Age of Universe relative to what?

[my_wan]

Then how do you think that your device can measure it independently of the simultaneity convention which is part of your coordinate choice? You seem to be arguing against the key point you are making.

Suppose we have a theory where the one-way speed of light in the +x direction is infinite and the one way speed of light in the -x direction is .5 c and the speed of light is c in the y and z directions and distances are unchanged wrt standard SR. If we have a standard coordinate system T,X,Y,Z in units where c=1 then our non-standard system is:
$t=T-X$
$x=X$
$y=Y$
$z=Z$

Do you see how this is nothing more than a change in simultaneity and how your device cannot distinguish between these two simultaneity conventions?

Of course, but look at what you have attributed the anisotropy to, the time component. What exactly do you mean to say when you attribute a metric (coordinate choice) to a coordinate variable that you are seeking to measure? In what way does the model used to define this variable as something distinct from the product of a coordinate choice? Even restricted solely to Galilean relativity, in what way does this model distinguish the coordinate designation from the coordinate choice induced location of kinetic energy? It sounds to me like what is being asked for, without explicitly saying so, is a measurement proving which of two meteors classical kinetic energy resides in. That's absurd even under purely Galilean relativity.

These questions are highly non-trivial and must be addressed to even ask the question. You cannot impose coordinate dependence just because of some vague notion that Newtonian kinetic energy must have some specific location, which it did not even prior to Einstein. So why then attempt to impose on classical mechanics a frame independent location that classical mechanics could not provide prior to Einstein?

If you want a better answer provide a better specification of what it is you want measure. Distance is relational construct, like kinetic energy, as is time. Do you wish explicitly postulate that space and time are measurably independent of the mechanistic constructs we measure it with? I do not get, after all the explanation provided, why you would then ask me to characterize a claim of a variance without squat of a description of what that variance relates to. Do you not see that your question implies, without specifying so, an attempt to get me to say I can physically measure a mechanistic difference between two coordinate choices? Do you not see that being a coordinate choice is not even a different distance, but merely a conversion like English to metric?

Yet under some circumstances the same question is an actual physical effect, rather than a coordinate choice, and leads to very real differences. So why do you not specify the circumstances if it can obviously go either way depending on those circumstances? Just like with $t$ coming back to a son older than yourself is a very real possibility. Are you trying to say, since a coordinate choice is not a physical effect, you can't possibly end up older than your son? Throwing out a raw variables (x,y,z,t) with 0 context and asking for an either/or is a strawman. A very boring strawman.

 

[my_wan]

Here is an interesting and perfectly valid perspective:
http://Newton.umsl.edu/run//traveler.html

 

[Dale]

Of course, but look at what you have attributed the anisotropy to, the time component.

Then propose an alternative which:
A) has a non isotropic one way speed of light
B) has an isotropic two way speed if light equal to c
C) does not attribute the anisotropy to the time component

I am not aware of any such alternative, which is why I have claimed that choosing the one way speed off light is the same as specifying your synchronization convention. But if you are aware of alternatives then I would be glad to learn.

 

[my_wan]

Then propose an alternative which:
A) has a non isotropic one way speed of light
B) has an isotropic two way speed if light equal to c
C) does not attribute the anisotropy to the time component

I am not aware of any such alternative, which is why I have claimed that choosing the one way speed off light is the same as specifying your synchronization convention. But if you are aware of alternatives then I would be glad to learn.

Though I doubt it the question sounds like you didn't get past the first sentence. It really makes no difference which variable you pick. Given the inverse relation between space and time isn't saying the time varied directionally the same as saying space varied directionally? When you said "nothing more than a change in simultaneity" in the original question, wasn't that equivalent to saying nothing more than a change in coordinate choices? Time and simultaneity are meaningless without a space over which it operates. So implicitly when you attached an operator to the time component you implied the inverse of that operator could equally well apply to the (x,y,z) components.

This is what I was getting at when I said "what do you mean [...]". This entails that the inverse of the anisotropic time component can equally be applied to the spatial component. Yet if applied to the spatial components it is wrong to also apply it to the time component in that same frame under SR. So which of those alternatives to you want to assume, or do you want to assume space and time are not precisely inversely related?

So that's 3 choices and the questions that's been asked of me didn't even explicitly specify one, even though I went over this already. Then when the test is objected to no specifications for what it is you presumed I thought the test was for in the first place. Yet I'm somehow supposed to psychically determine how to answer these questions again without any specification or acknowledgment, rebuttal, etc., of my repeated explanations.

Tell me I am allowed to assume GR and I'll tell exactly what I would expect the test I described to be able to accomplish, both in terms of anisotropic clock, distance, and a measure of $c \neq 299,792,458 m/s$. But just say "it" can't be done tells me squat about what "it" is. What others have posited as anisotropy doesn't in itself make the claim any more meaningful than saying 1 inch $\neq$ 2.54cm because 1 < 2.54. Yet you still expect me to make an absolute claim about 1 and 2.54 without saying squat about what those 2 numbers represent. BS.

 

[Dale]

Given the inverse relation between space and time isn't saying the time varied directionally the same as saying space varied directionally?

Not if you want to keep the two-way speed of light isotropic and equal to c.

So implicitly when you attached an operator to the time component you implied the inverse of that operator could equally well apply to the (x,y,z) components.

Go ahead and try it and see what happens to the two-way speed of light. Maybe you will find an example where you can change the spatial coordinates and not change the two-way speed of light, I don't have a rigorous proof that it cannot be done, I just have never seen it.

