Information Laws
Ed Fredkin
that they capture aspects of a dynamic world in static formulae.
Newton's Laws are good examples. On the other hand, digital
informational processes (what goes on in the innards of a computer)
are also subject to certain laws. For example, information in the
memory of a computer must have a means of its representation (bits and
bytes and computer files). The meaning of the information is defined
by the computer processes associated with that information. An MPEG2
file can be interpreted by a program to be a movie and an MP3 file can
be interpreted to be a song. A bunch of bits cannot have a meaning
independent of some associated digital informational process; whether
that process takes place in a computer or in the mind of an observer.
We might define rigorous concepts of "temporal process", "purpose" and
"meaning" by associating binary data formats with the processes that
interpret and modify those bits.
Every dynamic physical system can be thought of as representing
information. From our perspective the amount of information has to do
with the precision with which we can imagine measurements to be made.
The meanings commonly involve positions and velocities. Physical
processes can cause positions and velocities to evolve over time.
Thus the meaning inherent in such physical systems is a consequence of
the physics that controls the temporal evolution. But this is
essentially the same kind of process, when viewed from an
informational perspective, as what goes on in the innards of a
computer. Thus it seems reasonable to conclude that there are laws of
physics that are related to laws of informational systems.
Newton's and Einstein's idea that things simply move is the physics
equivalent of biologists of that era stating that "like begets like"
as a sufficient explanation for why chickens hatch out of chicken
eggs. Today, with our newly gained familiarity with digital
informational processes, we need not doubt that positions and
velocities involves information, just as the genetic material in a
chicken egg involves digital information. There is no doubt that the
temporal evolution of the positions and velocities of a particle is
analogous to a digital informational process, just as the
transformation of a fertilized chicken egg into a chick ready to hatch
also is a temporal process similar to a digital informational process.
Surprisingly, biology has led the way by showing us the possibility of
understanding the complex processes of genetics and life as the
evolution of systems subject to the laws of digital informational
systems. On the other hand the temporal evolution of the simplest
dynamical systems, such as particles in motion in the absence of a
field, remain shrouded in the mystic belief that "…things simply
move…".
Something should be obvious to everyone who can think about this while
keeping in mind the perspective of digital informational processes yet
de novo from the historical perspective of physics. Relativity (both
Newtonian and Einsteinian) is an informational impossibility. It
doesn't matter that the math works. It doesn't matter that we haven't
been able to measure absolute direction or absolute motion. It
doesn't matter how esthetically pleasing it is to assume the existence
of fundamental continuous symmetries. It is time to wake up to the
fact that there is no possibility of any informational process
carrying out: what physics purports to do with rotation, translation
and other continuous symmetries. That these symmetries must be
violated by microscopic discrete processes is actually the only
reasonable conclusion one can reach based on what we now know about
informational processes. To have believed otherwise can be understood
and excused; to doggedly persist is tantamount to believing in magic.
Ed F
Tim Tyler
> Something should be obvious to everyone who can think about this while
> keeping in mind the perspective of digital informational processes yet
> de novo from the historical perspective of physics. Relativity (both
> Newtonian and Einsteinian) is an informational impossibility. It
> doesn't matter that the math works. It doesn't matter that we haven't
> been able to measure absolute direction or absolute motion. It
> doesn't matter how esthetically pleasing it is to assume the existence
> of fundamental continuous symmetries. It is time to wake up to the
> fact that there is no possibility of any informational process
> carrying out: what physics purports to do with rotation, translation
> and other continuous symmetries. That these symmetries must be
> violated by microscopic discrete processes is actually the only
> reasonable conclusion one can reach based on what we now know about
> informational processes. To have believed otherwise can be understood
> and excused; to doggedly persist is tantamount to believing in magic.
Physics doesn't have to obey our notions of what is
"informationally possible".
It defines our notions of what is "informationally possible" - but
there's no reason to /assume/ that it is *limited* by them.
If you adopt some sort of "Finite Nature" principle - then it does
indeed follow that relativity is incorrect - but such a principle
is not supported by any evidence.
Doubting that "Finite Nature" is true is a bit different from
believing in magic - IMO.
We have no evidence to speak of - in either direction - about the
finiteness of nature.
Consequently, I reckon we should keep an open mind on the issue.
--
__________
|im |yler http://timtyler.org/ t...@tt1lock.org Remove lock to reply.
akha...@yahoo.com
One still has to provide a convincing interpretation of physical
notions in terms of information dynamics.
> Newton's and Einstein's idea that things simply move is the physics
> equivalent of biologists of that era stating that "like begets like"
> as a sufficient explanation for why chickens hatch out of chicken
> eggs. Today, with our newly gained familiarity with digital
> informational processes, we need not doubt that positions and
> velocities involves information, just as the genetic material in a
> chicken egg involves digital information. There is no doubt that the
> temporal evolution of the positions and velocities of a particle is
> analogous to a digital informational process, just as the
> transformation of a fertilized chicken egg into a chick ready to hatch
> also is a temporal process similar to a digital informational process.
