Interesting paper on consciousness, computation and MWI
David Nyman
Bruno Marchal
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To view this discussion on the web visit https://groups.google.com/d/msg/everything-list/-/AyiMzznp-hIJ.
To post to this group, send email to everyth...@googlegroups.com.
To unsubscribe from this group, send email to everything-li...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en.
Stephen P. King
I have found what I believe is a flaw in the reasoning in the paper.
On pages 5-6 we find:
" In Section 5, I attempt to apply this reasoning to the case of an infinite lifetime. I find that, on the one hand, in discovering his current moment out of an infinite ensemble of moments, the observer should gain an infinite amount of information. But, on the other hand, I argue that such a state of affairs is not logically possible. Thus I conclude that an infinite conscious lifetime is not possible in principle."
I disagree with this conclusion because the ability to 'discover' ones current moment out of an infinite ensemble of moments would require the ability to access the computational resources needed to run the computation of the search algorithm on the infinite ensemble. In this case it is required that an infinite quantity of resources be available in a finite or infinitesimal duration. The author does mention some aspects of the problem in computational terms but the issue of resources does not seem to have been noticed. I find it strange that computations can be treated as if they are not subject to the laws of physics that included prohibitions on perpetual motion machines. There is no such thing as a free computation. The content of our Observer moments is finite due to computational resource limitations not because of some universal prior measure.
Onward!
Stephen
Pierz
be a lot harder to pin down where the specious logic is – especially
when it’s all dressed up in a mathematical formalism that may be
inaccessible to the non-mathematician/logician. However the problem
with the arguments relating to consciousness in this paper is not so
hard to pin down, and indeed Stephen King is on the right track with
his objection.
Eastmond argues that an infinite conscious lifetime is impossible
because, in ‘finding oneself’ at a particular point in that lifetime,
one would have to gain an infinite amount of knowledge, which is
absurd. He concludes that such an infinite lifetime is in principle
impossible. The flaw lies in the way the author glosses over the
notion of “gaining information”. In examining the problem, he treats
this “gaining of information” as if it occurred magically the moment
one finds oneself at a certain point in a lifetime, but in fact such
information has to be acquired by a concrete computation. For example,
if I am to gain information about my current lifetime position, I need
to examine a calendar and compare this to stored or acquired knowledge
about my date of birth. In the case of an infinite lifetime, the size
of the computation required is arbitrarily large (but finite) in the
case of an infinite lifetime with a lower bound (a life time with a
starting point), or simply uncomputable in the case of an infinite
lifetime with no lower bound. This is the same as saying that one
cannot calculate the age of a person who has always existed. The fact
that such a person’s age is uncomputable does not however mean that
such a person cannot exist.
The favoured theory in modern cosmology suggests that the universe is
spatially infinite. How then do we calculate the position of our
planet in this universe? If astronomers had infinite access to the map
of the universe, they could still never calculate our position,
because the calculation would be infinite. Given that time is known to
be interconvertible with space, it follows that the same logic would
apply to locating an event on an infinite timeline. The situations are
mathematically indistinguishable, yet this does not prove away the
spatially infinite universe theory.
In an infinite lifetime with no lower bound, we can never know our
age, and the amount of information ‘gained’ when we find ourselves at
a point of time in such a lifetime is a function of how much
information we can process (concrete processing limitations) and the
amount of information available to us about our position. Whichever is
smaller forms the limit.
There is also a flaw in the reasoning in relation to the proposed
conscious computer which resets itself in order to generate repeated
(and therefore infinite) conscious moments. We must remember that the
information gain is made by the conscious entity and must form part of
its conscious computation. Otherwise where is the supposed gain
occurring? All we would have is an objective description of a
perfectly mathematically conceivable situation – an infinite set of
values for the set of conscious moments, or an infinitely long string
to define a moment within that set. So the computer must gain the
information. But it cannot do so if it continually resets. The
invocation of thermodynamics does not help if the computer cannot
access information about entropy. It can escape this problem with an
endless incrementing loop, but then it needs an infinite memory to
store this growing string. Its computational limitations inevitably
force its incrementing register to 'clock over' (like Y2K) at some
point, causing it to repeat itself.
So unless we grant the possibility of an infinite mind/computer, an
infinite lifetime necessarily entails the repeat of conscious
experience (just as cosmologists grant that the spatially infinite
universe with locally finite information must entail a Nietzschean
infinite recurrence). Such a lifetime is perfectly imaginable. Indeed,
theoretically an infinite lifetime with no repetition is possible with
infinite computational resources.
With these flaws the remaining argument regarding the impossibility of
a deterministic and conscious computer need not even be addressed,
since they are built on unsound foundations.
Bruno Marchal
I agree. Eastmond still fails to see that computationalism entails an
indexical approach to time and reality. This makes experiences
relational and relative, even if "correctly" felt as absolute "here-
and-now" by the person subject.
Bruno
>
>
> On Aug 25, 8:12 am, David Nyman <david.ny...@gmail.com> wrote:
>> This paper presents some intriguing ideas on consciousness,
>> computation and
>> the MWI, including an argument against the possibility of
>> consciousness
>> supervening on any single deterministic computer program (Bruno
>> might find
>> this interesting). Any comments on its cogency?
>>
>> http://arxiv.org/abs/gr-qc/0208038
>>
>> David
>
Stephen P. King
>
> On 31 Aug 2011, at 04:55, Pierz wrote:
>
>> Sophistry has a smell. Sometimes an argument smells of it, but it may
>> when it�s all dressed up in a mathematical formalism that may be
>> inaccessible to the non-mathematician/logician. However the problem
>> with the arguments relating to consciousness in this paper is not so
>> hard to pin down, and indeed Stephen King is on the right track with
>> his objection.
>>
>> Eastmond argues that an infinite conscious lifetime is impossible
>> one would have to gain an infinite amount of knowledge, which is
>> absurd. He concludes that such an infinite lifetime is in principle
>> impossible. The flaw lies in the way the author glosses over the
>> this �gaining of information� as if it occurred magically the moment
>> one finds oneself at a certain point in a lifetime, but in fact such
>> information has to be acquired by a concrete computation. For example,
>> if I am to gain information about my current lifetime position, I need
>> to examine a calendar and compare this to stored or acquired knowledge
>> about my date of birth. In the case of an infinite lifetime, the size
>> of the computation required is arbitrarily large (but finite) in the
>> case of an infinite lifetime with a lower bound (a life time with a
>> starting point), or simply uncomputable in the case of an infinite
>> lifetime with no lower bound. This is the same as saying that one
>> cannot calculate the age of a person who has always existed. The fact
>> such a person cannot exist.
>>
>> The favoured theory in modern cosmology suggests that the universe is
>> spatially infinite. How then do we calculate the position of our
>> planet in this universe? If astronomers had infinite access to the map
>> of the universe, they could still never calculate our position,
>> because the calculation would be infinite. Given that time is known to
>> be interconvertible with space, it follows that the same logic would
>> apply to locating an event on an infinite timeline. The situations are
>> mathematically indistinguishable, yet this does not prove away the
>> spatially infinite universe theory.
>>
>> In an infinite lifetime with no lower bound, we can never know our
>> a point of time in such a lifetime is a function of how much
>> information we can process (concrete processing limitations) and the
>> amount of information available to us about our position. Whichever is
>> smaller forms the limit.