 

[ghwellsjr]

I have been trying really hard to understand your experiment, my_wan, and then you made this comment in post #81:

The setup in http://arxiv.org/abs/1103.6086 is sufficiently close to what I proposed to qualify the general idea, and expressed in reference [7] (Phys. Rev. D 45, 403–411 (1992)) therein. It summed up the point quiet well with:

I read both these papers and discovered that the first one was not claiming to measure the one-way speed of light and the second one which was claiming to measure the one-way speed of light, was later discredited, and when I pointed that out to you, you responded with another paper http://arxiv.org/abs/1102.4837, which I thought you were doing to buttress your position but it shows why it is impossible to measure the one-way speed of light because all such measurements involve "close paths".

So why do you keep referencing papers that you later denounce? Where is the paper that supports your claim that you know how to measure the one-way speed of light? I'm looking for the paper that you agree with 100% and will not denounce after I have read it and point out a discrepancy between it and your position.

Apparently, you think your measurement does not involve close paths, is this correct? Is this why all these papers have nothing to do with your idea (like the last one you referenced in post #107 which has absolutely nothing to do with anything in this thread)? Should I go back and try to understand your experiment to see if it has close paths?

 

[my_wan]

Not if you want to keep the two-way speed of light isotropic and equal to c.

Go ahead and try it and see what happens to the two-way speed of light. Maybe you will find an example where you can change the spatial coordinates and not change the two-way speed of light, I don't have a rigorous proof that it cannot be done, I just have never seen it.

Just choose any frame as if it is the valid frame and define the constancy of c an observational illusion resulting from the inverse relation between space and time. Coming back to a son older than yourself then looks like the consequence of that anisotropy. Hence time still has to inversely relate to space regardless of which frame or coordinate choice you choose.

It's a pointless exercise in labeling a certain coordinate choice physically real while changing the frame (coordinate choice) under which it is meaningful. Yet it seems seems as though that what people often do when their talking about exceeding c so they can travel many light years faster. Speed c already let's you get there at the same time you left, yet it's sometimes not excepted as "real" because people appear to interpret it as though time dilation just give the illusion that you got there the same time you left. As if Earth is the real frame of reference.

So the fact that you can define these anisotropies in c and call the failure to measure it, like getting many light years in moments, as an illusion created by time dilation is both trivial and pointless.

 

[my_wan]

I have been trying really hard to understand your experiment, my_wan, and then you made this comment in post #81:

I read both these papers and discovered that the first one was not claiming to measure the one-way speed of light and the second one which was claiming to measure the one-way speed of light, was later discredited, and when I pointed that out to you, you responded with another paper http://arxiv.org/abs/1102.4837, which I thought you were doing to buttress your position but it shows why it is impossible to measure the one-way speed of light because all such measurements involve "close paths".

The first papers were not originally referenced by me, nor was I aware of them specifically till they were referenced here. The commonality exist only in the use of geometry rather than clock synchronization as the basis for the measurement. I thought that was sufficiently close unil it became obvious, from Perez et al that the comparison of speeds is going to match even if the speed of light differs.

So why do you keep referencing papers that you later denounce? Where is the paper that supports your claim that you know how to measure the one-way speed of light? I'm looking for the paper that you agree with 100% and will not denounce after I have read it and point out a discrepancy between it and your position.

I did not denounce the paper. The "close path" objection is valid. It is valid simply because if the the speed of light changed to some value v the comparing v/v still gives you 1 just like c/c.

Apparently, you think your measurement does not involve close paths, is this correct? Is this why all these papers have nothing to do with your idea (like the last one you referenced in post #107 which has absolutely nothing to do with anything in this thread)? Should I go back and try to understand your experiment to see if it has close paths?

You present the "close paths" disproof, pretend I'm denouncing a paper I am not, then state the reason I neither rebutted the "close paths" nor denounced the paper cited and ask me if I mean what I've been saying all this time.

Here is a picture:

 

[my_wan]

Now looking at that picture, is it not obvious that given some rotation differing speeds of light will change the amount of light detected at the detector?
Is it not obvious that "closed paths" are not being used?

 

[Dale]

Just choose any frame as if it is the valid frame and define the constancy of c an observational illusion resulting from the inverse relation between space and time. Coming back to a son older than yourself then looks like the consequence of that anisotropy. Hence time still has to inversely relate to space regardless of which frame or coordinate choice you choose.

Huh? Can you show what you mean here with an example?

 

[Dale]

Now looking at that picture, is it not obvious that given some rotation differing speeds of light will change the amount of light detected at the detector?
Is it not obvious that "closed paths" are not being used?

Is it not obvious to you that the picture itself depends on the synchronization convention?

 

[my_wan]

Huh? Can you show what you mean here with an example?

Do you mean to ask me to give an example of the fact that the physics is indepent of the coordinate choice? Newtonian physics restricted validity to a particular coordinate choice. Though Galilean transforms were allowed to translate between coordinate choices it was not always generally appreciated that these transforms allowed coordinate independence formulations in Newton's time. Ostensibly this wasn't a priority since it was so easy to presume simultaneity and space were absolute measurables. Relativity required these transforms to take center stage because the absolutes could not be maintained. Yet even with purely Galilean transforms the same coordinate independence required by relativity actually makes classical physics simpler.

Take the dilation factor $\gamma$. In SR $\gamma$ apply to time from one perspective and space from another viewing the exact same physical system. It makes no physical difference whether you define the capacity at near c to travel to Alpha Centauri in a couple of hours a result of time dilation or spatial contraction, yet mathematically you can't both by $\gamma$ from a single frame and get the right answer. Either is fine, both is wrong. Just like it makes no difference classically which Galilean frame you chose so long as you mathematically maintain that choice, or explicitly provide the transform. Just like it makes no difference which of two meteors you assign the kinetic energy to, but you can't assign the total to both.

Is that example enough, simply choose $\gamma$ to operate on space in one case, and on time in an physically equivalent case?

Is it not obvious to you that the picture itself depends on the synchronization convention?

