Talk:Relativity of simultaneity/Archive 1

Archive 1

Archive 2

Development of the final Lorentz Transformations

I find the paragraph good, but hardly relevant for this article. Isn't it a good idea to try to move it to a more appropriate place? Harald88 02:46, 7 January 2006 (UTC)

Yes, it could go somewhere else and be more relevant, but my guess it would upset some "owners" of the pages on Lorentz transformations and get hacked to pieces. I assume no one really wants to see the first order transformations on the LT page. Also, to me it is important that you get from stage one to stage two as Larmor did by including length contraction and time dilation. I think this makes it clear why relativity of simultaneity is different from time dilation, but I could be wrong. Maybe it is not clear. E4mmacro 12:46, 7 January 2006 (UTC)

Poincaré synchronisation procedure

The following discussion may not be historically correct; please correct me where I'm wrong. I may be duplicating things that are already in the article.

Synchronisation procedure with lightpulses

The first image shows a schematic depiction of the clock synchronisation proceduce that was examined by Poincaré in 1900.

The red lines represent world lines in newtonian space and time; in the context of Poincaré's examination these are world lines of motion with respect to an unmoving immutable luminiferous ether.

The yellow lines represent pulses of light (or pulses of any other electromagnetic wave) that are sent back and forth among a group of clocks that do not have a velocity relative to each other. If the two-way transit time of the signals remains the same, then it has been verified that the clocks do not have a velocity relative to each other. If the clocks to not have a velocity relative to each other, then there is no Doppler shift.

I will refer to this procedure as 'Poincaré synchronisation', since Poincaré was the first to examine it, obtaining important insights.

Poincaré first pointed out the obvious bit: that the clock sychronisation procedure gives no indication of what the clock's velocity is with respect to the luminiferous ether. The transit time for each leg of the time dissemination is unknown. The fact that after 10 seconds the signals arrive simultaneously back to the clock in the middle is in itself inconclusive.

If the indvidual clocks would be able to measure their own velocity with respect to the luminiferous ether, then the operators of the procedure would be able to identify a single plane of simultaneity as the absolute plane of simultaneity. However, since experiments to measure the Earth's velocity with respect to the luminiferous ether had consistently yielded a null result, the conclusion had to be that operators of a lightpulses based synchronisation procedure had no choice but to make do with relative simultaneity.

Synchronisation procedure with lightpulses. The procedure is inconclusive as to the position of the absolute plane of simultaneity.

It should be noted that this Poincaré (year 1900) relativity of simultaneity is presented as exclusively a property of the employed synchronisation procedure. If time is disseminated by use of portable clocks, then - assuming newtons absolute flow of time is true - the absolute plane of simultaneity can still be identified.

The non-obvious outcome of Poincaré's examination is where he looks at the problem of what group of velocities will fit into Poincaré synchronisation procedure. Poincaré points out that the solutions, if taken as a mathematical group, do not form the same group as velocities related by Galilean transformation.

The group of velocities that are solutions to the Poincaré synchronisation form another group, that is now called the Poincaré group.

Later, when examining the work of Lorentz, Poincaré notices that the Lorentz transformations can be written in a more symmetrical form, and then they are the transformations of a group, that is closely related to the Poincaré group.

If the Lorentz transformations hold good not only for electromagnetic phenomena, but for all of physics, and for all velocities, then time disseminaton with portable clocks will yield exactly the same result as Poincaré synchronisation. If the Lorentz transformations hold good for all physics and all velocities, then they imply that all observations will be in accordance with relativity of inertial motion.

I have made animations depicting time dissemination with portable clocks, shuttling back and forth, to illustrate the parallels. The animations in a Sandbox article, that I use to explain relativistic physics to myself. --Cleonis | Talk 13:19, 20 January 2006 (UTC)

That is a good presentation perhaps some of it can be put inthe article space. And I now notice that the Poincare relativity of simulataneity is not well covered in this article. I'll look and see if I can add some stuff of his 1898 paper. Harald88 14:42, 9 March 2006 (UTC)

Poincare and Lorentz

I hope you do not complicate the article too much. When I first wrote it, I was trying to show only that relativity of simultaneity is not the same thing as time dilation. And I don't think Poincare in 1900 was thinking of groups at all.

Since Poincare was talking of a first order theory, it is possible to detect motion of the reference frame order of v^2/c^2. There is more than one way to do that and portable clocks is one of them. I assume Poincare did not consider it or talk of it, becasue there were no such clocks accurate enough in existence. With clocks synchronised the way I describe, one can re-synchronise them, using a different clock as the master and find a small discrepency of order v^2/c^2. I am now starting to wonder if Poincare actually advocated a slightly different procedure where one did take this second step, and then re-adjusted both clocks.

Historical note, which you guys in Europe may be able to answer more easily than me. I have very little information on what Lorentz said about his local time. I get the impression from Poincare that he is only explaining what is "well known", but it is not clear. Can you find where Lorentz introduced it - I udnerstand it is in the book "Versuch ..." about 1895 (see Lorentz page). Larmor 1897, gives no specific reference though he is clearly following Lorentz. I wonder if it was well known at the time. Can you get Lorentz 1895 and check if he has any earlier references. It may be the Lorentz-synchronisation procedure.

And (another topic) what about Proc Amst Acad. (English Edition), VI (1903), p 809? Whittaker claims that Lorentz gave the correct transformations in that paper as well. Lorentz had given them in 1899 with a multiplying factor undetermined (I callled it $\ell$ on the Poincare page), so maybe this 1903 still leaves that factor undetermined. Still would like to know what is difference between Lorentz 1903 and 1904.

Also, on re-reading Poincare 1900 (where he is expounding on Lorentz's theory) I find that Poincare apparently misunderstands what Lorentz meant by local time: the same misunderstanding that I have discussed with Harald in a few places. i.e. Lorentz writes t' = t - vx*/c^2, where x* = x - vt (x is the coordinate relative to the fixed origin in rest frame), whereas in the place where Poincare uses the local time he seems to think Lorentz's local time is t - vx/c^2, (which gives dt'/dt $\neq$ 1, a time dilation, and in fact the local time in Voigt's transformation). (It seems that) Poincare has actually used t' = t - vx*/c^2 - (v^2/c^2)t. Maybe he thought he could drop the last term? E4mmacro 21:39, 1 February 2006 (UTC)

I'm afraid I can't help you with those details. Most of what I know is not from reading the original papers, but second hand knowledge.--Cleonis | Talk 22:35, 1 February 2006 (UTC)

The question of why and when Poincaré described the Poincaré group

I copy and paste from above:

And I don't think Poincare in 1900 was thinking of groups at all. E4mmacro 21:39, 1 February 2006 (UTC)

Actually, I have filled in blanks in my knowledge there. Poincaré had at some point in is mathematical explorations described the Poincaré group, and my understanding is that that was prior to Poincaré's examination of Lorentz's work.

It is my understanding that on examining Lorentz's work, Poincaré recognized that when the transformations that Lorentz used at the time are rearranged to a more symmetrical form, then they correspond to a subgroup of the more general Poincaré group.

I have assumed that Poincaré documented the mathematical properties of the Poincaré group in the course of his examination of clock synchronisation procedures. It looks plausible, I don't see any other scenario, but I do not have the original papers for corroboration of that assumption.

I really need to get hold of Poincaré's original papers, especially his essay 'the measure of time'. I should be going to the library, and ask what possibilities there are for obtaining works by Poincaré.--Cleonis | Talk 22:41, 1 February 2006 (UTC)

Hi sorry for my silence (I'm too busy), but about Poincare articles: many are now directly or indirectly linked from the poincare page. Harald88 23:17, 1 February 2006 (UTC)

Thank you. In the meantime, after some googling, it has dawned on me that probably the name 'Poincaré group' was assigned after Poincaré's death, to honor his memory. So the interpretation then becomes that Poincaré recognized group properties of the symmetrical form of the Lorentz transformations just straight away, and not necessarily because of prior examination of that type of mathematical forms. --Cleonis | Talk 23:35, 1 February 2006 (UTC)

"Straight away" is a bit of an exaggeration. Lorentz wrote his transformations in 1899, second half of that paper, full-second order version but still containing the undetermined factor $\ell =f(v)$, which Poincare realised in 1905 had to be 1 to form th group (after Lorentz had already decided \ell = 1 in 1904). Whittaker says Lorentx repeated the Lorentz transformations in 1903. Perhaps Poincare never noticed the 1899 and 1903 papers. We know by the time of Poincare's first letter to Lorentz 1905 that he still hadn't seen (after seeing the 1904 paper) that he had to re-arrange them, and that \ell = 1. So it took even Poincare quite some time to see it. E4mmacro 05:57, 2 February 2006 (UTC)

Seeing the group property

In his 1905 paper Einstein points out that the transformations that he derived form a group. I take it that with 'group' Einstein refers to a symmetry group as in Group theory
It is unclear whether Einstein had learned from reading Poincaré of seeing a set of transformations with an operation of composition of transformations as constituting a mathematical group.