>
> Surprisingly, biology has led the way by showing us the possibility of
> understanding the complex processes of genetics and life as the
> evolution of systems subject to the laws of digital informational
> systems. On the other hand the temporal evolution of the simplest
> dynamical systems, such as particles in motion in the absence of a
> field, remain shrouded in the mystic belief that "?things simply
> move?".
There is no mystic in Newton's mechanics. In computer simulation (like
lattice gas CA) the "things simply move" because we postulate the
continuation of motion "in the absence of a field" like it was done by
Newton and nothing more. The postulated framework is well defined in
both cases and quite different: continuous or discrete.
> Something should be obvious to everyone who can think about this while
> keeping in mind the perspective of digital informational processes yet
> de novo from the historical perspective of physics. Relativity (both
> Newtonian and Einsteinian) is an informational impossibility. It
> doesn't matter that the math works. It doesn't matter that we haven't
> been able to measure absolute direction or absolute motion. It
> doesn't matter how esthetically pleasing it is to assume the existence
> of fundamental continuous symmetries. It is time to wake up to the
> fact that there is no possibility of any informational process
> carrying out: what physics purports to do with rotation, translation
> and other continuous symmetries. That these symmetries must be
> violated by microscopic discrete processes is actually the only
> reasonable conclusion one can reach based on what we now know about
> informational processes. To have believed otherwise can be understood
> and excused; to doggedly persist is tantamount to believing in magic.
Nothing is obvious (even causality does not have absolute value in
physical theory for now). Any theory is just a set of postulates and
what is following from them. People do want to see something practical
in physics and it does matter that the math works. Not too many
physicists do care about what is "esthetically pleasing" to "radical
finitists" (you have seen it in the last 40 years).
To challenge the paradigm in physics it is important to provide not
only compelling theory but also experimental facts, which would
contradict current theories. Such facts probably are not limited to
direct observation of extinction of continuous symmetries (a little
bit more developed discrete picture would give something else).
Alex
Ed Fredkin
> Ed Fredkin <edfr...@yahoo.com> wrote or quoted:
>
> > Something should be obvious to everyone who can think about this while
> > keeping in mind the perspective of digital informational processes yet
> > de novo from the historical perspective of physics. Relativity (both
> > Newtonian and Einsteinian) is an informational impossibility. It
> > doesn't matter that the math works. It doesn't matter that we haven't
> > been able to measure absolute direction or absolute motion. It
> > doesn't matter how esthetically pleasing it is to assume the existence
> > of fundamental continuous symmetries. It is time to wake up to the
> > fact that there is no possibility of any informational process
> > carrying out: what physics purports to do with rotation, translation
> > and other continuous symmetries. That these symmetries must be
> > violated by microscopic discrete processes is actually the only
> > reasonable conclusion one can reach based on what we now know about
> > informational processes. To have believed otherwise can be understood
> > and excused; to doggedly persist is tantamount to believing in magic.
>
> Physics doesn't have to obey our notions of what is
> "informationally possible".
>
> It defines our notions of what is "informationally possible" - but
> there's no reason to /assume/ that it is *limited* by them.
When we discover a law, such as the law of conservation of momentum,
it is reasonable to assume that physics does have to obey that law and
further that it is indeed *limited* by that law. It doesn't take a
belief in the "Finite Nature" principle to come to the conclusion that
there are laws of physics that are related to the informational nature
of systems. For example, we know that systems that have reversible
laws, must contain enough information in the state of the system to
both determine the future state and the past state. This is a kind of
mathematical fact, certainly true in this universe and like many
mathematical facts, true in other kinds of non-physical systems that
we can imagine.
We can think of information as something that is represented by the
arrangement of things where there is a process that gives meaning to
the information. We can define a process as any system where the
arrangement of things evolves over time. We are certainly talking
about things within the scope of physics.
"Magic" in the sense used here means something we imagine for which
there is no physically realizable process. In other words, "magic"
refers to things we might imagine which violate accepted laws of
physics. An example would be a 300 HP automobile engine that is a
perpetual motion machine; needing no fuel or other source of energy.
Another law of macroscopic physics is that, in principle, one
informational process can model aspects of a physical process. Given
a solar system we can always build or program an orrery to arbitrary
precision.
However, what cannot be done is to build (or even imagine) our orrery
without the informational equivalent of a single fixed reference
frame. It's true that we have been unable to detect such a thing
(other than the CBR). Yet the idea of promoting the observation that
"...any unnaccelerated reference frame will do..." to the dogma that
"...there cannot be a single fixed reference frame..." is cute but not
necessarily required as a consequence of any set of experimentally
determined observations.
>
> If you adopt some sort of "Finite Nature" principle - then it does
> indeed follow that relativity is incorrect - but such a principle
> is not supported by any evidence.