Point it out to me, because I'm lost unless your want to make some absurd classical assumptions that cannot even stand scrutiny from a purely classical perspective.
Does it require the synchronization of two separate clocks? I say no, one single clock defining RPM, and one single yardstick defining the distance light has to travel to get detected before getting blocked. If, with sufficient resolution, you measure the speed of light as it travels straight down a gravitational potential then I fully expect it to appear as though $c > c$. Pointed up a gravitational potential I fully expect it to measure $c < c$ with the same apparatus. If you think I expect c to differ from c in an otherwise inertial frame by virtue of some medium lacking inertial effects it would be absurd.

Suppose instead of choosing between operating on space or time with some $\gamma$ you chose a frame in which some fraction operated spatially and some fraction on time. This is allowed, but the problem is that is such cases $\gamma_1 + \gamma_2$ cannot add up to the original $\gamma_{total}$. Is this unique to SR? No. If you choose a Galilean frame in which some fraction of $e_k$ is apportioned to both meteors then $e_k + e_k$ cannot add up to $e_{k(total)}$ as defined by the Galilean frame associated with either meteor.

It still seems to me that it is implicitly assumed that classical physics is coordinate independent, without recognizing that the formulation of it is in fact explicitly coordinate dependent. Then requiring me to physically measure the effects of a coordinate choice, due not to coordinate dependence, but rather a coordinate dependent formulation, in order to prove a one-way speed measure. I say no, it's an absurd distortion of logic. Yet the one-way measure is there nonetheless, even if it's not going to give a value other than c in an inertially uniform space.

 

[Dale]

Is that example enough, simply choose $\gamma$ to operate on space in one case, and on time in an physically equivalent case?

I was asking for a concrete example of a coordinate system which:
A) has a non isotropic one way speed of light
B) has an isotropic two way speed if light equal to c
C) does not attribute the anisotropy to the time component

I don't believe that such a system exists, but you seem to think that you could derive one from my example of post 105 simply by "the inverse relation between space and time". I don't get what you are saying with that, so a concrete example of such a coordinate transformation would be helpful.

 

[Dale]

Point it out to me

So, assume that we use the synchronization convention of post 105. Then the geometry of your tube changes as shown in the attachment.

 

[my_wan]

I was asking for a concrete example of a coordinate system which:
A) has a non isotropic one way speed of light
B) has an isotropic two way speed if light equal to c
C) does not attribute the anisotropy to the time component

I don't believe that such a system exists, but you seem to think that you could derive one from my example of post 105 simply by "the inverse relation between space and time". I don't get what you are saying with that, so a concrete example of such a coordinate transformation would be helpful.

In post #117 I give a very real and measurable example, involving a gravitational potential. Here's my problem with your request in general: When challenged to define a way to measure a one-way speed I mentioned how meaningless it was, but was challenged anyway. Then I'm accused of taking the physical meaning as somehow absolute. Then references are brought up to rebut it, so I allow variation to make it more clear. I'm then accused of accepting the reference as if it fully represented what I described. When I explained why it wasn't I was accused of rejecting the validity of the references. I explained again. In the meantime I was accussed of rejecting my own references, which were not even my own, but posted by others, nor did I reject their validity.

So, I can damn near guarantee that as soon as I submit to your request I will be accused of claiming it is absolutely real, and accused of trying to measure the this made up coordinate choice, then accused of rejecting my own coordinate your trying to demand I make up.

Now, if I've damn near written a book here trying to explain that a coordinate choice is not a physical choice. Yet your busy trying to goat me into constructing some BS coordinate choice for what? Unless the whole purpose is to somehow try to pin these BS accusations on me that I am somehow trying to defend the absolute physical reality of some BS coordinate choice. If there was a point I would do it anyway, yet it has no more of a point than a coordinate choice that puts the Earth at the center of the solar system.

So, assume that we use the synchronization convention of post 105. Then the geometry of your tube changes as shown in the attachment.

And so why did you divert the debate, with a complete lack of a response, pages back where I explained why this was a moot issue? I'll say it again: The fact that you can choose another equally valid coordinate choice, i.e., choose a differing synchronization convention that is consistent with SR that gets the same results with Einstein's synchronization convention is just another non-physical coordinate choice. It is NOT a required coordinate choice to get the same physical prediction, only Galilean coordinates are required for that, even though the predictions are the same.

So unless you want to claim that this non-physical coordinate choice (synchronization convention) you have chosen is in fact a physical choice then so what. Only then you are stuck trying to explain why a purely Galilean coordinate choice gives the same answers. Hence this whole, it bends to create the illusion that a Galilean coordinate choice valid implies that a coordinate choice is a physical thing in itself.

The only challenge I signed up for was not to prove any coordinate choice was a physical thing, only that with a single clock and a single tape measure a one way speed of light could be measured. It is not my problem if you want to insist on a specific coordinate choice from which you decide it's absolute physical meaning is derived.

 

[my_wan]

Basically implicit in that last post of yours is the claim that the geometry you chose to represent it is somehow the absolute geometry of the system. That's about as much abuse of the principles of relativity as can be dished out. No such geometric distortions is required for valid predictions.

 

[Dale]

Then I'm accused of taking the physical meaning as somehow absolute. Then references are brought up to rebut it, so I allow variation to make it more clear. I'm then accused of accepting the reference as if it fully represented what I described. When I explained why it wasn't I was accused of rejecting the validity of the references. I explained again. In the meantime I was accussed of rejecting my own references, which were not even my own, but posted by others, nor did I reject their validity.

So, I can damn near guarantee that as soon as I submit to your request I will be accused of claiming it is absolutely real, and accused of trying to measure the this made up coordinate choice, then accused of rejecting my own coordinate your trying to demand I make up.

None of that was me. At this point I was just trying to find out if such a coordinate system is even possible (which I still doubt) and also understand your "inverse relationship" bit which doesn't make much sense to me from your verbal descriptions, despite repeated attempts.

