Aether said:
The images are scanned photos of each page of the paper, so a text editor is not going to work. Is there a better format, like pdf that would be easier for you to view?
I think pdf would work much better.
Until there is a confirmed detection of a locally preferred frame, then LET and SR are empirically equivalent; if there is ever such a detection, then LET takes charge.
I would probably not care if that were the end of the story, however SR is not the end of the relativity story, not by a long shot. It is a starting point for everything else; a local approximation to reality. There are big problems on the largest scales with dark matter, dark energy, and on the smallest scales with the unification of QM, EM, and gravity, and more. That's what I'm really interested in, but applying SR can be something like trying to climb a greased flagpole (that's something that we sometimes try to do here in the US), take VSL as an example; SR forbids it by definition. It is flat wrong to claim that experiments constrain the speed of light to be forever constant; so I'm looking for the most efficient ways to argue that, and to model that.
Opinions vary - I find that trying to apply non-SR theories is like trying to "climb a greased flagpole". Metaphorically, anyway, I've never actually tried to do that :-)
In my opinion, if LET wants to accomplish something, it has to at least suggest some experiments which might allow one to detect some sort of preferred frame. If it is just another formulation of SR, it's probably not going to catch on, unless it is simpler to teach (but I suspect that the current formalism is much simpler). It may have a small niche for those who can't adjust their personal philosphies to deal with SR.
It is not likely that there are other than a few typographical errors in M-S. This is what Kostelecky & Mewes, Phys. Rev. D, 66, 056005-4 (2002) have to say about M-S: "In this simple example, the transformation T^\mu_\nu leaves invariant the rods and clocks, while \Lambda^\mu_\nu leaves invariant the speed of light. Both are equally valid. In the frames related by T^\mu_\nu, observers agree on rod lengths and clock rates but disagree on the velocity of light. Moreover, the velocity of light is no longer isotropic as measured by these rods and clocks. In contrast, observers related by Lorentz transformations agree that light propagates isotropically with speed 1 but may disagree on rod lengths and clock rates. The description is a matter of coordinate choice, and one can move freely from one to the other using T^\mu_\nu, \Lambda^\mu_\nu, and their inverses."
Well here's my take on isotropy via a physical example, giving some examples about what is involved in making such a coordinate choice.
Let us suppose that we decide that it is perfectly OK to use an arbitrary clock synchronization to determine speeds, and that we decide to synchronize our clocks by noontime, when the sun is directly overhead. (This is a continuous version of the "time zones" used in the US).
Now, let's compare airplanes flying east-west and west-east with our new clock synchronization methods. We find that airplines flying west travel much faster than the same airpanes flying east, even after we correct for the prevailing winds (which are significant, but I want to ignore this issue).
When we measure the speed of light, we find that it actually arrives before it left with this defintion of synchronization going west - making it have a negative speed (ouch). And it (light) is very pokey going east, traveling verrry slowly.
We also find that the physical expression of momentum depends on the direction one is moving.
With our old definiton of speed, in stil air we could say that the airplanes were going 600 mph east, and 600 mph west, and when they collided, they fell straight down to the ground with no net average velocity.
WIth our new definition of speed, the speeds of identical airplanes flying in still air east-west and west-east are *not* the same. Let's make this concrete, and say that the airplines are going something like 200 mph east, and 1000 mph west in our new system of measurement.
But these airplanes still fall straight down to the ground when they collide (well, that's idealized, but their pieces don't have any net average velocity, and if we could build the airplanes strong enough so that they didn't break apart, we would observe them falling straight down).
Now if we look at two identical airplines colliding with the same velocity, using our new system of synchroniation we find that when they have the same mass and speed, they do not have the same momentum, and that airplanes moving "at the same speed" (with our NEW defijntion of speed) in opposite directions don't fall straight down when they collide.
THe point of this exercise is that clock synchronizations don't really make new physics, which is exactly what the authors you quote are saying.
[add] By this I mean that there are no different experimental predictions. Clearly, Newton's laws have a different appearance when we adopt a non-isotropic clock synchronization method. But the behavior of the actual colliding masses (airplanes in this example) is unchanged.
[end add]
Furthermore, working with clocks synchronized in an anisotropic manner yields anisotropic behavior of physical bodies (like airplanes) as well as anisotropic behavior of light. The clock sychronization that makes light act isotropically is the same clock synchronization that makes airplanes and other physical bodies act isotropically (i.e. come to rest when equal masses at equal velocities coming from opposite directions have an inelastic collision).