When the physicist says "The best answer we can give is that reality is a vector in Hilbert space", it shows that he cannot be cured.Evgenii
![]() | (a1) |
is replaced in the quantum case by the theorem of composite amplitudes
![]() |
--Stathis Papaioannou
On 12 Sep 2019, at 06:45, Alan Grayson <agrays...@gmail.com> wrote:
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/b9ee8443-27de-4136-8ab8-a6367a5bddae%40googlegroups.com.
On 12 Sep 2019, at 06:45, Alan Grayson <agrays...@gmail.com> wrote:Sean Carroll assumes that reality is a vector in Hilbert space. But he used Mechanism (implicitly) but forget computer science. That reality is a vector in Hilbert space has to be derived from the relative statistic on all computations realised in arithmetic.To say that someone who thinks differently than you is in mental decline teaches us something about you, no about Carroll.Bruno
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
LC
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/fe4b02a2-9fcb-4126-b2ad-fb9982f20fc1%40googlegroups.com.
What's the argument for such a claim? Morevover, I don't believe a universe of finite age, such as ours which everyone more or less agrees began some 13.8 BYA, can be spatially infinite. Here I'm referring to our bubble, not some infinite substratum from which it might have arose. AG
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/f85775c7-7914-47e4-83eb-1142b1b58249%40googlegroups.com.
On 13 Sep 2019, at 04:26, Alan Grayson <agrays...@gmail.com> wrote:
On Thursday, September 12, 2019 at 11:01:54 AM UTC-6, Alan Grayson wrote:
On Thursday, September 12, 2019 at 7:45:22 AM UTC-6, Lawrence Crowell wrote:On Thursday, September 12, 2019 at 4:20:46 AM UTC-5, Philip Thrift wrote:
On Wednesday, September 11, 2019 at 11:45:41 PM UTC-5, Alan Grayson wrote:Many Worlds is where people go to escape from one world of quantum-stochastic processes. They are like vampires, but instead of running away from sunbeams, are running away from probabilities.@philipthriftThis assessment is not entirely fair. Carroll and Sebens have a paper on how supposedly the Born rule can be derived from MWI I have yet to read their paper, but given the newsiness of this I might get to it. One advantage that MWI does have is that it splits the world as a sort of quantum frame dragging that is nonlocal. This nonlocal property might be useful for working with quantum gravity,I worked a proof of a theorem, which may not be complete unfortunately, where the two sets of quantum interpretations that are ψ-epistemic and those that are ψ-ontological are not decidable. There is no decision procedure which can prove QM holds either way. The proof is set with nonlocal hidden variables over the projective rays of the state space. In effect there is an uncertainty in whether the hidden variables localize extant quantities, say with ψ-ontology, or whether this localization is the generation of information in a local context from quantum nonlocality that is not extant, such as with ψ-epistemology. Quantum interprertations are then auxiliary physical axioms or postulates. MWI and within the framework of what Carrol and Sebens has done this is a ψ-ontology, and this defines the Born rule. If I am right the degree of ψ-epistemontic nature is mixed. So the intriguing question we can address is the nature of the Born rule and its tie into the auxiliary postulates of quantum interpretations. Can a similar demonstration be made for the Born rule within QuBism, which is what might be called the dialectic opposite of MWI?To take MWI as something literal, as opposed to maybe a working system to understand QM foundations, is maybe taking things too far. However, it is a part of some open questions concerning the fundamentals of QM. If MWI, and more generally postulates of quantum interpretations, are connected to the Born rule it makes for some interesting things to think about.LCIf you read the link, it's pretty obvious that Carroll believes the many worlds of the MWI, literally exist. AGCarroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued.The idea comes from Tegmark, and I agree with you, it necessitate more than an infinite universe. It requires also some assumption of homogeneity.
Of course, (for those who are aware of Gödel 1931 and Turing 1936), arithmetic contains all computations, which entails, when assuming mechanism, an infinity of each os us.
That explains both where the appearance of universe come from, and the quantum mechanical type of formalism. In “many-world”, the “many” makes sense, but the term “world” is not well defined and should not been taken literally. It is more histories than worlds per se.Bruno
What's the argument for such a claim? Morevover, I don't believe a universe of finite age, such as ours which everyone more or less agrees began some 13.8 BYA, can be spatially infinite. Here I'm referring to our bubble, not some infinite substratum from which it might have arose. AG--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
On 13 Sep 2019, at 00:44, Lawrence Crowell <goldenfield...@gmail.com> wrote:
On Thursday, September 12, 2019 at 11:44:51 AM UTC-5, Philip Thrift wrote:
On Thursday, September 12, 2019 at 8:45:22 AM UTC-5, Lawrence Crowell wrote:On Thursday, September 12, 2019 at 4:20:46 AM UTC-5, Philip Thrift wrote:
On Wednesday, September 11, 2019 at 11:45:41 PM UTC-5, Alan Grayson wrote:Many Worlds is where people go to escape from one world of quantum-stochastic processes. They are like vampires, but instead of running away from sunbeams, are running away from probabilities.@philipthriftThis assessment is not entirely fair. Carroll and Sebens have a paper on how supposedly the Born rule can be derived from MWI I have yet to read their paper, but given the newsiness of this I might get to it. One advantage that MWI does have is that it splits the world as a sort of quantum frame dragging that is nonlocal. This nonlocal property might be useful for working with quantum gravity,I worked a proof of a theorem, which may not be complete unfortunately, where the two sets of quantum interpretations that are ψ-epistemic and those that are ψ-ontological are not decidable. There is no decision procedure which can prove QM holds either way. The proof is set with nonlocal hidden variables over the projective rays of the state space. In effect there is an uncertainty in whether the hidden variables localize extant quantities, say with ψ-ontology, or whether this localization is the generation of information in a local context from quantum nonlocality that is not extant, such as with ψ-epistemology. Quantum interprertations are then auxiliary physical axioms or postulates. MWI and within the framework of what Carrol and Sebens has done this is a ψ-ontology, and this defines the Born rule. If I am right the degree of ψ-epistemontic nature is mixed. So the intriguing question we can address is the nature of the Born rule and its tie into the auxiliary postulates of quantum interpretations. Can a similar demonstration be made for the Born rule within QuBism, which is what might be called the dialectic opposite of MWI?To take MWI as something literal, as opposed to maybe a working system to understand QM foundations, is maybe taking things too far. However, it is a part of some open questions concerning the fundamentals of QM. If MWI, and more generally postulates of quantum interpretations, are connected to the Born rule it makes for some interesting things to think about.LCQBism is not the dialectical opposite of MWI. This is:@philipthriftThe MWI and this path integral interpretation are both ψ-ontic and are thus not opposite.I agree. I would even add that with Feynman path formalism, the reduction of the wave packet does no more make sense. Feynman said it in his little book on light: he consider the Wave reduction as a confusion and appeal to magic (footnote at the end of the second chapter).Bruno
LC--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
Carroll is in irreversible mental decline. He's lost contact with reality. Sad case. I stand by my assessment. He doesn't even understand basic linear algebra, and that his "state vector" has no unique representation, and thus the mythical interpretation of the superposition of the wf is totally illusional. AG
Not for those of us who watch horseraces! Applied to QM, the wf becomes irrelevant when the measurement occurs. Wave packet reduction, by which I assume you mean "collapse", is nothing more than a bookkeeping device. AG
On 13 Sep 2019, at 04:26, Alan Grayson <agrays...@gmail.com> wrote:
On Thursday, September 12, 2019 at 11:01:54 AM UTC-6, Alan Grayson wrote:
On Thursday, September 12, 2019 at 7:45:22 AM UTC-6, Lawrence Crowell wrote:On Thursday, September 12, 2019 at 4:20:46 AM UTC-5, Philip Thrift wrote:
On Wednesday, September 11, 2019 at 11:45:41 PM UTC-5, Alan Grayson wrote:Many Worlds is where people go to escape from one world of quantum-stochastic processes. They are like vampires, but instead of running away from sunbeams, are running away from probabilities.@philipthriftThis assessment is not entirely fair. Carroll and Sebens have a paper on how supposedly the Born rule can be derived from MWI I have yet to read their paper, but given the newsiness of this I might get to it. One advantage that MWI does have is that it splits the world as a sort of quantum frame dragging that is nonlocal. This nonlocal property might be useful for working with quantum gravity,I worked a proof of a theorem, which may not be complete unfortunately, where the two sets of quantum interpretations that are ψ-epistemic and those that are ψ-ontological are not decidable. There is no decision procedure which can prove QM holds either way. The proof is set with nonlocal hidden variables over the projective rays of the state space. In effect there is an uncertainty in whether the hidden variables localize extant quantities, say with ψ-ontology, or whether this localization is the generation of information in a local context from quantum nonlocality that is not extant, such as with ψ-epistemology. Quantum interprertations are then auxiliary physical axioms or postulates. MWI and within the framework of what Carrol and Sebens has done this is a ψ-ontology, and this defines the Born rule. If I am right the degree of ψ-epistemontic nature is mixed. So the intriguing question we can address is the nature of the Born rule and its tie into the auxiliary postulates of quantum interpretations. Can a similar demonstration be made for the Born rule within QuBism, which is what might be called the dialectic opposite of MWI?To take MWI as something literal, as opposed to maybe a working system to understand QM foundations, is maybe taking things too far. However, it is a part of some open questions concerning the fundamentals of QM. If MWI, and more generally postulates of quantum interpretations, are connected to the Born rule it makes for some interesting things to think about.LCIf you read the link, it's pretty obvious that Carroll believes the many worlds of the MWI, literally exist. AGCarroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued.The idea comes from Tegmark, and I agree with you, it necessitate more than an infinite universe. It requires also some assumption of homogeneity.Of course, (for those who are aware of Gödel 1931 and Turing 1936), arithmetic contains all computations, which entails, when assuming mechanism, an infinity of each os us. That explains both where the appearance of universe come from, and the quantum mechanical type of formalism. In “many-world”, the “many” makes sense, but the term “world” is not well defined and should not been taken literally. It is more histories than worlds per se.Bruno
What's the argument for such a claim? Morevover, I don't believe a universe of finite age, such as ours which everyone more or less agrees began some 13.8 BYA, can be spatially infinite. Here I'm referring to our bubble, not some infinite substratum from which it might have arose. AG--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
> Carroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued. What's the argument for such a claim?
> Morevover, I don't believe a universe of finite age, such as ours which everyone more or less agrees began some 13.8 BYA, can be spatially infinite.
