On Mon, Feb 27, 2012 Bruno Marchal <marc...@ulb.ac.be> wrote:
> the UD is not theoretical. It is easy to implement it, and an UD has been > implemented in 1991, and run for one week. > And it ran all possible computer programs? I doubt that. My point is that you cannot fly with nothing but the blueprints of a 747, you need to implement that information in matter and perhaps the same is true for your dovetail machine. > In step seven, I suppose that we live in a sufficiently big universe > running that UD, forever. > Then you are assuming what you are trying to derive from nothing but pure numbers, you are assuming the existence of matter and enough of it to build your gigantic dovetail machine. > You can make matter behave in any way relatively to virtual reality that > you are building > Yes, and that's exactly why I don't see how you could tell if you were in a simulation or not, the outcome of any experiment you perform would be consistent with being in a simulation. > It is a consequence of the first person indeterminacy. > It's not indeterminate it's just a silly question, the question you're really asking in this 1p 2p 3p stuff is "If I change what is the probability I will remain the same?". > Comp makes arithmetic a theory of everything. > That could only be true if consciousness is everything because consciousness is what comp is all about, and that is a non trivial assumption. > from your paper: The reason is that comp forbids to associate inner > experiences with the physical processing related to the computations > corresponding (with comp) to those experiences. > That is not entirely correct, comp forbids to associate inner experiences with the particular type of physical processing going on, both intelligence and consciousness depend on the logical arrangement of the processors and it does not matter if the processing is electronic or mechanical or chemical or biological, but physical processing of some sort might still be necessary and if so then matter is just as fundamental as numbers. So maybe only numbers are fundamental and maybe both matter and numbers are fundamental; I think we both agree that matter alone can not be fundamental. >> intelligent behavior implies consciousness. >> > > > I agree, but "intelligent behavior" is not something that you can > defined. It is left to personal appreciation. > Yes, there is no infallible test for intelligence much less for consciousness, so we must use the Turing Test because imperfect though it may be it is better than nothing. > > Just tell me how you predict what *you will feel to see* when you throw > a dice in a universe running a universal dovetailer. > If the John K Clark program is running on your universal dovetailed machine then there is no way I can make a incorrect prediction about what the John K Clark of this instant will see because he will see everything that there is to see, the machine runs all programs that can be, but of course once he sees something new he will no longer be the John K Clark of that instant. So there is a version of me where exactly one second after I type a period at the end of this sentence I will receive news that I've just been elected Pope. But...well, apparently not this version of John K Clark. > > "surprising" is subjective. > And so is consciousness. > if asked before the experiment about his personal future location, the > experiencer must confess he cannot predict with certainty the personal > outcome of the experiment. > If he believes in comp he will predict that he will be in Washington and Moscow > If he believes in comp he will predict that the (3p) will be in W and M, > Yes, except you don't need that "3p" caveat, he will predict he will be in W and M period. > but he will predict also that he will feel to be in one and only one place > Yes absolutely, he will predict he will feel to be in one and only one place, therefore if he believes in comp, that is to say if he is logical, he will predict there will be more than one "he". Do you have a problem with that? I don't. > > You miss the 1p and 3p difference. > The entire question makes no sense! You ask the Helsinki man to predict if he will go to Moscow or Washington, but the question is not well formed. You keep asking if "you" will be in Moscow or Washington and usually that's a question that can only have one answer because up to now there is only one chunk of matter that behaves in a Brunomarchalian way, but there is no law of logic or physics that demands that always be true, and in this thought experiment it is specifically stated that it is NOT true. So both the Washington and the Moscow man will remember being the Helsinki man, neither the Washington nor the Moscow man will remember being each other, and the probability the Helsinki man will remember being the Washington man but not the Moscow man is zero because the Helsinki man only remembers being the Helsinki man . I think at least part of the difficulty has to do with language, if duplicating chambers become common the English language will have to modify the way it uses the words "I", "you", and "He" and "she". As Douglas Addams said about time travel "The major problem is quite simply one of grammar, and the main work to consult in this matter is Dr. Dan Streetmentioner's Time Traveler's Handbook of 1001 Tense Formations. > > With comp, being certain is impossible > What you really mean is that with comp you SHOULD never be certain not that you can't be, in actuality people can be certain for the looniest of reasons; if the 911 hijackers were not certain they would have 77 virgins after their death they would not do what they did. You can't always be certain AND correct, and it doesn't matter is comp is true or not. > Non-comp is not contradictory. If you show this, you would prove that > comp is true, which only an inconsistent machine can do. > Non-comp may not be contradictory but all the human practitioners of non-comp most certainly are, every single one, no exceptions. > > You did agree we might define God, by the reason why we are here. > No I did not agree, that is necessary but not sufficient. A God needs to be a person, a infinite person perhaps, a superhero of superheroes, but a person. Arithmetic is not a person, super or otherwise. And as the purpose of words is communication and in our culture nearly everybody thinks the word "God" means a being that created the universe I think it wise to use another word to describe arithmetic. > Take the number 0, <a billion of zero> then the first digit of > chaitin-omega, then <a billion of zero again>, then the second digit of > chaitin-omega, etc. This number is not computable, but is compressible. > OK, you are entirely correct. Good example. Thanks for the correction. Non-compressible numbers are not computable but the reverse is not necessarily so. > With comp, the material brain is indeed infinite, but this assume comp, > that is, I can truncate that brain at a finite level and survive the finite > digital substitution. > You can certainly do truncations of that sort but comp does not demand infinity, some of the newest theories in physics postulates that both time and space are quantized, if this turns out to be true I don't see it causing any problems for comp. > the