I agree that FN implies that relativity is incorrect, but I do not
agree that it is necessary to assume FN in order to question
contemporary concepts of motion.
>
> Doubting that "Finite Nature" is true is a bit different from
> believing in magic - IMO.
Of course.
>
> We have no evidence to speak of - in either direction - about the
> finiteness of nature.
I disagree here. There's plenty of evidence; it's just not recognized
as such. While the FN assumption awaits new experimental evidence in
order to convince the physics community, if that evidence is found and
views change, it won't take long for everyone to wonder why they
didn't figure it out sooner. There's plenty of evidence, it's just
that we've built an elaborate mathematical scaffold over it that
obscures its true nature.
>
> Consequently, I reckon we should keep an open mind on the issue.
Insofar as the question of the rigorous belief in the reality of
continuous symmetries I agree on the "...open mind..." part, but not
on the "...keep an open mind...". We have all trained ourselves to
not have an open mind on this subject. With an truly open mind we
could realize now that all apparently continuous symmetries must be
violated. Minds are so closed as to exclude, as not worth doing, any
experiments to look for such violation, despite the fact that such
experiments might entail nothing more than re-analyzing existing data.
Ed F
ueb
..
> Physics doesn't have to obey our notions of what is
> "informationally possible".
> It defines our notions of what is "informationally possible" - but
> there's no reason to /assume/ that it is *limited* by them.
> If you adopt some sort of "Finite Nature" principle - then it does
> indeed follow that relativity is incorrect - but such a principle
> is not supported by any evidence.
"Classical" relativity is indeed incorrect - but for the assumption
of mystic matter that allegedly determines the structure of time & space.
The consequence of this wrong assumption is a fundamental determinism
that has already been falsified. In fact, matter is nothing else than
the (geometrical) state of time & space.
In that, information is an outcome of matter, and tied on it. One can
of course simulate properties of time & space feeding the computer
with the belonging information. But this information is not complete. -
I know what I speak from, because I did it myself successfully.
> Doubting that "Finite Nature" is true is a bit different from
> believing in magic - IMO.
> We have no evidence to speak of - in either direction - about the
> finiteness of nature.
Above simulations could indicate it (if really "true").
> Consequently, I reckon we should keep an open mind on the issue.
Agree.
Ed Fredkin
>
> > Something should be obvious to everyone who can think about this while
> > keeping in mind the perspective of digital informational processes yet
> > de novo from the historical perspective of physics. Relativity (both
> > Newtonian and Einsteinian) is an informational impossibility. It
> > doesn't matter that the math works. It doesn't matter that we haven't
> > been able to measure absolute direction or absolute motion. It
> > doesn't matter how esthetically pleasing it is to assume the existence
> > of fundamental continuous symmetries. It is time to wake up to the
> > fact that there is no possibility of any informational process
> > carrying out: what physics purports to do with rotation, translation
> > and other continuous symmetries. That these symmetries must be
> > violated by microscopic discrete processes is actually the only
> > reasonable conclusion one can reach based on what we now know about
> > informational processes. To have believed otherwise can be understood
> > and excused; to doggedly persist is tantamount to believing in magic.
>
> Physics doesn't have to obey our notions of what is
> "informationally possible".
>
> It defines our notions of what is "informationally possible" - but
> there's no reason to /assume/ that it is *limited* by them.
When we discover a law, such as the law of conservation of momentum,
>
> If you adopt some sort of "Finite Nature" principle - then it does
> indeed follow that relativity is incorrect - but such a principle
> is not supported by any evidence.
I agree that FN implies that relativity is incorrect, but I do not
agree that it is necessary to assume FN in order to question
contemporary concepts of motion.
>
> Doubting that "Finite Nature" is true is a bit different from
> believing in magic - IMO.
Of course.
>
> We have no evidence to speak of - in either direction - about the
> finiteness of nature.
I disagree here. There's plenty of evidence; it's just not recognized
as such. While the FN assumption awaits new experimental evidence in
order to convince the physics community, if that evidence is found and
views change, it won't take long for everyone to wonder why they
didn't figure it out sooner. There's plenty of evidence, it's just
that we've built an elaborate mathematical scaffold over it that
obscures its true nature.
>
> Consequently, I reckon we should keep an open mind on the issue.
Insofar as the question of the rigorous belief in the reality of
continuous symmetries I agree on the "...open mind..." part, but not
on the "...keep an open mind...". We have all trained ourselves to
not have an open mind on this subject. With an truly open mind we
could realize now that all apparently continuous symmetries must be
violated. Minds are so closed as to exclude, as not worth doing, any
experiments to look for such violations, despite the fact that such
Tim Tyler
> Tim Tyler <t...@tt1lock.org> wrote in message news:<HxHur...@bath.ac.uk>...