>>
>> There is also a flaw in the reasoning in relation to the proposed
>> conscious computer which resets itself in order to generate repeated
>> (and therefore infinite) conscious moments. We must remember that the
>> information gain is made by the conscious entity and must form part of
>> its conscious computation. Otherwise where is the supposed gain
>> occurring? All we would have is an objective description of a
>> values for the set of conscious moments, or an infinitely long string
>> to define a moment within that set. So the computer must gain the
>> information. But it cannot do so if it continually resets. The
>> invocation of thermodynamics does not help if the computer cannot
>> access information about entropy. It can escape this problem with an
>> endless incrementing loop, but then it needs an infinite memory to
>> store this growing string. Its computational limitations inevitably
>> force its incrementing register to 'clock over' (like Y2K) at some
>> point, causing it to repeat itself.
>>
>> So unless we grant the possibility of an infinite mind/computer, an
>> infinite lifetime necessarily entails the repeat of conscious
>> experience (just as cosmologists grant that the spatially infinite
>> universe with locally finite information must entail a Nietzschean
>> infinite recurrence). Such a lifetime is perfectly imaginable. Indeed,
>> theoretically an infinite lifetime with no repetition is possible with
>> infinite computational resources.
>> With these flaws the remaining argument regarding the impossibility of
>> a deterministic and conscious computer need not even be addressed,
>> since they are built on unsound foundations.
>
> I agree. Eastmond still fails to see that computationalism entails an
> indexical approach to time and reality. This makes experiences
> relational and relative, even if "correctly" felt as absolute
>
> Bruno
>
> http://iridia.ulb.ac.be/~marchal/
>
>
>>
>>
>> On Aug 25, 8:12 am, David Nyman <david.ny...@gmail.com> wrote:
>>> This paper presents some intriguing ideas on consciousness,
>>> computation and
>>> the MWI, including an argument against the possibility of consciousness
>>> supervening on any single deterministic computer program (Bruno
>>> might find
>>> this interesting). Any comments on its cogency?
>>>
>>> http://arxiv.org/abs/gr-qc/0208038
>>>
>>> David
>>
>
>
Hi Pierz,
Thank you for your comments and elaborations. Your remarks also
shows that the measure problem is important to understand. BTW, there is
a very good discussion of this in David Deutsch's new book 'The
Beginning of Infinity'.
HI Bruno,
Could you elaborate on how the indexification of time 'makes'
experiences relational and relative?
Onward!
Stephen
Russell Standish
comments is that the argument is a variant of the one given on this
list some time ago by Jaques Mallah. But there are some differences.
One of the main problems is that the measure on observer moments
cannot be uniform, as this paper relies upon to draw the
contradiction. The measure of younger moments must be much higher than
older age moments, in just such a way that one should expect to have
an observer moment within the normal lifespan of a human. I explain
this in my book. Jacques Mallah successfully argued that I exaggerate
it in my book (one shouldn't expect to be a baby necessarily) - in my
next version, this statement is modified.
I will still need to study it some more. I've been involved in a bit
of a ruckus over of the Fabric of Reality mailing list. See
http://www.hpcoders.com.au/blog
Can I plug the existence of Fabric of Alternative Reality here? See
http://www.hpcoders.com.au/foar.html
--
----------------------------------------------------------------------------
Prof Russell Standish Phone 0425 253119 (mobile)
Principal, High Performance Coders
Visiting Professor of Mathematics hpc...@hpcoders.com.au
University of New South Wales http://www.hpcoders.com.au
----------------------------------------------------------------------------
Russell Standish
I've done a partial read of this paper, and already in section 5 I see
a problem.
In section 5, Eastmond attempts to derive a paradox from the
assumption of an infinite number of observer moments in a lifetime (as
might be the case with quantum immortality, for example).
He starts with a mapping between the lifetime of OMs and the rational
numbers between 0 & 1. Then he argues that in observing one's current
observer moment, determining which half of the unit interval the OM is
mapped to gives 1 bit of information. Further subdividing the interval
gives, of course, more bits of information. He then concludes that an
infinite number of bits of information is needed to specify the OM.
The paradox is derived by using Cantor's argument to show that there
are an uncountable number of infinite length bitstrings, many more
than the OMs.
The problem with this argument is that all rational numbers, when
expressed in base2, ultimately end in a repeating tail. In decimal
notation, we write dots above the digits that repeat. Once the
recurring tail has been reached, no further bits of information is
required to specify the rational number. Another way of looking at it
is that all rational numbers can be specified as two integers - a
finite amount of information.
I notice this paper is an 02 arXiv paper, so rather old. It hasn't
been through peer review AFAICT. There was a bit of a critique of it
on Math Forum, but that degenerated pretty fast.
Cheers
Stephen P. King
> On Wed, Aug 24, 2011 at 03:12:31PM -0700, David Nyman wrote:
>> This paper presents some intriguing ideas on consciousness, computation and
>> the MWI, including an argument against the possibility of consciousness
>> supervening on any single deterministic computer program (Bruno might find
>> this interesting). Any comments on its cogency?
>>
>> http://arxiv.org/abs/gr-qc/0208038
>>
>> David
>>
> I've done a partial read of this paper, and already in section 5 I see
> a problem.
>
> In section 5, Eastmond attempts to derive a paradox from the
> assumption of an infinite number of observer moments in a lifetime (as
> might be the case with quantum immortality, for example).
>
> He starts with a mapping between the lifetime of OMs and the rational
> observer moment, determining which half of the unit interval the OM is
> mapped to gives 1 bit of information. Further subdividing the interval
> gives, of course, more bits of information. He then concludes that an
> infinite number of bits of information is needed to specify the OM.
>
> The paradox is derived by using Cantor's argument to show that there
> are an uncountable number of infinite length bitstrings, many more
> than the OMs.
Exactly why are there not a continuum of OMs? It seems to me if we
parametrize the cardinality of distinct OMs to *all possible*
partitionings of the tangent spaces of physical systems (spaces wherein
the Lagrangians and Hamiltonians exist) then we obtain at least the
cardinality of the continuum. It is only if we assume some arbitrary
coarse graining that we have a countable set of OMs.
>
> The problem with this argument is that all rational numbers, when
> expressed in base2, ultimately end in a repeating tail. In decimal
> notation, we write dots above the digits that repeat. Once the
> recurring tail has been reached, no further bits of information is
> required to specify the rational number. Another way of looking at it
> is that all rational numbers can be specified as two integers - a
> finite amount of information.
I must dispute this claim because that reasoning in terms of 'two
integer' encoding of rationals ignores the vast and even infinite
apparatus required to decode the value of an arbitrary pair of
'specified by two integers' values. The same applies to the notion of
digital information. Sure, we can think that the observed universe can
be represented by some finite collection of finite bit strings, but this
is just the result of imposing an arbitrary upper and lower bound on the
resolution of the recording/describing machinery. There is no ab initio
reason why that particular upper/lower bound on resolution exists in the
first place.
>
> I notice this paper is an 02 arXiv paper, so rather old. It hasn't
> been through peer review AFAICT. There was a bit of a critique of it
> on Math Forum, but that degenerated pretty fast.
>
> Cheers
Ideas are sometimes like vine or a single malt whiskey that must age
before its bouquet is at its prime.
Onward!
Stephen
Russell Standish
>
> Exactly why are there not a continuum of OMs? It seems to me if
> we parametrize the cardinality of distinct OMs to *all possible*
> partitionings of the tangent spaces of physical systems (spaces
> wherein the Lagrangians and Hamiltonians exist) then we obtain at
> least the cardinality of the continuum. It is only if we assume some
> arbitrary coarse graining that we have a countable set of OMs.
I do not assume an arbitrary coarse graining, but do think that each OM
must contain a finite amount of information. This implies the set of
OMs is countable.