On Thu, Sep 12, 2019 at 10:26 PM Alan Grayson <agrays...@gmail.com> wrote:> Carroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued. What's the argument for such a claim?Of course it's been proven! It's simple math, there are only a finite number of ways the atoms in your body, or even the entire OBSERVABLE universe, can be arranged so obviously if the entire universe is infinite then there is going to have to be copies, an infinite number of them in fact. Max Tegmark has even calculated how far you'd have to go to see such a thing.Your closest identical copy is 10^12 light years away. About 10^76 light years away there is a sphere of radius 100 light-years identical to the one centered here, so everything we see here during the next century will be identical to those of our counterparts over there. And 10^102 light years away the is a exact copy of our entire observable universe. And all this is true regardless of if the Many Worlds Interpretation of Quantum Mechanics is correct or not, it only depends on the universe being spatially infinite.
On Friday, September 13, 2019 at 5:24:11 AM UTC-6, Bruno Marchal wrote:On 13 Sep 2019, at 04:26, Alan Grayson <agrays...@gmail.com> wrote:
On Thursday, September 12, 2019 at 11:01:54 AM UTC-6, Alan Grayson wrote:
On Thursday, September 12, 2019 at 7:45:22 AM UTC-6, Lawrence Crowell wrote:On Thursday, September 12, 2019 at 4:20:46 AM UTC-5, Philip Thrift wrote:
On Wednesday, September 11, 2019 at 11:45:41 PM UTC-5, Alan Grayson wrote:Many Worlds is where people go to escape from one world of quantum-stochastic processes. They are like vampires, but instead of running away from sunbeams, are running away from probabilities.@philipthriftThis assessment is not entirely fair. Carroll and Sebens have a paper on how supposedly the Born rule can be derived from MWI I have yet to read their paper, but given the newsiness of this I might get to it. One advantage that MWI does have is that it splits the world as a sort of quantum frame dragging that is nonlocal. This nonlocal property might be useful for working with quantum gravity,I worked a proof of a theorem, which may not be complete unfortunately, where the two sets of quantum interpretations that are ψ-epistemic and those that are ψ-ontological are not decidable. There is no decision procedure which can prove QM holds either way. The proof is set with nonlocal hidden variables over the projective rays of the state space. In effect there is an uncertainty in whether the hidden variables localize extant quantities, say with ψ-ontology, or whether this localization is the generation of information in a local context from quantum nonlocality that is not extant, such as with ψ-epistemology. Quantum interprertations are then auxiliary physical axioms or postulates. MWI and within the framework of what Carrol and Sebens has done this is a ψ-ontology, and this defines the Born rule. If I am right the degree of ψ-epistemontic nature is mixed. So the intriguing question we can address is the nature of the Born rule and its tie into the auxiliary postulates of quantum interpretations. Can a similar demonstration be made for the Born rule within QuBism, which is what might be called the dialectic opposite of MWI?To take MWI as something literal, as opposed to maybe a working system to understand QM foundations, is maybe taking things too far. However, it is a part of some open questions concerning the fundamentals of QM. If MWI, and more generally postulates of quantum interpretations, are connected to the Born rule it makes for some interesting things to think about.LCIf you read the link, it's pretty obvious that Carroll believes the many worlds of the MWI, literally exist. AGCarroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued.The idea comes from Tegmark, and I agree with you, it necessitate more than an infinite universe. It requires also some assumption of homogeneity.Our universe is, on a large scale, homogeneous. But it can't be infinite since it has only been expanding for finite time, 13.8 BY. I had a discussion with Brent about this some time ago, and he claimed finite in time doesn't preclude infinite in space. I strongly disagree. Perhaps I am missing something. Wouldn't be the first time. AG
On Thu, Sep 12, 2019 at 10:26 PM Alan Grayson <agrays...@gmail.com> wrote:> Carroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued. What's the argument for such a claim?Of course it's been proven! It's simple math, there are only a finite number of ways the atoms in your body, or even the entire OBSERVABLE universe, can be arranged so obviously if the entire universe is infinite then there is going to have to be copies, an infinite number of them in fact. Max Tegmark has even calculated how far you'd have to go to see such a thing.
Your closest identical copy is 10^12 light years away. About 10^76 light years away there is a sphere of radius 100 light-years identical to the one centered here, so everything we see here during the next century will be identical to those of our counterparts over there. And 10^102 light years away the is a exact copy of our entire observable universe. And all this is true regardless of if the Many Worlds Interpretation of Quantum Mechanics is correct or not, it only depends on the universe being spatially infinite.
> Morevover, I don't believe a universe of finite age, such as ours which everyone more or less agrees began some 13.8 BYA, can be spatially infinite.I see no reason in principle why something can't be finite along one dimension and infinite along another dimension.
John K Clark
I don't think there is an implied disconnect between our measurements of the CMBR and what an observer would measure in parts we have no access to. It was everywhere hot and dense, and very very small.
If it were infinite at that time, its temperature would have been near absolute zero. AG
On Friday, September 13, 2019, Alan Grayson <agrays...@gmail.com> wrote:
On Friday, September 13, 2019 at 4:42:00 PM UTC-6, Jason wrote:On Fri, Sep 13, 2019 at 8:25 AM Alan Grayson <agrays...@gmail.com> wrote:
On Friday, September 13, 2019 at 5:24:11 AM UTC-6, Bruno Marchal wrote:On 13 Sep 2019, at 04:26, Alan Grayson <agrays...@gmail.com> wrote:
On Thursday, September 12, 2019 at 11:01:54 AM UTC-6, Alan Grayson wrote:
On Thursday, September 12, 2019 at 7:45:22 AM UTC-6, Lawrence Crowell wrote:On Thursday, September 12, 2019 at 4:20:46 AM UTC-5, Philip Thrift wrote:
On Wednesday, September 11, 2019 at 11:45:41 PM UTC-5, Alan Grayson wrote:Many Worlds is where people go to escape from one world of quantum-stochastic processes. They are like vampires, but instead of running away from sunbeams, are running away from probabilities.@philipthriftThis assessment is not entirely fair. Carroll and Sebens have a paper on how supposedly the Born rule can be derived from MWI I have yet to read their paper, but given the newsiness of this I might get to it. One advantage that MWI does have is that it splits the world as a sort of quantum frame dragging that is nonlocal. This nonlocal property might be useful for working with quantum gravity,I worked a proof of a theorem, which may not be complete unfortunately, where the two sets of quantum interpretations that are ψ-epistemic and those that are ψ-ontological are not decidable. There is no decision procedure which can prove QM holds either way. The proof is set with nonlocal hidden variables over the projective rays of the state space. In effect there is an uncertainty in whether the hidden variables localize extant quantities, say with ψ-ontology, or whether this localization is the generation of information in a local context from quantum nonlocality that is not extant, such as with ψ-epistemology. Quantum interprertations are then auxiliary physical axioms or postulates. MWI and within the framework of what Carrol and Sebens has done this is a ψ-ontology, and this defines the Born rule. If I am right the degree of ψ-epistemontic nature is mixed. So the intriguing question we can address is the nature of the Born rule and its tie into the auxiliary postulates of quantum interpretations. Can a similar demonstration be made for the Born rule within QuBism, which is what might be called the dialectic opposite of MWI?To take MWI as something literal, as opposed to maybe a working system to understand QM foundations, is maybe taking things too far. However, it is a part of some open questions concerning the fundamentals of QM. If MWI, and more generally postulates of quantum interpretations, are connected to the Born rule it makes for some interesting things to think about.LCIf you read the link, it's pretty obvious that Carroll believes the many worlds of the MWI, literally exist. AGCarroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued.The idea comes from Tegmark, and I agree with you, it necessitate more than an infinite universe. It requires also some assumption of homogeneity.Our universe is, on a large scale, homogeneous. But it can't be infinite since it has only been expanding for finite time, 13.8 BY. I had a discussion with Brent about this some time ago, and he claimed finite in time doesn't preclude infinite in space. I strongly disagree. Perhaps I am missing something. Wouldn't be the first time. AGI think what you may be missing is that in popular (but misleading) accounts of the BB they often say everything originated from a point, rather than everywhere at once. To say "everything came from a point" is at best only valid for describing the observable universe (or any finite portion of the universe) but is invalid to extrapolate it to the whole universe, which may be spatially infinite.I am not assuming our universe began from a mathematical point, but I do assume that 13.8 BYA it was very very small, the observable and unobservable parts.Why do you assume this? Most cosmologists make no such assumption. Under the concordance (standard assumed) model of cosmology, space is infinite.
I don't think there is an implied disconnect between our measurements of the CMBR and what an observer would measure in parts we have no access to. It was everywhere hot and dense, and very very small.There's no observational motivation for the universe being very very small at the beginning. It could have been small, large or infinite, for all we know.
If it were infinite at that time, its temperature would have been near absolute zero. AGI think you're working under the assumption that some finite amount of energy was injected into space at one particular point. This is not what the big bang theory says, rather all space (everywhere there was space), was equally hot and dense.
Inflation modifies the picture a bit where the vacuum of space expands rapidly due to its high energy density (which suggests a negative pressure). Under the equations of GR, such a state would expand itself exponentially. Eventually parts of this vacuum decay to a lower energy density, and this dump of energy into space gives us the early hot stage of the big bang.We don't know how big this initial inflating space was, but if inflation is right, most of the universe is still experiencing exponential growth.Each pocket universe may be finite in volume, but extends infinitely in the time direction.
As Alan Guth explains, GR can warp things in the internal view of each pocket universe such that the time and space dimensions flip, the infinite time dimension within the pocket universe can give rise to the appearance of infinite space, and the finite space appears as finite time:
Jason
Each pocket universe may be finite in volume, but extends infinitely in the time direction.
