Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.physics.relativity > #380138 > unrolled thread

Re: Who Killed Theoretical Physics?

Back to article view | Back to sci.physics.relativity

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: Who Killed Theoretical Physics? Tom Roberts <tjroberts137@sbcglobal.net> - 2016-03-25 10:58 -0500
    Re: Who Killed Theoretical Physics? David Waite <waitedavid1618@yahoo.com> - 2016-03-25 09:22 -0700
      Re: Who Killed Theoretical Physics? Tom Roberts <tjroberts137@sbcglobal.net> - 2016-03-25 14:43 -0500
        Re: Who Killed Theoretical Physics? David Waite <waitedavid1618@yahoo.com> - 2016-03-26 08:53 -0700
          Re: Who Killed Theoretical Physics? Tom Roberts <tjroberts137@sbcglobal.net> - 2016-03-31 21:21 -0500
            Re: Who Killed Theoretical Physics? The Starmaker <starmaker@ix.netcom.com> - 2016-03-31 23:38 -0800
    Re: Who Killed Theoretical Physics? "Dono," <sa_ge@comcast.net> - 2016-03-26 06:49 -0700

#380138 — Re: Who Killed Theoretical Physics?

On 3/21/16 3/21/16 - 8:39 PM, Dono, wrote:
> On Monday, March 21, 2016 at 5:58:18 PM UTC-7, Eric Baird wrote:
>> In the case of the spinning star surrounded by a distortional field
>> component that makes light move faster "with" the direction of nearest star
>> surface than "against" it,
>
> No, stubborn imbecile, it is the COORDINATE light speed that differest in the
> two directions. [...]

Actually it is deeper than that. This is GR.

Consider a distant source beyond a rapidly spinning massive spherical object, 
right in line with the observer, such that light from the source can skim past 
the surface of the object, and is gravitationally focused so the observer can 
see rays skimming by each side of the object. Pulse the light source, and the 
observer will see the light from the co-rotating side arrive before the light 
from the anti-rotating side; light that skims past either pole arrives in between.

So it seems that light following those different non-local paths has different 
speeds, even though at each point along each path the local speed is c.

Or perhaps not, as if one carefully measured the spatial length of those paths 
one would obtain different values (yes, the rotation of the object affects 
this). Clearly light traveling along different-length paths can arrive at 
different times while traveling at c. I have not done the calculation to see if 
the spatial length differences are c times the flight-time differences for these 
paths (I suspect it is close but not exact).


> The light speed locally is STILL c.

Yes. But non-local measurements can obtain a different answer (e.g. the Shapiro 
time delay, and the gedanken above).

Perhaps the most definitive prediction for a non-local measurement is to 
consider a light source and mirror located radially apart in Schwarzschild 
spacetime. One can integrate the metric to determine their spatial separation, 
and then integrate the appropriate null geodesic to determine the round-trip 
flight time; the ratio differs slightly from c. In Schw. spacetime the 
anisotropy in such non-local TWLS depends on both the orientation of the light 
path and the distance between source and mirror. Current technology cannot do 
this experiment, as we cannot measure distance accurately enough without using 
light.... I posted an article showing this calculation to this newsgroup, 
sometime in the early 2000's; its subject probably included "anisotropic" and 
"light speed".


Tom Roberts

[toc] | [next] | [standalone]

#380140

On Friday, March 25, 2016 at 8:58:52 AM UTC-7, tjrob137 wrote:
> Or perhaps not, as if one carefully measured the spatial length of those paths 
> one would obtain different values 

The answer would literally depend on how the observers distance coordinates are defined because with this scenario involving spacetime curvature, no globally rectilinear coordinates even exist on which to set a preferred standard for his distance coordinates. Remote from him he has to make a somewhat arbitrary choice concerning his coordinates behavior which effects how he interprets that result.

[toc] | [prev] | [next] | [standalone]

#380151

On 3/25/16 3/25/16 - 11:22 AM, David Waite wrote:
> On Friday, March 25, 2016 at 8:58:52 AM UTC-7, tjrob137 wrote:
>> Or perhaps not, as if one carefully measured the spatial length of those
>> paths one would obtain different values
>
> The answer would literally depend on how the observers distance coordinates
> are defined because with this scenario involving spacetime curvature, no
> globally rectilinear coordinates even exist on which to set a preferred
> standard for his distance coordinates. Remote from him he has to make a
> somewhat arbitrary choice concerning his coordinates behavior which effects
> how he interprets that result.