>
> >
> >The problem with this argument is that all rational numbers, when
> >expressed in base2, ultimately end in a repeating tail. In decimal
> >notation, we write dots above the digits that repeat. Once the
> >recurring tail has been reached, no further bits of information is
> >required to specify the rational number. Another way of looking at it
> >is that all rational numbers can be specified as two integers - a
> >finite amount of information.
>
> I must dispute this claim because that reasoning in terms of
> 'two integer' encoding of rationals ignores the vast and even
> infinite apparatus required to decode the value of an arbitrary pair
> of 'specified by two integers' values.
Both the human brain, and computers are capable of handling rational
numbers exactly. Neither of these are infinite apparatuses. If you're
using an arbitrary precision integer representation (eg the software
GMP), the only limitation to storing the rational number (or decoding
it, as you put it) is the amount of memory available on the computer.
The amount of information needed to represent any rational number is
finite (although may be arbitarily large, as is the case for any
integer). Only real numbers, in general, require infinite
information. Such numbers are known as uncomputable numbers.
> The same applies to the
> notion of digital information. Sure, we can think that the observed
> universe can be represented by some finite collection of finite bit
> strings, but this is just the result of imposing an arbitrary upper
> and lower bound on the resolution of the recording/describing
> machinery. There is no ab initio reason why that particular
> upper/lower bound on resolution exists in the first place.
>
It rather depends what we mean by universe. An observer moment, ISTM,
is necessarily a finite information object. Moving from one observer
moment to the next must involve a difference of at least one bit, in
order for there to be an evolution in observer moments. A history, or linear
sequence of observable moments, must therefore be a countable set of
OMs, but this could be infinite. A collection of such histories would
be a continuum.
A world (or universe), in my view, is given by a bundle of histories
satisfying a finite set of constraints. As such, an infinite amount of
information in the histories is irrelevant ("don't care bits"). But if
you'd prefer to identify the world with a unique history, or even as
something with independent existence outside of observation, then
sure, it may contain an infinite amount of information.
> >
> >I notice this paper is an 02 arXiv paper, so rather old. It hasn't
> >been through peer review AFAICT. There was a bit of a critique of it
> >on Math Forum, but that degenerated pretty fast.
> >
> >Cheers
>
> Ideas are sometimes like vine or a single malt whiskey that must
> age before its bouquet is at its prime.
>
Partly I was wondering how much effort to put into it. Unfortunately,
it appears that the author's email addresses are no longer valid, as
it would be very interesting to have him engage in our discussions.
> Onward!
>
> Stephen
>
> --
> You received this message because you are subscribed to the Google Groups "Everything List" group.
> To post to this group, send email to everyth...@googlegroups.com.
> To unsubscribe from this group, send email to everything-li...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/everything-list?hl=en.
--
Stephen P. King
On Mon, Sep 19, 2011 at 01:14:04PM -0400, Stephen P. King wrote:Exactly why are there not a continuum of OMs? It seems to me if we parametrize the cardinality of distinct OMs to *all possible* partitionings of the tangent spaces of physical systems (spaces wherein the Lagrangians and Hamiltonians exist) then we obtain at least the cardinality of the continuum. It is only if we assume some arbitrary coarse graining that we have a countable set of OMs.I do not assume an arbitrary coarse graining, but do think that each OM must contain a finite amount of information. This implies the set of OMs is countable.
Umm, how does the finiteness of the elements of a set X induce finiteness of X? I may have missed this in my studies of set theory.
The problem with this argument is that all rational numbers, when expressed in base2, ultimately end in a repeating tail. In decimal notation, we write dots above the digits that repeat. Once the recurring tail has been reached, no further bits of information is required to specify the rational number. Another way of looking at it is that all rational numbers can be specified as two integers - a finite amount of information.I must dispute this claim because that reasoning in terms of 'two integer' encoding of rationals ignores the vast and even infinite apparatus required to decode the value of an arbitrary pair of 'specified by two integers' values.Both the human brain, and computers are capable of handling rational numbers exactly. Neither of these are infinite apparatuses. If you're using an arbitrary precision integer representation (eg the software GMP), the only limitation to storing the rational number (or decoding it, as you put it) is the amount of memory available on the computer.
True, but that misses my point. Brains and Computers are not entities existing in an otherwise empty universe; we have to consider a multiplicity of mutually observing and measuring entities and the internal interpretational and representational structures thereof. Consider a simple digital camera. The images that the camera can capture are limited by the pixel resolution of the camera, this is a constraint induced by the physical design of the camera. The camera itself, as a physical object, is not limited in the detail of its properties by those intrinsic constraints. We must take care to not assume that the limits of the observational or measurement process is not assumed to be that of the system that is making the observations/measurement.
While a the measured properties of an object A as determined by object B is limited to the resolving abilities of B, this in no way is a constraint on the properties of A. To consider the properties of A one at least might consider the set of all possible measurements of A and one might notice that these involve real valued variations, say of relative position, and thus the total set of measurable information of A is infinite, at least in principle. BTW, this is one reason why the Hilbert space of a realistic QM system is infinite, even modulo linear superposition!
The amount of information needed to represent any rational number is finite (although may be arbitarily large, as is the case for any integer). Only real numbers, in general, require infinite information. Such numbers are known as uncomputable numbers.
Surely Reality is not limited to the rationals! Are we to be crypto-Pythagoreans, claiming to believe that only the rationals exist, yet still using pi, e and other irrationals without question??? If Nature is computational, does it not make sense that its computations /information accessing and processing might not be limited to the rationals?
The same applies to the notion of digital information. Sure, we can think that the observed universe can be represented by some finite collection of finite bit strings, but this is just the result of imposing an arbitrary upper and lower bound on the resolution of the recording/describing machinery. There is no ab initio reason why that particular upper/lower bound on resolution exists in the first place.It rather depends what we mean by universe. An observer moment, ISTM, is necessarily a finite information object. Moving from one observer moment to the next must involve a difference of at least one bit, in order for there to be an evolution in observer moments. A history, or linear sequence of observable moments, must therefore be a countable set of OMs, but this could be infinite. A collection of such histories would be a continuum.
I consider an Observer moment to be the content of experience on an ideal non-anthropomorphic observer that might obtain in a minimum quantity of time, thus there is a maximum quantity of energy involved, as per the energy-time uncertainty relation (which is controversial as time is not an observable per se!). If we assume that this observer is constrained by the laws of QM then its ability to communicate its information/knowledge to another via emission and/or absorption events is finite, it is "quantized", but its observational content is only constrained by the Heisenberg Uncertainty relation, a relation that does not put an upper bound on any single observable, it only constrains simultaneous measurement of pairs of canonically conjugate variables.
A world (or universe), in my view, is given by a bundle of histories
satisfying a finite set of constraints. As such, an infinite amount of
information in the histories is irrelevant ("don't care bits"). But if
you'd prefer to identify the world with a unique history, or even as
something with independent existence outside of observation, then
sure, it may contain an infinite amount of information.
I like the Kripkean notion of a world, where W is the non-empty set. SInce W is not constrained by some prior notion of elementhood, one is free to use any self-consistent set theory to define W. One can even adopt the anti-foundation axiom and have some real fun!
I notice this paper is an 02 arXiv paper, so rather old. It hasn't been through peer review AFAICT. There was a bit of a critique of it on Math Forum, but that degenerated pretty fast. CheersIdeas are sometimes like vine or a single malt whiskey that must age before its bouquet is at its prime.Partly I was wondering how much effort to put into it. Unfortunately, it appears that the author's email addresses are no longer valid, as it would be very interesting to have him engage in our discussions.