Yes, time extends infinitely into the future, and so does space, unless the universe is closed. But then he claims time and space flip when viewed externally to the bubble. But how does one get outside the bubble to observe it? Not possible AFAICT. AG
On Friday, September 13, 2019 at 4:08:23 PM UTC-6, John Clark wrote:On Thu, Sep 12, 2019 at 10:26 PM Alan Grayson <agrays...@gmail.com> wrote:> Carroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued. What's the argument for such a claim?Of course it's been proven! It's simple math, there are only a finite number of ways the atoms in your body, or even the entire OBSERVABLE universe, can be arranged so obviously if the entire universe is infinite then there is going to have to be copies, an infinite number of them in fact. Max Tegmark has even calculated how far you'd have to go to see such a thing.What I think you're missing (and Tegmark) is the possibility of UNcountable universes. In such case, one could imagine new universes coming into existence forever and ever, without any repeats. Think of the number of points between 0 and 1 on the real line, each point associated with a different universe. AGYour closest identical copy is 10^12 light years away. About 10^76 light years away there is a sphere of radius 100 light-years identical to the one centered here, so everything we see here during the next century will be identical to those of our counterparts over there. And 10^102 light years away the is a exact copy of our entire observable universe. And all this is true regardless of if the Many Worlds Interpretation of Quantum Mechanics is correct or not, it only depends on the universe being spatially infinite.But our universe is NOT spatially infinite if its been expanding for finite time, starting very small, as can be inferred from the temperature of the CMBR. AG
>> Your closest identical copy is 10^12 light years away. About 10^76 light years away there is a sphere of radius 100 light-years identical to the one centered here, so everything we see here during the next century will be identical to those of our counterparts over there. And 10^102 light years away the is a exact copy of our entire observable universe. And all this is true regardless of if the Many Worlds Interpretation of Quantum Mechanics is correct or not, it only depends on the universe being spatially infinite.
> The trouble with all such arguments is that they miss the fact that our initial conditions might have been very special, of measure zero.
If it has been expanding for finite time, its spatial extent must be finite,
like a huge hypersphere. All the models I see pictorially illustrated, have it much much smaller than it presently is. AG
I don't think there is an implied disconnect between our measurements of the CMBR and what an observer would measure in parts we have no access to. It was everywhere hot and dense, and very very small.There's no observational motivation for the universe being very very small at the beginning. It could have been small, large or infinite, for all we know.It was opaque just before 380,000 years, when the CMBR emerged, precisely because it was hugely hot and dense, so much so that light could get out. Is this not observational evidence? AG
If it were infinite at that time, its temperature would have been near absolute zero. AGI think you're working under the assumption that some finite amount of energy was injected into space at one particular point. This is not what the big bang theory says, rather all space (everywhere there was space), was equally hot and dense.I am not assuming what you allege.
Yes, all space was hot and dense, but much smaller in spatial extent than today. Just play the movie backward. AG
Inflation modifies the picture a bit where the vacuum of space expands rapidly due to its high energy density (which suggests a negative pressure). Under the equations of GR, such a state would expand itself exponentially. Eventually parts of this vacuum decay to a lower energy density, and this dump of energy into space gives us the early hot stage of the big bang.We don't know how big this initial inflating space was, but if inflation is right, most of the universe is still experiencing exponential growth.Each pocket universe may be finite in volume, but extends infinitely in the time direction.Then why is there general agreement that the age of our bubble is 13.8 BY?
As Alan Guth explains, GR can warp things in the internal view of each pocket universe such that the time and space dimensions flip, the infinite time dimension within the pocket universe can give rise to the appearance of infinite space, and the finite space appears as finite time:Time might, and probably does extend into an infinite future, but not into an infinite past. Otherwise, cosmologists wouldn't agree that the age of our bubble is 13.8 BY. What do you think that measurement means? AG
As Alan Guth explains, GR can warp things in the internal view of each pocket universe such that the time and space dimensions flip, the infinite time dimension within the pocket universe can give rise to the appearance of infinite space, and the finite space appears as finite time:Jason
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/ca6b23fc-0946-4bee-9170-ba6026e88360%40googlegroups.com.
> What I think you're missing (and Tegmark) is the possibility of UNcountable universes. In such case, one could imagine new universes coming into existence forever and ever, without any repeats. Think of the number of points between 0 and 1 on the real line, each point associated with a different universe. AG
>> true regardless of if the Many Worlds Interpretation of Quantum Mechanics is correct or not, it only depends on the universe being spatially infinite.> But our universe is NOT spatially infinite if its been expanding for finite time,
On Saturday, September 14, 2019 at 7:18:40 AM UTC-6, John Clark wrote:On Fri, Sep 13, 2019 at 11:22 PM Alan Grayson <agrays...@gmail.com> wrote:> What I think you're missing (and Tegmark) is the possibility of UNcountable universes. In such case, one could imagine new universes coming into existence forever and ever, without any repeats. Think of the number of points between 0 and 1 on the real line, each point associated with a different universe. AGThere is no reason to think physics needs all the real numbers and considerable evidence to think it does not. To my mind the strongest evidence is that a physical Turing Machine is incapable of even approximating most real numbers, I happened to have posted a proof of this yesterday on the "Observation versus assumption" thread.
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CA%2BBCJUjvW5pQ7v4Ex490SfUFOcBeUF5EJkxQzsVYyFHXgHmbXQ%40mail.gmail.com.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CA%2BBCJUjvW5pQ7v4Ex490SfUFOcBeUF5EJkxQzsVYyFHXgHmbXQ%40mail.gmail.com.
--Stathis Papaioannou
On Fri, Sep 13, 2019 at 11:22 PM Alan Grayson <agrays...@gmail.com> wrote:> What I think you're missing (and Tegmark) is the possibility of UNcountable universes. In such case, one could imagine new universes coming into existence forever and ever, without any repeats. Think of the number of points between 0 and 1 on the real line, each point associated with a different universe. AGThere is no reason to think physics needs all the real numbers and considerable evidence to think it does not.
>> There is no reason to think physics needs all the real numbers and considerable evidence to think it does not. To my mind the strongest evidence is that a physical Turing Machine is incapable of even approximating most real numbers, I happened to have posted a proof of this yesterday on the "Observation versus assumption" thread.> Physics doesn't need all the real numbers, just some of them, say any continuous range of any variable; like the mass of the electron.
> Einstein's field equations use PI, and so do Maxwell's equations.
and inflation preserved the homogeneity. This is what Guth was trying to solve with inflation, among other problems, such as no detectable monopoles. This entire logic breaks down if one assumes an infinite universe at the time of inflation.
In this case, the infinite universe was always homogeneous even though it was never causally connected.
Further, how could it have been so hot 380,000 years after the BB if it wasn't dense at that time?
An infinite universe right after the BB would be COOL,
event | temperature (K) | scale factornow / scale factorthen | time |
---|---|---|---|
strong forces freeze out | 1027 | 3.7 * 1026 | 10-35 s |
weak forces freeze out | 1015 | 3.7 * 1014 | 10-10 s |
protons, neutrons freeze out | 1013 | 3.7 * 1012 | 0.0001 s |
neutrinos decouple | 3 * 1010 | 1.1 * 1010 | 1 s |
electrons freeze out | 6 * 109 | 2.2 * 109 | 100 s |
primordial 2H, 4He form | 9 * 108 | 3.3 * 108 | 2-15 minutes |
event | temperature (K) | scale factornow / scale factorthen | time |
---|---|---|---|
photons decouple, atoms form | 3000 | 1091 | 377000 years |
first stars | 60 | 10.4 | 109 years |
today | 2.73 | 1 | 1.378 * 1010 years |
and COOLER after 380,000 years had elapsed. All of the foregoing makes a decent case for a universe which was very very tiny right after the BB. AG
On Sat, Sep 14, 2019 at 9:33 AM Alan Grayson <agrays...@gmail.com> wrote:>> There is no reason to think physics needs all the real numbers and considerable evidence to think it does not. To my mind the strongest evidence is that a physical Turing Machine is incapable of even approximating most real numbers, I happened to have posted a proof of this yesterday on the "Observation versus assumption" thread.> Physics doesn't need all the real numbers, just some of them, say any continuous range of any variable; like the mass of the electron.The electron doesn't have a continuous range of mass.
And mass is the force on a object divided by its acceleration, but acceleration is the change in speed per unit of time and speed is the change in positional distance per unit of time, so if neither time or space is continuous then mass can't be either.
and inflation preserved the homogeneity. This is what Guth was trying to solve with inflation, among other problems, such as no detectable monopoles. This entire logic breaks down if one assumes an infinite universe at the time of inflation.Correct, using inflation and previous causal connectedness does not produce for homogeneity of temperature to all parts of the universe if the universe is infinite.
At best it can only extend to some finite region of that universe. But once you are working in an inflationary model, you already have accepted there is a large scale where the universe is not homogenous (pocket regions vs. the rapidly inflating regions of vacuum).
In this case, the infinite universe was always homogeneous even though it was never causally connected.That is another possibility that avoids inflation as an explanation of homogeneity: To simply assume everything at all places began at the same temperature and density.
Further, how could it have been so hot 380,000 years after the BB if it wasn't dense at that time?Actually the universe was not very dense at the time of 380,000 years. It was billions of times more sparse than Earth's atmosphere. Each time the scale factor halves going backwards in time, the temperature doubles, and the density increases by a factor of 8 (2 cubed). You can follows this backwards at least until the temperature is about 10^27 K, far far hotter and denser than 380,000 years, back to a time just a fraction of a second after inflation ended.
I prefer to go with what we think we know, rather than with a model which is completely speculative. AG
and inflation preserved the homogeneity. This is what Guth was trying to solve with inflation, among other problems, such as no detectable monopoles. This entire logic breaks down if one assumes an infinite universe at the time of inflation.Correct, using inflation and previous causal connectedness does not produce for homogeneity of temperature to all parts of the universe if the universe is infinite.So far, as I just stated, our best evidence
does NOT suggest an infinite universe. AG
At best it can only extend to some finite region of that universe. But once you are working in an inflationary model, you already have accepted there is a large scale where the universe is not homogenous (pocket regions vs. the rapidly inflating regions of vacuum).I don't see why assuming inflation implies acceptance of large parts of the UNobservable universe which is NOT homogeneous. AG
In this case, the infinite universe was always homogeneous even though it was never causally connected.That is another possibility that avoids inflation as an explanation of homogeneity: To simply assume everything at all places began at the same temperature and density.If so, why did Guth think homogeneity needed an explanation? On its face, thermal equilibrium for a non causally connected universe seems improbable. AG
Further, how could it have been so hot 380,000 years after the BB if it wasn't dense at that time?Actually the universe was not very dense at the time of 380,000 years. It was billions of times more sparse than Earth's atmosphere. Each time the scale factor halves going backwards in time, the temperature doubles, and the density increases by a factor of 8 (2 cubed). You can follows this backwards at least until the temperature is about 10^27 K, far far hotter and denser than 380,000 years, back to a time just a fraction of a second after inflation ended.Yes, it was far hotter and denser just after the BB, than at 380,000 years.
But contrary to what you allege above and below, it must have far hotter and denser at 380,000 years, than it is today, 2.7 deg K, so hot and dense that it was opaque to light.
