Thanks Bruno I want to have a good think about your answers and also the eighth step in your paper. I think it is the most difficult for me and yet I sense its somehow. Schmidhuber assumes a great programmer runs the UD but you effectively dispense with him. If a universal turing machine necessarilly exists platonically which is capable of running UD's that can simulate our minds then our experience of reality follows. Yet I still feel that somehow this will be confusing the map of the territory with the "reality", the equations of physics with the physically real.
Best wishes Nick On Dec 23, 2:15 pm, Bruno Marchal <marc...@ulb.ac.be> wrote: > Hi Nick, > > On 23 Dec 2009, at 01:55, Nick Prince wrote: > > > > > > > My background is in mathematical physics but I am trying to read up a > > bit of this new stuff as I go along. Thank you for being patient with > > me. > > > However, I think you have confirmed some things - let me know if any > > of these is fundamentally wrong. > > > I want to put aside the platonic arithmetical UD for the moment but > > will come back to that. > > > Just supposing that there is as yet NO UD operating. Comp says it is > > possible to build a "concrete" one and in step 7 of your paper you say > > that the UD could provide infinitely many possible consistent > > extensions of me. (I am thinking of descriptions of simulated worlds > > with me in them as bit strings) I quote you from the sane paper: > > > "Then, it follows from the six preceding steps that it will generate > > all possible Turing machine states, infinitely often (why?), which (by > > comp) includes all your virtual reconstitutions corresponding to > > (hopefully) consistent extensions of yourself, in all possible > > (locally) emulable environments or computational histories. And this, > > with comp, even in the case you consider that your ‘‘generalised > > brain’’ (the ‘‘whatever’’ which is needed to be emulated by a DU > > digital body/brain to survive) is the whole Milky Way galaxy. And we > > don’t need any Science Fiction like devices to make this concrete, if > > we make exception of the robust universe." > > > Actually the kind of "teleportation" I am interested in, for reasons > > as you will see is the usual simple one which takes us from moment to > > moment. I am being teleported into the next observer moment all the > > time ( if this is because I'm already being computed by a UD then as I > > say lets just ignore this possibility for now as you did in your > > paper). If someone is blown to bits, then we have lost the chance to > > make a decent copy of them. However, Once the “concrete” UD is run > > then it computes all possible futures for all possible virtual > > extensions. Then there will be an (infinitely many) extension(s) for > > the blown to bits person. The blowing to bits is just the equivalent > > of the annhiallation part of your earlier steps. So here we have the > > basic quantum immortality thing coming in again. > > OK. TO be sure it is the "older" comp immortality, and it is an open > problem if the quantum interference and immortality *is* a result of > the comp interference and immortality. It looks like that, and up to > now the math confirms formally the resemblance. > > > However, if it > > takes the UD a long time to generate sufficient extensions then the > > delay will be considerable before the blown to bits man continues > > consciousness - although to him it will seem instantaneous. From 3d > > person, well - they see the delay. > > Yes. (would they live long enough) > > > > > Now is the interesting bit. Because this future UD creates all > > possible extensions of all possible states of the blown to bits man > > then what’s to stop him finding continuation with a consistent > > extension prior to the blowing up! > > Sure. (that happens all the time, and that's why we have to justify > the apparent stable laws from that). > > > In other words every observer > > moment of his life (not just the one just before being blown up - but > > any of them) could just as easily be followed by a suitable one in > > the virtual UD rather than one in the initial run of the universe. > > Absolutely. Would a real *singular* concrete material universe exist, > the probability to stay in that universe is zero. > > > In > > conclusion, from our ist person point of view we do not know whether > > our next observer moment will occur in the “real” universe or in a > > simulated one- this is 1-indeterminacy again. If the UD can simulate > > all possible observer moments then it will have those associated with > > our very first sense of consciousness and hence we will have very > > quickly slipped, without knowing it, into the UD’s virtual world. We > > never noticed any delay of course but there may have been a huge time > > difference assuming Russell’s time postulate has meaning here! > > It has meaning, because it is neither physical time nor subjective > time, but just the natural numbers with the successor operation, or > the number of steps taken by the UD to reach the computational states. > And the step 8 explains why, even if a real physical time exists, it > just cannot compete with the UD "time". It is no more than 0, 1, 2, > 3, ... or a set having a computable bijection with N. > > > > > Hence if a UD is possible, > > Well, the mathematical existence of the UD is a logical consequence of > Church thesis + Turing's theorem in computer science. > There is number U such that for all x and y, phi_U(<x, y>) = phi_x(y). > U can emulate x on y. > Once