The book La valeur de la Science contains the essay La mesure du temps and my browser does not find the name Lorentz in that essay. Chapter 8 of that book is about. 'La principe de relativité' and Poincaré discusses Lorentz' concept of local time and the synchronisation procedure using electromagnetic pulses in one breath. --Cleonis | Talk 15:41, 2 February 2006 (UTC)

What does that last part have to do with group theory? Poincare stressed it in his June 1905 paper. I doubt that Einstein learned any group theory, thus apparently he copied that from Poincare. Harald88 22:22, 2 February 2006 (UTC)

It is known that Einstein and a group of friends studied and discussed Poincaré's book La science et l'hypothese (1902). In Chapter 4 and in Chapter 5 Poincaré discusses the concept of operations constituting a group in the context of geometrical systems. That kind of elementary concepts of geometry were possibly also part of the mathematics curriculum of the ETH in Zurich, so it is inconclusive. --Cleonis | Talk 11:18, 4 February 2006 (UTC)

Cleonis's intro

See [1]

"In newtonian physics, it is assumed that there is a meaningful concept of absolute simultaneity, and the success of newtonian physics indicated the concept of absolute simultaneity was justified."

...Yes, but: presented like that, it risks to put a reader on a false track. In Newtonian physics, "absolute" is meant as described in dictionary.com, 7a.

And that is probably not the way Cleonis meant it! There is relative simultaneity and absolute simulataneity, and Newton assumed those to be equal; making all simultaneity "absolute" in the way Cleonis probably meant it.

Thus, following Lorentz and Langevin, Newton's absolute simultaneity concept is not affected by special relativity; relative simultaneity just differs from absolute simultanety.

We are used to such things with time in daily life: "local" time is not equal to "universal time", but they are not incompatible, instead they happen to be differently defined. Harald88 22:18, 2 February 2006 (UTC)

It is unclear what you are referring to. I am referring to an operational definition of time-keeping. If you put an atomic clock onboard an airoplane that circumnavigates the world. (Such as the GlobalFlyer in which Steve Fosset flew non-stop around the world) then on arrival the clock that has travelled will be seen not to have counted the same amount of time as non-travelling clocks (With a second flight with circumnavigation in opposite direction, gravitational effects can be corrected for). The newtonian expectation is that on arrival the onboard clock is seen to have counted the same amount of time. The newtonian expectation is that the amount of time that is counted is independent of the path that is travelled.

Please define operationally your concepts of 'relative simultaneity' and 'absolute simultaneity' in the context of newtonian physics on one hand, and in the context of relativistic physics on the other hand. --Cleonis | Talk 23:07, 2 February 2006 (UTC)

What is unclear about the dictionary definition? Here once more:
Relating to measurements or units of measurement derived from fundamental units of length, mass, and time.
Fundamental units depend on one's definition; and such change with time as well with application.
The operational definition of time keeping as used on the earth is "universal time"; it's not "relativistic" time, as that simply doesn't work on a rotating earth.

Using the dictionary definition, which corresponds to the traditional one:
- in Newtonian mechanics I already explained above, the simultaneities are the same (equal but not identical);
- in relativistic physics according to Einstein, there's only "relative"; while according to Lorentz, Langevin, Ives and Builder, "absolute" would refer to the unknown but physical "absolute" reference frame; but "relative" would refer to the operational definition.(Note; I didn't check if they all used this definition; it's possible that for example Poincare used again another one).
Since Wikipedia cites old literature, any confusions should be avoided. Harald88 23:38, 2 February 2006 (UTC)

You have a point in remarking that 'absolute' and 'operational' are not neccessarily the same. I was thinking along the following lines: if the flow of time is absolute in the sense of being independent of anything else, then it is expected that a procedure can be set up that will converge on a unique plane of the simultaneity that will be converged on by every observer, no matter his relative velocity to other observers. Some people may opt to call that unique plane of simultaneity 'the absolute plane of simultaneity'. --Cleonis | Talk 23:55, 2 February 2006 (UTC)

Replace "expected" by "imaginable", and I agree with you. In practice this can't be done (except if one defines for example the CBR "frame" as that), so that "universal time" has taken its operational role instead. Harald88 18:53, 3 February 2006 (UTC)

Summarizing: the words 'expected' and 'imaginable' are here used in relation to what the laws of physics allow. Given the newtonian assumptions, it is imaginable (be it technologically very challenging) that a procedure can converge on a frame-indipendent plane of simultaneity

As you state, there is also the level of what is imaginable as technologically attainable.

I have been using the expressions 'absolute simultaneity' and 'relative plane of simultaneity' in relation to what the formulated laws of phycics allow, what is technologically attainable has not figured in how I have used those expressions. --Cleonis | Talk 21:59, 3 February 2006 (UTC)

It is a little disapointing to me to see the very things I was trying to separate - simultaneity and clock rate - tied up together again. I claim these are different things, and the example that Einstein uses, two light flashes at the ends of the train station, with the train carriage moving past it, shows that they are different things. There is no clock in Einstein's example, there is nothing about clock rates. His argument shows that simultaneity is relatives (and incidentally that length must be relative).

I had hoped that the page would overcome the unthinking assumption that clock rates and simultaneity are the same thing, or they are inextricably linked. They may be related in fact; my point was they were not necessarily related. Looks like I failed. E4mmacro 07:48, 6 February 2006 (UTC)

The Geoid as surface of simultaneity

I copy and paste from above:

The operational definition of time keeping as used on the earth is "universal time"; it's not "relativistic" time, as that simply doesn't work on a rotating earth. Harald88 23:38, 2 February 2006 (UTC)

As you know the Geoid is a surface of simultaneity, it is an isochronic surface. Wherever you are on Earth, if you are at sea-level height then a local clock that is co-rotating with the Earth counts time at the same rate as all other sea-level clocks on Earth. The reason that for sea-level clocks on Earth no relativistic procedures are in place is the simple fact that no time dilation occurs between sea-level clocks worldwide. Time dilation occurs for clocks that do not follow the same path as corotating-with-the-Earth clocks.
The fact that the Earth is rotating cancels out for co-rotating clocks; the fact that the Earth is rotating does not affect the Geoid being a global surface of simultaneity.

However, the Earth's plane of simultaneity is tilted with respect to the plane of simultaneity of the center of mass of the Solar system. What is simultaneous in the terrestrial time-keeping is not necessarily simultaneous in the center-of-mass-of-the-Solar-System time-keeping. Possibly, astronomers already have to take that into account. --Cleonis | Talk 00:38, 3 February 2006 (UTC)

As you may know, the simultaneity at the geoid is such that the geoid is moving (rotating) relative to the used frame of simultaneity. This is because no rotating frame can be used for synchronization, except for by correcting for that rotation. Harald88 18:57, 3 February 2006 (UTC)

I wrote the article about the Sagnac effect, so yeah, I know. Of course the Earth's rotation needs to be accounted for in a synchronisation procedure using signals, but that is not specifically a relativistic issue.

As you point out, a synchronisation procedure is only self-consistent if its reference frame is an inertial frame. Correcting for the Sagnac effect means that the procedure refers to the local inertial frame instead of to some rotating frame. --Cleonis | Talk 21:37, 3 February 2006 (UTC)

I am reluctant to get involved in the time dilation issue, since I think it is logically different from relativity of simultaneity and I don't really want to encourage the idea that this page should be about time dilation, but: Didn't Einstein (1905) say that a clock at the equator runs slower than one at the poles? Is that what we are talking about? An why has no one tested this. The atomic clock in Washington, at a different latitude from that at NPL in Britain should eventually get out of sync, from time dilation? E4mmacro 20:39, 6 February 2006 (UTC)

Absolute temperature and absolute simultaneity

I think the following analogy between absolute temperature and absolute simultaneity applies well.

Recapitulating: temperature can be defined in terms of fundamental constants, and in that sense temperature is absolute. There is a unique temperature that is assigned the value of zero: zero Kelvin. A second point defines a scale. The melting point of ice at a pressure of one bar defines the scale of Celcius.

You can transmit the specifications of that procedure to another planet somewhere else in the Galaxy, and thus they can duplicate the Kelvin scale. So temperature is absolute, in the standard dictionary (part 7a and 7b) sense.

If - theoretically - a procedure can be set up that allows distant observers to duplicate a physically unique plane of simultaneity, then the plane of simultanety is absolute. (Of course, the level of accuracy of the procedure is dependent on the level of technology that is available.)

If the very theory predicts that there can be no procedure that allows observers to establish a frame-independent standard, then only a concept of relativity of the plane of simultaneity (dependent on relative inertial motion), is available. --Cleonis | Talk 12:50, 3 February 2006 (UTC)

Indeed, light flashes and moving trains do not illustrate special relativity

I copy and paste from above:

It is a little disapointing to me to see the very things I was trying to separate - simultaneity and clock rate - tied up together again. I claim these are different things, and the example that Einstein uses, two light flashes at the ends of the train station, with the train carriage moving past it, shows that they are different things. [...] E4mmacro 07:48, 6 February 2006 (UTC)

I agree that the example of light flashes and moving trains does not enforce the concepts of special relativity. A physics teacher who would present the example of light flashes and moving trains as an example of special relativity would be a poor physics teacher. Good physics teaching avoids the light flashes and moving trains example.

I guess Einstein was not a good teacher of relativity. E4mmacro 23:56, 16 April 2006 (UTC)

Also, there is babylonian confusion here. I think you are using the expression 'relativity of simultaneity' in a fundamentally different way from how I am using it. For the time being I will switch to the expression 'full blown relativity of simultaneity'.