Yes.

As long as he uses the same method for all paths, the path-lengths will differ, 
even though their values depend on which method is used. (He must select a 
method that can be applied to all paths.)

But I see I was not comprehensive enough in describing the physical situation.

Implicit (but unstated) in the description is that the source and observer are 
far away, in asymptotically flat spacetime. As no mention of source motion wrt 
the observer was made, they can be assumed to be at rest in a single Minkowski 
frame out there, and those background coordinates do provide such a preferred 
standard, as long as the massive object has only weak gravity at its surface 
(e.g. no more than 10X the sun).

GR is very "slippery", and I may not have nailed down everything. But I think 
this is sufficient to resolve the original issues.


Tom Roberts

[toc] | [prev] | [next] | [standalone]

#380210

On Friday, March 25, 2016 at 12:43:38 PM UTC-7, tjrob137 wrote:
> On 3/25/16 3/25/16 - 11:22 AM, David Waite wrote:
> > On Friday, March 25, 2016 at 8:58:52 AM UTC-7, tjrob137 wrote:
> >> Or perhaps not, as if one carefully measured the spatial length of those
> >> paths one would obtain different values
> >
> > The answer would literally depend on how the observers distance coordinates
> > are defined because with this scenario involving spacetime curvature, no
> > globally rectilinear coordinates even exist on which to set a preferred
> > standard for his distance coordinates. Remote from him he has to make a
> > somewhat arbitrary choice concerning his coordinates behavior which effects
> > how he interprets that result.
> 
> Yes.
> 
> As long as he uses the same method for all paths, the path-lengths will differ, 
> even though their values depend on which method is used. (He must select a 
> method that can be applied to all paths.)
> 
> But I see I was not comprehensive enough in describing the physical situation.
> 
> Implicit (but unstated) in the description is that the source and observer are 
> far away, in asymptotically flat spacetime. As no mention of source motion wrt 
> the observer was made, they can be assumed to be at rest in a single Minkowski 
> frame out there, and those background coordinates do provide such a preferred 
> standard, as long as the massive object has only weak gravity at its surface 
> (e.g. no more than 10X the sun).
> 
> GR is very "slippery", and I may not have nailed down everything. But I think 
> this is sufficient to resolve the original issues.
> 
> 
> Tom Roberts

If the light paths are effected enough that he's receiving the light signals from the same event at different times from different directions then he can have no truly rectilinear coordinates covering the full paths. Lets say you have a kerr spacetime. If he is sufficiently remote a common choice would likely be Boyer Lindquist coordinates. But that's not the only choice. You might do a transformation to the r coordinate of something like r→r'*(1+GM/2rc²)² because you maybe you like isotropic coordinates for the Schwarzschild solution and that would reduce it to that as the rotation parameter is taken to zero. Now consider a bit of radial path. Δr and Δr' would be different depending on the standard chosen. Now both coordinate choices in this scenario agree on a sort of "ruler length" along a path for a constant time slice which would be a constant time slice integral for ds, but then one could always change to something like inward horizon penetrating coordinates Vs outward horizon penetrating coordinates or really anything which means that a constant time slice using one set won't be a constant time slice for some other set of coordinates. All we can really agree on doing as a standard would be to use coordinates that make the metric look like special relativity's remote where the observer is, but beyond that there's a lot of arbitrary choice goes into what you use and therefor how you interpret the end result.

[toc] | [prev] | [next] | [standalone]

#380591

On 3/26/16 3/26/16   10:53 AM, David Waite wrote:
> On Friday, March 25, 2016 at 12:43:38 PM UTC-7, tjrob137 wrote:
>> Implicit (but unstated) in the description is that the source and observer
>> are far away, in asymptotically flat spacetime. As no mention of source
>> motion wrt the observer was made, they can be assumed to be at rest in a
>> single Minkowski frame out there, and those background coordinates do
>> provide such a preferred standard, as long as the massive object has only
>> weak gravity at its surface (e.g. no more than 10X the sun).
>
> If the light paths are effected enough that he's receiving the light signals
> from the same event at different times from different directions then he can
> have no truly rectilinear coordinates covering the full paths.