I am just saying that it does seem to be cooling as it expands,
and the curvature data seems to imply smallness just after the BB.
Moreover, applying the Cosmological principle, it couldn't have been homogeneous on large scale in the finite observable region, and at the same time infinite and non-homogeneous in regions we can't observe. AG
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/666ba0a0-918d-422f-b2ed-f0ded39b82b1%40googlegroups.com.
and inflation preserved the homogeneity. This is what Guth was trying to solve with inflation, among other problems, such as no detectable monopoles. This entire logic breaks down if one assumes an infinite universe at the time of inflation.Correct, using inflation and previous causal connectedness does not produce for homogeneity of temperature to all parts of the universe if the universe is infinite.So far, as I just stated, our best evidenceThere's no evidence either way, as far as I am aware, which is why i is still considered an open question. If you can point me to some evidence I would be interested.does NOT suggest an infinite universe. AGWhat are you calling as the universe here? How are you defining it?
At best it can only extend to some finite region of that universe. But once you are working in an inflationary model, you already have accepted there is a large scale where the universe is not homogenous (pocket regions vs. the rapidly inflating regions of vacuum).I don't see why assuming inflation implies acceptance of large parts of the UNobservable universe which is NOT homogeneous. AGBecause decay events of the vacuum do not happen everywhere at once, this leads to isolated "pocket universes" separated by exponentially expanding space. The inhomogenity I am referring to are the different parts of the vacuum in different energy states.In this case, the infinite universe was always homogeneous even though it was never causally connected.That is another possibility that avoids inflation as an explanation of homogeneity: To simply assume everything at all places began at the same temperature and density.If so, why did Guth think homogeneity needed an explanation? On its face, thermal equilibrium for a non causally connected universe seems improbable. AGIt came for free, with the other explanations. On its own, I am not sure it would be justified to trade one assumption for another, but inflation replaced 4 or 5 assumptions with a single one, which is its main strength.Further, how could it have been so hot 380,000 years after the BB if it wasn't dense at that time?Actually the universe was not very dense at the time of 380,000 years. It was billions of times more sparse than Earth's atmosphere. Each time the scale factor halves going backwards in time, the temperature doubles, and the density increases by a factor of 8 (2 cubed). You can follows this backwards at least until the temperature is about 10^27 K, far far hotter and denser than 380,000 years, back to a time just a fraction of a second after inflation ended.Yes, it was far hotter and denser just after the BB, than at 380,000 years.Okay.But contrary to what you allege above and below, it must have far hotter and denser at 380,000 years, than it is today, 2.7 deg K, so hot and dense that it was opaque to light.I'm not sure how this is contrary to what I say above and below... I agree it was hotter and denser the farther back you go.
I am just saying that it does seem to be cooling as it expands,Yes.and the curvature data seems to imply smallness just after the BB.What curvature data are you referring to? The latest Planck data say the curvature is flat to within the limits of our measurement accuracy. Is there a new result that indicates positive curvature?
Moreover, applying the Cosmological principle, it couldn't have been homogeneous on large scale in the finite observable region, and at the same time infinite and non-homogeneous in regions we can't observe. AGIt all comes down to scale. At the scale of stars or galaxies, the universe is non homogeneous, on the scale of super clusters and above it is, but at larger scales of inflating vacuums and pocket universes, again it is non homogeneous, but perhaps if you zoom out far enough the picture becomes homogeneous again. The non-homogeneous part I am referring to can be seen as the spiky image, a rendering of eternal inflation: https://www.preposterousuniverse.com/blog/2011/10/21/the-eternally-existing-self-reproducing-frequently-puzzling-inflationary-universe/
Jason
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
and inflation preserved the homogeneity. This is what Guth was trying to solve with inflation, among other problems, such as no detectable monopoles. This entire logic breaks down if one assumes an infinite universe at the time of inflation.Correct, using inflation and previous causal connectedness does not produce for homogeneity of temperature to all parts of the universe if the universe is infinite.So far, as I just stated, our best evidenceThere's no evidence either way, as far as I am aware, which is why i is still considered an open question. If you can point me to some evidence I would be interested.does NOT suggest an infinite universe. AGWhat are you calling as the universe here? How are you defining it?I am referring to our bubble, which arose with the BB, and refers to the observable and UNobservable regions (not to the possibly infinite substrate from which it arose). AG
At best it can only extend to some finite region of that universe. But once you are working in an inflationary model, you already have accepted there is a large scale where the universe is not homogenous (pocket regions vs. the rapidly inflating regions of vacuum).I don't see why assuming inflation implies acceptance of large parts of the UNobservable universe which is NOT homogeneous. AGBecause decay events of the vacuum do not happen everywhere at once, this leads to isolated "pocket universes" separated by exponentially expanding space. The inhomogenity I am referring to are the different parts of the vacuum in different energy states.In this case, the infinite universe was always homogeneous even though it was never causally connected.That is another possibility that avoids inflation as an explanation of homogeneity: To simply assume everything at all places began at the same temperature and density.If so, why did Guth think homogeneity needed an explanation? On its face, thermal equilibrium for a non causally connected universe seems improbable. AGIt came for free, with the other explanations. On its own, I am not sure it would be justified to trade one assumption for another, but inflation replaced 4 or 5 assumptions with a single one, which is its main strength.Further, how could it have been so hot 380,000 years after the BB if it wasn't dense at that time?Actually the universe was not very dense at the time of 380,000 years. It was billions of times more sparse than Earth's atmosphere. Each time the scale factor halves going backwards in time, the temperature doubles, and the density increases by a factor of 8 (2 cubed). You can follows this backwards at least until the temperature is about 10^27 K, far far hotter and denser than 380,000 years, back to a time just a fraction of a second after inflation ended.Yes, it was far hotter and denser just after the BB, than at 380,000 years.Okay.But contrary to what you allege above and below, it must have far hotter and denser at 380,000 years, than it is today, 2.7 deg K, so hot and dense that it was opaque to light.I'm not sure how this is contrary to what I say above and below... I agree it was hotter and denser the farther back you go.And smaller as well? (BTW, "smaller" can't be a property of a spatially infinite universe.)
It had to have gotten smaller to explain its present homogeneity. I want to avoid the assumption that homogeneity can arise spontaneously in a causally DIS-connected universe, the one we observe.
And I don't believe that at 380,000 years it was less dense than our atmosphere (as you earlier alleged). AG
I am just saying that it does seem to be cooling as it expands,Yes.and the curvature data seems to imply smallness just after the BB.What curvature data are you referring to? The latest Planck data say the curvature is flat to within the limits of our measurement accuracy. Is there a new result that indicates positive curvature?"Flat" means curvature is exactly zero; that is, flat like a Euclidean plane. But if we measure slightly positive, which I think is the case, it must be a closed hyperspace, but HUGE. Physicists tend to equate "almost flat", which if true would mean a huge spherical hyperspace, with Euclidean flat. This is a persistent error. AGWhat I don't understand is why, a universe with accelerating expansion, must be open, like a saddle.
Why can't a spherical hyperspace retain its closure if its expansion is accelerating? AG
Moreover, applying the Cosmological principle, it couldn't have been homogeneous on large scale in the finite observable region, and at the same time infinite and non-homogeneous in regions we can't observe. AGIt all comes down to scale. At the scale of stars or galaxies, the universe is non homogeneous, on the scale of super clusters and above it is, but at larger scales of inflating vacuums and pocket universes, again it is non homogeneous, but perhaps if you zoom out far enough the picture becomes homogeneous again. The non-homogeneous part I am referring to can be seen as the spiky image, a rendering of eternal inflation: https://www.preposterousuniverse.com/blog/2011/10/21/the-eternally-existing-self-reproducing-frequently-puzzling-inflationary-universe/I would forget about inflating vacuums and pocket universes, which are totally speculative,
and focus on what we can observe -- which, on a large scale, is homogeneous. Why trash the Cosmological Principle by appealig to unobservable phenomena? AG
and COOLER after 380,000 years had elapsed. All of the foregoing makes a decent case for a universe which was very very tiny right after the BB. AGI still see no connection between the temperature at time 380,000 years, and the size of the universe. Can you do more to explain more why you think there is a relation? I can see how you might relate the initial temperature and density at an earlier time to the temperature and density after 380,000 years, but I am not seeing how you relate the size of the universe to either the temperature or density at time 380,000 years.Oh, because the temperature is decreasing from just after the BB to 380,000 years, we need a very small universe to inflate to explain the current homogeneity. Otherwise the present large scale homogeneity is only explicable by appealing to highly improbable chance in a causally disconnected universe, our present universe. AG
On Sun, Sep 15, 2019 at 7:45 AM Alan Grayson <agrays...@gmail.com> wrote:and COOLER after 380,000 years had elapsed. All of the foregoing makes a decent case for a universe which was very very tiny right after the BB. AGI still see no connection between the temperature at time 380,000 years, and the size of the universe. Can you do more to explain more why you think there is a relation? I can see how you might relate the initial temperature and density at an earlier time to the temperature and density after 380,000 years, but I am not seeing how you relate the size of the universe to either the temperature or density at time 380,000 years.Oh, because the temperature is decreasing from just after the BB to 380,000 years, we need a very small universe to inflate to explain the current homogeneity. Otherwise the present large scale homogeneity is only explicable by appealing to highly improbable chance in a causally disconnected universe, our present universe. AG
Inflation requires a minimum starting size (which can be microscopic), and minimum duration of inflation (which can be as little as ~100 doublings) taking as little as 10^-35 seconds, but as far as I know these are only the minimums to be congruent with observations. Inflation, by no means requires the preinflation universe to be tiny, nor the time period of inflation to be short. Either the preinflation size could be unboundedly large, or the inflation duration could be unboundedly long.Jason
At best it can only extend to some finite region of that universe. But once you are working in an inflationary model, you already have accepted there is a large scale where the universe is not homogenous (pocket regions vs. the rapidly inflating regions of vacuum).I don't see why assuming inflation implies acceptance of large parts of the UNobservable universe which is NOT homogeneous. AGBecause decay events of the vacuum do not happen everywhere at once, this leads to isolated "pocket universes" separated by exponentially expanding space. The inhomogenity I am referring to are the different parts of the vacuum in different energy states.In this case, the infinite universe was always homogeneous even though it was never causally connected.That is another possibility that avoids inflation as an explanation of homogeneity: To simply assume everything at all places began at the same temperature and density.If so, why did Guth think homogeneity needed an explanation? On its face, thermal equilibrium for a non causally connected universe seems improbable. AGIt came for free, with the other explanations. On its own, I am not sure it would be justified to trade one assumption for another, but inflation replaced 4 or 5 assumptions with a single one, which is its main strength.Further, how could it have been so hot 380,000 years after the BB if it wasn't dense at that time?Actually the universe was not very dense at the time of 380,000 years. It was billions of times more sparse than Earth's atmosphere. Each time the scale factor halves going backwards in time, the temperature doubles, and the density increases by a factor of 8 (2 cubed). You can follows this backwards at least until the temperature is about 10^27 K, far far hotter and denser than 380,000 years, back to a time just a fraction of a second after inflation ended.Yes, it was far hotter and denser just after the BB, than at 380,000 years.Okay.But contrary to what you allege above and below, it must have far hotter and denser at 380,000 years, than it is today, 2.7 deg K, so hot and dense that it was opaque to light.I'm not sure how this is contrary to what I say above and below... I agree it was hotter and denser the farther back you go.And smaller as well? (BTW, "smaller" can't be a property of a spatially infinite universe.)Smaller is not implied by the big bang model, only things being "previously closer".