you can emulate all x, you can dovetail on all emulations > possible, including those with oracles in some rings. > > > then only the first observer moment(s) -or > > fraction of our conscious lives - were ever lived in a “basic/real” > > universe at all. The rest is all simulation. The very existence of a > > UD implies that we are in a simulation as Nick Bostrom has suggested. > > Hmm... OK. To be precise, we belong to the infinities of computations > which differ below our level of substitution, and which are similar > above. This makes the apparent physical universe a priori NOT Turing > emulable, or not entirely Turing emulable. But here too, QM confirms > this: we cannot emulate with a Turing machine a beam of electrons all > prepared in the state "up+down" going through an up/down measuring > apparatus, (without emulating the couple "observer + electron", but > then we have to classically emulate all the "parallel worlds"). > > > > > If they exist platonically then it's all simulation and > > computationalism must be necessary rather than contingent. > > OK. And some people are already convinced at that stage. But from a > logical point of view, a materialist could argue that the seven step > just shows that if comp is true, then the material universe is just > not enough big, or not enough robust to implement the running of the > UD. UDA would just show that materialism would imply our universe is > "small". > The 8th step has two purposes: > - one is showing that the very idea of linking consciousness to > physical activity is a non sense (despite the strong appeal for such > an idea, even through comp at first sight), > - and the second is to prevent the above move toward a small universe. > It shows that with comp, we have to link consciousness, not to > material physical activity, nor even to any singular universal machine > computation, but to an infinity of computations, and we have to > understand that by "computation" we don't mean anything physical. A > computation is the doing of any universal machine: it is defined by > the "pure" additive-multiplicative relations between positive > integers---- or by anything recursively equivalent (as defined in > theoretical computer science), like the "pure" abstractive-applicative > relations between the SK-combinators. > > > > > It's a very fumbling line of thinking but it helps me to learn about > > things as I go along. > > Cool. Ask any questions. Best, > > Bruno > > > > > On Dec 22, 6:41 pm, Bruno Marchal <marc...@ulb.ac.be> wrote: > >> On 22 Dec 2009, at 18:48, Nick P wrote: > > >>>>> Hence by it generating all possible emulations of > >>>>> stages of my life) that I could just as easily experience for my > >>>>> next > >>>>> OM as opposed to the one i would expect to experience on the > >>>>> current > >>>>> wetware (or whateverware I'm running on if we are in fact already > >>>>> software constructs in a simulation). > > >>>> This is weird. From some "absolute", non machine accessible view > >>>> point, you can expect anything. Perhaps. > > >>> Assuming that comp is true, then I am not sure why you think it is > >>> weird. Perhaps I have not explained myself very well. First of all > >>> please check that my understanding of computationalism is correct. > >>> By > >>> comp I mean (loosly)that I am assuming that any conscious being > >>> can be > >>> simulated on some form of computer. Currently my consciousness is > >>> running on the substrate provided by my brain (hardware). If the > >>> underlying reality is a much more fundamental (unknown) substrate > >>> then > >>> fine because this shouldn’t invalidate what I’m saying. Now If I > >>> want > >>> to be teleported from Brussels to Moskow then sufficient > >>> information > >>> must be coded for my reconstitution later on. This may or may not be > >>> possible because it may turn out that the accessing of my final > >>> state > >>> in Brussels destroys my brain before the detailed brain state was > >>> properly copied. Worse still, suppose someone loses whatever coded > >>> data they did have of me such that the reconstitution becomes > >>> impossible. What I am trying to say is that if comp is true then at > >>> least I can be confident that some consistent extension of me could > >>> exist in the future provided the robust physical universe you > >>> speak of > >>> exists such that a suitable UD can actually be built. > > >> OK. (and then step 8 explains why the initial universe is no more > >> useful, the arithmetical UD is enough). > >> Also, it is perhaps always one next 1-observer moment, but also > >> always an infinity of 3-observer moments. The UD is terribly > >> redundant, and anything it does, it will repeat it infinitely > >> often. A > >> compactification of it looks really like the border of the Mandelbrot > >> set. The closer you look, the more complex it appears. > >> http://www.youtube.com/watch?v=g9iOORSU9zk > > >>> Once built then > >>> there would exist at least one consistent extension of me (including > >>> the milky way if this level of entanglement is to be necessary to > >>> adequately ensure this is the most probable next state of my > >>> consciousness) in the UD which will enable me to experience my next > > ... > > read more »- Hide quoted text - > > - Show quoted text -- Hide quoted text - > > - Show quoted text - -- You received this message because you are subscribed to the Google Groups "Everything List" group. To post to this group, send email to everything-l...@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.