I did a search and believe that his email is now east...@yahoo.com. I am sending a CC of this post to him in hope of a response.
Onward!
Stephen
meekerdb
> [SPK]
> I consider an Observer moment to be the content of experience on an ideal
> non-anthropomorphic observer that might obtain in a minimum quantity of time, thus there
> is a maximum quantity of energy involved, as per the energy-time uncertainty relation
> (which is controversial as time is not an observable per se!). If we assume that this
> observer is constrained by the laws of QM then its ability to communicate its
> information/knowledge to another via emission and/or absorption events is finite, it is
> "quantized", but its observational content is only constrained by the Heisenberg
> Uncertainty relation, a relation that does not put an upper bound on any single
> observable, it only constrains simultaneous measurement of pairs of canonically
> conjugate variables.
You seem to be invoking the Heisenberg uncertainty backwards. What is says is:
delta-t*delta-E > hbar
not "<". So if you make delta-t small then you force delta-E > hbar/delta-t. The HUP
puts a *lower* bound on E. Or perhaps you are saying that since and observer has only a
finite amount of energy there is a limit on how big delta-E can be and hence delta-t >
hbar/max[delta-E] and this provides a lower bound on the duration of an Observer Moment.
? Of course as you note time is not an observable in QM, but one can construct quantum
mechanical clocks that provide a local measure of time and the HUP applies to them.
Brent
Stephen P. King
Thank you for pointing this out. You are correct in that I was
considering that since an observer has only a finite amount of energy
... But the same situation would occur if the observer has only a finite
duration within which to make an observation...
Onward!
Stephen
Russell Standish
> On 9/21/2011 6:41 AM, Russell Standish wrote:
> >On Mon, Sep 19, 2011 at 01:14:04PM -0400, Stephen P. King wrote:
> >> Exactly why are there not a continuum of OMs? It seems to me if
> >>we parametrize the cardinality of distinct OMs to *all possible*
> >>partitionings of the tangent spaces of physical systems (spaces
> >>wherein the Lagrangians and Hamiltonians exist) then we obtain at
> >>least the cardinality of the continuum. It is only if we assume some
> >>arbitrary coarse graining that we have a countable set of OMs.
> >I do not assume an arbitrary coarse graining, but do think that each OM
> >must contain a finite amount of information. This implies the set of
> >OMs is countable.
> [SPK]
> Umm, how does the finiteness of the elements of a set X induce
> finiteness of X? I may have missed this in my studies of set theory.
That is not what I said. Firstly, I said the set of OMs are countable, which
includes the lowest transfinite cardinal aleph_0. Also, there is more
to it. Perhaps I wasn't explicit about the fact that I consider two
OMs with the same information content to be identical. Ie, the
contained information uniquely identifies the OM.
In that case, the set of all OM can be mapped 1-1 to the set of finite
binary strings [0,1]* (I think that's how it is written). That set is
countable, so the set of all OMs must be too.
> >> I must dispute this claim because that reasoning in terms of
> >>'two integer' encoding of rationals ignores the vast and even
> >>infinite apparatus required to decode the value of an arbitrary pair
> >>of 'specified by two integers' values.
> >Both the human brain, and computers are capable of handling rational
> >numbers exactly. Neither of these are infinite apparatuses. If you're
> >using an arbitrary precision integer representation (eg the software
> >GMP), the only limitation to storing the rational number (or decoding
> >it, as you put it) is the amount of memory available on the computer.
> [SPK]
> True, but that misses my point. Brains and Computers are not
> entities existing in an otherwise empty universe; we have to
> consider a multiplicity of mutually observing and measuring entities
> and the internal interpretational and representational structures
> thereof. Consider a simple digital camera. The images that the
> camera can capture are limited by the pixel resolution of the
> camera, this is a constraint induced by the physical design of the
> camera. The camera itself, as a physical object, is not limited in
> the detail of its properties by those intrinsic constraints. We must
> take care to not assume that the limits of the observational or
> measurement process is not assumed to be that of the system that is
> making the observations/measurement.
>
Since the observable world is defined by the observer, one can't
really not take the observer into account. One can perhaps get higher order
cardinalities by looking at the boundary of that which is common to
all observers. For concreteness, consider the UD trace UD* in Bruno's
work. UD* is isomorphic to the reals - you would have to define
something like that to be your world to get uncountable things.
>
> >The amount of information needed to represent any rational number is
> >finite (although may be arbitarily large, as is the case for any
> >integer). Only real numbers, in general, require infinite
> >information. Such numbers are known as uncomputable numbers.
> [SPK]
> Surely Reality is not limited to the rationals! Are we to be
> crypto-Pythagoreans, claiming to believe that only the rationals
> exist, yet still using pi, e and other irrationals without
> question??? If Nature is computational, does it not make sense that
> its computations /information accessing and processing might not be
> limited to the rationals?
>
No, again, I didn't say that. I think of reality as being the set of
knowable things, which is necessarily countable. Various computable
numbers such as e, pi etc are definitely knowable.
John Eastmond was the one to bring up the rationals by means of a
bijection from a set of OMs. I was pointing to flaws in his use of
rational numbers (they're still countable, for instance).
Stephen P. King
> On Wed, Sep 21, 2011 at 10:08:55AM -0400, Stephen P. King wrote:
>> On 9/21/2011 6:41 AM, Russell Standish wrote:
>>> On Mon, Sep 19, 2011 at 01:14:04PM -0400, Stephen P. King wrote:
>>>> Exactly why are there not a continuum of OMs? It seems to me if
>>>> we parametrize the cardinality of distinct OMs to *all possible*
>>>> partitionings of the tangent spaces of physical systems (spaces
>>>> wherein the Lagrangians and Hamiltonians exist) then we obtain at
>>>> least the cardinality of the continuum. It is only if we assume some
>>>> arbitrary coarse graining that we have a countable set of OMs.
>>> I do not assume an arbitrary coarse graining, but do think that each OM
>>> must contain a finite amount of information. This implies the set of
>>> OMs is countable.
>> [SPK]
>> Umm, how does the finiteness of the elements of a set X induce
>> finiteness of X? I may have missed this in my studies of set theory.
> That is not what I said. Firstly, I said the set of OMs are countable, which
> includes the lowest transfinite cardinal aleph_0. Also, there is more
> to it. Perhaps I wasn't explicit about the fact that I consider two
> OMs with the same information content to be identical. Ie, the
> contained information uniquely identifies the OM.
>
> In that case, the set of all OM can be mapped 1-1 to the set of finite
> binary strings [0,1]* (I think that's how it is written). That set is
> countable, so the set of all OMs must be too.
Thank you for this explanation. There is just something about this
that is still unsettling to me. I will ponder it further.
[SPK]
How do we extend this to a countable number of seperate observers
and their interactions and communications with each other. It seems to
me, and I may be wrong, that this generates a diagonalization. The
bisimulation algebra that was developed is not closed in non-symmetric
cases.
**
Summary of basic properties:
A = A ~ A real identity bisimulation rule
B ~ C not= C ~ B non-commutativity rule; conjugate of
bisimulation not equal to itself
A ~ A = A ~ B ~ A law of real identity bisimulation (when
conjugate equal to itself)
Corollary:
A ~ A not= A ~ B ~ C ~ A by law of real identity bisimulation
A ~ A = A ~ B ~ C ~ B ~ A retractable path independence; by law of
real identity bisimulation
A ~ C not= A ~ B ~ C non-closure
**
Am I missing something?
>>> The amount of information needed to represent any rational number is
>>> finite (although may be arbitarily large, as is the case for any
>>> integer). Only real numbers, in general, require infinite
>>> information. Such numbers are known as uncomputable numbers.