copy is not a zombie, but is someone else, believing wrongly that it > is you. > Then maybe I'm wrong in believing I'm me too; and maybe Homer didn't write the Iliad and the Odyssey, maybe those poems were written by another blind poet named "Homer" who was born in Smyrna and lived in 850 BC. > You might say that those who does not believe in comp are inconsistent or > stupid, but you have not provide argument for this. > Somebody who said they believe members of this list that they have never seen are conscious but a computer that can write good poetry, make up a funny joke, and invent a quantum theory of gravity is not conscious is not being consistent; do I really need to point out why? > > So in Helsinki you are asked where you will FEEL to be after having been > read and cut in Helsinki and reconstituted in W and in M. > Normally the meaning of the word "you" in questions is clear and causes no ambiguity because usually your body is the only thing in the universe that behaves like you, but in this thought experiment we have duplicating chambers and that is not normal and normal English was never built to handle such situations. > accepting comp, and thus accepting the certainty of surviving a classical > comp teleportation, here is what you can say in Helsinki: It is certain > that I will feel to be in one city. > Yes. > it is impossible that I will feel to be simultaneously in both city. > Yes. >It is certain that my memory will contain either "I am in W" or "I am in M" > Yes. > It is certain that I will be unable to justify why "I am the one in the > city I am observing right now". > Nonsense. The reason I feel I'm in Moscow not Washington is that I have memories of being in Moscow not Washington, so I'm the Moscow guy, those memories are what made me the Moscow guy, if it wasn't for them I'd still be the Helsinki guy; if the situation had been reversed (and it has been as a matter of fact) then I'd be the Washington guy. I know nothing about the Washington guy but we both know all about the Helsinki guy although the Helsinki guy knows nothing about us because you can remember the past but not the future. I just don't see any deep dark mystery in any of this. > Conclusion: if I am duplicated (at the right level, etc.) then I cannot > predict in advance with any certainty what I will precisely feel after the > experiment. > Conclusion: It doesn't matter if I say Washington or Moscow, I cannot make a incorrect prediction. > You might also think about the iteration of such experiences. You are in > Helsinki, and you are cut in W and M, then both of you come back and both > of you do again that experience, and so on. After n such experiences, all > of you will have 2^n -1 doppelgangers, and if they decide to count among > themselves the number of having reach M or W, they will see that it obeys > to the Newton binomial coefficients (the Pascal triangle), confirming the > feeling of the vast majority of the population that the probability of > going in W, or in M, is given by the binomial distribution, or its Gaussian > approximation (if n is big), confirming the early feeling that P = 1/2. > I don't see having a large voting population tells us anything; the statistics that work for nouns are not the same as the statistics that work for adjectives like me or you. If a thousand iPods are synchronized and playing the same song there is only one song playing, and if you destroy one of the iPods the music does not stop. And a thousand Turing machines playing the same identical AI program are not producing a thousand conscious beings but only one. > Don't equalize non computable and random, many notion in math are non > computable yet totally not random > I found some quotes I liked by Gregory Chaitin, the discoverer of the Omega number, on this subject: "I didn't really prove that mathematics is random; I came up with a definition of randomness which has this strange property: the most interesting thing about it is that you can never prove that something satisfies this definition---even though most things do.[...] Then, some years later, I realized that there was an area of mathematics that I could construct, or I could discover, where in fact mathematical truth had no structure, was completely random, in that area. So this is a part of mathematics which is a black hole, where individual questions cannot be answered. You can make statistical statements about the answers. The answers will be one thing or another 50% of the time. [...] I found an area of mathematics, or I constructed an area of mathematics, where in fact God does play dice, where mathematical truth is accidental, where things are true for no reason. This is in pure mathematics itself. So in this part of mathematics, in fact, things are maximally random, things have absolutely no structure, mathematical truth is completely accidental, it's a worst case. It's sort of a nightmare for the rational mind". Chaitin also said something on a different subject that I agree with completely: "In a way, computer technology can be thought of as technology for dealing with souls. Software is like a soul. You take a machine and it's dead, when you finish constructing a machine. But when you put software into it, it comes to life. And this software can move from machine to machine, it's as if you were passing a soul from one body to another. It's a crazy analogy, but there is a little bit of truth in this crazy viewpoint. So what is the soul? Well, maybe it's information! Software is information, it's not physical." " When I was young, I was a materialist. You know, I thought everything was physics---in principle psychology, human society... in theory it was all a big, complicated problem in physics. But I don't believe that anymore, and I'll give you the example that changed my mind, the reason that I've become an anti-reductionist. The example is, look at a computer! A computer has software and it has hardware, and these are two different levels of reality. The hardware level is physics, it's engineering. When you see the hardware, that's when the machine breaks down, for example. But you want that to be invisible. The computer engineer, the computer designer, wants to hide the physics, so that he can simulate this fantasy world of software, which in a way is the Platonic world of ideas. And this example really convinced me that when you go to a higher level, the lower level may be irrelevant. So look at us. We're built out of DNA, we're built out of chemicals, chemicals that have an explanation that's physics. But DNA creates a higher level of abstraction which is biological information. And our thinking is at still a higher level, the mind, the brain. And it may be that really the lower level, the level of physics and chemistry, is irrelevant, the same way that the computer engineer tries to make the physics of transistor and computer hardware be invisible to the programmer." John K Clark -- You received this message because you are subscribed to the Google Groups "Everything List" group. To post to this group, send email to everything-list@googlegroups.com. To unsubscribe from this group, send email to everything-list+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/everything-list?hl=en.