A concept of full blown relativity of simultaneity is enforced by the combination of the following two things.
1) All procedures, no matter what physics phenomenon is employed, that an observer A sets up, yield results that are consistent with each other. All procedures that an observer B sets up yields results that are consistent with each other too, but if observer A and observer B have a velocity with respect to each other, then the two sets of results do not match.
2)The phenomenon of time dilation.

Again, I emphasize that in my opinion the example of flashes of light and moving trains is very ill suited for illustrating special relativity. I would never use it.

Let me describe an example that in my opinion does illustrate special relativity.
Let an atomic clock be put onboard an aeroplane that circumnavigates the world at a constant velocity, (such GlobalFlyer, in which Steve Fosset made a non-stop flight around the world.)Let the GlobalFlyer fly at constant altitude, with a constant velocity. Then the atomic clock onboard the Global flyer will on arival be seen to have counted 207 nanoseconds less or more time, depending on whether the flight was eastwards around the world or westwards around the world. (The time count does need to be corrected for gravitational effects.)

What is special about the circumnavigation is that a loop is closed. Because a loop is closed there is an opportunity of comparing things that could otherwise not be compared.

These 207 nanoseconds are consistent with the following:
It takes a signal propagating at lightspeed 0.1336 second to circumnavigate the world. In 0.1336 second the Earth's surface at the equator moves 62 meters. Light takes 207 nanoseconds to travel 62 meters.

Signals propagating at lightspeed are the most extreme case of a more general class of cases. Special relativity predicts for time disseminaton around the world that all forms of time dissemination, independent of the velocity of the time bearer, will end up with those 207 nanoseconds that need to be corrected for. --Cleonis | Talk 10:32, 6 February 2006 (UTC)

Addendum:because of the loop-closing this example does not illustrate relativity of simultaneity, it only illustrates time dilation and the phenomenon of all procedures yielding the same result (in this case: all procedures requiring the same correction: 207 nanoseconds). --Cleonis | Talk 10:42, 6 February 2006 (UTC)

Sorry but, as you yourself stated above, the Sagnac effect is not specific for relativity. IMO, your example is completely messed up, perhaps because you confused relativity of simultaneity with time dilation. Indeed, relativity of simultaneity has been argued to be a special case, only for inertial frames.

The only things that matter for calculating the number of nanoseconds difference are speed and altitude of the Global Flyer relative to the ECI frame at sea level; and those were not even provided! Harald88 18:30, 6 February 2006 (UTC)

[continuation from intermezzo on my Talk page]

Cleon, I now worked it out, so that I now can quantify it. 

I may have made an error somewhere, but I'm pretty sure that the following is pretty much correct.

Assuming a Global Flyer at a cruising speed of 250 m/s, and taking Earth radius 6378 km, I obtain a total flying time around the equator of 160300 s; and an earth rotation speed of 465 m/s.

From that I find that a radio signal that is sent Westward will be back 207.0 ns "too soon", and one that was sent Eastward, 207.1 ns "too late".

The time dilation factor of earth clocks on the equator (we take those as reference standard),

using the approximation that sqrt(1-v^2/c^2) = (1 - 0.5v^2/c^2):

t_0 / t = (1 - 1.20125E-12).

BTW, it may be useful to point out that this implies that relative to non-rotating clocks, these earth clocks "time dilate" by 103.5 ns per rotation (=half of the 207 ns).

Next we can compute the time dilation factors t_E /t and t_W /t for the Global Flyer flying east and flying west, relative to the earth clocks :

Those are resp. -1.64E-12 and +0.95E+12 relative to the earth clocks.

In the end, I obtain:

dt_E = -263 ns (263 ns retarded)

dt_W = +152 ns (152 ns gained)

Of course, when we slowly transport a clock relative to the equator, we will nearly get your above claimed time dilation or advance relative to earth clocks, because they are synchronized in the ECI frame. For example, taking 2.5 m/s ground speed Westward:

dt_W = +1.2882E-14 * 16030000 = + 206.5 ns ( = almost 207 ns advanced on Earth clocks.)

Time dilation and simultaneity must not be mixed up.

Cheers, Harald88 22:00, 8 February 2006 (UTC)

Not trying to illustrate Special Relativity

I think I see where we differ Cleonis. I see the new page has become another illustration of special relativity and the clock paradox and so on. We have many such pages and I doubt that we need another. I was not trying to illustrate special relativity, and in my view it is irrelevant whether the train and light flashes that Einstein used is the best way to do so. I am trying to illustrate relativity of simultaneity (ROS), and in my opinion, the train is perfect for that. I thought there a was need for a page which discussed ROS, and made it clear (from the history) that ROS is a different thing (at least in logic) from clocks running at different rates. Probably you disagree with me on that; probably you think the two things are inextricably linked (as I said in my version, "inextricably linked" is what I was disputing). I find the revised page, while perfectly true and admirable, confusing in exactly the way I was trying to avoid. IOW, I think we are trying to illustrate different things. E4mmacro 20:31, 6 February 2006 (UTC)

BTW, Heaven forbid I would ever be allowed to teach SR, but I have to say I would use the train example - a great thought provoker. Not only does it show something odd about simultaneity in moveing frames, but also that there is something odd about lengths - i.e. the ends of the train coincided with the ends of the platform at simultaneous moments in one frame but not in the other. Hence the train is the same length as the platform in one frame, but not in the other. But again, this is a discussion that belongs elsewhere, in my opinion. E4mmacro 20:49, 6 February 2006 (UTC)

Your intention was, I understand, to write an article that discusses the concept of relativity of simultaneity from a theory-independent point of view. I hadn't expected that, because I do not see a way to discuss full blown relativity of simultaneity in a theory-independent way.

I agree that the version I wrote only duplicates material that is in the many articles about special relativity.

It may be possible to discuss a watered down version of full blown relativity of simultaneity theory-independent, but I don't see any purpose being served by discussing a watered down version of full blown relativity of simultaneity.

I agree that attempts must be made to avoid discussing fine points of time dilation on this Talk page.

In my opinion the train example does not in itself enforce the concepts of time dilation and length contraction. Only if the second postulate of the 1905 special relativity article is granted is the special relavity enforced (the second postulate being the postulate that light emitted from a single source will propagate away with precisely velocity c from both observer A and observer B, even if observer A and B have a velocity with respect to each other. --Cleonis | Talk 23:45, 6 February 2006 (UTC)

The 1905 Einstein prediction that didn't last.

I copy and paste from above:

I am reluctant to get involved in the time dilation issue, since I think it is logically different from relativity of time dilation and I don't really want to encourage the idea that this page should be about time dilation, but: Didn't Einstein (1905) say that a clock at the equator runs slower than one at the poles? [...] E4mmacro 20:39, 6 February 2006 (UTC)

That is indeed in the 1905 (electrodynamcs of moving bodies) article. Einstein couldn't possibly foresee it, but it turned out there is more to time dilation than that.

The general theory of relativity describes a curvature of space-time around bodies of mass. One of the aspects of this curvature is rate of time. Clocks that are deeper in a gravitational well run slower than clocks higher up in the gravitational well. Clocks on the North pole are about 20 kilometers closer to the Earth's center of mass than clocks on the Equator. The two effects cancel out precisely, and world-wide all clocks at sea-leavel (or any other continuous geopotential surface) count time at the same rate. This canceling out of velocity time dilation and gravitational time dilation is not a coincidence, of course. The solid Earth is ductile. If the rate of time would not be equal everywhere, there would be shear stresses, and over millions of years these shears stresses mold the shape of the Earth, until eventually the shear stresses are resolved. --Cleonis | Talk 00:06, 7 February 2006 (UTC)

Ok, I never knew that, never considered the gravity time dilation in this context (thought, or failed to think - the gravity part was the same for all the surface clocks). Time dilation introduced stresses? I will have to think about that. Is this the same or similar to saying the effective gravity force (i.e. Newtonian - minus the centrifugal force) is the same everywhere on the Earth's surface? Because the Earth deforms until this condition is met. And it follows from this that the clock rates are the same everywhere? E4mmacro 06:32, 7 February 2006 (UTC)

The newtonian view is uncomparable with the relativistic view, but both predict the same shape for the planet Earth. I wrote the wikipedia article about the Eötvös effect I thing the newtonian story is pretty much there.

About the relation between gravitional time dilation and shear stresses in the Earth:
Special relativity describes that the path with the least spatial length is the path with the most lapse of proper time. It is invariably seen that object on which no mechanical force is acting follow a path in space that is the path with the most lapse of proper time. Any path that is longer than the path that is spatially the shortest path has more lapse of proper time than the spatially shortest path.

In order to make an object follow a path that is not the spatially shortest path, a mechanical force must be exerted. This animation illustrates that.

So let's take the Himalaya mountains. Those mountains are relatively young and they are being pushed up as a result of two tectonic plates pushing against each other. The height of the Himalaya is a big deviation from the Geoid The himalaya are higher in the gravitational well, corresponding to a smaller amount of proper time elapsing.

Any system tends towards the state where the amount of time that elapses is the maximum possible. (This corresponds to the thermodynamic concept that any system tends towards a state where all of the energy that can dissipate is dissipated.)