> > Ed Fredkin <edfr...@yahoo.com> wrote or quoted:
> > > Something should be obvious to everyone who can think about this while
> > > keeping in mind the perspective of digital informational processes yet
> > > de novo from the historical perspective of physics. Relativity (both
> > > Newtonian and Einsteinian) is an informational impossibility. It
> > > doesn't matter that the math works. It doesn't matter that we haven't
> > > been able to measure absolute direction or absolute motion. It
> > > doesn't matter how esthetically pleasing it is to assume the existence
> > > of fundamental continuous symmetries. It is time to wake up to the
> > > fact that there is no possibility of any informational process
> > > carrying out: what physics purports to do with rotation, translation
> > > and other continuous symmetries. That these symmetries must be
> > > violated by microscopic discrete processes is actually the only
> > > reasonable conclusion one can reach based on what we now know about
> > > informational processes. To have believed otherwise can be understood
> > > and excused; to doggedly persist is tantamount to believing in magic.
> >
> > Physics doesn't have to obey our notions of what is
> > "informationally possible".
> >
> > It defines our notions of what is "informationally possible" - but
> > there's no reason to /assume/ that it is *limited* by them.
[...]
> "Magic" in the sense used here means something we imagine for which
> there is no physically realizable process. In other words, "magic"
> refers to things we might imagine which violate accepted laws of
> physics. An example would be a 300 HP automobile engine that is a
> perpetual motion machine; needing no fuel or other source of energy.
>
> Another law of macroscopic physics is that, in principle, one
> informational process can model aspects of a physical process. Given
> a solar system we can always build or program an orrery to arbitrary
> precision.
>
> However, what cannot be done is to build (or even imagine) our orrery
> without the informational equivalent of a single fixed reference
> frame. [...]
Why not?
Quantum computers were invented to simulate quantum physics.
Similarly - though you can't simulate continuum physics with a
conventional digital computer - if continuum physics is true then
there may exist infinitely more powerful machines - machines
that can be used to simulate continuum physics without any
"magic" (your definition) being required.
Cedric Beny
> It is time to wake up to the
> fact that there is no possibility of any informational process
> carrying out: what physics purports to do with rotation, translation
> and other continuous symmetries. That these symmetries must be
> violated by microscopic discrete processes is actually the only
> reasonable conclusion one can reach based on what we now know about
> informational processes.
What if, instead of taking a classical computer as my model of what an
informational process is, I take a quantum computer. The state of a
quantum computer does display continuous symmetries.
You may object that observations of qubits always yield discrete
quantities. I would answer that this is because we (humans) process
information in a classical, decoherent, way. But truly fundamental
laws of physics should work for "anyone", as well for a quantum
observer.
Such a quantum observer could be any quantum computer with the task of
achieving some goal in an fully autonomous way. This machine, loaded
with the proper quantum program should be able to use an efficient
model of the physics of the system it will have to interact with.
In this context your conclusion doesn't seem obvious at all. With
enough imagination one could probably think of many other exemples.
Cedric Beny
daniel B miller
[...]
> Similarly - though you can't simulate continuum physics with a
> conventional digital computer - if continuum physics is true then
> there may exist infinitely more powerful machines - machines
> that can be used to simulate continuum physics without any
> "magic" (your definition) being required.
Didn't someone prove that in theory, the 3-body problem is capable of
universal computation? If time is continuous, I imagine such a computer
could operate at infinite speed.
I have a problem with the idea that every little interaction between
three bits of matter is capable of an infinite amount of computation.
It smacks of perpetual motion -- getting something for nothing.
Same goes for Quantum Computation, though I realize present theory
implies it should be possible. My hunch is that there will turn out to
be fundamental limits to what we can do with matter and energy; these
limits may be defined in discrete terms, or in continuous terms but with
caveats such as limited precision measurements (ie, the uncertainty
principle comes to mind, though it's in the wrong context here).
I propose that any such (fundamental) limits in our ability to extract
information from the system will turn out to be equivalent to some sort
of discrete (finite) formalism. In other words, 'noise' in the system
is due to its limited ability to contain and process information, which
is the same thing as saying the system is finite.
Finally, finite == discrete.
-dbm
daniel B miller
...
>
> Nothing is obvious (even causality does not have absolute value in
> physical theory for now). Any theory is just a set of postulates and
> what is following from them. People do want to see something practical
> in physics and it does matter that the math works. Not too many
> physicists do care about what is "esthetically pleasing" to "radical
> finitists" (you have seen it in the last 40 years).
>
> To challenge the paradigm in physics it is important to provide not
> only compelling theory but also experimental facts, which would
> contradict current theories. Such facts probably are not limited to
> direct observation of extinction of continuous symmetries (a little
> bit more developed discrete picture would give something else).
>
some of the core features of physical systems, we will be in a much
better position to make the sorts of arguments we are making. As Ed has
mentioned many times, this field of inquiry is really just getting started.