If you had proposed a coordinate system I would certainly test it to see if it met those three conditions. If it did, I would have to revise my position, but if it did not then I would point it out and repeat my belief that it is not possible and suggest that you should revise your position.

The only challenge I signed up for was ... only that with a single clock and a single tape measure a one way speed of light could be measured.

But this is what you have not done. The synchronization convention I used would result in the exact same experimental result from your device, but the one way speed of light is infinite in the +x direction and 1/2 c in the -x direction under that synchronization convention. So your device does not measure the one way speed of light. If you assume the one-way speed of light to be anything from 1/2 c to infinity then your device will confirm that assumption. This is because the spatial geometry depends on the synchronization convention.

When challenged to define a way to measure a one-way speed I mentioned how meaningless it was, but was challenged anyway.

Sorry, I missed this. If you agree that it is meaningless to try to measure the one-way speed of light then we are in agreement. This whole conversation, from post 50 onward, was only a reaction to your claim of post 50 that your device could indeed measure the one way speed of light. If you now agree that the measurement is meaningless then we can end here with an apology from me for missing that comment.

 

[Dale]

Basically implicit in that last post of yours is the claim that the geometry you chose to represent it is somehow the absolute geometry of the system.

There certainly is no such claim. The only claim is that the spatial geometry depends on the synchronization convention, which it clearly does.

This claim shouldn't be a surprising claim, it is right in line with standard SR fare like length contraction.

 

[my_wan]

There certainly is no such claim. The only claim is that the spatial geometry depends on the synchronization convention, which it clearly does.

This claim shouldn't be a surprising claim, it is right in line with standard SR fare like length contraction.

Ok then. Is this a claim that Cartesian coordinates are dependent on Einstein's synchronization convention?

As far as meaningless, along the same lines I said the only people who could possibly care is the Einstein is wrong crowd, and that such claims were based on some form on the a physical reality of coordinate choices.

 

[Dale]

Ok then. Is this a claim that Cartesian coordinates are dependent on Einstein's synchronization convention?

I don't know what you mean by this.

Considered as a 4D object independent of any coordinate system your rotating tube is a double helix. The claim that the tube is straight requires a very specific "slicing" of that helix. If you slice it on any other hypersurface then it is no longer straight.

This includes weird synchronization conventions discussed here, but it also includes any inertial frame (Einstein synchronization) where the COM of the tube is not at rest. Even in other inertial frames the tube is not straight. This is why I mentioned the other thread way back in post 52.

As far as meaningless, along the same lines I said the only people who could possibly care is the Einstein is wrong crowd

I care and I am not among the Einstein is wrong crowd.

 

[my_wan]

I don't know what you mean by this.

Considered as a 4D object independent of any coordinate system your rotating tube is a double helix. The claim that the tube is straight requires a very specific "slicing" of that helix. If you slice it on any other hypersurface then it is no longer straight.

This includes weird synchronization conventions discussed here, but it also includes any inertial frame (Einstein synchronization) where the COM of the tube is not at rest. Even in other inertial frames the tube is not straight. This is why I mentioned the other thread way back in post 52.

I care and I am not among the Einstein is wrong crowd.

I'll address the coordinate issue further down, but I think there are some interpretation problems, on for both of us, that I should mention first.

I feel like my position points have been misrepresented in terms of some kind of "weird synchronization convention" in which I have little notion of the details of how this "weird convention" is supposed to be constructed, and can imagine a bewildering number of absurd but . Fine by itself if it was well defined, but I feel stuck in the position of trying to second guess some model when all I can do is try to qualify when certain notions presented can be contextually valid and when they can't. Hence it's essentially like trying to address a moving target. This is possibly creating an illusion of flip-flopping.


At the same time I think it's likely that some people are trying to interpret what I'm saying in the context of some preconceived notions of the context in which my claims are intended to have meaning. This entails that the fine critiques here are in no better shoes than I am in terms of how to proceed. The critiques cannot be blamed any more than myself. Now, how to proceed?

>>>I'll now attempt a (almost certainly incomplete) complete characterization of the issues.

Yes you are absolute correct that the claim that the tube is straight requires a very specific set of coordinate choices, as any coordinate choice involves certain topological assumptions. My main point is that choosing between a flat or curved topology doesn't in itself have physical meaning. You have chosen to point out that the Euclidean geometry I have chosen is equivalent to the curved topology you are mapping it with in this case. This is true enough by itself, but my point is that it makes no difference. You get the same answers irrespective of whether your coordinate choices involves a curved topology or not. Hence the fact that one perfectly valid coordinatization choice empirically maps to a second perfectly valid coordinatization choice does not prove one or the other coordinate choice is the one valid choice, or even that one is a required presumption of the other. In fact the coordinate choice in itself has no physical meaning at all.

Now, this issue of coordinate independence is complicated by certain notions of synchronization conventions. Yet if you accept that a coordinate choice is not in itself a choice of physical parameters then a valid synchronization convention is determined by empirical consistency with the coordinate choice used. If a synchronization convention is valid then we can fully expect to transform our coordinatization into equally valid curved topologies without implying any specific actual reality to a curved verses a flat coordinate choice.

The Experiment:
So now it's down to the question of what the experiment I described entails. Certainly, if and only if (IIF) Einstein's synchronization is physically valid generally (as I certainly fully expect), and I suggest a one-way light speed measurement that is (by definition) inconsistent with Einstein's synchronization, then something is absolutely wrong with the interpretation of the experimental design.

However, the question is not strictly whether Einstein's synchronization is valid or not, but whether I must presuppose that it is in order to perform such an experiment. To this question the answer is no, because I only a priori assumed flat Euclidean space with purely Galilean transforms using the ratio of a single clock to a single ruler. Hence, IIF the domain of validity of Einstein's synchronization can be truncated under these circumstances then the experiment will demonstrate that. The failure to demonstrate any inconsistency with Einstein's synchronization is therefore NOT a product of a presumption of Einstein's synchronization, but would merely be the result of the empirical validity of Einstein's synchronization.