Sure ("truly rectilinear coordinates" = Minkowski coordinates). But that is NOT 
required. What is required is that they be Minkowski coordinates at source and 
observer, and valid in between (with presumably non-Minkowski metric components 
near the massive object), at least along the paths. As long as gravitation is 
"weak" along all paths of interest, that can hold. After all, this is how one 
calculates gravitational lensing and the Shapiro delay (etc.) via the PPN formalism.


> [... more detail than necessary ...]
> All we can really agree on doing as
> a standard would be to use coordinates that make the metric look like special
> relativity's remote where the observer is, but beyond that there's a lot of
> arbitrary choice goes into what you use and therefor how you interpret the
> end result.

If one attempts to deduce the coordinate speed of light all along the path, this 
is true. But if one is only interested in the observed time-difference of 
light-pulse arrivals for different paths, that is an observable and thus 
independent of coordinate choice (to compute it one basically integrates over 
each path, and the metric components involved ensure that the result is invariant).

That's my point -- there's more here than mere coordinate dependence. The fact 
that the arrival time differences are observable requires it.

	And when light from a single pulse arrives at different times
	over essentially the same path [#], it's hard to avoid the
	conclusion that the speed of light over those non-local
	paths does indeed vary. Even though the LOCAL speed of light
	does NOT vary -- this is just evidence of spacetime curvature.

	[#] consider a source-to-observer distance of several parsecs
	with the light paths passing on different sides of a rotating
	star 10 times larger than the sun.


Tom Roberts

[toc] | [prev] | [next] | [standalone]

#380599

Who Killed Theoretical Physics? richard feynman did.

[toc] | [prev] | [next] | [standalone]

#380200

On Friday, March 25, 2016 at 8:58:52 AM UTC-7, tjrob137 wrote:
> On 3/21/16 3/21/16 - 8:39 PM, Dono, wrote:
> > On Monday, March 21, 2016 at 5:58:18 PM UTC-7, Eric Baird wrote:
> >> In the case of the spinning star surrounded by a distortional field
> >> component that makes light move faster "with" the direction of nearest star
> >> surface than "against" it,
> >
> > No, stubborn imbecile, it is the COORDINATE light speed that differest in the
> > two directions. [...]
> 
> Actually it is deeper than that. This is GR.
> 
> Consider a distant source beyond a rapidly spinning massive spherical object, 
> right in line with the observer, such that light from the source can skim past 
> the surface of the object, and is gravitationally focused so the observer can 
> see rays skimming by each side of the object. Pulse the light source, and the 
> observer will see the light from the co-rotating side arrive before the light 
> from the anti-rotating side; light that skims past either pole arrives in between.
> 
> So it seems that light following those different non-local paths has different 
> speeds, even though at each point along each path the local speed is c.
> 
> Or perhaps not, as if one carefully measured the spatial length of those paths 
> one would obtain different values (yes, the rotation of the object affects 
> this). Clearly light traveling along different-length paths can arrive at 
> different times while traveling at c. I have not done the calculation to see if 
> the spatial length differences are c times the flight-time differences for these 
> paths (I suspect it is close but not exact).
> 


Yes, I am fully aware of the above (see the explanation for the Shapiro delay, the time of arrival is a function of the distance traveled: it is the distance traveled that is influenced by the gravitating body, not the light speed. I tried to keep it simple for Eric Baird and it seems that it shut him up.


> 
> > The light speed locally is STILL c.
> 
> Yes. But non-local measurements can obtain a different answer (e.g. the Shapiro 
> time delay, and the gedanken above).
> 
> Perhaps the most definitive prediction for a non-local measurement is to 
> consider a light source and mirror located radially apart in Schwarzschild 
> spacetime. One can integrate the metric to determine their spatial separation, 
> and then integrate the appropriate null geodesic to determine the round-trip 
> flight time; the ratio differs slightly from c. In Schw. spacetime the 
> anisotropy in such non-local TWLS depends on both the orientation of the light 
> path and the distance between source and mirror. Current technology cannot do 
> this experiment, as we cannot measure distance accurately enough without using 
> light.... I posted an article showing this calculation to this newsgroup, 
> sometime in the early 2000's; its subject probably included "anisotropic" and 
> "light speed".
> 
> 
> Tom Roberts

Yes, this would be an interesting experiment

[toc] | [prev] | [standalone]

Back to top | Article view | sci.physics.relativity

csiph-web