-dbm
Ed Fredkin
This is my (Ed F's) 3rd attempt trying to get Google to pick this up
"Tim Tyler" <t...@tt1lock.org> wrote in message news:HxHur...@bath.ac.uk...
> Ed Fredkin <edfr...@yahoo.com> wrote or quoted:
>
> > Something should be obvious to everyone who can think about this while
> > keeping in mind the perspective of digital informational processes yet
> > de novo from the historical perspective of physics. Relativity (both
> > Newtonian and Einsteinian) is an informational impossibility. It
> > doesn't matter that the math works. It doesn't matter that we haven't
> > been able to measure absolute direction or absolute motion. It
> > doesn't matter how esthetically pleasing it is to assume the existence
> > of fundamental continuous symmetries. It is time to wake up to the
> > fact that there is no possibility of any informational process
> > carrying out: what physics purports to do with rotation, translation
> > and other continuous symmetries. That these symmetries must be
> > violated by microscopic discrete processes is actually the only
> > reasonable conclusion one can reach based on what we now know about
> > informational processes. To have believed otherwise can be understood
> > and excused; to doggedly persist is tantamount to believing in magic.
>
> Physics doesn't have to obey our notions of what is
> "informationally possible".
>
> It defines our notions of what is "informationally possible" - but
> there's no reason to /assume/ that it is *limited* by them.
When we discover a law, such as the law of conservation of momentum,
it is reasonable to assume that physics does have to obey that law and
further that it is indeed *limited* by that law. It doesn't take a
belief in the "Finite Nature" principle to come to the conclusion that
there are laws of physics that are related to the informational nature
of systems. For example, we know that systems that have reversible
laws, must contain enough information in the state of the system to
both determine the future state and the past state. This is a kind of
mathematical fact, certainly true in this universe and like many
mathematical facts, true in other kinds of non-physical systems that
we can imagine.
We can think of information as something that is represented by the
arrangement of things where there is a process that gives meaning to
the information. We can define a process as any system where the
arrangement of things evolves over time. We are certainly talking
about things within the scope of physics.
"Magic" in the sense used here means something we imagine for which
there is no physically realizable process. In other words, "magic"
refers to things we might imagine which violate accepted laws of
physics. An example would be a 300 HP automobile engine that is a
perpetual motion machine; needing no fuel or other source of energy.
Another law of macroscopic physics is that, in principle, one
informational process can model aspects of a physical process. Given
a solar system we can always build or program an orrery to arbitrary
precision.
However, what cannot be done is to build (or even imagine) our orrery
without the informational equivalent of a single fixed reference
frame. It's true that we have been unable to detect such a thing
(other than the CBR). Yet the idea of promoting the observation that
"...any unnaccelerated reference frame will do..." to the dogma that
"...there cannot be a single fixed reference frame..." is cute but not
necessarily required as a consequence of any set of experimentally
determined observations.
>
> If you adopt some sort of "Finite Nature" principle - then it does
> indeed follow that relativity is incorrect - but such a principle
> is not supported by any evidence.
I agree that FN implies that relativity is incorrect, but I do not
agree that it is necessary to assume FN in order to question
contemporary concepts of motion.
>
> Doubting that "Finite Nature" is true is a bit different from
> believing in magic - IMO.
Of course.
>
> We have no evidence to speak of - in either direction - about the
> finiteness of nature.
I disagree here. There's plenty of evidence; it's just not recognized
as such. While the FN assumption awaits new experimental evidence in
order to convince the physics community, if that evidence is found and
views change, it won't take long for everyone to wonder why they
didn't figure it out sooner. There's plenty of evidence, it's just
that we've built an elaborate mathematical scaffold over it that
obscures its true nature.
>
> Consequently, I reckon we should keep an open mind on the issue.
Insofar as the question of the rigorous belief in the reality of
continuous symmetries I agree on the "...open mind..." part, but not
on the "...keep an open mind...". We have all trained ourselves to
not have an open mind on this subject. With a truly open mind we
Gerard Westendorp
> Tim Tyler wrote:
> [...]
>
>> Similarly - though you can't simulate continuum physics with a
>> conventional digital computer - if continuum physics is true then
>> there may exist infinitely more powerful machines - machines
>> that can be used to simulate continuum physics without any
>> "magic" (your definition) being required.
>
>
> Didn't someone prove that in theory, the 3-body problem is capable of
> universal computation? If time is continuous, I imagine such a computer
> could operate at infinite speed.
>
> I have a problem with the idea that every little interaction between
> three bits of matter is capable of an infinite amount of computation. It
> smacks of perpetual motion -- getting something for nothing.
As human beings, it seems our knowledge must always be discrete. After
all, we have to be able to express our knowledge in words, which can
be represented as ASCII, which is finite and discrete.
So our *knowledge* is always discrete and finite.
