On Wed, Feb 19, 2020 at 11:20 PM Bruno Marchal <[email protected]> wrote:
> On 18 Feb 2020, at 23:14, Bruce Kellett <[email protected]> wrote: > > On Wed, Feb 19, 2020 at 12:05 AM Bruno Marchal <[email protected]> wrote: > >> On 18 Feb 2020, at 02:37, Bruce Kellett <[email protected]> wrote: >> >> And if the probabilities are to be objective >> >> They have to be at least first person plural. >> > > 'Objective', as I use the word here, means 'interpersonally agreed'. In > your terminology, that would be 1pp since there is no 3p in many-worlds. > > We should be able to make bet. But that is the case if instead of >> duplicating the H-guy, you duplicate the H-Guy + the person (the witness) >> with whom he make a bet. In that case, if you bet “W”, and the witness bet >> “M" then in W you win the bet, and in M you lost the bet. In the irate >> case, you recover the idea that by using the Pascal Triangle, you can >> maximise your benefits, and this shows that we can use the Dutch Argument >> to define some probabilities in simple duplication scenario (to be sure, >> the real case will be in arithmetic where such simple case scenario can be >> shown to never occur, and that is why the math is a bit more sophisticated >> there). >> > > If I understand you here, I think this is wrong. In the iterated case, the > 1pp probabilities are those calculated on each branch, and they are all > valid — > > > I don’t think so. On most branches the probability (or the indeterminacy) > comes from the fact that most of the 1p(p) histories will be > algorithmically incompressible, and thus highly non predictable, and > behave, for n great, as random sequence. > That is almost certainly true. But it is also completely irrelevant to the issues at hand. “W or M” will be the only always correct prediction, and P = 1/2 will match > well the ignorance in a set with a measure converging to the Gaussian. > You build in your initial prejudices and ignore the data that most individuals will actually observe in this scenario. As stated, it is the 1p probabilities that are of interest, not some "God's-eye-view". Most 1p or 1pp will just predict white noise, > The sequence of results that they observe will be random, of course. > like we predict that a sheaf of light get divided by two when going > through an half silvered mirror. > But that comment forces a probability of 50%, which is not necessarily what is observed. You attempt a sleight-of-hand by using this analogy. The worlds/histories departing from the normal distribution get > infinitesimally rare in the limit > And the histories that deviate from my observed probability of p (whatever that might be), also get infinitesimally rare in the limit (assuming that p is the correct probability). You are simply ignoring the mathematical proof that this is the case. Your insistence that 0.5 is the correct probability leads you into serious mathematical error. (and Dital Mechanism explains why we have to consider that limit, mainly > the invariance of the first person experience for the delays of > reconstitution in the arithmetical Universal Dovetailer. > Keep the dovetailer for the birds....... there is no 3p view (God's-eye-view) to contradict them. > > > They are aware of the protocol, and by definition, the protocol is > respected, so they do have some 3p idea of what is going on. > In other words, it is not just about the 1p view, personal indeterminacy. You want to be able to assume a particular answer and then insist that everyone gets just this answer. That does not work in science. In the WM duplication, then the copies are able to meet and compare > diaries, so things are different, > > > That is what I am talking about. Eventually this will justify both the > quantum MW and its formalisme. > Except that the inhabitants of different quantum worlds can never meet. So the possibility of meeting in the WM duplication case makes it irrelevant for discussion of Everett. but I am interest in the Many-worlds case, not classical single-world > duplication. > > > With mechanism, the quantum Many-Worlds (or any physics) has to be > explained by the single-body (or single population of bodies) > duplication/multiplication occurring (virtually, arithmetically) in > arithmetic. > With mechanism pigs fly about all over the place. Mechanism is a failed idea. Bruce If interested I can explain more. A lot of people miss that the notion of > computation (i.e. the notion of universal machine and their executions) is > a purely arithmetical notion. Gödel is the first to have shown this, but he > did not realise what he as doing because he missed, in 1931, the Church > Turing thesis, as he explained himself. Gödel will accept it later when > reading Turing. > > Bruno > -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAFxXSLQE%2BoDOYi1wV%2BdMPnK3gVCwKLzH0CWrJAtN523JdXO2sg%40mail.gmail.com.