In order for the Himalaya to remain that high, an upwards force must be provided. For every kilometer the Himalaya has risen to the skies, other material has been pushed down, creating buoyancy for the mountains on top of it. When a mountain range is neutrally buoyant on top of the Earth's mantle, then there are no shear stresses. --Cleonis | Talk 11:54, 7 February 2006 (UTC)

Another way to put it, is that hydrodynamic equilibrium is reached when the kinetic energy is countered by an equal amount of potential energy (easy to derive, and perhaps useful to show once more that we have no need of fictive forces for calculating water levels in rotating buckets). Cheers, Harald88 20:56, 8 February 2006 (UTC)

I exaggerated when I wrote that the newtonian outlook is different. There are newtonian counterparts to the GR outlook. I developed this theme in the following articles: Mercury mirror, Oblate spheroid, and Rotational-vibrational coupling. (Sorry, this is way off-topic, I just couldn't resist plugging my articles.) --Cleonis | Talk 22:08, 8 February 2006 (UTC)

Return to discussing Relativity of Simultaneity

I copy and paste from above:

I was not trying to illustrate special relativity, and in my view it is irrelevant whether the train and light flashes that Einstein used is the best way to do so. I am trying to illustrate relativity of simultaneity (ROS), and in my opinion, the train is perfect for that. [...]
[...] I think we are trying to illustrate different things. E4mmacro 20:31, 6 February 2006 (UTC)

To pinpoint what you want to illustrate, I will discuss the moving trains example, but with sound signals instead of light signals; both light and sound are described with wave-mechanics.
Using sound signals means that the physics of special relativity is avoided.
My estimation is that the discussion below does not illustrate what you want to illustrate. My question to you is to pinpoint what physics you want to incorporate in the illustration.

Let there be a train station, a moving train, and let sound signals be used instead of light signals; let the air be stationary with respect to the train station. For the purpose of having round numbers:
The speed of sound is 300 m/s
The velocity of the train is 150 m/s.
The train station is 300 meters long.
The train is 90 meters long.

At a moment in time designated t=0, the middle of the train is at the middle of the train station. At that moment a pulse of sound is emitted from the center of the train, propagating in all directions. The sound emitter is placed on the outside of the train; the sound signals travels through the outside air. At the moment the ends of the train recieve the pulse of sound they respond by answering back with a pulse of sound.

The pulses of sound emitted from the middle of the train are recieved simultaneously at the ends of the train station after 0.5 second (simultaneous in the train station frame).

The pulse of sound travelling to the front of the train is travelling at a velocity of 150 m/s relative to the train, so it takes the pulse of sound 0.3 second to reach the front of the train. The pulse of sound travelling to the rear of the train takes 0.1 second to get there.

If the train would be stationary with respect to the air then a round trip of a pulse of sound (from middle of the train to front/rear and return) would take in all 0.3 second. (As measured by the clock in the middle.) In the above example the round trip takes in all 0.4 second (as measured by the clock in the middle), instead of the 0.3 second that it would take if the train is stationary. So the operators of the train can calculate the velocity of the train on the basis of the amount of lengthening of the round trip time of the acoustic signals. If the train is stationary the round trip time is 0.3 second, if the train is moving at 150 m/s the round trip time is 0.4 second. --Cleonis | Talk 21:05, 7 February 2006 (UTC)

Return to discussing Relativity of Simultaneity (2)

Impatient as I am, I am immediately following up.

I will now discuss the train station and moving train and signals example, for the case of signaling with pulses of light, combined with the second postulate of Einstein's 1905 article
Postulate:
Light emitted from a single source will propagate away with precisely velocity c from both observer A and observer B, even if observer A and B have a velocity with respect to each other.

In itself, the second postulate allows for three possibilities:

1. Avoiding an assumption of length contraction by attributing all effects to an assumed time dilation
2. Assuming symmetrical time dilation and length contraction, leading to the symmetrical form of the Lorentz transformation. (Which is what Poincaré and Einstein did.)
3. Avoiding an assumption of time dilation by attributing all effects to an assumed length contraction.

All three of the above possibilies can be formulated in a mathematically consistent way. Whether that mathematical formulation is consistent with observations is another matter. All three possibilities have in common that they imply a concept of full blown relativity of simultaneity.

E4mmacro wrote:

I would use the train example - a great thought provoker. Not only does it show something odd about simultaneity in moveing frames, but also that there is something odd about lengths - i.e. the ends of the train coincided with the ends of the platform at simultaneous moments in one frame but not in the other. Hence the train is the same length as the platform in one frame, but not in the other. But again, this is a discussion that belongs elsewhere, in my opinion. E4mmacro 20:49, 6 February 2006 (UTC)

Michael, it appears to me that your claims and statements are consistent with the third of the three points above.
From a mathematical point of view it is possible to formulate a theory of physics that incorporates the second postulate of Einstein's 1905 article, but without any time dilation. --Cleonis | Talk 02:20, 8 February 2006 (UTC)

The article states that Einstein gave an example in 1905 of simultaneity using moving trains. I do not find it it his June 1905 paper. Where can it be found? The article needs a reference for this. TY, green 193.108.45.131 13:51, 12 March 2006 (UTC)

Separating what is logically separate

I copy and paste from above:

It is a little disappointing to me to see the very things I was trying to separate - simultaneity and clock rate - tied up together again. I claim these are different things, [...] E4mmacro 07:48, 6 February 2006 (UTC)

What I have been trying to convey is that the example of moving trains and light flashes does not automatically imply the concept of length contraction. I had hoped that my efforts would help to overcome the unthinking assumption that length contraction and simultaneity are the same thing, or they are inextricably linked. --Cleonis | Talk 20:17, 9 February 2006 (UTC)

Perhaps I need to explain the comment about relativity of length (or delete it). Einstein's talks of a train which is the same length as an embankment when it is moving. Lightning flashes hit each end of the train and the embankment and the images of these falshes hiting the ends of the train and embankment meet at the mid point of the embankment. Hence the flashes were simultaneous in the ground frame and the moving train was the same length as the embankment (according to the ambnakment POV). On the moving the train, the images of the lightning do not meet at the center of the train but towards the back of the train (the train has moved on, so its center does not coincide with the centre of the embankment when the images meet). Hence the flashes were non-simultaneous in the moving train frame of reference. But notice also that, on the train, they now know that the train is longer than the embankment. This is not necessary for the discussion of simulatneity; it just shows that measurement of length is also relative to the reference frame.

La mesure du Temps

I have that article of Poincare, which is perhaps the oldest one that explains the subject (and that without bringing up time dilation issues); but I don't have it in English translation. For now I'll translate relevant parts myself; but for Wikipedia I guess it will next have to either replaced by an official translation, or by paraphrasing. Does anyone know of an English translation? Harald88 16:49, 9 March 2006 (UTC)

Reduced to a stub

I was only recently alerted to the state of this article. It is overall a rant givng credit to Lamor for what is really Einstein's discovery. Starting with the second paragraph is head off on the wrong path. Time dilation and the relativity of simultaneity is intimately intertwined (along with the Lorentz contraction).

An examination of how the Lorentz transformations leads to this effect is needed. A discussion of the train exercise, or even a horizontal light clock would also be nice. However, Lamor and the history of the derivation of the Lorentz transformations are quite off-topic here.

Those are the reasons that this article has been reduced to a stub. I call on others to support this unless they can create a good article on this topic. --EMS | Talk 22:23, 30 August 2006 (UTC)

You make claims that you better support with evidence. As it stands, the evidence of this article is apparently undeniable. The only way to hide that information is to vandalise it by deletion ("stubbing").

I agree that some of the text doesn't directly refer to simultaneity, and that it may be improved in places. However, deletion of a whole article just because one or two editors don't like some of the information cannot be tolerated in Wikipedia. Thus I undo what I regard as vandalism. Harald88 23:03, 30 August 2006 (UTC)

You seem to somewhat get my point. Basically everything after the first paragraph does nothing to describe the relativity of simultaneity. Instead it is a diatribe declaring that it and time dilation are seperate pheonmena. That assertion is fundamentally false. To consider the relativity of simultaneity in the absense of time dilation is a good thing to do, but this discourse about history in which Einstein's train exercise (which BTW was done 1905 and not 1960) is only an aside is not appropriate at all. Also be aware that I have never heard Lamor knowning of the relativity of simultaneity before Einstein did. Indeed, Einstein's description of how it arises from the universal constancy of c is how it got introduced to the world.

I will leave the article in its current state for now. I am about to go on vacation, and so starting an edit war is useless now. However, be advised that this is not a case of needing to impove the article in a place or two. Instead it needs a serious rewrite, and I stongly feel that the stubified version is far superior to what is here now. --EMS | Talk 00:25, 31 August 2006 (UTC)

Looking at the article again, I notice some subtle mistakes but I think that they can easily be dealt with, without a "complete rewrite". Relativity surely doesn't work with only relativity of simultaneity. My suggestion of how to improve the structure of the article follows below. Harald88 09:19, 2 September 2006 (UTC)

Sorry that you think it is a diatribe. I thought it was a clear, calm explanation of a logical difference between relativity of simultaneity (ROS) and time dilation (TD) - that the two things could in logic be unconnected - though in reality they are linked as described by the Lorentz transformation. I am sure Larmor never used the phrase "relativity of simultaneity", but the fact is he wrote the Lorentz transformations (in 1897) which contained the information or implication that two events simultaneous in one frame were non-simultaneous in a moving frame. I didn't mean to claim Larmor invented relativity of simultaneity - Lorentz did that in 1895 or so. What Larnor discovered or predicted or expressed was the physical effect of time dilation. The fact that the ROS and TD were introduced at different times by different people seems relevant to the claim that they are logically different things (which happen in fact to be linked).