>> [SPK]
>> Surely Reality is not limited to the rationals! Are we to be
>> crypto-Pythagoreans, claiming to believe that only the rationals
>> exist, yet still using pi, e and other irrationals without
>> question??? If Nature is computational, does it not make sense that
>> its computations /information accessing and processing might not be
>> limited to the rationals?
>>
> No, again, I didn't say that. I think of reality as being the set of
> knowable things, which is necessarily countable. Various computable
> numbers such as e, pi etc are definitely knowable.
[SPK]
e and pi are Computable by an endless computation... Here I have a
problem, the computation involves the use of an indefinite quantity of
resources... Not a very 'physical' situation.
> John Eastmond was the one to bring up the rationals by means of a
> bijection from a set of OMs. I was pointing to flaws in his use of
> rational numbers (they're still countable, for instance).
>
>
[SPK]
I hope that he has a comment on this point.
Onward!
Stephen
Bruno Marchal
On 21 Sep 2011, at 12:41, Russell Standish wrote:
> On Mon, Sep 19, 2011 at 01:14:04PM -0400, Stephen P. King wrote:
>>
>> Exactly why are there not a continuum of OMs? It seems to me if
>> we parametrize the cardinality of distinct OMs to *all possible*
>> partitionings of the tangent spaces of physical systems (spaces
>> wherein the Lagrangians and Hamiltonians exist) then we obtain at
>> least the cardinality of the continuum. It is only if we assume some
>> arbitrary coarse graining that we have a countable set of OMs.
>
> I do not assume an arbitrary coarse graining, but do think that each
> OM
> must contain a finite amount of information. This implies the set of
> OMs is countable.
OK. But note that in this case you are using the notion of 3-OM (or
computational state), not Bostrom notion of 1-OM (or my notion of
first person state).
The 3-OM are countable, but the 1-OMs are not.
>
>>
>>>
>>> The problem with this argument is that all rational numbers, when
>>> expressed in base2, ultimately end in a repeating tail. In decimal
>>> notation, we write dots above the digits that repeat. Once the
>>> recurring tail has been reached, no further bits of information is
>>> required to specify the rational number. Another way of looking at
>>> it
>>> is that all rational numbers can be specified as two integers - a
>>> finite amount of information.
>>
>> I must dispute this claim because that reasoning in terms of
>> 'two integer' encoding of rationals ignores the vast and even
>> infinite apparatus required to decode the value of an arbitrary pair
>> of 'specified by two integers' values.
>
> Both the human brain, and computers are capable of handling rational
> numbers exactly. Neither of these are infinite apparatuses. If you're
> using an arbitrary precision integer representation (eg the software
> GMP), the only limitation to storing the rational number (or decoding
> it, as you put it) is the amount of memory available on the computer.
>
> The amount of information needed to represent any rational number is
> finite (although may be arbitarily large, as is the case for any
> integer). Only real numbers, in general, require infinite
> information. Such numbers are known as uncomputable numbers.
OK. This of course does not prevent a machine to discover and handle
many non computable numbers. She can even generate them all, like in
finite self-duplication experiences.
>
>> The same applies to the
>> notion of digital information. Sure, we can think that the observed
>> universe can be represented by some finite collection of finite bit
>> strings, but this is just the result of imposing an arbitrary upper
>> and lower bound on the resolution of the recording/describing
>> machinery. There is no ab initio reason why that particular
>> upper/lower bound on resolution exists in the first place.
>>
>
> It rather depends what we mean by universe. An observer moment, ISTM,
> is necessarily a finite information object. Moving from one observer
> moment to the next must involve a difference of at least one bit, in
> order for there to be an evolution in observer moments. A history,
> or linear
> sequence of observable moments, must therefore be a countable set of
> OMs, but this could be infinite. A collection of such histories would
> be a continuum.
OK. And they define the structure of the 1-OMs.
>
> A world (or universe), in my view, is given by a bundle of histories
> satisfying a finite set of constraints. As such, an infinite amount of
> information in the histories is irrelevant ("don't care bits").
It might be for the 1-OM measure problem.
Bruno
> But if
> you'd prefer to identify the world with a unique history, or even as
> something with independent existence outside of observation, then
> sure, it may contain an infinite amount of information.
>
>>>
>>> I notice this paper is an 02 arXiv paper, so rather old. It hasn't
>>> been through peer review AFAICT. There was a bit of a critique of it
>>> on Math Forum, but that degenerated pretty fast.
>>>
>>> Cheers
>>
>> Ideas are sometimes like vine or a single malt whiskey that must
>> age before its bouquet is at its prime.
>>
>
> Partly I was wondering how much effort to put into it. Unfortunately,
> it appears that the author's email addresses are no longer valid, as
> it would be very interesting to have him engage in our discussions.
>
>> Onward!
>>
>> Stephen
>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Everything List" group.
>> To post to this group, send email to everything-
>> li...@googlegroups.com.
>> To unsubscribe from this group, send email to everything-li...@googlegroups.com
>> .
>> For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
>> .
>
> --
>
> ----------------------------------------------------------------------------
> Prof Russell Standish Phone 0425 253119 (mobile)
> Principal, High Performance Coders
> Visiting Professor of Mathematics hpc...@hpcoders.com.au
> University of New South Wales http://www.hpcoders.com.au
> ----------------------------------------------------------------------------
>
> --
> You received this message because you are subscribed to the Google
> Groups "Everything List" group.
> To post to this group, send email to everyth...@googlegroups.com.
> To unsubscribe from this group, send email to everything-li...@googlegroups.com
> .
> For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
> .
>
Russell Standish
>
> OK. But note that in this case you are using the notion of 3-OM (or
> computational state), not Bostrom notion of 1-OM (or my notion of
> first person state).
> The 3-OM are countable, but the 1-OMs are not.
Could you explain more why you think this? AFAICT, Bostrom makes no
mention of the cardinality of his OMs.
Bruno Marchal
On 01 Oct 2011, at 09:31, Russell Standish wrote:
> On Thu, Sep 22, 2011 at 07:02:28PM +0200, Bruno Marchal wrote:
>>
>> OK. But note that in this case you are using the notion of 3-OM (or
>> computational state), not Bostrom notion of 1-OM (or my notion of
>> first person state).
>> The 3-OM are countable, but the 1-OMs are not.
>
> Could you explain more why you think this? AFAICT, Bostrom makes no
> mention of the cardinality of his OMs.
I don't think that Bostrom mentions the cardinality of his OMs,
indeed. I don't think that he clearly distinguish the 1-OMs and the 3-
OMs either. By "3-OM" I refer to the computational state per se, as
defined relatively to the UD deployment (UD*). Those are clearly
infinite and countable, even recursively countable.
The 1-OMs, for any person, are not recursively countable, indeed by an
application of a theorem of Rice, they are not even 3-recognizable. Or
more simply because you cannot know your substitution level. In front
of some portion of UD*, you cannot recognize your 1-OMs in general.
You cannot say "I am here, and there, etc." But they are (non
constructively) well defined. "God" can know that you are here, and
there, ... And the measure on the 1-OMs should be defined on those
unrecognizable 1-OMs.