I know many popular accounts of the BB say the universe was once smaller, but this is sloppy writing. They are referring to some fixed part of the universe being smaller, such as the observable part. But to say the universe in total was smaller is to assume one knows if it is infinite or finite, open/flat or closed. This is not known, so no accurate account of the BB would implicitly assume it to be known.It had to have gotten smaller to explain its present homogeneity. I want to avoid the assumption that homogeneity can arise spontaneously in a causally DIS-connected universe, the one we observe.But to explain that rather than assuming it, then you need inflation, but below you call this "totally speculative". Which is it?And I don't believe that at 380,000 years it was less dense than our atmosphere (as you earlier alleged). AGThe present density of the universe is about 5 hydrogen atoms per cubic meter. At the time of 380,000 years, things were ~1100x closer together (the scale factor is ~1/1100) compared to today. This is a simple calculation of the temperature difference. If it's 2.73K now, and it was 3000K then, then the scale factor growth from then to now is 3000/2.73 = 1098.If each dimension changed by a factor of 1,000, this means the density back then would have been 1,000 x 1,000 x 1,000 or a billion times what it is now. So instead of 5 hydrogen atoms per cubic meter, you get 5 billion hydrogen atoms per cubic meter. This is many many orders of magnitude less dense than atmospheric pressure. A cubic meter of air at sea level weighs 1.3 kilograms ( https://hypertextbook.com/facts/2000/RachelChu.shtml ). Compare this weight to the weight of 5 billion hydrogen atoms. A Hydogen atom weighs 1.67 × 10^-24 g, 5 billion of them would get you to 8.37 × 10^-18 kilograms.So I was wrong, it wasn't a billion times less dense, it was closer to a billion billion times less dense than the atmosphere.I am just saying that it does seem to be cooling as it expands,Yes.and the curvature data seems to imply smallness just after the BB.What curvature data are you referring to? The latest Planck data say the curvature is flat to within the limits of our measurement accuracy. Is there a new result that indicates positive curvature?"Flat" means curvature is exactly zero; that is, flat like a Euclidean plane. But if we measure slightly positive, which I think is the case, it must be a closed hyperspace, but HUGE. Physicists tend to equate "almost flat", which if true would mean a huge spherical hyperspace, with Euclidean flat. This is a persistent error. AGWhat I don't understand is why, a universe with accelerating expansion, must be open, like a saddle.The shape (closed, flat, open), depends on how much gravitating stuff is in the universe compared to how much anti-gravitating stuff is in the universe, and the current expansion rate and density. A closed universe implies gravitational attraction wins out in the end and things eventually collapse. In a universe where there is more gravitating stuff then anti gravitating stuff, the speed of expansion ought to be slowing down. If it is slowing down in a way that only after infinite time the expansion rate = 0 (loosely analogous to throwing something upwards at exactly the escape velocity) then the geometry is flat. But the only way for the universe expansion to be accelerating now is if the anti-gravity stuff exceeds the gravitating stuff. In this case, (should the condition persist), then the universe will not recollapse (can't be closed), nor will it come to a rest after infinite time (can't be flat), so the alternative is that it must be open.
Why can't a spherical hyperspace retain its closure if its expansion is accelerating? AGMathematically you can of course imagine an ever expanding hypersphere, but the reason it is not possible physically is comes down to general relativity, which informs of us of a relationship between the spatial curvature and the ultimate fate of the universe. So if anti-gravity stuff wins out such that the universe expands forever in an accelerating or constant rate, then GR requires that the spatial curvature be negative. It would not allow for a positive curvature.
Moreover, applying the Cosmological principle, it couldn't have been homogeneous on large scale in the finite observable region, and at the same time infinite and non-homogeneous in regions we can't observe. AGIt all comes down to scale. At the scale of stars or galaxies, the universe is non homogeneous, on the scale of super clusters and above it is, but at larger scales of inflating vacuums and pocket universes, again it is non homogeneous, but perhaps if you zoom out far enough the picture becomes homogeneous again. The non-homogeneous part I am referring to can be seen as the spiky image, a rendering of eternal inflation: https://www.preposterousuniverse.com/blog/2011/10/21/the-eternally-existing-self-reproducing-frequently-puzzling-inflationary-universe/I would forget about inflating vacuums and pocket universes, which are totally speculative,They're more or less a direct consequence of inflation. Inflation is a little bit more than totally speculative. I would go so far to saying it is at least weakly confirmed.and focus on what we can observe -- which, on a large scale, is homogeneous. Why trash the Cosmological Principle by appealig to unobservable phenomena? AGThe cosmological principle is not a firm rule or law, it is a rule of thumb which works under the assumption that the same laws operate everywhere and same conditions hold everywhere, and therefore things should be roughly the same everywhere. Inflation tells us that at certain scales the conditions are not the same everywhere, so we should not expect everything to seem homogeneous at those scales.
Jason
I know many popular accounts of the BB say the universe was once smaller, but this is sloppy writing. They are referring to some fixed part of the universe being smaller, such as the observable part. But to say the universe in total was smaller is to assume one knows if it is infinite or finite, open/flat or closed. This is not known, so no accurate account of the BB would implicitly assume it to be known.It had to have gotten smaller to explain its present homogeneity. I want to avoid the assumption that homogeneity can arise spontaneously in a causally DIS-connected universe, the one we observe.But to explain that rather than assuming it, then you need inflation, but below you call this "totally speculative". Which is it?And I don't believe that at 380,000 years it was less dense than our atmosphere (as you earlier alleged). AGThe present density of the universe is about 5 hydrogen atoms per cubic meter. At the time of 380,000 years, things were ~1100x closer together (the scale factor is ~1/1100) compared to today. This is a simple calculation of the temperature difference. If it's 2.73K now, and it was 3000K then, then the scale factor growth from then to now is 3000/2.73 = 1098.If each dimension changed by a factor of 1,000, this means the density back then would have been 1,000 x 1,000 x 1,000 or a billion times what it is now. So instead of 5 hydrogen atoms per cubic meter, you get 5 billion hydrogen atoms per cubic meter. This is many many orders of magnitude less dense than atmospheric pressure. A cubic meter of air at sea level weighs 1.3 kilograms ( https://hypertextbook.com/facts/2000/RachelChu.shtml ). Compare this weight to the weight of 5 billion hydrogen atoms. A Hydogen atom weighs 1.67 × 10^-24 g, 5 billion of them would get you to 8.37 × 10^-18 kilograms.So I was wrong, it wasn't a billion times less dense, it was closer to a billion billion times less dense than the atmosphere.I am just saying that it does seem to be cooling as it expands,Yes.and the curvature data seems to imply smallness just after the BB.What curvature data are you referring to? The latest Planck data say the curvature is flat to within the limits of our measurement accuracy. Is there a new result that indicates positive curvature?"Flat" means curvature is exactly zero; that is, flat like a Euclidean plane. But if we measure slightly positive, which I think is the case, it must be a closed hyperspace, but HUGE. Physicists tend to equate "almost flat", which if true would mean a huge spherical hyperspace, with Euclidean flat. This is a persistent error. AGWhat I don't understand is why, a universe with accelerating expansion, must be open, like a saddle.The shape (closed, flat, open), depends on how much gravitating stuff is in the universe compared to how much anti-gravitating stuff is in the universe, and the current expansion rate and density. A closed universe implies gravitational attraction wins out in the end and things eventually collapse. In a universe where there is more gravitating stuff then anti gravitating stuff, the speed of expansion ought to be slowing down. If it is slowing down in a way that only after infinite time the expansion rate = 0 (loosely analogous to throwing something upwards at exactly the escape velocity) then the geometry is flat. But the only way for the universe expansion to be accelerating now is if the anti-gravity stuff exceeds the gravitating stuff. In this case, (should the condition persist), then the universe will not recollapse (can't be closed), nor will it come to a rest after infinite time (can't be flat), so the alternative is that it must be open.That's what the books say. But suppose the universe was a spherical expanding hyperspace at some point in its history, closed, and then the expansion rate started to increase. Would that closed universe somehow "tear" and become open? AG
Why can't a spherical hyperspace retain its closure if its expansion is accelerating? AGMathematically you can of course imagine an ever expanding hypersphere, but the reason it is not possible physically is comes down to general relativity, which informs of us of a relationship between the spatial curvature and the ultimate fate of the universe. So if anti-gravity stuff wins out such that the universe expands forever in an accelerating or constant rate, then GR requires that the spatial curvature be negative. It would not allow for a positive curvature.I find this rather dubious. Can you show me how GR requires this? AG
Moreover, applying the Cosmological principle, it couldn't have been homogeneous on large scale in the finite observable region, and at the same time infinite and non-homogeneous in regions we can't observe. AGIt all comes down to scale. At the scale of stars or galaxies, the universe is non homogeneous, on the scale of super clusters and above it is, but at larger scales of inflating vacuums and pocket universes, again it is non homogeneous, but perhaps if you zoom out far enough the picture becomes homogeneous again. The non-homogeneous part I am referring to can be seen as the spiky image, a rendering of eternal inflation: https://www.preposterousuniverse.com/blog/2011/10/21/the-eternally-existing-self-reproducing-frequently-puzzling-inflationary-universe/I would forget about inflating vacuums and pocket universes, which are totally speculative,They're more or less a direct consequence of inflation. Inflation is a little bit more than totally speculative. I would go so far to saying it is at least weakly confirmed.and focus on what we can observe -- which, on a large scale, is homogeneous. Why trash the Cosmological Principle by appealig to unobservable phenomena? AGThe cosmological principle is not a firm rule or law, it is a rule of thumb which works under the assumption that the same laws operate everywhere and same conditions hold everywhere, and therefore things should be roughly the same everywhere. Inflation tells us that at certain scales the conditions are not the same everywhere, so we should not expect everything to seem homogeneous at those scales.The same laws must operate everywhere; otherwise we can't do physics. But obviously, within those laws, whatever they are, different events can occur in different locations. AGJason
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/c7b2d6c8-41f4-4588-a668-a1ae824a2c06%40googlegroups.com.