It could be that nature is also discrete, and that we can find a nice
discrete model that approximates it. We can of course never
completely describe nature, because our model would have to contain
itself. (Nature is always "bigger" than a model.)
It could also be that nature is continuous, but that it can be
approximated by a discrete model. In an important class of this case
we can refine the approximation indefinitely. I am referring
of course to differential calculus, and to more tricky methods of
taking limits, such as renormalisation.
Finally, it could be that nature is "uncountable", [ "continuous"
is a wrong word here, because it implies that we can use calculus,
which implies that we can approximate with discrete models]. In other
words, you need an amount of information to approximately describe
nature that is in principle not possible to contain in any finite set.
In that case, we as humans will never succeed in understanding nature.
But we already know that we understand nature to a certain degree,
in the form of present day physics. The question is, how much
further can we go?
Finally^2, it could be that nature is discrete, but certain aspects
of it cannot
be successfully approximated by a model that is smaller than it.
Smaller in the sense that is has less bytes of information.
Einstein : "The least understandable
aspect of nature is that it is understandable".
Anyway, I did not know about the 3 body problem being capable of
universal computation.
Cool.
A related idea is that if you had infinite accuracy in certain
data, you could (in classical physics at least) deduce almost
anything you want from the knowledge of a single atom. This
atom would be ever so slightly influenced by past events, such
as the death of Caesar. If Caesar had not been murdered, each
molecule on earth would move slightly differently.
As you mention, in practice this does not work. Something to
think about:
Suppose you had a huge budget to buy the best acoustic
equipment, huge amounts of time, could you in principle
hear a conversation of 2 people on the other side
of the planet? Or does quantum mechanics, or perhaps
thermodynamics put *fundamental* limits to accuracy?
Gerard
Tim Tyler
> Dan wrote:
> > Tim Tyler wrote:
> >> Similarly - though you can't simulate continuum physics with a
> >> conventional digital computer - if continuum physics is true then
> >> there may exist infinitely more powerful machines - machines
> >> that can be used to simulate continuum physics without any
> >> "magic" (your definition) being required.
> >
> > Didn't someone prove that in theory, the 3-body problem is capable of
> > universal computation? If time is continuous, I imagine such a computer
> > could operate at infinite speed.
> >
> > I have a problem with the idea that every little interaction between
> > three bits of matter is capable of an infinite amount of computation. It
> > smacks of perpetual motion -- getting something for nothing.
>
> As human beings, it seems our knowledge must always be discrete. After
> all, we have to be able to express our knowledge in words, which can
> be represented as ASCII, which is finite and discrete.
>
> So our *knowledge* is always discrete and finite.
The premise that we have to be able to express our knowledge in words
seems unjustifable to me.
There are clearly other channels of communication available to us.
Tim Tyler
> Tim Tyler wrote:
> [...]
A continuous universe - which we could not take full computational
advantage of - due to practical difficulies in producing the small-scale
accurate engineering needed - would be a bit different from the world the
discrete modellers are hoping for, though.
In such a world discrete models would offer no advantages over continuous
ones.
akha...@yahoo.com
[...]
>
> Similarly - though you can't simulate continuum physics with a
> conventional digital computer - if continuum physics is true then
> there may exist infinitely more powerful machines - machines
> that can be used to simulate continuum physics without any
> "magic" (your definition) being required.
Tim,
May I try to continue your smooth reasoning?
--------------
So far, while quantum and continuous computers are "immature",
discrete computation has huge advantage. But we need to catch up on
discrete physics theory...
If we could come to realistic picture in more than one approach,
people will use Occam's razor to choose what "magic" is right...
--------------
Only if significant physical distinction could be pointed out, the
experiment will influence the verdict (it would happen sooner than
later).
Alex
akha...@yahoo.com
[...]
> There's plenty of evidence, it's just
> that we've built an elaborate mathematical scaffold over it that
> obscures its true nature.
> >
> > Consequently, I reckon we should keep an open mind on the issue.
>
> Insofar as the question of the rigorous belief in the reality of
> continuous symmetries I agree on the "...open mind..." part, but not
> on the "...keep an open mind...". We have all trained ourselves to
> not have an open mind on this subject. With a truly open mind we
> could realize now that all apparently continuous symmetries must be
> violated. Minds are so closed as to exclude, as not worth doing, any
> experiments to look for such violations, despite the fact that such
> experiments might entail nothing more than re-analyzing existing data.
>
> Ed F
Could you be more specific and propose those experiments?
What kind of existing data should be re-analyzed?
Would the conventional interpretation of that data be quite controversial?
Do you know how to "open mind" and make a step from philosophy to physics?
Alex
ueb
> Suppose you had a huge budget to buy the best acoustic
> equipment, huge amounts of time, could you in principle
> hear a conversation of 2 people on the other side
> of the planet?
No. That is limited by resistance noise, and the limit of
a good equipment has already been adapted to this fundamental
limit.