Are the 1-OMs countable? In the quote above, I say that they are not
countable. What I meant by this is related to the measure problem,
which cannot be made on the states themselves, but, I think, on the
computational histories going through them, and, actually, on *all*
computational histories going through them. This includes the dummy
histories which duplicate you iteratively through some processes
similar to the infinite iteration of the WM self-duplication. Even if
you don't interact with the output (here: W or M) or the iteration,
such computations multiplies in the non-countable infinity. (I am
using implictly the fist person indeterminacy, of course). Those
computation will have the shape:
you M
you M
you W
you M
You W
You W
You W
You M
ad infinitum
This gives a white noise, which is not necessarily available to you,
but it still multiplies (in the most possible dumb way) your
computational histories. Such infinite computations, which are somehow
dovetailing on the reals (infinite sequence of W and M) have a higher
measure than any finite computations and so are good candidates for
the "winning" computations. Note that such an infinite background
noise, although not directly accessible through your 1-OMs, should be
experimentally detectable when you look at yourselves+neighborhood
below the substitution level, and indeed QM confirms this by the many
(up + down) superposition states of the particles states in the
(assumed to be infinite) multi-universes.
This might be also confirmed by some possible semantics for the logic
of the first person points of view (the quantified logic qS4Grz1, qX1*
have, I think, non countable important models).
3-OMs are relatively simple objects, but 1-OMs are more sophisticated,
and are defined together with the set of all computations going
through their correspondent states.
To be sure, I am not entirely persuaded that Bostrom's 1-OMs makes
sense with digital mechanism, and usually I prefer to use the label of
first person experiences/histories. With the rule Y = II, that is: a
bifurcation of a computations entails a doubling of the measure even
on its "past" (in the UD steps sense), this makes clear that we have a
continuum of infinite histories.
Again, this is made more complex when we take amnesia and fusion of
histories) into consideration.
I hope this helps a bit. In my opinion, only further progress on the
"hypostases" modal logics will make it possible to isolate a
reasonable definition of 1-OMs, which obviously is a quite intricate
notion.
Bruno
>
> --
>
> ----------------------------------------------------------------------------
> Prof Russell Standish Phone 0425 253119 (mobile)
> Principal, High Performance Coders
> Visiting Professor of Mathematics hpc...@hpcoders.com.au
> University of New South Wales http://www.hpcoders.com.au
> ----------------------------------------------------------------------------
>
meekerdb
>
> On 01 Oct 2011, at 09:31, Russell Standish wrote:
>
>> On Thu, Sep 22, 2011 at 07:02:28PM +0200, Bruno Marchal wrote:
>>>
>>> OK. But note that in this case you are using the notion of 3-OM (or
>>> computational state), not Bostrom notion of 1-OM (or my notion of
>>> first person state).
>>> The 3-OM are countable, but the 1-OMs are not.
>>
>> Could you explain more why you think this? AFAICT, Bostrom makes no
>> mention of the cardinality of his OMs.
>
> I don't think that Bostrom mentions the cardinality of his OMs, indeed. I don't think
> clearly infinite and countable, even recursively countable.
>
> The 1-OMs, for any person, are not recursively countable, indeed by an application of a
> theorem of Rice, they are not even 3-recognizable. Or more simply because you cannot
> know your substitution level. In front of some portion of UD*, you cannot recognize your
> 1-OMs in general. You cannot say "I am here, and there, etc." But they are (non
> constructively) well defined. "God" can know that you are here, and there, ...
Wouldn't that require that all the infinite UD calculations be completed before all the
"you" could be indentified?
But aside from the quantum level, doesn't the measure problem have the same drawback and
Boltzman's brains. Shouldn't I find myself in a world where everyone is Brent Meeker?
Russell Standish
>
> On 01 Oct 2011, at 09:31, Russell Standish wrote:
>
> >On Thu, Sep 22, 2011 at 07:02:28PM +0200, Bruno Marchal wrote:
> >>
> >>OK. But note that in this case you are using the notion of 3-OM (or
> >>computational state), not Bostrom notion of 1-OM (or my notion of
> >>first person state).
> >>The 3-OM are countable, but the 1-OMs are not.
> >
> >Could you explain more why you think this? AFAICT, Bostrom makes no
> >mention of the cardinality of his OMs.
>
> I don't think that Bostrom mentions the cardinality of his OMs,
> indeed. I don't think that he clearly distinguish the 1-OMs and the
> as defined relatively to the UD deployment (UD*). Those are clearly
> infinite and countable, even recursively countable.
>
> The 1-OMs, for any person, are not recursively countable, indeed by
> an application of a theorem of Rice, they are not even
> 3-recognizable. Or more simply because you cannot know your
> substitution level. In front of some portion of UD*, you cannot
> recognize your 1-OMs in general. You cannot say "I am here, and
> there, etc." But they are (non constructively) well defined. "God"
> can know that you are here, and there, ... And the measure on the
> 1-OMs should be defined on those unrecognizable 1-OMs.
>
I'm still struggling to understand what you mean by 1-OM here. Are you
talking about the infinite histories making up UD*? There are an
uncountable number of these, it is true.
But then, I wouldn't call these OMs. An OM must surely be related to
the set of all such histories passing through your current "here and
now". Such things, I am convinced, must be countable, implying that
each such sets histories is a continuum.
Bruno Marchal
On 10/1/2011 8:15 AM, Bruno Marchal wrote:On 01 Oct 2011, at 09:31, Russell Standish wrote:On Thu, Sep 22, 2011 at 07:02:28PM +0200, Bruno Marchal wrote:OK. But note that in this case you are using the notion of 3-OM (orcomputational state), not Bostrom notion of 1-OM (or my notion offirst person state).The 3-OM are countable, but the 1-OMs are not.Could you explain more why you think this? AFAICT, Bostrom makes nomention of the cardinality of his OMs.I don't think that Bostrom mentions the cardinality of his OMs, indeed. I don't think that he clearly distinguish the 1-OMs and the 3-OMs either. By "3-OM" I refer to the computational state per se, as defined relatively to the UD deployment (UD*). Those are clearly infinite and countable, even recursively countable.The 1-OMs, for any person, are not recursively countable, indeed by an application of a theorem of Rice, they are not even 3-recognizable. Or more simply because you cannot know your substitution level. In front of some portion of UD*, you cannot recognize your 1-OMs in general. You cannot say "I am here, and there, etc." But they are (non constructively) well defined. "God" can know that you are here, and there, ...
Wouldn't that require that all the infinite UD calculations be completed before all the "you" could be indentified?
And the measure on the 1-OMs should be defined on those unrecognizable 1-OMs.Are the 1-OMs countable? In the quote above, I say that they are not countable. What I meant by this is related to the measure problem, which cannot be made on the states themselves, but, I think, on the computational histories going through them, and, actually, on *all* computational histories going through them. This includes the dummy histories which duplicate you iteratively through some processes similar to the infinite iteration of the WM self-duplication. Even if you don't interact with the output (here: W or M) or the iteration, such computations multiplies in the non-countable infinity. (I am using implictly the fist person indeterminacy, of course). Those computation will have the shape:you Myou Myou Wyou MYou WYou WYou WYou Mad infinitumThis gives a white noise, which is not necessarily available to you, but it still multiplies (in the most possible dumb way) your computational histories. Such infinite computations, which are somehow dovetailing on the reals (infinite sequence of W and M) have a higher measure than any finite computations and so are good candidates for the "winning" computations. Note that such an infinite background noise, although not directly accessible through your 1-OMs, should be experimentally detectable when you look at yourselves+neighborhood below the substitution level, and indeed QM confirms this by the many (up + down) superposition states of the particles states in the (assumed to be infinite) multi-universes.
But aside from the quantum level, doesn't the measure problem have the same drawback and Boltzman's brains. Shouldn't I find myself in a world where everyone is Brent Meeker?
