It still seems to me that the environment is irrelevant, in that given comp the brain or computer can be cut off and inputs (in principle) mimicked - even if those inputs are due to quantum entanglement. Hence the MGA at leasts starts on a sound footing, with the entire computation including inputs prescribed from the outset. But the balance of opinion seems to be that this isn't the case fro some reason - remind me why?
On 11 May 2015 at 03:57, Bruno Marchal <marc...@ulb.ac.be> wrote: > > On 09 May 2015, at 03:48, smitra wrote: > > On 01-05-2015 17:59, Bruno Marchal wrote: >> >>> On 30 Apr 2015, at 17:07, smitra wrote: >>> >>>> On 30-04-2015 09:19, Bruce Kellett wrote: >>>> >>>>> Stathis Papaioannou wrote: >>>>> >>>>>> On 30 April 2015 at 13:20, Bruce Kellett <bhkell...@optusnet.com.au> >>>>>> wrote: >>>>>> >>>>>>> The way I understand it, nothing happens in Platonia. Which is to >>>>>>>> say >>>>>>>> nothing ever happens. The real question is why we think stuff is >>>>>>>> 'happening'. Well, OK - the hallucination that stuff is happening >>>>>>>> is what is >>>>>>>> happening. >>>>>>>> >>>>>>> So explain the hallucination. Why does that 'happen'. Note that >>>>>>> 'happen' is >>>>>>> a temporal term. >>>>>>> >>>>>> I have the feeling that I have been alive for years, but I would >>>>>> still >>>>>> have this feeling if I had only been alive for seconds. There does >>>>>> not >>>>>> have to be a physical, causal connection between the observer moments >>>>>> of my life for them to form a subjective temporal sequence. The >>>>>> sequence is implied by their content. >>>>>> >>>>> The brain in the vat is always possible. We cannot rule out solipsism >>>>> either. >>>>> Julian Barbour, in his book 'The End of Time' tried to abolish time >>>>> altogether because of the difficulties of defining time in general >>>>> relativity. He replaced time as a parameter with the notion of 'time >>>>> capsules' present in every point of phase space. >>>>> It is not really clear whether this idea was successful or not. It has >>>>> not attracted a great following. >>>>> But if any such idea is to make sense, the observer moments do have to >>>>> be connected by quite strong causal laws so that the sequence of >>>>> moments tells a coherent story. Or else each moment tells a different >>>>> story, and we are back with 'Last Tuesdayism' or solipsism. >>>>> I don't think Fred Hoyle's account works either. It feels like a 'many >>>>> minds' collapse interpretation of quantum mechanics. >>>>> Bruce >>>>> >>>> You can use the formalism developed in this article: >>>> http://arxiv.org/abs/1305.1615 >>>> If we take finite time steps corresponding to a computational step, >>>> then an observer momement is defined by specifying some operator: >>>> sum over {in} of| j1,j2j3,...jn><i1,i2,i3...in| >>>> where the jk are functions of the i1,...,in. >>>> This then simplify specifies that a computation proceeds from an >>>> initial state defined by the sequence of numbers i1,i2 etc. to the next >>>> step defined by the numbers j1,j2 etc. >>>> The summation has some finite range, so the algorithm is not defined >>>> precisely. On the other hand, the fact that the summand contains more than >>>> just a single term means that the state of the system is not well defined. >>>> The more terms there are in the summation, the better defined the >>>> computation becomes, but the state of the system becomes less well >>>> defined. >>>> A computation that is complex enough to represent what the brain is >>>> capable of will contain an astronomically large number of terms; whatever >>>> consciousness is and how it works, from experience we know that what we >>>> feel and think doesn't contain enough information to nail down exactly >>>> what the brain is doing. >>>> This means that in a MWI picture, it is wrong to represent the >>>> branches as single lines, they are bundles consisting of an astronomically >>>> large number of lines, the correlation contained in them contain a vast >>>> amount of information, more than what you need to define what computation >>>> is actually being performed at any instant. >>>> Anyway, I think that Bruno should consider deriving physics from >>>> starting with defining observer moments as matrix elements O = sum over i >>>> of |j><i| and then physics should be derived by introducing more degrees >>>> of freedom and then finding a generator of O. So, you invent a universe >>>> described by a Hamiltonian so that running the laws of physics starting >>>> from some initial conditions will allow you to properly represent O. >>>> Then one considers that particular representation that requires the >>>> least amount of information given some O. Then one should consider also >>>> minimizing that information over the possible ways of defining O (note >>>> that O being defined by a summation indicates that O itself doesn't know >>>> what state it is in). >>>> >>> That seems interesting, but is there not still treachery here, copying >>> of physics? How will we take into account the G G* distinction? I >>> have first to justisfy such O from the material hypostases. But the >>> shadow of what you say is already there: I mean the ket-bra Ii><jI. >>> Of course, any intermediate work will help! >>> >> >> Putting aside the precise details, what I think is a common mistake is to >> assume that we're dealing with precisely defined computational states. As >> the MGA shows, that leads to problems. But there is no need to make that >> assumption, one can also assume that you need to consider a set of such >> states, each element of which is mapped to elements of another set. That >> way you have both a notion of the computational state and the algorithm >> that is being run, albeit that both are imprecisely defined. But in the >> context of complex systems that would need to be specified using a huge >> amount of information, like our brain, they can both be quite well defined. >> >> This is analogous to how despite the uncertainty relations, you can say >> where your car is and what its velocity is. But if you were to try to >> derive this fact from quantum mechanics in the case of of a quantum >> mechanical car that is completely isolated (so that we don;t invoke >> decoherence), then you have to work with wave packets of finite widths and >> then take the appropriate limits. You would not get the correct result if >> you were to assume that, say, velocity is always exactly defined. >> >> Similarly, I believe that you need to work with a finite uncertainty >> about the exact states that would specify the computational states, and >> then consider the analogue of the classical/macroscopic limit of systems >> specified by an infinite amount of information, but where you still have a >> finite uncertainty (that perhaps tends to zero in a relative sense). >> > > I agree. The finite specification is given by the substitution level, and > the current approximate state, and consciousness supervene on a cloud of > continuations, structured by the definition of the points of view. > > Just above the level, we might most plausibly still survive, but perhaps > for a shorter time, then comes the problem, the recurrent dreams, the > headache, and you go back to the doctor to fix it hopefully. > > Consciousness is really an abstractor from the non relevant thing to take > a decision, and it sums on all the non relevant things below the subst > level, which still have a role in the relative measure; Not a role of > content (unless we are comp-quantum-machine) but a role in the relative > weight of that content. > > There is a sort of self-diffraction. The more you know who you are are, > the more possibility you get, as you identify yourself less and less with > token details, and so you augment your spectrum of continuations in > arithmetic. > > Bruno > > > >> Saibal >> >> -- >> 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 everything-list+unsubscr...@googlegroups.com. >> To post to this group, send email to everything-list@googlegroups.com. >> Visit this group at http://groups.google.com/group/everything-list. >> For more options, visit https://groups.google.com/d/optout. >> > > http://iridia.ulb.ac.be/~marchal/ > > > > > -- > 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 everything-list+unsubscr...@googlegroups.com. > To post to this group, send email to everything-list@googlegroups.com. > Visit this group at http://groups.google.com/group/everything-list. > For more options, visit https://groups.google.com/d/optout. > -- 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 everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