The claim that I must empirically invalidate Einstein's synchronization in order not to presume it's validity a priori is simply not tenable. Though I think most of us know how absurd things could get if Einstein's synchronization was empirically invalidated by this experiment, and worth testing if for no other reason. As uninterested as I am, based on the apparently obvious validity of Einstein's synchronization, the empirical results are more meaningful than our sensibilities.

Ending Questions:
[1] If it is maintained that I have a priori presumed Einstein's synchronization in this experimental design, please explain in what way such a priori as assumptions where embedded in the design prior to obtaining results.

[2] If not [1] and it is maintained that the consistency of results with Einstein's synchronization automatically entails that this synchronization convention was a priori assumed, please explain how the empirical consequences of an experiment entails an a priori presumption of those results.

Otherwise it cannot be maintained that the empirical validity of Einstein's synchronization entails the a priori assumption of its validity, or that this argument is sufficient to claim the experimental design I outlined contains such a priori assumptions. Of course you may beg to differ, but please at least address these issues in the rebuttal, if for no other reason than to articulate why they are irrelevant.

 

[Dale]

I feel like my position points have been misrepresented in terms of some kind of "weird synchronization convention" in which I have little notion of the details of how this "weird convention" is supposed to be constructed, and can imagine a bewildering number of absurd but . Fine by itself if it was well defined, but I feel stuck in the position of trying to second guess some model when all I can do is try to qualify when certain notions presented can be contextually valid and when they can't. Hence it's essentially like trying to address a moving target. This is possibly creating an illusion of flip-flopping.

That is why I provided an explicit example in post 105 and asked you for a counter-example when you objected. Post 105 was a specific concrete example of a possible "weird synchronization convention" under which the one-way speed of light was not c.

Personally, I think that you are not flip-flopping but that you have just not worked through this completely so you are unaware of some of the issues and background.

You have chosen to point out that the Euclidean geometry I have chosen is equivalent to the curved topology you are mapping it with in this case. This is true enough by itself, but my point is that it makes no difference. You get the same answers irrespective of whether your coordinate choices involves a curved topology or not.

This statement is only true if the questions are such that the answers are coordinate independent. The question of the one-way speed of light is not such a question.

You have spoken about "physical meaning", "physically valid", and "physically real". I generally stay away from such terms. However, I would submit to you that if you believe that a coordinate choice has no physical meaning then a question whose answer depends on the coordinate choice should also be designated as having no physical meaning.

[1] If it is maintained that I have a priori presumed Einstein's synchronization in this experimental design, please explain in what way such a priori as assumptions where embedded in the design prior to obtaining results.

The assumption is embedded in the shape of the device, as I have shown above.

Of course you may beg to differ, but please at least address these issues in the rebuttal, if for no other reason than to articulate why they are irrelevant.

I didn't address [2] since it started out "if not [1]", and I assert [1].

 

[my_wan]

That you assume I haven't worked through these issues is a reasonable assumption, however much I might disagree. Of course it's not too hard to be wrong when reworked enough. My biggest deficit is not the lack of working through these issues myself, but in bothering with articulating how to express what I have worked through in terms of every possible conceptual model variation others might hold value in. I don't think any of us can fully appreciate the conceptual differences people might hold, with or without actual empirical incongruence.

I made a long post and deleted it to get to the core issue below.

The assumption is embedded in the shape of the device, as I have shown above.

Yet your shape requires a coordinate choice in which the time varies at each point in the space of the experiment. Yet if you assume it is completely flat, i.e., a purely Galilean frame with a constant t across the whole space of the apparatus, the empirical results remain with or without the results justifying your coordinate dependent topological curvature. Hence you have not shown that the shape you describe is anything more than an artifact of your coordinate choice, i.e., a coordinate dependent claim.

That is why I started with a purely Galilean frame, a global absolute t, one clock, and one ruler to avoid this a priori assumption. I therefore get the same experimental results IIF (if and only if) Einstein's synchronization is strictly valid. Hence the a priori assumption that the shape you defined is a valid coordinate choice is determined by the empirical outcome, not on my presumption that a universal time doesn't exist.

I didn't address [2] since it started out "if not [1]", and I assert [1].

Yes you did, thanks. Yet it is a repeat of a previous statement, so the above response repeats the unnecessary coordinate dependence of your shape assumption, via a position dependent t. Whereas I started with a flat space and a global t and ONLY justify the validity of your coordinate choice IIF the empirical results agree. Regardless of how absurd it would be to presume it wouldn't empirically agree.

However reasonable the presumption, it is you who is making the a priori assumption that it will empirically agree in order to claim I required those presumption in order to get the empirical results to agree. Where is it in the desin (not coordinate choice). It's not in the variable t, since that is assumed globally uniform. It's not in the shape, since I'm using global flatness as the basis for comparing empirical results to. It's not in the synchronization of a pair of clocks, since I only have one clock. It's not in the presumption that two length measurements are equal, because I only have one straight length to measure.

So the rebuttal requires something more than a coordinate dependent claim that the topology is curved.

 

[my_wan]

That you assume I haven't worked through these issues is a reasonable assumption, however much I might disagree. Of course it's not too hard to be wrong and reworking more is always warranted. My biggest deficit is not the lack of working through these issues myself, but in bothering with articulating how to express what I have worked through in terms of every possible conceptual model variation others might hold value in. I don't think any of us can fully appreciate the conceptual differences people might hold, with or without actual empirical incongruence.

I made a long post and deleted it to get to the core issue below.

The assumption is embedded in the shape of the device, as I have shown above.

Yet your shape requires a coordinate choice in which the time varies at each point in the space of the experiment. Yet if you assume it is completely flat, i.e., a purely Galilean frame with a constant t across the whole space of the apparatus, the empirical results remain with or without the results justifying your coordinate dependent topological curvature. Hence you have not shown that the shape you describe is anything more than an artifact of your coordinate choice, i.e., a coordinate dependent claim.