I can see why you were confused - if you think time dilation is necessarily the same as relativity of simultaneity, then you think Larnor's discovery of the first must be the discovery of the second. Or if you think the Lorentz transformations are all there is to relativity, then you must take the fact that Larmor wrote them correctly in 1897 as a much larger claim than I think I was making.

The 1960 date refers to the publication date of the edition of the book I was refering to, not a claim of when Einstein wrote it - it identifies the book as given in the reference list.

What specific statement in the article is challenged as factually inaccurate? E4mmacro 05:39, 31 August 2006 (UTC)

Basic stuff

It is common in physics language that an expression can acquire a specificness that goes farther than the literal words. For example: the expression 'conservation of momentum' is by convention understood as referring to the Principle of conservation of momentum.

The same is the case with the expression 'relativity of simultaneity'. That expression is by convention understood as referring to the relativity of simultaneity that follows logically from assuming that the Lorentz transformations are fundamental. The threesome: time dilation, length contraction, and relativity of simultaneity are inseparably interlinked in following logically from the Lorentz transformations.

Of course, in the context of a thought experiment that proceeds in newtonian space and time, without the Lorentz transformations, it is possible to elicit a vague resemblance of relativity of simultaneity. If you set up the Einstein synchronisation procedure in newtonian space and time you can elicit a vague resemblance of relativity of simultaneity, but it's not comparable to the relativistic relativity of simultaneity.

The following can be taken as a starting point: there is no logical progression from the Einstein synchronisation procedure to special relativity. Contemplating the Einstein synchronisatio procedure does not lead to the concepts of time dilation, and length contraction.

What does lead to the threesome time dilation, length contraction and relativity of simultaneity is either of the following (mathematically equivalent) starting points:

• assuming the Lorentz transformations are fundamental
• assuming invariance of the spacetime interval is fundamental
• assuming the two postulates of Einstein's 1905 article.

(With some ingenuity other variations can be generated, I'm sure)

There is no context in which the assertion "relativity of simultaneity and time dilation are independent phenomena" makes any sense. It does not make sense in the context of newtonian space and time, for in newtonian space and time there is no time dilation anyway, and it does not make sense in relativistic spacetime, for in the context of relativistic spacetime the threesome time dilation, length contraction, and relativity of simultaneityare inseparably interlinked.

Some things are so fundamental that it is unreasonable to demand referencing. For example: Newtons third law and conservation of momentum are mathematically equivalent. There is hardly something more basic than that: any physicist knows that. Likewise the threesome: time dilation, length contraction relativity of simultaneity; that's the very basics of special relativity. --Cleonis | Talk 23:02, 31 August 2006 (UTC)

Cleon, I'm afraid that I disagree with you that the concept "relativity of simulataneity" is dictated by a theory that was proposed after the concept was established. Thus I don't fully follow what you mean. IMO it would be more constructive to just discuss what parts of the article you want to be amended, and why. Harald88 09:45, 1 September 2006 (UTC)

I guess this might become a futile argument about definitions, but consider the following:

"{ROS] is by convention understood as referring to the relativity of simultaneity that follows logically from assuming that the Lorentz transformations are fundamental." I have an idea that this means that Einstein's wonderfully clear example of the train and the lightening flashes fails your test. He did not mention clocks; he did not mention length contraction; he does not show ROS follows from the Lorentz transformations. He wanted merely to illustrate one thing, which he did very neatly. That one thing was that ROS follows from assuming there are no instantaneous signals. IMO his explanation is forcefull precisely because it does NOT mention the other things. E4mmacro 07:32, 1 September 2006 (UTC)

The trains-and-flashes procedure fails the test

I copy and paste from above:

"ROS is by convention understood as referring to the relativity of simultaneity that follows logically from assuming that the Lorentz transformations are fundamental." I have an idea that this means that Einstein's wonderfully clear example of the train and the lightening flashes fails your test. E4mmacro 07:32, 1 September 2006 (UTC)

Yes, I regard the train-and-flashes-of-light example as uninformative for special relativity. The train-and-flashes-of-light example is neither necessary nor sufficient to logically imply special relativity. It fails the test.

So from a historical point of view there is a transition. There is - in a limited sense - a pre-relativistic concept of relativity of simultaneity, and there is a relativistic concept of relativity of simultaneity, and while superficially similar, they are profoundly different concepts. The train-and-flashes-of-light example obscures the distinction between pre-relativistic RoS and relativistic RoS, which is why I judge the train-and-flashes-of-light procedure to be totally unsuitable for the purpose of education.

The difference:
In the context of pre-relativistic RoS, technological sophistication takes care of the problem: instead of using pulses of light for the procedure to disseminate time, portable clocks can be used. Thus, time can be disseminated faithfully. This implies that in the pre-relativistic context there is in fact no expectation of relativity of simultaneity, the limitation is expected to be merely technical in nature.
In the case of relativistic RoS, there is no technological remedy; disseminating time with portable clocks would result in exactly the same outcome as disseminating time with pulses of light.

Therefore I argue that the convention is good and should be used as the basis for composing the article: that the expression 'Relativity of Simultaneity' is understood as referring to the relativity of simultaneity that follows logically from assuming that the Lorentz transformations are fundamental. --Cleonis | Talk 18:11, 1 September 2006 (UTC)

I argue partly the same, and partly the opposite: for an article that is separate from the SRT article, it makes sense that it mostly not deal with the same subject matter. But I do agree with you and EMS that not much should be said about time dilation in this article, as it's only related to the subject matter in the context of SRT, and that is also explainend in the SRT articel - as indeed it should be.

IMO you make a good point about how to make this article more transparent: a clear distinction should be made between the pre-SRT relativity of simultaneity (only light pulses), and the more fundamental RoS (incl. time dilation). That may make the reading much more clear.

Note: I'm not convinced that the technology limit is of a fundamental difference: at that time clocks were just not accurate enough, and it can't be excluded that someday some kind of quantum technology could "end" relativity of simultaneity after all. There are always new technologies that open up new ways to challenge an hypothesis. Harald88 23:26, 1 September 2006 (UTC)

PS: I still favour including your illustration above, under "Poincaré synchronisation procedure" Harald88 00:24, 2 September 2006 (UTC)

About thought experiments and thought demonstrations.

To my knowledge, the routine way of conducting thought experiments and thought demonstrations is to assume infinite technological sophistication. A thought experiment may go as follows: if a spacecraft not only uses fuel stored in tanks, but uses every part of itself as fuel, down to the last atoms, and the conversion of mass to potential energy to kinetic energy is 100% efficient, what is the maximum velocity the spacecraft can achieve? The purpose of a thought experiment is to investigate exclusively what limits are set by a given set of physical postulates. By contrast: it is inherently unfertile to present the following thought experiment: "What is theoretically possible if in the future a theory is discovered that we currently have no knowledge of?" --Cleonis | Talk 08:00, 2 September 2006 (UTC)

--Answers-- 1: v=csquared/m and m=0 2: A mathematical extrapolation of the theory to some kind of absurdity.WFPMWFPM (talk) 15:54, 2 October 2008 (UTC)

I agree, which is why I referred to QM: sofar I still don't fully grasp the pro-and-contra arguments, but it is alleged by some that it imposes an "absolute" frame that may be detectable with sufficient sophistication. If that would happen, it would invalidate the PoR as valid for all laws of physics (Poincare 1904) and limit relativity to non-quantum phenomena - which was the scope as given by Einstein in 1905. If some editor has enough knowledge about QM experiments of simultaneity, he/she could make a very interesting contribution to this article. Harald88 08:59, 2 September 2006 (UTC)

I feel I was over-confident when I named those diagrams 'Poincaré synchronisation procedure diagrams'. As one would expect, Poincaré's thoughts continued to evolve. I now doubt whether those diagrams are a fair representation of Poincaré's thoughts (at any stage of his development.) The diagrams illustrate the concept of using pulses of light for synchronisation. If those diagrams are used, I feel they should have a more neutral name. --Cleonis | Talk 08:00, 2 September 2006 (UTC)

The muon synchronisation procedure

I have written that I regard the trains-and-flashes-of-light procedure as totally unsuitable for the purpose of education. I'd like to describe a procedure that in my opinion does qualify.

Instead of pulses of light, pulses of muons are employed. The supply of muons is accelerated to very close to the speed of light and then beamed to the receiver. At the emitter the luminosity of the muon beam is measured. The receiver has a detector that measures the energy of the received muons, and the luminosity of the beam is measured. The loss of muons to muon decay during transit is a measure for the transit time of the muons. The measurement of the energy gives the velocity of the muons, and together with the transit time that gives the distance between the emitter and the receiver. Applying the muon synchronisation procedure establishes a plane of simultaneity.