> Or does quantum mechanics, or perhaps
> thermodynamics put fundamental limits to accuracy?
You may _describe_ these limits quantum-mechanically or
thermodynamically.
Ulrich
Ed Fredkin
<akha...@yahoo.com> wrote in message
news:5bdca6d3.04051...@posting.google.com...
> "Ed Fredkin" <edfr...@yahoo.com> wrote in message
news:<10a516k...@news.supernews.com>...
> [...]
> > There's plenty of evidence, it's just
> > that we've built an elaborate mathematical scaffold over it that
> > obscures its true nature.
> > >
> > > Consequently, I reckon we should keep an open mind on the issue.
> >
> > Insofar as the question of the rigorous belief in the reality of
> > continuous symmetries I agree on the "...open mind..." part, but not
> > on the "...keep an open mind...". We have all trained ourselves to
> > not have an open mind on this subject. With a truly open mind we
> > could realize now that all apparently continuous symmetries must be
> > violated. Minds are so closed as to exclude, as not worth doing, any
> > experiments to look for such violations, despite the fact that such
> > experiments might entail nothing more than re-analyzing existing data.
> >
> > Ed F
>
> Could you be more specific and propose those experiments?
One could look for violation of rotation symmetry by plotting the
initial angles of quark jets, plotted in terms of Right Ascension and
Declination (RA & Dec). This requires knowing the true angle (from
true north) of the beam line, the date and time (UTC) and the latitude
and longitude of the experiment. Other data could also be analyzed in
the same way. The clue would be a lack of angular symmetry in the
initial angles of the jets (or other fundamental processes) might be
noted (with respect to the fixed stars).
>
> What kind of existing data should be re-analyzed?
Data from LEP at CERN, SLAC (such as from BABAR), CDF at Fermilab etc.
>
> Would the conventional interpretation of that data be quite controversial?
Today, when one looks for anisotropies, they mean with reference to
the beam line. If one found absolute angular anisotropy (in terms of
RA and Dec it would amaze people, but they would scramble to find any
possible interpretation other than the obvious one: microscopic
violation of rotation symmetry.
>
> Do you know how to "open mind" and make a step from philosophy to physics?
While I see plenty of existing evidence, only new, unexpected and "in
your face" experimental evidence, easily replicated, has a chance of
"opening minds" about violations of apparently continuous symmetries.
> Alex
Ed F
Cedric Beny
> It is time to wake up to the
> fact that there is no possibility of any informational process
> carrying out: what physics purports to do with rotation, translation
> and other continuous symmetries. That these symmetries must be
> violated by microscopic discrete processes is actually the only
> reasonable conclusion one can reach based on what we now know about
> informational processes.
What if, instead of taking a classical computer as my model of what an
informational process is, I take a quantum computer. The state of a
quantum computer does display continuous symmetries.
You may object that observations of qubits always yield discrete
quantities. I would answer that this is because we (humans) process
information in a classical, decoherent, way. But truly fundamental
laws of physics should work for "anyone", as well for a quantum
observer.
Such a quantum observer could be any quantum computer with the task of
achieving some goal in an fully autonomous way. This machine, loaded
with the proper quantum program should be able to use an efficient
model of the physics of the system it will have to interact with.
I see no reason why our fundamental notion of informational process
should be classical rather than quantum.
Cedric Beny
akha...@yahoo.com
> > Could you be more specific and propose those experiments?
>
> One could look for violation of rotation symmetry by plotting the
> initial angles of quark jets, plotted in terms of Right Ascension and
> Declination (RA & Dec). This requires knowing the true angle (from
> true north) of the beam line, the date and time (UTC) and the latitude
> and longitude of the experiment. Other data could also be analyzed in
> the same way. The clue would be a lack of angular symmetry in the
> initial angles of the jets (or other fundamental processes) might be
> noted (with respect to the fixed stars).
> >
> > What kind of existing data should be re-analyzed?
>
> Data from LEP at CERN, SLAC (such as from BABAR), CDF at Fermilab etc.
The army of experimentalists in high energy physics is relatively
prepared to "open mind" to discrete ideas. They are processing
terabytes of (mostly angular) data obtained from discrete electronic
detectors (to "mess up" those pure discrete data with all kind of
modern theories later on).
It would be much easy to deal with small/chip experiments (not with
high energy generals).
Anyone would need justification. So far, I have not seen anything
convincing on the topic (what could be observable, how, why).
Alex
Ed Fredkin
However, access is controlled and not readily available to those not
directly involved. This makes perfect sense in that a large group of
physicists (along with many others) have worked for many years to design and
construct the experiment. They should obviously get first crack at all of
the data vis-à-vis such issues as showing symmetry violation in B and B bar
data. Nevertheless, that is not the main issue in why I could expect firm
resistance to access given that I would like to look for violations of
absolute angular isotropy. Years ago I had almost arranged to obtain quark
jet data from LEP when the powers that be learned what I intended to look
for in the data. Their position was that, if they allowed me to do so, it
could be so embarrassing to CERN in general, that it could threaten future
funding. They actually took the time to convince me that they were likely
to be correct.