Sean Carroll's many-selves
On Sunday, September 15, 2019 at 3:34:10 PM UTC-6, Philip Thrift wrote:Sean Carroll's many-selvesAnd the good news is ... the one in this world is going bald. AG
I know many popular accounts of the BB say the universe was once smaller, but this is sloppy writing. They are referring to some fixed part of the universe being smaller, such as the observable part. But to say the universe in total was smaller is to assume one knows if it is infinite or finite, open/flat or closed. This is not known, so no accurate account of the BB would implicitly assume it to be known.It had to have gotten smaller to explain its present homogeneity. I want to avoid the assumption that homogeneity can arise spontaneously in a causally DIS-connected universe, the one we observe.But to explain that rather than assuming it, then you need inflation, but below you call this "totally speculative". Which is it?And I don't believe that at 380,000 years it was less dense than our atmosphere (as you earlier alleged). AGThe present density of the universe is about 5 hydrogen atoms per cubic meter. At the time of 380,000 years, things were ~1100x closer together (the scale factor is ~1/1100) compared to today. This is a simple calculation of the temperature difference. If it's 2.73K now, and it was 3000K then, then the scale factor growth from then to now is 3000/2.73 = 1098.If each dimension changed by a factor of 1,000, this means the density back then would have been 1,000 x 1,000 x 1,000 or a billion times what it is now. So instead of 5 hydrogen atoms per cubic meter, you get 5 billion hydrogen atoms per cubic meter. This is many many orders of magnitude less dense than atmospheric pressure. A cubic meter of air at sea level weighs 1.3 kilograms ( https://hypertextbook.com/facts/2000/RachelChu.shtml ). Compare this weight to the weight of 5 billion hydrogen atoms. A Hydogen atom weighs 1.67 × 10^-24 g, 5 billion of them would get you to 8.37 × 10^-18 kilograms.So I was wrong, it wasn't a billion times less dense, it was closer to a billion billion times less dense than the atmosphere.I am just saying that it does seem to be cooling as it expands,Yes.and the curvature data seems to imply smallness just after the BB.What curvature data are you referring to? The latest Planck data say the curvature is flat to within the limits of our measurement accuracy. Is there a new result that indicates positive curvature?"Flat" means curvature is exactly zero; that is, flat like a Euclidean plane. But if we measure slightly positive, which I think is the case, it must be a closed hyperspace, but HUGE. Physicists tend to equate "almost flat", which if true would mean a huge spherical hyperspace, with Euclidean flat. This is a persistent error. AGWhat I don't understand is why, a universe with accelerating expansion, must be open, like a saddle.The shape (closed, flat, open), depends on how much gravitating stuff is in the universe compared to how much anti-gravitating stuff is in the universe, and the current expansion rate and density. A closed universe implies gravitational attraction wins out in the end and things eventually collapse. In a universe where there is more gravitating stuff then anti gravitating stuff, the speed of expansion ought to be slowing down. If it is slowing down in a way that only after infinite time the expansion rate = 0 (loosely analogous to throwing something upwards at exactly the escape velocity) then the geometry is flat. But the only way for the universe expansion to be accelerating now is if the anti-gravity stuff exceeds the gravitating stuff. In this case, (should the condition persist), then the universe will not recollapse (can't be closed), nor will it come to a rest after infinite time (can't be flat), so the alternative is that it must be open.That's what the books say. But suppose the universe was a spherical expanding hyperspace at some point in its history, closed, and then the expansion rate started to increase. Would that closed universe somehow "tear" and become open? AGI'm not sure, it is a good question if the shape can evolve over time. Perhaps someone on this list more familiar with GR can answer.Why can't a spherical hyperspace retain its closure if its expansion is accelerating? AGMathematically you can of course imagine an ever expanding hypersphere, but the reason it is not possible physically is comes down to general relativity, which informs of us of a relationship between the spatial curvature and the ultimate fate of the universe. So if anti-gravity stuff wins out such that the universe expands forever in an accelerating or constant rate, then GR requires that the spatial curvature be negative. It would not allow for a positive curvature.I find this rather dubious. Can you show me how GR requires this? AGIt is made clear in the Friedmann equations, which are derived from the field equations of GR:Where they contain a parameter k which is either -1, 0, or 1, corresponding to the curvature of space. The value of k determines how the expansion rate changes over time.
Jason
--Moreover, applying the Cosmological principle, it couldn't have been homogeneous on large scale in the finite observable region, and at the same time infinite and non-homogeneous in regions we can't observe. AGIt all comes down to scale. At the scale of stars or galaxies, the universe is non homogeneous, on the scale of super clusters and above it is, but at larger scales of inflating vacuums and pocket universes, again it is non homogeneous, but perhaps if you zoom out far enough the picture becomes homogeneous again. The non-homogeneous part I am referring to can be seen as the spiky image, a rendering of eternal inflation: https://www.preposterousuniverse.com/blog/2011/10/21/the-eternally-existing-self-reproducing-frequently-puzzling-inflationary-universe/I would forget about inflating vacuums and pocket universes, which are totally speculative,They're more or less a direct consequence of inflation. Inflation is a little bit more than totally speculative. I would go so far to saying it is at least weakly confirmed.and focus on what we can observe -- which, on a large scale, is homogeneous. Why trash the Cosmological Principle by appealig to unobservable phenomena? AGThe cosmological principle is not a firm rule or law, it is a rule of thumb which works under the assumption that the same laws operate everywhere and same conditions hold everywhere, and therefore things should be roughly the same everywhere. Inflation tells us that at certain scales the conditions are not the same everywhere, so we should not expect everything to seem homogeneous at those scales.The same laws must operate everywhere; otherwise we can't do physics. But obviously, within those laws, whatever they are, different events can occur in different locations. AGJason
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
On 13 Sep 2019, at 15:24, Alan Grayson <agrays...@gmail.com> wrote:
On Friday, September 13, 2019 at 5:24:11 AM UTC-6, Bruno Marchal wrote:
On 13 Sep 2019, at 04:26, Alan Grayson <agrays...@gmail.com> wrote:
On Thursday, September 12, 2019 at 11:01:54 AM UTC-6, Alan Grayson wrote:
On Thursday, September 12, 2019 at 7:45:22 AM UTC-6, Lawrence Crowell wrote:
On Thursday, September 12, 2019 at 4:20:46 AM UTC-5, Philip Thrift wrote:
On Wednesday, September 11, 2019 at 11:45:41 PM UTC-5, Alan Grayson wrote:
Many Worlds is where people go to escape from one world of quantum-stochastic processes. They are like vampires, but instead of running away from sunbeams, are running away from probabilities.@philipthrift
This assessment is not entirely fair. Carroll and Sebens have a paper on how supposedly the Born rule can be derived from MWI I have yet to read their paper, but given the newsiness of this I might get to it. One advantage that MWI does have is that it splits the world as a sort of quantum frame dragging that is nonlocal. This nonlocal property might be useful for working with quantum gravity,I worked a proof of a theorem, which may not be complete unfortunately, where the two sets of quantum interpretations that are ψ-epistemic and those that are ψ-ontological are not decidable. There is no decision procedure which can prove QM holds either way. The proof is set with nonlocal hidden variables over the projective rays of the state space. In effect there is an uncertainty in whether the hidden variables localize extant quantities, say with ψ-ontology, or whether this localization is the generation of information in a local context from quantum nonlocality that is not extant, such as with ψ-epistemology. Quantum interprertations are then auxiliary physical axioms or postulates. MWI and within the framework of what Carrol and Sebens has done this is a ψ-ontology, and this defines the Born rule. If I am right the degree of ψ-epistemontic nature is mixed. So the intriguing question we can address is the nature of the Born rule and its tie into the auxiliary postulates of quantum interpretations. Can a similar demonstration be made for the Born rule within QuBism, which is what might be called the dialectic opposite of MWI?To take MWI as something literal, as opposed to maybe a working system to understand QM foundations, is maybe taking things too far. However, it is a part of some open questions concerning the fundamentals of QM. If MWI, and more generally postulates of quantum interpretations, are connected to the Born rule it makes for some interesting things to think about.LC
If you read the link, it's pretty obvious that Carroll believes the many worlds of the MWI, literally exist. AG
Carroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued.
The idea comes from Tegmark, and I agree with you, it necessitate more than an infinite universe. It requires also some assumption of homogeneity.
Our universe is, on a large scale, homogeneous. But it can't be infinite since it has only been expanding for finite time, 13.8 BY.
I had a discussion with Brent about this some time ago, and he claimed finite in time doesn't preclude infinite in space. I strongly disagree. Perhaps I am missing something. Wouldn't be the first time. AG
Of course, (for those who are aware of Gödel 1931 and Turing 1936), arithmetic contains all computations, which entails, when assuming mechanism, an infinity of each os us.I really don't see how you make that jump.
And what exactly does "assuming mechanism" mean? AG
That explains both where the appearance of universe come from, and the quantum mechanical type of formalism. In “many-world”, the “many” makes sense, but the term “world” is not well defined and should not been taken literally. It is more histories than worlds per se.BrunoWhat's the argument for such a claim? Morevover, I don't believe a universe of finite age, such as ours which everyone more or less agrees began some 13.8 BYA, can be spatially infinite. Here I'm referring to our bubble, not some infinite substratum from which it might have arose. AG
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/f85775c7-7914-47e4-83eb-1142b1b58249%40googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/46e1d02d-2609-4316-8818-781a6db0814a%40googlegroups.com.