The advantage of the muon synchronisation procedure is that it extracts all information that can be extracted. By contrast, if the synchronisation procedure uses only flashes of light then it is unknown how long the flashes of light have been in transit. To only use flashes of light means that you obtain only a fraction of the information that can be extracted. --Cleonis | Talk 19:47, 1 September 2006 (UTC)

That sounds to me another good example (next to the historical train example), except for two points:

* If presented like that, it doesn't clearly demonstrate the relativity of simultaneity (in contrast to Einstein's train); it could be misunderstood to suggest absolute simultaneity instead!

* If no such example exists in literature, it may be considered WP:OR.

Indeed, the verifiability for other editors and readers will be hampered if we can't cite it.

BTW, it's unclear to me why you claim that with the train example "it is unknown how long the light has been in transit" - for in fact, we do know that as well as with the muons: we know the (average) speed of light as well as the distances. Harald88 23:38, 1 September 2006 (UTC)

First: the muon synchronisation procedure extracts more ínformation than the pulses-of-light synchonisation procedure. The muon synchronisation procedure in and of itself is inconclusive, it does not demonstrate anything. This shows that the trains-and-flashes-of-light example (which extracts less of the available information) doesn't demonstrate anything in and of itself.
In order to obtain a conclusion, the procedure must be contemplated in conjunction with an assumption about the nature of space an time.
If you contemplate the muon synchronisation procedure in conjunction with assuming newtonian space and time, then you do not arrive at a concept of relativity of simultaneity.
If you contemplate the muon synchronisation procedure in conjunction with assuming that the Lorentz transformations are fundamental, then you arrive at the threesome: time dilation, length contraction and relativity of simultaneity. --Cleonis | Talk 08:19, 2 September 2006 (UTC)

In physics, the decay of muons is a standard example, presented very frequently, and synchronisation procedure is a very frequently used example. I regard it as a small step to combine the two; very straightforward synthesis of established knowledge. I have not encountered an actual occurence of muon synchronisation procedure in the books or articles that I have read. Verifiability is not an issue: muon decay is rock-bottom physics knowledge. --Cleonis | Talk 08:19, 2 September 2006 (UTC)

In the case of the trains-and-flashes-of-light example it is never specified whether the experimentors know the length of the train. In stating the example, the number of carriages is never specified; no need to. If you specifically state that the experimentors have no knowledge of the length of the train, the content of the thought demonstation is not affected. The trains-and-flashes-of-light example is a special case of a more general class. The general class is that you want to synchronize a number of co-moving clocks without prior knowledge of the spatial distance between the clocks. --Cleonis | Talk 08:19, 2 September 2006 (UTC)

The decay of a muon is a random event. You can do synchronization with the local timimg of single photons, but are left with nothing but uncertainty with single muons. Remember that in SR light moves at c in all inertial reference frames. So by dividing the out-and-back time of a photon by 2c, you can find out the distance to the reflection point. Admitedly this procedure dovetails with SR itself, but if the procedure is producing inconsistent results between observers, then SR is challenged. So far, SR has not been challenged in this way. --EMS | Talk 03:11, 4 September 2006 (UTC)

EMS, I it unclear why you suddenly you begin talking about individual muons. The muon synchronisation procedure that I described works with a supply of muons, the procedure works with beams of muons. I can arbitrarily put down a number: billions of muons in each pulse. The decay rate of muons is known with high precision. If half the muons make it to the other side, then the experimentor knows the how long the muon beam was in transit. Why on earth did you start thinking about individual muons? --Cleonis | Talk 16:30, 4 September 2006 (UTC)

Education??

I hadn't realised, at least thought of, the importance of education as a function of wikipedia. To me the best eduaction challenges and looks at things from different angles which is not what wikipedia should do is it? Since that might be original research. Personally, I like all the false steps, groping in the dark approaches, that are made before a subject is finally put in its most coherent form. I like the history. If we delete all the history could it go anywhere? E4mmacro 23:59, 1 September 2006 (UTC)

Encyclopedia are known to be important educational tools; Wikipedia can become one by excellence because of its NPOV policy, which implies expressing a plurality of opinions instead of imposing a single opinion. And history of science is for sure an important part of encyclopedia content. Harald88 00:15, 2 September 2006 (UTC)

I don't see your text as being informative about the relativity of simultaneity, which is its subject. Instead I see a historical diatribe which neglects discussion about with the relativity of simultaneity is, and text aiding in its comprehension. Even worse, it fails to give Einstein his due as the person who connected the relativity of simultaneity with the second postulate of SR. (I do admit that Lamor's early discoveries of relativistic phenomena are improperly neglected, but at the same time strongly counsel you to remember the it is Einstein who put all this stuff onto a sound theoretical foundation, and therefore deserves the primary credit for this phenomenon.) --EMS | Talk 03:17, 4 September 2006 (UTC)

That would be just as bad, as that would fail to give Poincare his due credit. However, I propose to spin most about Larmor off to an article about Larmor. Harald88 07:10, 4 September 2006 (UTC)

restructuring the article

Based on the above discussions I propose to restructure the article as follows:

1. Dawning of the relativity of simultaneity (incl. the above light diagrams)

2. Relativity of simultaneity and special relativity (incl. both the train and the muon examples?)

3. Relativity of simultaneity and quantum mechanics (with mention of a recent experiment in Geneva)

The above structure is at the same time topical as well as, roughly, chronological. Topic no.3 is a challenge as few people seem understand it, but some past editors of QM articles may be asked to get involved. Harald88 09:22, 2 September 2006 (UTC) PS: recently I bought "Quantum Non-locality and Relativity"by Tim Maudlin which should be an interesting read. It happens to focus on relativity of simultaneity and, possibly, its incompatibility with QM. Harald88 17:25, 2 September 2006 (UTC)

Dawning of the concept of relativity of simultaneity.

A 'what if' scenario:
What if somewhere around 1900 there was group of bright physics students, not yet familiar with the Maxwell equations or any other theories of electromagnetism. All they know is newtonian mechanics, their expectation pattern is newtonian space and time.

This group of students is asked to contemplate the Einstein synchronisation procedure. They point out that the procedure is insufficient for the purpose of singling out the true plane of simultaneity, the procedure fails to single out the galilean invariant plane of simultaneity that would be found with a more exacting procedure.
When the group of student physically perform the synchronisation procedure, the instrument readings do not compel them to reconsider their newtonian expectations.

(Next step in the 'what if' scenario: introduction of technology that was actually only invented decades after 1900.)
The group of students is asked to contemplate the muon synchronisation procedure. Given their expectation that space and time are newtonian, they will point out that the muon synchronisation procedure's design meets the requirement of singling out the galilean invariant plane of simultaneity that according to newtonian assumptions must exist.
When the group of students physically perform the muon synchronisation procedure, they find to their surprise that the instrument readings cannot be accounted for terms of any newtonian theory. They find that the muon procedure performed onboard a moving train singles out a particular plane of simultaneity, but not the same one as is singled out on the platform. They find that the plane of simultaneity that is singled out is independent of the velocity of the muons in each measurement run, suggesting they have hit something fundamental.
In due course, it will dawn on them that there is in fact another expectation pattern for motion in space and time that enables them to account for the results of the muon synchronisation procedure: the set of transformations that we know as the Lorentz transformations.

Actual history
Poincaré published a incisive critique of assumptions about space and time, the essay 'la mesure du temps', in which he discussed the inadequacy of using light signals to establish an unequivocal plane of simultaneity. He then proceeded to discuss the obvious remedy: usage of portable clocks. Poincaré argued that it cannot a priori be excluded that clocks at different locations count time at a different rate, or that clocks that have a velocity with respect to each other count time at different rates. If that should be the case, argued Poincaré, then science will be compelled to abandon an expectation that measurement can single out an underlying true plane of simultaneity. The true plane of simultaneity, argued Poincaré, may be inaccessible to inspection. (I'm cutting corners everywhere here, but you get my drift, I think.)

It is certainly possible that in writing the essay 'la mesure du temps', Poincaré had Lorentz's thought provoking theory of electron motion in mind. A cross-fertilisation of ideas seems quite a plausible scenario here. The really bold step is often referred to as the 'molecular force hypothesis'. As I understand it, the 'molecular force hypothesis' in its widest sense is equivalent to postulating that the Lorentz transformations are fundamental.

It is often claimed that 'the Einstein synchronisation procedure is thought provoking'. But in and of itself the Einstein synchronisation procedure is totally innocuous. Provocativeness is conferred to the Einstein synchronisation procedure by contemplating it in conjunction with the brazen postulate that the Lorentz transformations are fundamental. --Cleonis | Talk 15:14, 3 September 2006 (UTC)

The cross-fertilization is not a possibility but a certainty. And something like your summary would IMO form a good ending of topic 1. Note that if Poincare wasn't lying, the actual history doesn't need a "what if" and started well before Poincare. Topic 1 typically begins with for example Poincare's mention of the 19th century "admission" of astronomers that the speed of light is isotropically c relative to themselves. Your diagrams above fit neatly as illustration of that conventional assumption. Harald88 21:42, 3 September 2006 (UTC)

I'm reading 'La mesure du temps' in french now, (which I can just manage). Poincaré points out that in the community of astronomers there is the following tacit assumption: that the speed of light is the same in all directions. I take that to mean the following assumption: the speed of light with respect to newtonian absolute space is the same in all directions. That, argues Poincaré, is a tacit assumption that Roemer made in inferring the velocity of light, a tacit assumption that every astronomer agreed with.