That is why I started with a purely Galilean frame, a global absolute t, one clock, and one ruler to avoid this a priori assumption. I therefore get the same experimental results IIF (if and only if) Einstein's synchronization is strictly valid. Hence the a priori assumption that the shape you defined is a valid coordinate choice is determined by the empirical outcome, not on my presumption that a universal time doesn't exist.

I didn't address [2] since it started out "if not [1]", and I assert [1].

Yes you did, thanks. Yet it is a repeat of a previous statement, so the above response repeats the unnecessary coordinate dependence of your shape assumption, via a position dependent t. Whereas I started with a flat space and a global t and ONLY justify the validity of your coordinate choice IIF the empirical results agree. Regardless of how absurd it would be to presume it wouldn't empirically agree.

However reasonable the presumption, it is you who is making the a priori assumption that it will empirically agree in order to claim I required those presumption in order to get the empirical results to agree. Where is it in the design (not coordinate choice). It's not in the variable t, since that is assumed globally uniform. It's not in the shape, since I'm using global flatness as the basis for comparing empirical results to. It's not in the synchronization of a pair of clocks, since I only have one clock. It's not in the presumption that two length measurements are equal, because I only have one straight length to measure.

So the rebuttal requires something more than a coordinate dependent claim that the topology is curved.

 

[Dale]

Can you decide which of your two previous posts you prefer and delete the other?

 

[Dale]

My biggest deficit is not the lack of working through these issues myself, but in bothering with articulating how to express what I have worked through

That is certainly possible. Your use of standard terminology is very non-standard and confusing. For example:

a coordinate dependent claim that the topology is curved.

Topology introduces concepts like continuity and connectedness, not distances, angles, or curvature. In order to get curvature you need a metric space, not just a topological space. So you would say that the manifold is curved since a manifold is a topological space with an associated metric, or you could even say that the metric is curved. However, the curvature of a manifold is not a coordinate dependent claim, it is coordinate independent. So this whole phrase is very confusing and non-standard.

Hence you have not shown that the shape you describe is anything more than an artifact of your coordinate choice, i.e., a coordinate dependent claim.

Correct, the shape I described is coordinate dependent. So is the shape you described. They are both individual cases of an infinite number of equally valid shapes, each of which depend on the coordinates chosen.

the above response repeats the unnecessary coordinate dependence of your shape assumption, via a position dependent t. Whereas I started with a flat space and a global t

When we say "A depends on B" we mean that if you change B then A also changes.

As I demonstrated, when you change coordinates (B) then the shape also changes (A). Therefore the shape (A) depends on the coordinates (B). The fact that you "started with a flat space and a global t" is completely irrelevant as to whether or not the shape depends on the coordinates. The flatness or globalness of A and B simply doesn't enter into the definition of "A depends on B".

 

[my_wan]

I went to bed after that last post, now I'm unable to delete either one. The connection timed out when posting so I checked to see if it posted before reposting and apparently didn't see it somehow. Anybody with the authority is welcome to delete either they so choose.

As I demonstrated, when you change coordinates (B) then the shape also changes (A). Therefore the shape (A) depends on the coordinates (B). The fact that you "started with a flat space and a global t" is completely irrelevant as to whether or not the shape depends on the coordinates. The flatness or globalness of A and B simply doesn't enter into the definition of "A depends on B".

Main Argument:
What you have done is a mapping of A to B, a coordinate transform, and then you say that "A depends on B". This is absolutely true for any valid mapping of A to B. The question is whether this mapping is valid in every respect, as we have every reason to believe it is. Yet, in starting with a flat space using Galilean transforms and a global t, I'm not required to presume a priori that the mapping is entirely valid in the context of one-way speeds or otherwise. Yet in order for you to claim A is a consequence of B you must make this presumption that I have not required.

The reason this presumption is not required is because if you start with strictly Cartesian coordinates with purely Galilean transforms then the validity of the mapping A to B depends on the character of the inertial properties of the space, such as Newtonian absolutes verses relative, or any empirically accessible property thereof like (unlikely) one-way light speeds. Change these relational properties in some empirically accessible way and the the mapping A to B must change accordingly. Hence the claim that "A depends on B" is essentially a post hoc ergo propter hoc argument.

Other Issues:
As I've stated, I see no real reason to be terribly interested in this experiment due to the primary issues it addresses are those that claim Einstein is wrong. Though it can also address inverse conceptual issues with how some perceive the distinctions between Einstein's relativity and Galilean relativity. In fact the thing SR changes has nothing to do with Newtonian mechanics, only the Newtonian assumption of some preferred Galilean frame that was only possible to justify as a result of the linearity of Newtonian time. Yet if time itself is the result of a mechanistic process it can't possibly be a universal constant.

In terms of the specific claims you hear, like one way speeds can't be measured due to the requirement of synchronizing a pair of clocks, it is relevant and takes away the last thread that the anti-Einstein crowd can hang onto. It also removes this notion that somehow relativity is fundamentally inconsistent with Newtonian mechanics, as distinct from the claims of absolute space and time, i.e., coordinate dependence. Classical thermodynamics remains coordinate dependent to this day, with loads of debate over extensive properties.

Let's look closer at the source of the "A depends on B" and clock synchronization claims.
PhilSci preprint linked: Clock synchronization, a universal light speed, and the terrestrial red-shift experiment
American Journal of Physics, Volume 51, Issue 9, pp. 795-797 (1983).

But[/PLAIN] the Hafele-Keating experiment [1] and muon decay experiments which measure time dilation [2] show that a universal time does not exist, and so the notion of separated synchronized clocks can have no a priori meaning. It follows that the speed of light can have no meaning until a definition of synchronized clocks is given. It is not simply that the speed cannot be measured; it can have no meaning.