On 13 Sep 2019, at 15:28, Alan Grayson <agrays...@gmail.com> wrote:
On Friday, September 13, 2019 at 5:18:50 AM UTC-6, Bruno Marchal wrote:On 13 Sep 2019, at 00:44, Lawrence Crowell <goldenfield...@gmail.com> wrote:
On Thursday, September 12, 2019 at 11:44:51 AM UTC-5, Philip Thrift wrote:
On Thursday, September 12, 2019 at 8:45:22 AM UTC-5, Lawrence Crowell wrote:On Thursday, September 12, 2019 at 4:20:46 AM UTC-5, Philip Thrift wrote:
On Wednesday, September 11, 2019 at 11:45:41 PM UTC-5, Alan Grayson wrote:Many Worlds is where people go to escape from one world of quantum-stochastic processes. They are like vampires, but instead of running away from sunbeams, are running away from probabilities.@philipthriftThis assessment is not entirely fair. Carroll and Sebens have a paper on how supposedly the Born rule can be derived from MWI I have yet to read their paper, but given the newsiness of this I might get to it. One advantage that MWI does have is that it splits the world as a sort of quantum frame dragging that is nonlocal. This nonlocal property might be useful for working with quantum gravity,I worked a proof of a theorem, which may not be complete unfortunately, where the two sets of quantum interpretations that are ψ-epistemic and those that are ψ-ontological are not decidable. There is no decision procedure which can prove QM holds either way. The proof is set with nonlocal hidden variables over the projective rays of the state space. In effect there is an uncertainty in whether the hidden variables localize extant quantities, say with ψ-ontology, or whether this localization is the generation of information in a local context from quantum nonlocality that is not extant, such as with ψ-epistemology. Quantum interprertations are then auxiliary physical axioms or postulates. MWI and within the framework of what Carrol and Sebens has done this is a ψ-ontology, and this defines the Born rule. If I am right the degree of ψ-epistemontic nature is mixed. So the intriguing question we can address is the nature of the Born rule and its tie into the auxiliary postulates of quantum interpretations. Can a similar demonstration be made for the Born rule within QuBism, which is what might be called the dialectic opposite of MWI?To take MWI as something literal, as opposed to maybe a working system to understand QM foundations, is maybe taking things too far. However, it is a part of some open questions concerning the fundamentals of QM. If MWI, and more generally postulates of quantum interpretations, are connected to the Born rule it makes for some interesting things to think about.LC
QBism is not the dialectical opposite of MWI. This is:@philipthriftThe MWI and this path integral interpretation are both ψ-ontic and are thus not opposite.I agree. I would even add that with Feynman path formalism, the reduction of the wave packet does no more make sense. Feynman said it in his little book on light: he consider the Wave reduction as a confusion and appeal to magic (footnote at the end of the second chapter).BrunoNot for those of us who watch horseraces! Applied to QM, the wf becomes irrelevant when the measurement occurs.
Wave packet reduction, by which I assume you mean "collapse", is nothing more than a bookkeeping device. AG
LC
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/fe4b02a2-9fcb-4126-b2ad-fb9982f20fc1%40googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/10ffd66d-76ce-474b-a4e4-4f858f0e5153%40googlegroups.com.
On 13 Sep 2019, at 15:40, Philip Thrift <cloud...@gmail.com> wrote:
On Friday, September 13, 2019 at 6:17:12 AM UTC-5, Alan Grayson wrote:Carroll is in irreversible mental decline. He's lost contact with reality. Sad case. I stand by my assessment. He doesn't even understand basic linear algebra, and that his "state vector" has no unique representation, and thus the mythical interpretation of the superposition of the wf is totally illusional. AGHe can play with math, like anyone else, but his fictions are a little too real for him.
In the landscape of fictions modeling quantum phenomena, his not only denies probability, but denies the 'self' (in the sense of consciousness bring a real thing).
@philipthrift
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/07640a35-36de-43ec-8338-1c50aeeb39f4%40googlegroups.com.
On 13 Sep 2019, at 23:44, Lawrence Crowell <goldenfield...@gmail.com> wrote:
On Friday, September 13, 2019 at 6:24:11 AM UTC-5, Bruno Marchal wrote:
On 13 Sep 2019, at 04:26, Alan Grayson <agrays...@gmail.com> wrote:
On Thursday, September 12, 2019 at 11:01:54 AM UTC-6, Alan Grayson wrote:
On Thursday, September 12, 2019 at 7:45:22 AM UTC-6, Lawrence Crowell wrote:On Thursday, September 12, 2019 at 4:20:46 AM UTC-5, Philip Thrift wrote:
On Wednesday, September 11, 2019 at 11:45:41 PM UTC-5, Alan Grayson wrote:Many Worlds is where people go to escape from one world of quantum-stochastic processes. They are like vampires, but instead of running away from sunbeams, are running away from probabilities.@philipthriftThis assessment is not entirely fair. Carroll and Sebens have a paper on how supposedly the Born rule can be derived from MWI I have yet to read their paper, but given the newsiness of this I might get to it. One advantage that MWI does have is that it splits the world as a sort of quantum frame dragging that is nonlocal. This nonlocal property might be useful for working with quantum gravity,I worked a proof of a theorem, which may not be complete unfortunately, where the two sets of quantum interpretations that are ψ-epistemic and those that are ψ-ontological are not decidable. There is no decision procedure which can prove QM holds either way. The proof is set with nonlocal hidden variables over the projective rays of the state space. In effect there is an uncertainty in whether the hidden variables localize extant quantities, say with ψ-ontology, or whether this localization is the generation of information in a local context from quantum nonlocality that is not extant, such as with ψ-epistemology. Quantum interprertations are then auxiliary physical axioms or postulates. MWI and within the framework of what Carrol and Sebens has done this is a ψ-ontology, and this defines the Born rule. If I am right the degree of ψ-epistemontic nature is mixed. So the intriguing question we can address is the nature of the Born rule and its tie into the auxiliary postulates of quantum interpretations. Can a similar demonstration be made for the Born rule within QuBism, which is what might be called the dialectic opposite of MWI?To take MWI as something literal, as opposed to maybe a working system to understand QM foundations, is maybe taking things too far. However, it is a part of some open questions concerning the fundamentals of QM. If MWI, and more generally postulates of quantum interpretations, are connected to the Born rule it makes for some interesting things to think about.LC
If you read the link, it's pretty obvious that Carroll believes the many worlds of the MWI, literally exist. AG
Carroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued.
The idea comes from Tegmark, and I agree with you, it necessitate more than an infinite universe. It requires also some assumption of homogeneity.
Of course, (for those who are aware of Gödel 1931 and Turing 1936), arithmetic contains all computations, which entails, when assuming mechanism, an infinity of each os us. That explains both where the appearance of universe come from, and the quantum mechanical type of formalism. In “many-world”, the “many” makes sense, but the term “world” is not well defined and should not been taken literally. It is more histories than worlds per se.BrunoIt would be best to separate MWI from the multiverse for at least the moment. There are several levels of the multiverse. MWI does define a high level multiverse, but MWI is not all multiverses.The first level has to do with what exists beyond the cosmological horizon and in particular if the spatial surface of spacetime is flat. This would be an infinite R^3 manifold. Since the level of complexity or the number of possible states is bounded by the size of the cosmology horizon, out about 13 billion light years, this means there are other regions that are copies of this world. This is just plain combinatorics.The type II multiverse, or maybe type IIA, is where a deSitter or FLRW spacetime with an inflationary vacuum at high energy is unstable and there are vacuum transitions in regions within it. These regions have a vacuum at a much lower energy and define what are sometimes called pocket worlds. Since this inflationary cosmology is in a hugely accelerated expansion then these pockets of vacuum instability are only local. There are some interesting questions. In particular is the boundary of any such pocket world pinched off, so to speak, to define a new detached and topologically complete spacetime? The boundary has quantum field information, and if this transitions to a spatial surface that is a sphere S^3 or R^3, then this quantum field information may play some role.The next is a type IIB multiverse, which is a multiplicity of these eternally inflating spacetimes. These may be emergent from entanglement wedges with AdS spacetimes.There is then the type III multiverse, which is the identification of the MWI as a multiverse. This is where things may get a bit odd. The problem is that quantum interpretations are physically questionable, or maybe metaphysics, and this would mean the type III multiverse is really a sort of metaphysics. The other multiverse scenarios may hold without MWI.
Then there is type IV multiverse that is Max Tegmark's mathematical universe hypothesis. This is so completely metaphysical that it is hard to take this as a serious scientific proposition.
LC
What's the argument for such a claim? Morevover, I don't believe a universe of finite age, such as ours which everyone more or less agrees began some 13.8 BYA, can be spatially infinite. Here I'm referring to our bubble, not some infinite substratum from which it might have arose. AG
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/f85775c7-7914-47e4-83eb-1142b1b58249%40googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/e8f736e0-cf55-4eaa-93da-145d2073bd73%40googlegroups.com.
On 14 Sep 2019, at 05:22, Alan Grayson <agrays...@gmail.com> wrote:
On Friday, September 13, 2019 at 4:08:23 PM UTC-6, John Clark wrote:On Thu, Sep 12, 2019 at 10:26 PM Alan Grayson <agrays...@gmail.com> wrote:> Carroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued. What's the argument for such a claim?Of course it's been proven! It's simple math, there are only a finite number of ways the atoms in your body, or even the entire OBSERVABLE universe, can be arranged so obviously if the entire universe is infinite then there is going to have to be copies, an infinite number of them in fact. Max Tegmark has even calculated how far you'd have to go to see such a thing.
What I think you're missing (and Tegmark) is the possibility of UNcountable universes. In such case, one could imagine new universes coming into existence forever and ever, without any repeats. Think of the number of points between 0 and 1 on the real line, each point associated with a different universe. AG
Your closest identical copy is 10^12 light years away. About 10^76 light years away there is a sphere of radius 100 light-years identical to the one centered here, so everything we see here during the next century will be identical to those of our counterparts over there. And 10^102 light years away the is a exact copy of our entire observable universe. And all this is true regardless of if the Many Worlds Interpretation of Quantum Mechanics is correct or not, it only depends on the universe being spatially infinite.But our universe is NOT spatially infinite if its been expanding for finite time, starting very small, as can be inferred from the temperature of the CMBR. AG
> Morevover, I don't believe a universe of finite age, such as ours which everyone more or less agrees began some 13.8 BYA, can be spatially infinite.
I see no reason in principle why something can't be finite along one dimension and infinite along another dimension.In general, one can of course have some dimensions finite and others infinite. But if our universe is finite in time since the BB, 13.8 BY, its spatial extent must be finite, since that's how long its been expanding. AG
John K Clark
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/5a270b8e-3bf2-4d34-b0e7-4e0daa3cebce%40googlegroups.com.
