure I can figure out what do you mean by "universe" here, nor
how could a universe be an input (nor an output actually).
> In other
> words, consider the shortest program which, given the universe as
> produces that precise subjective lifetime record as output;
There is no algorithm for finding that shortest programs. (part of the
price of Church thesis like in the diagonalization posts).
> if the length
> of that program is L, then this universe contributes 1/2^L to the
> of that subjective lifetime.
> Note that I am not trying to start from the universe and decide what
> first-person stream of consciousness is; rather, I compute the
> degree to which the universe instantiates any first-person stream of
> consciousness. However, this does in effect answer the first question,
> since we can consider all possible streams of consciousness, and
> determine which one(s) the universe mostly adds measure to.
Only with an oracle for the halting problem. Still, I don't understand
what you mean (in this context) by "this universe".
Also, minimal complexity is a prior akin to ... classical physics. Even
quantum mechanician does not really need it, because they can derive
such minimization from the quantum phase randomization, as Feynman
discovered (see a concise explanation of this in my 2001 paper
"Computation, Consciousness and the Quantum" (cf url below)).
> These would
> be the ones that we would informally say that the universe
> Now, let me illustrate how this would be applied to the situation in
> question, and some other thought experiments. Specifically, let us
> imagine three hypothetical streams of consciousness: B goes through
> until the moment the subject is annihilated in Brussels, then stops.
> W goes through life as does B but continues with the life experiences
> from Washington. And M is like W, going through life until the event
> in Brussels but then continuing with the events in Moscow.
> Normally we only consider first-person experiences like M and W when
> we discuss this experiment, where the consciousness "jumps" to Moscow
> or Washington respectively, but it is also useful to consider B, which
> corresponds to dying in Brussels.
Dying? (so you reintroduce "cul-de-sac" assumptions like if a person
could be aware of not being aware, so here you clearly reject comp or
its consequences. But this is problematic given that you are using
Kolmogorov measure, which although not computational, is a
> Let me first deal with a trivial case to illustrate one of the issues
> arise when we compare first-person experiences that stop at different
> times. Imagine a conventional lifetime where a person lives to a ripe
> old age of 90. Now imagine the truncated version of that which we
> cut off arbitrarily at age 50. Obviously the universe will contribute
> significant measure to both of these first-person experience streams.
> Which one will get more?
> I would suggest that it is actually the 90 year old lifespan which
> will have more measure. The reason is because any program to turn the
> third-person record of all events into a meaningful, compact record of
> the lifetime experience is going to have to deal with the enormous gap
> between the fundamental events of physics, which happen at the Planck
Here you are presupposing a physical universe (so again you are in
trouble with comp or at least with the UD-Argument). The Planck scale
is very different if you accept the whole quantum idea or if you take
some deterministic account of it (a la t'hooft). Comp saves us for
making that kind of choice, I think.
> and the fundamental events of consciousness, which although
> small to us are at an enormously larger scale compared to physics.
> This means that the program to do this conversion is going to have to
> be intensively data driven; it will have to identify tenuous and rather
> amorphous patterns of physical events, in order to translate them into
> the neurophysiological events that we would want to record.
> Given this structure, the (approximate) moment of physical death will
> be easily recognized, as it is that moment when the structure which the
> program has been built to track disappears. The simplest program is
> going to be one that has its own built-in, implicit stopping rule.
This is not obvious.
> In contrast, a program which stops at some arbitrary time, like age 50,
> is goint to have to be larger, because we are going to have to build in
> the stopping rule.
Very few bits more are needed here (like a counter + a "if-then-else"
> And given that the average human lifetime is enormous
> when expressed in the most natural physical units, the Planck time,
> it means that expressing the time to stop is going to take substantial
> program space.
Some very big number needs substantial program space, but much bigger
number than that will not. A random string like
"9579210055822309227007732586654794" is longer to describe than 10^100.