This might be also confirmed by some possible semantics for the logic of the first person points of view (the quantified logic qS4Grz1, qX1* have, I think, non countable important models).3-OMs are relatively simple objects, but 1-OMs are more sophisticated, and are defined together with the set of all computations going through their correspondent states.To be sure, I am not entirely persuaded that Bostrom's 1-OMs makes sense with digital mechanism, and usually I prefer to use the label of first person experiences/histories. With the rule Y = II, that is: a bifurcation of a computations entails a doubling of the measure even on its "past" (in the UD steps sense), this makes clear that we have a continuum of infinite histories.Again, this is made more complex when we take amnesia and fusion of histories) into consideration.I hope this helps a bit. In my opinion, only further progress on the "hypostases" modal logics will make it possible to isolate a reasonable definition of 1-OMs, which obviously is a quite intricate notion.Bruno------------------------------------------------------------------------------Prof Russell Standish Phone 0425 253119 (mobile)Principal, High Performance CodersVisiting Professor of Mathematics hpc...@hpcoders.com.auUniversity of New South Wales http://www.hpcoders.com.au----------------------------------------------------------------------------
Bruno Marchal
Only those going through some computational state being mine, from my
points of view. It looks like a relativistic cone, except that futures
might be more numerous than past. Now the 1-OM is the subjective part
of this: it is an indexical, whose logics will obey the modalities
having a connection with truth (like Bp & p, and Bp & Dy & p).
>
> But then, I wouldn't call these OMs. An OM must surely be related to
> the set of all such histories passing through your current "here and
> now".
Yes. And there are non countably many such histories.
> Such things, I am convinced, must be countable, implying that
> each such sets histories is a continuum.
The states are countable, but not the (3-)states + the neighborhhood
of (infinite) computations that you are mentioning yourselves.
Not sure if I see where is the problem. It seems that you have
answered it. The 1-OMs *are* set of histories, but with a particular 3-
state, single out in the indexical way, and which will play the role
of the "Bp". The "& p" will force the logic of the computational
extensions to be different.
Bruno
Russell Standish
>
> The states are countable, but not the (3-)states + the neighborhhood
> of (infinite) computations that you are mentioning yourselves.
> Not sure if I see where is the problem. It seems that you have
> answered it. The 1-OMs *are* set of histories, but with a particular
> role of the "Bp". The "& p" will force the logic of the
> computational extensions to be different.
The way I was talking about it, there is a 1:1 correspondence between
the 3-states and the sets of histories making up the 1-OM. In that
case the cardinality of 1-OM is the same as that of the 3-states -
which you have already admitted is countable.
Perhaps I'm missing something? I don't quite get the "indexical" bit
for instance.
Cheers
Brian Tenneson
"It is my contention that the only way out of this dilemma is to deny the
initial assumption that a classical computer running a particular
program can
generate conscious awareness in the first place."
What about the possibility of allowing for a "large number" of conscious
moments that would, in a limit of some sort, approximate continuous,
conscious awareness? In my mind, I liken the comparison to that of a
radioactive substance and half-life decay formulas. In truth, there are
finitely many atoms decaying but the half-life decay formulas never
acknowledge that at some point the predicted mass of what's left
measures less than one atom. So I'm talking about a massive number of
calculated conscious moments so that for all intents and purposes,
continuous conscious awareness is the observed result.
Earlier on page 17...
"its program must
only generate a finite sequence of conscious moments."
Cheers
Brian
Bruno Marchal
On 04 Oct 2011, at 01:00, Russell Standish wrote:
> On Mon, Oct 03, 2011 at 05:31:21PM +0200, Bruno Marchal wrote:
>>
>> The states are countable, but not the (3-)states + the neighborhhood
>> of (infinite) computations that you are mentioning yourselves.
>> Not sure if I see where is the problem. It seems that you have
>> answered it. The 1-OMs *are* set of histories, but with a particular
>> 3-state, single out in the indexical way, and which will play the
>> role of the "Bp". The "& p" will force the logic of the
>> computational extensions to be different.
>
> The way I was talking about it, there is a 1:1 correspondence between
> the 3-states and the sets of histories making up the 1-OM. In that
> case the cardinality of 1-OM is the same as that of the 3-states -
> which you have already admitted is countable.
OK, I see your point. You are right on this, and I should have
perhapssaid "set of sets of histories". This is related to the
possible semantics of the first person logics (S4Grz1).
The 1-OMs are mutiplied by the computations going through it, making
it as great as the continuum of those computations going through, and
that can be understood intuitively by UDA-like reasoning. This come
from the rule Y = I. To have the measure on the 1-OMs, we have to
count the computations going through, not the states themselves. So an
1-OMs can be defined just by one computations (perhaps infinite), but
this does not entirely work, and that is why we have to take into
account the structure imposed by the logic to which the first person
obeys.
>
> Perhaps I'm missing something? I don't quite get the "indexical" bit
> for instance.
Examples of "indexical" are terms like "now, me, here, I, there".
Their meaning or referent depends on the locutor or of its current
(also an indexical) situation.
The 3-I of a machine, is an indexical, rather well handled by the
description of the machine as handled by the machine, like with
Gödel's beweisbar "B". The precise (arithmetical contento of "B"
varies from one machine to another, but if the machine verifies some
conditions (rich, ideally correct, etc.) it obeys the same modal
logics (G, G*).
The "1-I" is similarly well captured by the conjunction of the 3-I and
truth (Bp & p). This is a drastic change, because that "I" is no
describable by the machine, it is *not* arithmetical, and it changes
the logic of self-reference (which becomes a logic of evolving states).
Sorry for having been unclear, but I continue to doubt about the
relevance of the term OMs terms. It did mislead me in my answer too
you, and I should stick on the person-views and their modalities.
Ask for any clarification. Nothing is really simple here. I will add
some info in my reply to Brent.
Bruno
Bruno Marchal
I think I agree with you. I think that such a view is the only
compatible with Digital Mechanism, but also with QM (without collapse).
Consciousness is never generated by the "running of a particular
computer". If we can survive with a digital brain, this is related to
the fact that we already "belong" to an infinity of computations, and
the artificial brain just preserve that infinity, in a way such that I
can survive in my usual normal (Gaussian) neighborhoods.
Bruno
Brian Tenneson
Digital brain. What's a brain? I ask because I'm betting it doesn't
mean a pile of gray and white matter.
Then you mention artificial brain. That's different from digital? Is
digital more nonphysical than artificial?
Bruno Marchal
On 04 Oct 2011, at 23:14, Brian Tenneson wrote:
> Hmm... Unfortunately there are several terms there I don't understand.
> Digital brain. What's a brain? I ask because I'm betting it doesn't
> mean a pile of gray and white matter.
Suppose that you have a brain disease, and you doctor propose to you
an artificial brain, and he does not hide that this mean he will copy
your brain state at the level of the molecules, processed by a
computer. he adds that you can choose between a mac or a pc.
Comp assumes that there is a level such that you can survive in the
usual clinical sense with such a digital brain like you can already
survive with an artificial pump at the place of the heart.
> Then you mention artificial brain. That's different from digital?
Well, it could be for those studying an analog version of comp. But
unless the analog system use actual infinities, it will be emulable by
a digital machine. The redundancy of the brains and its evolution
pleads for the idea that the brain is indeed digitally emulable.
> Is
> digital more nonphysical than artificial?
Not a priori, at all. Sellable computers are digital and physical.
Today the non physical universal machines are still free, and can be
found in books or on the net. You might find a lot by looking toward
yourself, but the study of computer science can accelerate that
discovery a lot.
Bruno
>> To post to this group, send email to everything-
>> li...@googlegroups.com.