On 14 Sep 2019, at 05:22, Alan Grayson <agrays...@gmail.com> wrote:
On Friday, September 13, 2019 at 4:08:23 PM UTC-6, John Clark wrote:On Thu, Sep 12, 2019 at 10:26 PM Alan Grayson <agrays...@gmail.com> wrote:> Carroll also believes that IF the universe is infinite, then there must exist exact copies of universes and ourselves. This is frequently claimed by the MWI true believers, but never, AFAICT, proven, or even plausibly argued. What's the argument for such a claim?Of course it's been proven! It's simple math, there are only a finite number of ways the atoms in your body, or even the entire OBSERVABLE universe, can be arranged so obviously if the entire universe is infinite then there is going to have to be copies, an infinite number of them in fact. Max Tegmark has even calculated how far you'd have to go to see such a thing.What I think you're missing (and Tegmark) is the possibility of UNcountable universes. In such case, one could imagine new universes coming into existence forever and ever, without any repeats. Think of the number of points between 0 and 1 on the real line, each point associated with a different universe. AGTegmark missed this?Deutsch did not, and in his book “fabric of reality”, he gave rather good argument in favour of Everett-type of multiverse having non countable universe. That makes sense with mechanism which give raise to a continuum (2^aleph_0) of histories, but the “equivalence class” brought by the measure can have lower cardinality, or bigger. Open problem, to say the least.
Your closest identical copy is 10^12 light years away. About 10^76 light years away there is a sphere of radius 100 light-years identical to the one centered here, so everything we see here during the next century will be identical to those of our counterparts over there. And 10^102 light years away the is a exact copy of our entire observable universe. And all this is true regardless of if the Many Worlds Interpretation of Quantum Mechanics is correct or not, it only depends on the universe being spatially infinite.But our universe is NOT spatially infinite if its been expanding for finite time, starting very small, as can be inferred from the temperature of the CMBR. AG> Morevover, I don't believe a universe of finite age, such as ours which everyone more or less agrees began some 13.8 BYA, can be spatially infinite.I see no reason in principle why something can't be finite along one dimension and infinite along another dimension.In general, one can of course have some dimensions finite and others infinite. But if our universe is finite in time since the BB, 13.8 BY, its spatial extent must be finite, since that's how long its been expanding. AGI agree with Grayson here. (Accepting a lot of premises, like the BB is the beginning of the physical reality, which I doubt).Bruno
John K Clark--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everyth...@googlegroups.com.
It had to have gotten smaller to explain its present homogeneity. I want to avoid the assumption that homogeneity can arise spontaneously in a causally DIS-connected universe, the one we observe.But to explain that rather than assuming it, then you need inflation, but below you call this "totally speculative". Which is it?And I don't believe that at 380,000 years it was less dense than our atmosphere (as you earlier alleged). AGThe present density of the universe is about 5 hydrogen atoms per cubic meter. At the time of 380,000 years, things were ~1100x closer together (the scale factor is ~1/1100) compared to today. This is a simple calculation of the temperature difference. If it's 2.73K now, and it was 3000K then, then the scale factor growth from then to now is 3000/2.73 = 1098.If each dimension changed by a factor of 1,000, this means the density back then would have been 1,000 x 1,000 x 1,000 or a billion times what it is now. So instead of 5 hydrogen atoms per cubic meter, you get 5 billion hydrogen atoms per cubic meter. This is many many orders of magnitude less dense than atmospheric pressure. A cubic meter of air at sea level weighs 1.3 kilograms ( https://hypertextbook.com/facts/2000/RachelChu.shtml ). Compare this weight to the weight of 5 billion hydrogen atoms. A Hydogen atom weighs 1.67 × 10^-24 g, 5 billion of them would get you to 8.37 × 10^-18 kilograms.So I was wrong, it wasn't a billion times less dense, it was closer to a billion billion times less dense than the atmosphere.I am just saying that it does seem to be cooling as it expands,Yes.and the curvature data seems to imply smallness just after the BB.What curvature data are you referring to? The latest Planck data say the curvature is flat to within the limits of our measurement accuracy. Is there a new result that indicates positive curvature?"Flat" means curvature is exactly zero; that is, flat like a Euclidean plane. But if we measure slightly positive, which I think is the case, it must be a closed hyperspace, but HUGE. Physicists tend to equate "almost flat", which if true would mean a huge spherical hyperspace, with Euclidean flat. This is a persistent error. AGWhat I don't understand is why, a universe with accelerating expansion, must be open, like a saddle.The shape (closed, flat, open), depends on how much gravitating stuff is in the universe compared to how much anti-gravitating stuff is in the universe, and the current expansion rate and density. A closed universe implies gravitational attraction wins out in the end and things eventually collapse. In a universe where there is more gravitating stuff then anti gravitating stuff, the speed of expansion ought to be slowing down. If it is slowing down in a way that only after infinite time the expansion rate = 0 (loosely analogous to throwing something upwards at exactly the escape velocity) then the geometry is flat. But the only way for the universe expansion to be accelerating now is if the anti-gravity stuff exceeds the gravitating stuff. In this case, (should the condition persist), then the universe will not recollapse (can't be closed), nor will it come to a rest after infinite time (can't be flat), so the alternative is that it must be open.Why can't a spherical hyperspace retain its closure if its expansion is accelerating? AGMathematically you can of course imagine an ever expanding hypersphere, but the reason it is not possible physically is comes down to general relativity, which informs of us of a relationship between the spatial curvature and the ultimate fate of the universe. So if anti-gravity stuff wins out such that the universe expands forever in an accelerating or constant rate, then GR requires that the spatial curvature be negative. It would not allow for a positive curvature.Moreover, applying the Cosmological principle, it couldn't have been homogeneous on large scale in the finite observable region, and at the same time infinite and non-homogeneous in regions we can't observe. AGIt all comes down to scale. At the scale of stars or galaxies, the universe is non homogeneous, on the scale of super clusters and above it is, but at larger scales of inflating vacuums and pocket universes, again it is non homogeneous, but perhaps if you zoom out far enough the picture becomes homogeneous again. The non-homogeneous part I am referring to can be seen as the spiky image, a rendering of eternal inflation: https://www.preposterousuniverse.com/blog/2011/10/21/the-eternally-existing-self-reproducing-frequently-puzzling-inflationary-universe/I would forget about inflating vacuums and pocket universes, which are totally speculative,They're more or less a direct consequence of inflation. Inflation is a little bit more than totally speculative. I would go so far to saying it is at least weakly confirmed.and focus on what we can observe -- which, on a large scale, is homogeneous. Why trash the Cosmological Principle by appealig to unobservable phenomena? AGThe cosmological principle is not a firm rule or law, it is a rule of thumb which works under the assumption that the same laws operate everywhere and same conditions hold everywhere, and therefore things should be roughly the same everywhere. Inflation tells us that at certain scales the conditions are not the same everywhere, so we should not expect everything to seem homogeneous at those scales.Jason
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CA%2BBCJUhB2L7sZX2qN6%2B6V5UNH6tm2GvTgTBCSjZ8QssNwbeeeA%40mail.gmail.com.
Why can't a spherical hyperspace retain its closure if its expansion is accelerating? AGMathematically you can of course imagine an ever expanding hypersphere, but the reason it is not possible physically is comes down to general relativity, which informs of us of a relationship between the spatial curvature and the ultimate fate of the universe. So if anti-gravity stuff wins out such that the universe expands forever in an accelerating or constant rate, then GR requires that the spatial curvature be negative. It would not allow for a positive curvature.I find this rather dubious. Can you show me how GR requires this? AGIt is made clear in the Friedmann equations, which are derived from the field equations of GR:Where they contain a parameter k which is either -1, 0, or 1, corresponding to the curvature of space. The value of k determines how the expansion rate changes over time.JasonMoreover, applying the Cosmological principle, it couldn't have been homogeneous on large scale in the finite observable region, and at the same time infinite and non-homogeneous in regions we can't observe. AGIt all comes down to scale. At the scale of stars or galaxies, the universe is non homogeneous, on the scale of super clusters and above it is, but at larger scales of inflating vacuums and pocket universes, again it is non homogeneous, but perhaps if you zoom out far enough the picture becomes homogeneous again. The non-homogeneous part I am referring to can be seen as the spiky image, a rendering of eternal inflation: https://www.preposterousuniverse.com/blog/2011/10/21/the-eternally-existing-self-reproducing-frequently-puzzling-inflationary-universe/I would forget about inflating vacuums and pocket universes, which are totally speculative,They're more or less a direct consequence of inflation. Inflation is a little bit more than totally speculative. I would go so far to saying it is at least weakly confirmed.and focus on what we can observe -- which, on a large scale, is homogeneous. Why trash the Cosmological Principle by appealig to unobservable phenomena? AGThe cosmological principle is not a firm rule or law, it is a rule of thumb which works under the assumption that the same laws operate everywhere and same conditions hold everywhere, and therefore things should be roughly the same everywhere. Inflation tells us that at certain scales the conditions are not the same everywhere, so we should not expect everything to seem homogeneous at those scales.The same laws must operate everywhere; otherwise we can't do physics. But obviously, within those laws, whatever they are, different events can occur in different locations. AGJason--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/c7b2d6c8-41f4-4588-a668-a1ae824a2c06%40googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CA%2BBCJUjF-JZ6qAJy0MoE1D%2BQegR_dhvACfVuM8%2B-P8VhRi-ASg%40mail.gmail.com.
On 16 Sep 2019, at 02:46, spudboy100 via Everything List <everyth...@googlegroups.com> wrote:My guess is among the physics community, most, would be mildly, skeptical of MWI, because it's a bridge too far to get evidence of, as yet and thus, unconcerned.
Having said this, many cosmologists are still having a cat fight about the Hubble Constant (The rate of cosmological expansion). my suspicion is, that once we get to the point of hanging truly gigantic telescopes on the periphery of the solar system, new discoveries will be made, and revisions to old laws of physics will be done. We'll gain a few definitive answers through observation, and we shall see that quantum in action at a vastly large scale. Relatedly, hey!, where's my dark matter? in fact, hey!, where's my fusion reactors. Ah! So much for the 'mentally fit' physicists and astronomers….
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/8643456.7738728.1568594779723%40mail.yahoo.com.
On Fri, Sep 13, 2019 at 08:25:06PM -0700, Alan Grayson wrote:
>
> Whether they're boring or not is irrelevant. As I previously posted, an
> uncountable infinity of universes is possible without any repeats. AG
>
Incorrect. Each world has a finite amount of information that defines
it, and consequently nonzero measure. If these worlds are drawn from
an uncountable infinite set, then there must be an uncountable number
of copies of each world.