It was only after the shift to the paradigm of special relativity that the pattern of expectation became that the velocity of light is c with respect to any and every inertially moving observational platform. That is: only after the introduction of special relativity was there among astronomers a shift to a notion that the speed of light is isotropically c with respect to themselves. --Cleonis | Talk 23:45, 3 September 2006 (UTC)

I think that you are confused: the astronomers had to pretend that the one way speed of light was c relative to the earth in order to be able to establish anything. I started to edit it, but instead I will first read tht part again, as I have forgotten the details. Harald88 07:08, 6 September 2006 (UTC)

Interesting point: if the solar system as a whole would have a significant velocity with respect to the luminiferous ether, then the velocity of light with respect to the solar system would be noticably non-isotropic. Astronomers inferred that the center of mass of the solar system has insignificant velocity with respect to the luminiferous ether. However, there was no challenge to the expectation that sufficiently sensitive equipment would detect a non-isotropy corresponding to the velocity of the Earth with respect to the center of mass of the Solar system. --Cleonis | Talk 08:36, 6 September 2006 (UTC)

It just occurred to me: Roemers observations of Jovian satellites disappearing behind Jupiter and reappearing again, can be regarded as messages from out there of 'this is happening now'. Indeed a tacit assumption of Roemer was that with respect to the solar system light has the same velocity in all directions. --Cleonis | Talk 08:56, 6 September 2006 (UTC)

Indeed, that is how I understood the cited paragraph by Poincare. I'll reinsert it after checking the context. It was only by convention (a "postulate") that the speed of the solar system was taken to be zero; Michelson explained that it was estimated that the true speed could be hundreds of km/s. Harald88 19:50, 6 September 2006 (UTC)

PS I now verified the context: he stated that astronomers demand that "time" must be defined in such a way that the equations of mechanics are as simple as possible." And indeed next the subject matter is that of simultaneity (par.VI), and he explains that for determining time at a distant place we need to take our refuge in convention; he repeats the problem at the end of par. X (if we hear one thunderstroke before another of lightnings at different locations, how can we know which one really happened first?). And he introduces the measurement of the speed of light in the context of the measurement of time as follows, here is the whole passage in English:

"When an astronomer tels me that a stellar phenomenon that his telescope reveals to him at this moment, nevertheless occurred fifty years ago, I try to find out what he tries to say and for this reason, I will ask him initially how he knows it, i.e. how he measured the speed of the light. He started by admitting that the light has a constant speed, and in particular that its speed is the same in all directions. That is a postulate without which no measurement of this speed could be attempted."

Interestingly, furtheron he also discusses the use of clock transport, the constancy and reliability of which he had put in doubt at the start of the article as well as the telegraphline synchronization procedure. Harald88 21:13, 6 September 2006 (UTC)

The clock synchronisation diagrams

Diagram 1
Synchronisation procedure with lightpulses

Diagram 2
Synchronisation procedure with lightpulses. Newtonian space and time is assumed. The velocity of the clocks with respect to the ether is 1/5 the speed of licht

I uploaded new versions of the diagrams, this time without reference to Poincaré. I now believe that my earlier attribution was not a fair representation of Poincaré's views.

About the procedure itself.
It matters whether the operators of the procedure have prior knowledge of the spatial distance between the clocks or not.

Suppose the distance between the clocks has been measured with perfect measuring rods. Let this distance be 5 units of spatial distance. Assume that all motion takes place with respect to a luminiferous ether. Let the clocks of diagram 2 have a velocity 1/5 the speed of light with respect to the ether. Then a two-way transit of light signals will take slightly more than 10 units of time (whereas the two-way transit as depicted in diagram 1 would take exactly 10 units of time.) (At 1/5 the speed of light with respect to the ether, the expected two-way transit time of light signals would be 10.05 units of time)

So, theoretically, if you can beforehand perfectly measure the spatial distance between the clocks, you should be able to calculate your velocity with respect to the luminiferous ether.
Of course, in practice there is no prior knowledge of the spatial distance between the clocks, in practice the two-way transit time of the light signals is used to obtain a value for the distance between the clocks.

Earlier I wrote that the Einstein synchronisation procedure does not in itself logically imply special relativity. There is a demand that narrows things down sufficiently: the additional demand: that no inconsistency arises. The Einstein synchronisation procedure, combined with the demand that no inconsistency should arise, is sufficient to imply special relativity. --Cleonis | Talk 17:27, 6 September 2006 (UTC)

Indeed, you are about to reinvent the M-M experiment: the speed of light is determined by two-way measurements in all directions, and we can only compare such. Harald88 21:31, 6 September 2006 (UTC)

Translation of the Poincaré quote

In the article, at the end of the section Development of "local time" there is a translated quote from Poincaré, with (a translation of ) the following sentence:
"[Le astronome] a commencé par admettre que la lumière a une vitesse constante, et en particulier que sa vitesse est la même dans toutes les directions."

There are a number of ways this sentence can be translated:
[The astronomer] has begun by allowing that the velocity of light is constant, [The astronomer] has begun by approving that the velocity of light is constant. [The astronomer] has begun by granting that the velocity of light is constant. Currently, the translation goes: "[An astronomer begins] with admitting that light has a constant speed, and particularly that its speed is the same in all directions.

I think that the verb 'to admit' is generally felt as a synonym to the verb 'to confess'. The french verb 'admettre' here is more in the spirit of an expression like 'granting admission'; some people may be wrongfooted by the 'admitting' in the current translation.
I propose the following translation:
[The astronomer] begins with entering the supposition that light has a constant speed, and particularly that its speed is the same in all directions.
--Cleonis | Talk 22:03, 6 September 2006 (UTC)

That's a bit cumbersome; what of "grants" or "supposes"?Harald88 22:06, 6 September 2006 (UTC)

"begins by assuming .." or "assumes" E4mmacro 22:16, 6 September 2006 (UTC)

Different things?

Quote form article: "For clocks to be completely in synchrony they must count units of time at the same rate, and they must simultaneously indicate the same time." I see there is no way you fellows will not make this page another exposition of special relativity. To me there is nothing about simultaneity that says clocks have to be sychronous or run at the same rate, or indeed that clocks need to be mentioned. But I've said it before, so I guess I'll leave it at that. E4mmacro 22:12, 6 September 2006 (UTC)

As a resort, you may consider making an article with the title, 'pre-relativistic relativity of simultaneity'.

I would agree that that was a good name for the original article I wrote. E4mmacro 07:21, 7 September 2006 (UTC)

Why should the scientific community usurp the expression 'relativity of simultaneity' as exclusively relativistic? If the expression 'relativity of simultaneity' is to be understood exclusively as 'relativistic relativity of simultaneity', the pre-relativistic musings are deprived of even a name to be referred to.

It isn't and it won't be, as we already include pre-relativistic (pragmatic) relativity of simultaneity; furthermore, quantummechanical relativity of simultaneity also needs to be discussed in an article with this title. Harald88 22:09, 7 September 2006 (UTC)

The difference in outlook is not just about how expressions are used, of course. As I understand E4mmacro, he feels that there is a continuity from the pre-relativistic musings to the relativistic concept of relativity of simulteneity. By contrast, I refer to the pre-relativistic considerations as mere musings, with no content worth of preserving for posterity. If I hazard a guess, it seems that for E4mmacro the very distinction between 'pre-relativistic' and 'relativistic' is far from a helpful one. --Cleonis | Talk 23:56, 6 September 2006 (UTC)

Contradictory paragraph?

The following paragraph was deleted (by cleonis) since it was said to contradict a previous paragraph.

"In 1900 Poincaré was more explicit about the "new rule". He remarked that Lorentz's "wonderful invention", local time, arose when clocks in a moving reference frame are synchronized by exchanging signals which are assumed to travel with the same speed $c\$ in both directions, a procedure now familar from the special theory of relativity and which he first developed. It is worth examining Poincaré's reasoning to see that relativity of simultaneity can be considered without accounting for time dilation and length contraction."

Can we have some more information about what is wrong with this paragraph? It looks basically correct to me. What does it contradict? I would delete the phrase "and which he first developed" as unnecessary, but not the whole paragraph. E4mmacro 07:19, 7 September 2006 (UTC)

The concept of "local time" arises in the context of Lorentz' theorem of corresponding states. Initially, Lorentz regarded this "local time" as purely a mathematical device to speed up calculations, and no physical meaning should be attributed to it. Lorentz has indicated that up to 1905 he expected that an actual clock would not show this "local time", but the true newtonian time.

You deleted a pargraph dealing with Lorentz's local time of 1895 (actually Larmor's slightly different version). That local time does not incorporate time dilation. If all clocks ran at the same rate, independent of motion, a clock set by Poincare's procedure would in fact indicate Lorentz's local time of 1895 (actually it indicates Larmor's 1897 local time which is close enough to the same thing). So Lorentz has no problem believing "Newtionian" clocks would indicate his 1895 local time when sychronised by the procedure described by Poincare. It is a convention (as Poincare indicates) and has nothing to do with time dilation. I can't see that Lorentz's disbelief in time dilation, which Lorentx put into his equatiopns later, is a reason for deleting the paragraph you did. E4mmacro 02:29, 8 September 2006 (UTC)

It was through the influence of Einstein's 1905 article that Lorentz changed his mind about that. Poincaré's shift to taking serious the possibility that the "local time" of the theorem of corresponding states is physically meaningful, was several years earlier than Lorentz' shift.