>> To unsubscribe from this group, send email to
>> everything-li...@googlegroups.com.
>> For more options, visit this group at
>> http://groups.google.com/group/everything-list?hl=en.
>>
>>
>
> --
> You received this message because you are subscribed to the Google
> Groups "Everything List" group.
> To post to this group, send email to everyth...@googlegroups.com.
> To unsubscribe from this group, send email to everything-li...@googlegroups.com
> .
> For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
> .
>
Brian Tenneson
It's interesting that you bring up computer science as I am doing a
career change right now and am going into computer science. I
eventually want to work in brain simulation. A lot of the ideas in
this group are relevant.
From the paper, I'll quote again (mainly for myself when I look back
at this message)
From page 17
"It is my contention that the only way out of this dilemma is to deny the
initial assumption that a classical computer running a particular program can
generate conscious awareness in the first place."
If the author is correct that would seem to drive a nail in the coffin
for the digital generation of conscious awareness though in some way
that might not prove that brain simulation is impossible. Perhaps
brain simulation would occur in such a way that the simulation is
never consciously self-aware but if that were the case, how good is
that simulation??
If my doctor wanted to replace my brain with an artificial brain, I
think I'd be scared out of my mind if LINUX wasn't an option hehe...
Thanks Bruno.
I know this might seem like a naive observation but the Bolshoi
universe simulation recently done on a supercomputer at UC Santa Cruz
in California produced some images of an early universe that had an
uncanny resemblance to the human brain. It gives me hope that it is
possible to simulate a brain on a classical computer. Perhaps the
details would involve highly complex neural networks; the hope would
be to rival the complexity of an actual brain.
Here is a link that includes video
http://hipacc.ucsc.edu/Bolshoi/
(Then of course we might get into some ethical quandaries regarding
the personhood of a simulated brain such as can we run any experiment
on it that we feel like running... is simulated suffering ethically
equivalent to actual suffering... and that sort of thing.)
>>> To post to this group, send email to everyth...@googlegroups.com.
>>> To unsubscribe from this group, send email to
>>> everything-li...@googlegroups.com.
>>> For more options, visit this group at
>>> http://groups.google.com/group/everything-list?hl=en.
>>>
>>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "Everything List" group.
>> To post to this group, send email to everyth...@googlegroups.com.
>> To unsubscribe from this group, send email to
>> everything-li...@googlegroups.com.
>> For more options, visit this group at
>> http://groups.google.com/group/everything-list?hl=en.
>>
>
> http://iridia.ulb.ac.be/~marchal/
>
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "Everything List" group.
> To post to this group, send email to everyth...@googlegroups.com.
> To unsubscribe from this group, send email to
Bruno Marchal
On 07 Oct 2011, at 01:59, Brian Tenneson wrote:
> Thanks Bruno for patiently explaining things.
>
> It's interesting that you bring up computer science as I am doing a
> career change right now and am going into computer science. I
> eventually want to work in brain simulation. A lot of the ideas in
> this group are relevant.
Thanks.
>
> From the paper, I'll quote again (mainly for myself when I look back
> at this message)
> From page 17
> "It is my contention that the only way out of this dilemma is to
> deny the
> initial assumption that a classical computer running a particular
> program can
> generate conscious awareness in the first place."
>
> If the author is correct that would seem to drive a nail in the coffin
> for the digital generation of conscious awareness though in some way
> that might not prove that brain simulation is impossible.
Yes. the expression is ambiguous.
> Perhaps
> brain simulation would occur in such a way that the simulation is
> never consciously self-aware but if that were the case, how good is
> that simulation??
That would lead to zombie. Still, I don't believe any particular
implementation of a computation "generates awareness" by itself
(neither in a physical universe, nor in a immaterial arithmetical
dovetailing). Awareness needs all implementation of all computations,
as it follows from the step 8 in UDA.
When I say yes to the doctor, I might survive in the usual sense, but
this does not mean that the artificial brain generates my
consciousness (which is more an heaven kind of object). but the
artificial brain, if well done enough, might make it possible for my
mind (existing only in hevan) to continue to manisfest itself here (on
earth), like my brain seems to already be able to do.
It is a subtle point, but if our bodies are machine, we provably have
an independent soul, and machines (silicon or carbon based) just makes
it possible fro a soul to manifest itself with respect to other souls
with reasonable probabilities.
>
> If my doctor wanted to replace my brain with an artificial brain, I
> think I'd be scared out of my mind if LINUX wasn't an option hehe...
> Thanks Bruno.
All right, but then everyone can get "your code source", and your fist
person indeterminacy might grow a lot. Expect to find your self in the
nightmarish fantasy of your neighbors. Be careful :)
>
> I know this might seem like a naive observation but the Bolshoi
> universe simulation recently done on a supercomputer at UC Santa Cruz
> in California produced some images of an early universe that had an
> uncanny resemblance to the human brain.
It is the filamentous web of cluster of galaxies, using information
from Hubble and COBE, I think. It is very impressive and shows how big
the physical cosmos is.
I think that comp implies that the cosmos is infinite. The cosmos is
the border of an infinite universal mind, and an infinity of
computations plays some role. But this is hard to prove, because comp
can also collapse, from the first person views, or renormalize, many
infinities. The cosmos is a priori infinite, but some weird
computational phenomenon collapsing infinities are hard to avoid
especially before we understand better why the 'white rabbits' are so
rare in our neighborhoods.
> It gives me hope that it is
> possible to simulate a brain on a classical computer. Perhaps the
> details would involve highly complex neural networks;
Don't forget the glial cells. They are 20 times more numerous than
neurons, and we know that they don't not only communicate (by chemical
waves instead of ionic electricity) between themselves, but they do
communicate with the neurons also (this plays a role notably in the
chronic pains). They might be needed for the conscious background.
> the hope would
> be to rival the complexity of an actual brain.
Good luck. It will be easier to copy highly "plastic" brain (like
baby's brain) and let them organize themselves that to actually copy
an adult brain, which contains tremendous amount of distributed
information.
>
> Here is a link that includes video
> http://hipacc.ucsc.edu/Bolshoi/
It is beautiful.
Have you look to this nice video (by SpaceRip):
http://www.youtube.com/watch?v=CEQouX5U0fc
Ah, but you can find impressive filamentous structure in the
Mandelbrot set too, and even without digging deep:
http://www.youtube.com/watch?v=9G6uO7ZHtK8
>
> (Then of course we might get into some ethical quandaries regarding
> the personhood of a simulated brain such as can we run any experiment
> on it that we feel like running... is simulated suffering ethically
> equivalent to actual suffering... and that sort of thing.)
With comp, simulated suffering is the same as suffering, and should be
forbidden, unless someone accept it for its own brain, and this before
doing the copy. (Like I think you have the right to kill or even
torture yourself, as far as you are not making other suffering).
The very complex case, is when artificial sexual dolls, like it
already exists in Japan, will fight for their right. They are
programmed to fake suffering, so we will meet the complex zombie
border there ... Some "intelligent" (conscious) machines will be
descendent of zombie toys.
Bruno
>> To post to this group, send email to everything-
>> li...@googlegroups.com.
>> To unsubscribe from this group, send email to
>> everything-li...@googlegroups.com.
>> For more options, visit this group at
>> http://groups.google.com/group/everything-list?hl=en.
>>
>>
>
> --
> You received this message because you are subscribed to the Google
> Groups "Everything List" group.
> To post to this group, send email to everyth...@googlegroups.com.
> To unsubscribe from this group, send email to everything-li...@googlegroups.com
> .
> For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
> .
>