(Italics original)

Now in principle, since universal time (like speed) is not just non-existent but lacks meaning altogether, entails that (like a coordinate choice) it makes no difference what synchronization convention you use, flat, curved, or whatever, but the physical results must be consistent regardless. Just like Galilean transforms do not need the extra baggage of absolute space and time, under which laws of physics are valid only for some choice of Galilean frame. Both the denial of Einstein's relativity and the denial that it is consistent with Galilean relativity given a locally constant c entails the same error.

This of course fully and absolutely justifies DaleSpam's presumptions that A maps to B. Yet does not justify the claim that, lacking prior knowledge the independence of space and time to universal absolute metric, of which not even c is an absolute constant, "A depends on B". The claim A depends on B itself depends on the absents of universal absolute metrics.

Now further down it states:

This[/PLAIN] one-way speed requires two clocks, and to be meaningful, the clocks must be synchronized.

Yet as this article already stated the synchronization convention cannot make a difference if the labeling of t has no meaning in itself whatsoever. The labeling of t is what defines the meaning, just as it is what defines the difference between a flat and a curved geometry. Thus if you choose the two clocks in the described test it corresponds to the assumption that t was consistent with a universal time over all space. Yet the most obvious presumable results is that no such universal time exist, i.e., no difference between one a two way light speeds. Yet synchronization of clock pairs was NOT required a priori that was inconsistent with a universal time. Neither can the failure to maintain a one-way speed anisotropy in an inertially flat plane of space be blamed on an a prior synchronization convention chosen that was inconsistent with universal space or time. I started with one synchronization convention, globally absolute, and fully expect results that disavow the global absolutes its presumptions were predicated on. Hence it is a true test, however trivial it for those who already see the obvious, that no such absolute metric exist.

This lack of absolute metrics applies equally as well to those who would assume that Einstein's synchronization convention must be incompatible with alternate synchronization conventions. This is the very meaning of the statement that synchronization definition is not simply non-existent in nature, it can have no meaning whatsoever without the definition. The synchronization convention has no more a preferred status, beyond the simplification of the formalism, than a coordinate choice or a preferred frame. This extends to the lack of preferred Galilean frames in classical physics as well. We have simply moved from preferred frames in classical physics to preferred synchronization conventions in Relativity, neither of which has any 'absolute' meaning. This goes for those supposed extensive properties in classical thermodynamics as well, where the mean field limits and associated state variables, and the ensembles derived thereof, are tied to a unique Galilean (preferred) frame.

 

[Dale]

What you have done is a mapping of A to B, a coordinate transform, and then you say that "A depends on B". This is absolutely true for any valid mapping of A to B.

OK, so you agree that the spatial shape of the device depends on your coordinate choice.

Yet, in starting with a flat space using Galilean transforms and a global t, I'm not required to presume a priori that the mapping is entirely valid in the context of one-way speeds or otherwise.

You may or may not presume that my mapping is valid, that is your perogative. The point is that you made an assumption ("global t") and the thing you said you were measuring depends on that assumption. Whether or not any other assumptions are valid, the thing you claim to be measuring depends on the assumption. You are therefore not measuring it but assuming it.

Here is an analogy of the error you are making. Suppose I was building a device to measure the two way speed of light. My device consists of a light source, detector, and clock all colocated and a mirror some distance away. I flash the bulb and measure the time to see the reflection. The distance to the mirror is given by d=2t/c and the measured speed of light is given by 2t/d. You complain "but your measurement depends on the value of c you use to calculate d, if you change that value you get a different result". I reply, "I am not required to presume a priori that any other value of c is valid, in fact, there is a large body of evidence showing that no other value of c is valid". You reply, "whether or not any other assumptions are valid, the value you claim to be measuring still depends on the assumption you are making, you are therefore not measuring it but assuming it".

 

[my_wan]

OK, so you agree that the spatial shape of the device depends on your coordinate choice.

Yes, exactly as my original claim stated.

You may or may not presume that my mapping is valid as you please.

Here you make it sound as if an operation assumption in the setup is an assumption of the converse. It's not, that's what the empirical results, which have yet to be obtained, are to determine.

The point is that you made an assumption ("global t") and the thing you said you were measuring depends on that assumption.

No, the absolute validity of that assumption, not the assumption itself, is what the empirical results depends on. If what I measure only depends on the assumption of an globally uniform t that is uniquely valid then that assumption is invalidated by the expected empirical results when those results fail to justify this assumption uniquely.

You cannot say the results of a "global t" assumption depends on that assumption when the results are fully expected to be entirely consistent with a non-global t. In other words the outcome is independent of your assumption of global t or not, just like physics is independent of your coordinate choice.

Whether or not any other assumptions are valid, the thing measured still depends on the assumption.

No. If the measurement depended on the global t assumption it by definition precludes justification of a non-global t as you chose, only it will. Instead it will fully justify the non-global t just as well, IIF as we know it will the speed c is constant one-way. The results are independent of choice of defining t, just like coordinate independence.

Repeat bottom line: I made the assumption of a global t, you made the assumption t varied over the same space, yet we get the same results. Hence the result is independent of the global verses non-global t assumption. Therefore it cannot be said that the measurement is dependent on a global t, since the actual results fully justify a non-global t. An assumption does not justify itself by justifying the opposite.

 

[my_wan]

Here is an analogy of the error you are making. Suppose I was building a device to measure the two way speed of light. My device consists of a light source, detector, and clock all colocated and a mirror some distance away. I flash the bulb and measure the time to see the reflection. The distance to the mirror is given by d=2t/c and the measured speed of light is given by 2t/d. You complain "but your measurement depends on the value of c you use to calculate d, if you change that value you get a different result". I reply, "I am not required to presume a priori that any other value of c is valid, in fact, there is a large body of evidence showing that no other value of c is valid". You reply, "whether or not any other assumptions are valid, the value you claim to be measuring still depends on the assumption you are making, you are therefore not measuring it but assuming it".