Big numbers can have quite short programs outputting them (cf the fairy
Also, I cannot understand a reasoning which depends on the chosen units.
> The conclusion is that, for a conventional life experience, the largest
> measure is contributed to first-person subjective records which span
> the whole range of lifetime from start to end. Arbitrary subsets of
> that subjective lifespan will have less measure than the whole thing.
> Now, back to B, W and M. As just discussed, the measure of B, which
> corresponds to a life which ends in Brussels, is likely to be
> substantial. However this hypothetical teleporting or copying machine
> works, we stipulate that the guy who started in Brussels is not there a
> moment later. It's likely that a straightforward program which has
> tracking neural events by virtue of their exceedingly slight impact
> on the Planck scale is going to be thrown off by this new process.
> Therefore a relatively simple program can output subjective record B.
> When we consider W and M (revivals in Washington and Moscow
> there is still probably a fairly small program that can produce those
> records. The main additional complication is that the program has to
> somehow be designed to be able to pick up the trail in the new city,
> to jump from recording neural events in Brussels to recording them in
> Washington or Moscow.
> This could be done in a couple of ways. The simplest would be to
> hard-code the location of the new version of the person, perhaps
> as a vector from the old one. But this could take quite a bit of
> information, as it might have to be accurate to the size of a synapse,
> a few nanometers. Probably a better way would be to track whatever
> physical event carries the causal signal from Brussels to the
> Presumably something has to travel between the cities, carrying the
> information about the person, to allow him to be reconstructed at the
> destination(s). Whatever this effect or principle is, we could write a
> program which was able to follow this signal, as the neural activities
> Belgium are being scanned or whatever. This would allow identifying
> location of the new instance of the person, without having to hard-code
> the precise coordinates. The third-person universe data would tell us
> where to look.
> This would probably not be too complicated a program, but it is
> nevertheless going to be substantially larger than program B.
I don't see how this follows. What if "I" am computationally determined
(in the normal histories) to do that duplication Moscow and Washington.
Then dying in Brussels (accepting your "non-comp") would, according to
your own reasoning above, make the B experience more exceptional,
contradicting your case.
> B only had
> to track neural events. W and M have to be able to track both neural
> events and whatever physical principles are utilized by this copying
> and transmission process. W and M are therefore going to have to have
> two different analysis methods, compared to one for B. They should not
> have to be twice as big, since they don't have to actually track
> during the transmission (we assume that thought is suspended during
I would say OK except that this idea implies the non-cul-de-sac
principle, which you need to throw out for giving sense to the notion
of dying in Brussels.
> they just have to figure out where the new brain is being built.
> But chances are it is going to make W and M quite a bit larger than B.
> Compared to each other, W and M are probably almost identical. The
> difference is that they make arbitrary different choices for which
> to follow in tracking the copying information, in order to find where
> new instance is located. Maybe one has a 1 bit where the other has a
> In all other respects the two programs are identical, and their measure
> should be the same.
> But, as noted above, B is substantially smaller, and as a result, B has
> a substantially larger measure. This means that the contribution of
> third-person record of universe events to a subjective, first-person
> experience that ends in Brussels is much larger than to an experience
> which continues in either Washington or Moscow.
> If we consider these as three hypothetical people, one who dies in
> Brussels, one who continues in Washington, and one who continues in
> Moscow, it is the first one who is instantiated to the substantially
> greatest degree by the operation of such a copying machine as we are
> considering. Informally, we could say that your most likely experience
> is that you will die in Brussels (bearing in mind the formal statement
> in the previous sentence). That is how I would analyze it based on
> computational principles.
You should make clearer when you talk on programs (finite piece of
code), program's states,
computations (finite or infinite redundant strings with some grammar)
and output or inputs
of programs. Those things, although related, are different and obeys
Honestly, here I have the feeling that you are using both comp and
non-comp at once.
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to firstname.lastname@example.org
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at