All I need is one unafraid member of the BABAR group. I'm willing to remain
totally silent unless the data shows something really interesting. If there
is such a person, he or she could send me an email to e...@cmu.edu and I can
arrange for all the rest. There's no necessity for anyone to be
embarrassed.
The nature of the analysis is simplicity itself. We can use the kinds of
algorithms that allow astronomers to calculate what star their telescopes is
pointing at. Given the initial angle of a particle in laboratory
coordinates, we want to translate that angle into RA and Dec. Then we want
to plot a 2D spherical histogram of numbers of particles in all small
angular areas covering the surface of a fixed celestial sphere. Given
issues such as the angular resolution available in the data, we might or
might not be able to see anisotropy even if it exists. I know that I'm
describing an experiment where essentially everyone is certain we will see
nothing significant, but it's worth a try. The cost is close to zero and
there's nothing unscientific about an experiment that challenges certain
basic assumptions.
Ed F
Eray Ozkural exa
> Consequently, I reckon we should keep an open mind on the issue.
Which brings to my mind, do you imagine an infinite complexity in a
finite volume, or have you been referring to something else?
Interestingly enough, Chaitin's Omega could not be real in a finite
universe! [*] Yet, it does not look like pure fiction, and
unfortunately as positivists, it does not seem good scientific
practice to posit a secondary, infinite, and even worse, invisible
substance. :)
I wonder, is that a truly deep puzzle of nature, or can we solve it on
this thread?
Best Regards,
--
Eray Ozkural
[*] Trivially so, since K(x) <= |x| + c, and if |x| is finite, we
can't have a physical Omega where x represents the universe.
Tim Tyler
> > Consequently, I reckon we should keep an open mind on the issue.
>
> Which brings to my mind, do you imagine an infinite complexity in a
> finite volume, or have you been referring to something else?
The usual way of characteising the continuous world view would be
something like an infinite information storage capacity in a finite
volume.
That might not necessarily translate to infinite complexity - e.g.
if the space in question happened to be empty.
Eray Ozkural exa
> Eray Ozkural exa <er...@bilkent.edu.tr> wrote or quoted:
> > Tim Tyler <t...@tt1lock.org> wrote in message news:<HxHur...@bath.ac.uk>...
>
> > > Consequently, I reckon we should keep an open mind on the issue.
> >
> > Which brings to my mind, do you imagine an infinite complexity in a
> > finite volume, or have you been referring to something else?
>
> The usual way of characteising the continuous world view would be
> something like an infinite information storage capacity in a finite
> volume.
>
> That might not necessarily translate to infinite complexity - e.g.
> if the space in question happened to be empty.
OK, I understand. Maybe you would like to know why I suggested such an
insane sounding thing at all.
The problem is basically the incompressibility of Omega. So, if we are
to believe in Godel and Chaitin's brand of realism, then Omega must
exist right here, right now. Now, we're coming to the dilemma of
realist philosophy. We are being offered Godel's realism, so
mathematical objects are real like a chair or table. So far, fine.
But, then Omega must exist, too. The problem is the following:
Platonism isn't a very scientific view. So, in my opinion, if Omega
*really* exists and not *psychologically*, it can exist in only one
sense: the physical. And if that is the case, then physics is indeed
inexhaustible. If the universe were then discrete, it would have to be
infinite volume. But if it is a finite volume, like we think ours is,
then the universe isn't discrete, it's continuous. The problem is that
this conclusion doesn't seem plausible, and is certainly at odds with
our current understanding of the nature.
Did you find this train of thought entertaining?
If the Aristotelian Realism I described is false, then some kind of
Computationalist Platonism is correct. For instance, there is an
eternal realm of mathematical objects that includes Omega (oh my
god!), and then there is this mere material, computable, world. The
problem is that this view is a little old-fashioned, and bent on
religious speculation. Imagine yourself among a photo of scientists on
one of those religious magazines: "The Scientists that Found God!", or
a sci-fi magazine: "The Matrix is Real!".
There are of course more sensible alternatives, such as a finite,
discrete world, with finite complexity, and *no* silly philosophical
interpretation. :)
Serg
> Eray Ozkural exa <er...@bilkent.edu.tr> wrote or quoted:
> > Tim Tyler <t...@tt1lock.org> wrote in message news:<HxHur...@bath.ac.uk>...
>
> something like an infinite information storage capacity in a finite
> volume.
>
I'm not sure about it, continuous model still can have finite
information storage capacity - for example in quantum mechanics.
Probaly more exact descripion of your thought would be : the state of
continuous sytem can not be stored in finite volume of digital
<computer> memory, but that is a tautology, which basycally means
"continuous is not discrete"