I have given very specific circumstances under which it is flat out wrong to presume the speed of light is constant, and involves any accelerated system. In fact the described measurement can in principle actually measure this non c of c.

 

[TrickyDicky]

my_wan, thanks a lot for your clear answers and the clever thought experiment in this thread from #50 till the last one, I fully agree with you. And you are the most patient person I know too.
Take a look at the parallel thread about one-way light speed measurement(you probably already have) where thanks to Pallen a reasonable conclusion has been reached in line with what you are saying:the posibility of measuring it is a condition of falsifiability of the SR theory.

 

[Dale]

Welcome back TrickyDicky!

 

[Dale]

No, the absolute validity of that assumption, not the assumption itself, is what the empirical results depends on.

This is not correct. I took your same device, made a different assumption, and got a device that would measure an infinite one-way speed of light. The result you obtain from the experiment depends on the assumption itself.

The unprocessed data that you are getting from your proposed experiment is some brightness measure as a function of the angular velocity. You then take some feature of that curve, e.g. the RPM of the peak brightness, and you mathematically convert that value to a measurement of the speed of light. In order to make that conversion you must make an assumption about the shape of the device as it rotates, and that assumption completely determines the value that you get for your conversion, anywhere from .5 c to infinity.

 

[my_wan]

This is not correct. I took your same device, made a different assumption, and got a device that would measure an infinite one-way speed of light. The result you obtain from the experiment depends on the assumption itself.

Wait a minute. You just claimed claimed to have gotten an empirical experimental result from your raw math! You must be badly misinterpreting what I mean by a "result".

There are two ways I can assume you did this. The first, and most reasonable, would allow you to take this theoretical infinite speed curve at differing RPMs and compare it to the actual measured curves, which cannot be had without performing the experiment, and determined their rate of divergence.

The second, and rather absurd, approach that you seem to be implying is tantamount to assigning t=0 for the emission and detection of any given photon. The empirical results, not your assumption which you can calculate with, is the change in the total number of photons as the RPM is steadily increased. The curve this is compared to is the expected rate at which the total number of photons detected per revolution decreases. How fast the two curves diverge is determined by the speed of light in that one Galilean frame for which RPM is defined.

The unprocessed data that you are getting from your proposed experiment is some brightness measure as a function of the angular velocity. You then take some feature of that curve, e.g. the RPM of the peak brightness, and you mathematically convert that value to a measurement of the speed of light. In order to make that conversion you must make an assumption about the shape of the device as it rotates, and that assumption completely determines the value that you get for your conversion, anywhere from .5 c to infinity.

In red badly missed the mark. Let me go through this again, as I try to unmangle the MaxEnt mixture of... I don't know what it is.

There is NO peak brightness, unless the contraption is not even turned on to perform any test. Peak brightness is 0 RPM if the rod is oriented in a certain direction.

You have a light source. It enters a hollow hole to be detected at the other end. If the light is not fast enough to get to the other end before the 1 cm^2 detector, the same size as the hole, moves past the area that the light will be detected.

I'm going to speak in terms of counting individual photons for the sake of simplicity.

Let's say the length of the pipe is 1 m. The hollow light tunnel is 1 cm^2, and is square not round, to linearize the change in detected photons as the RPM increases, assuming an infinite c. Now, if the speed of light is infinite, the decrease in the number of photons being detected per revolution is simply a function of the amount of time the total paths to the detector remains available. From this you mathematically plot on a curve in 1 RPM increments between 1 and say 30,000 RPM, under the assumption that the speed of light in infinite. This is 30,000 data points, and this is merely the reference curve the actual results will be compared against. No actual results are available as yet.

Now we need the empirical data curve to compare this to. For this you start at 1 RPM and measure number of photon detected per revolution. Then step through the next 29,999 RPM increments. This is your empirical data curve, which you cannot possibly have done on your computer, pencil and paper, or whatever. By having a regression of data points this way it's possible to get exquisitely sensitive measurements from relatively dirty data, much like the pioneer anomaly data capable of resolving an effect on the order of the Hubble expansion within the solar system. 30,000 data points is almost certainly overkill, but so what.

Now the divergence, over this entire range of RPM, from the infinite speed reference curve defines the speed of light. The noise in any given data point can then be washed out by fitting it to the progression of data. No one data point, corresponding to any given RPM, has any real importance whatsoever. Even less relevant is some "peak brightness" at any RPM.

The notion that your computer (or pencil) generated these necessarily conditions to obtain these empirical results is absurd.

>>>>
I say I'm not interested in this. This remains so in terms of your basic one-way light speed arguments, but to my knowledge no anisotropy in light speed has ever been measured in terms of GR either. Though it is perfectly allowed in principle by GR. A WAG can be posited that this could even play some role in the recent FTL neutrinos, though not something that is fitting to derail here with. Nonetheless, this design can in principle measure GR induced anisotropies in c, which have been acknowledged by and since Einstein. Though no method has to my knowledge been suggested to measure it directly.

 

[Dale]

The second, and rather absurd, approach that you seem to be implying is tantamount to assigning t=0 for the emission and detection of any given photon.

Not t=0, but Δt=0 for photons moving in the +x direction. That is a synchronization convention corresponding to an infinite one way speed of light in the +x direction.

Regarding the rest of your post, fine there is not a peak brightness, but there is still a brightness as a function of the RPM and you calculate the corresponding "measurement" of the one way speed of light from that function. The calculation you use to obtain the value of c depends not only on the measured brightness v RPM function but also on the assumed geometry of your device at different speeds (which is coordinate dependent). So the same brightness v RPM curve can be made to fit any one-way speed of light under an appropriate choice of coordinates.