The crucial shift is the step of attributing physical meaning to the "local time". Attributing physical meaning to the "local time" is one and the same thing as taking time dilation into account.

If one attributes physical meaning to Larmor's 1897 local time or Poincare's synchronisation procedure one is lead to contradictions - there is something wrong. One can remove the contradictions by introducing length contraction and time dilation. I suppose that is what you mean by saying "one and the same thing". I would say `Attributing physical meaning to the "local time" leads inevitably to the idea of time dilation'. E4mmacro 02:44, 8 September 2006 (UTC)

I think that you are stating why these concepts had no currency before 1905 (with the notable exception of the Lorentz contraction), when Einstein showed that all three sprang from a single source. It does you no good to rationalize about what these people should have meant. Instead this focus on the pre-history of the relativity of simultanety is a complete distraction and should be removed from this article. --EMS | Talk 03:07, 8 September 2006 (UTC)

I have to say that the remark "these concepts had no currency before 1905" is a historical judgement which I don't agree with. But to each his own reading of history. But it should not be a relativity priority dispute article and I never meant it to be. See below for why I included the history. E4mmacro 05:56, 8 September 2006 (UTC)

Can I rephrase you statement as follows?

"Lorentz' "local time" arises when clocks with a velocity with respect to the luminiferous ether are synchronized by exchanging signals which are assumed to propagate through the luminiferous ether with the same speed $c$ in both directions."

What is undisputed is that the result of that synchronisation procedure coincides with Lorentz's "local time".

If one begins with attributing physical meaning to Lorentz' "local time" then this attribution of physical meaning is subsequently conferred to the result of the synchronisation procedure.

On the other hand, if one begins with assuming that Lorentz' "local time" is just a mathematical tool for speeding up calculations, there is no reason to attribute physical meaning to the result of the synchronisation procedure.

Overall conclusion: relativity of simultaneity (in the sense of a concept with physical meaning) can only be considered in conjunction with accounting for time dilation and length contraction. It is the last sentence of the paragraph "relativity of simultaneity can be considered without accounting for time dilation and length contraction." that contradicts earlier statements.

I think I have pleaded this case to excess now. For the time being, I rest my case. --Cleonis | Talk 10:35, 7 September 2006 (UTC)

Ok I have restored the sentence indicating what Poincare actually did. i.e. he was silly enough to consider ROS without length contraction and time dilation. I think that is an interesting fact. E4mmacro 02:27, 8 September 2006 (UTC)

It's not clear what in the above explanation is taken as reference for "the same speed in both directions" - presumably the moving frame for if it were the ether, no local time could be construed. There can be little doubt that Poincare's use of "local time" was physical; it was a source of misunderstanding between the two men. In any case, the above clarification may be useful, but I don't think that it is acceptable to replace a strongly sourced account by an account that appears to be weakly sourced, and that could be WP:OR. Instead, errors in the sourced account should be tackled (either errors in the source, as supported by other sources, or errors in the account, as supported by the source). Probably what is disputed is the last sentence: "It is worth examining Poincaré's reasoning to see that relativity of simultaneity can be considered without accounting for time dilation and length contraction." Harald88 22:06, 7 September 2006 (UTC)

Totally disputed

I have restored this banner becuase this article has deteriorated since I last saw it. The lead of a week ago was much better than the current one. At least I could make a reasonable stub out it alone! (I remain tempted to revert it again to that state, BTW.) The added text, while well intentioned, is just so much hand-waving to me. I don't see a simple, coherent statement here of what the relativity of simultaneity is anywhere in this article (although there is a good attempt at one mid-way through the article, but that is not where it belongs).

This article is a piece of trash that needs to be rewritten, not patched up. In the meantime, it should either be stubified or deleted. --EMS | Talk 02:29, 8 September 2006 (UTC)

That's not the way it is done in Wikipedia - I could for the same reason stub or delete the GRT article, or Rod Ball could stub or delete the Ehrenfest paradox article.

Instead, one can contest and/or mofify the passages that one thinks should be removed or rewritten, with an explanation why. For example, the new intro is IMO faulty - but I won't stub the article because of it! Harald88 12:46, 9 September 2006 (UTC)

I don't think this article is entirely a piece of trash, but I definitely agree that it needs lots of rewriting. I've added a lead section as a first step in this process. Historical narrative is not the desired form of an encyclopedia entry. Some history may be appropriate, but it needs to be pared down a lot, and should definitely come after the modern understanding of this idea is explained fully. There should be one section at most on the history. E4mmacro wrote a nice journal article on Larmor's place in history, but it doesn't need to be reprinted here in full. MOBle 04:45, 8 September 2006 (UTC)

That doesn't appear very useful to me: it would risk to become just a possibly superfluous appendix to the Special relativity article.

An alternative would be to rename it "History of Relativity of simultaneity".

But as far as Larmor is concerned, the pertinence of that section in an article on simultaneity is unclear to me as well. Harald88 12:46, 9 September 2006 (UTC)

I have to agree the new material seems like "so much hand-waving", and I liked my original first paragraph, now gone. If others do not like the history as much as I do, that's fine. I repeated the history here for a reason - to try to make a point that discussion of RoS from the modern point of view of special relativity tends to make the subject more confusing than it need be. The history of the re-writes of this page re-inforces that opinion for me, but I accept that others see it differently. E4mmacro 05:45, 8 September 2006 (UTC)

In spite of my objections to the article as a whole in its current form, I do agree that the prior lead paragraph was a good one. After all, I retained it in full when I attempted to stubify this article. Basically, the example given communicated the gist of what the relativity of simultaneity is all about. At the moment, the lead is improving again, but it still has a ways to go before it returns to the standard set by that earlier paragraph.

This is not an easy topic to write about, and I apologize for having enough time to complain about this article but not enough time to work on it. I very much point out to people that the mechanics of this effect is what needs to be discussed here. I have placed links to it in comments and other articles anticipating a discussion of the mechanics of the relativity of simultaneity instead of its history. I also repeat that the history part of this article is entirely misleading in that it leaves out Einstein and his classic train example (found in On the Electrodynamics of moving bodies, the 1905 article which introduced special relativity). BTW - If the current history text exists in a journal aarticle, then the article should be referenced instead of the text being placed here. I also suspect a copyright violation exists, but that may depend somewhat on the rights of E4mmacro under the copyright transfer agreement he made to have the journal article published. --EMS | Talk 13:50, 8 September 2006 (UTC)

I didn't mean to imply that text of the article had been cut and pasted to this page. I just meant that the history text in this page doesn't need to be as extensive as E4mmacro's article. My intention is to consolidate the history into one section placed at the bottom, and to reproduce the train example as the main substance of the article. MOBle 17:48, 8 September 2006 (UTC)

Cleon was opposed to the train example, but from his last comments I guess that now he would go along with that. For now I'm neutral about it; probably it depends on how it is put. Harald88 12:46, 9 September 2006 (UTC)

It sounds like things are beginning to converge towards a place that I like. Please go ahead and edit this article to suit yourself, and let's see how it works out. I will be happy to remove the {{totallyDisputed}} tag as soon as this article is in somewhat decent shape. --EMS | Talk 20:31, 8 September 2006 (UTC)

Change of tags

I have replaced the {{totally disputed}} tag with a {{POV-section}} tag for the older and contested material. I feel that the article is finally stating what the relativity of simultaneity is in at least a reasonable manner. The "Clock synchonization" section I question however. It is still just hand-waving, and makes the same point as the prior text. This should be replaced by a discussion of the Einstein synchronization procedure. Curiously, some illustations that are idea for doing this are now in the "History of the idea" section. In any case, the article is getting better, and I am happy to rearrange the tags to reflect this. --EMS | Talk 03:31, 11 September 2006 (UTC)

EMS, you said somewhere the history section appeared to be written to deny Einstein the credit he deserves. From the above comment I wonder if you think the Einstein synchronisation procedure is different from the Poincare synchronisation procedure. If so that might explain your comment. As far as I can see Poincare's procedure is the same as Einstein's and by the normal rules of priority would be better called the Poincare procedure. If you disagree your reasons might throw some light on what exactly the problem is. I can see how history might be irrelevant, but am very surprised to see you tag the history as non-neutral - I seriously don't get it. I don't see any opinions expressed in that section. E4mmacro 07:53, 11 September 2006 (UTC)

On another reading I do see an opinion at the very end of the history section, where it claims it is worth repeating the first prediction of time dilation in 1897. If one thinks time dilation is a seperate issue from RoS, there could be no reason to mention time dilation, other than to show Einstein wasn't the first (or even the second) to think of it or anything like it. That might be construed as anti-Einstein, I guess. On the other hand, Cleonis spent a long time arguing that RoS could not be treated without mentioning time dilation and length contraction. E4mmacro 08:07, 11 September 2006 (UTC)

Contested statements

Here is one statement that appears to be misinformed; but if I overlooked some passage, please provide a quotation.

"The special theory of relativity, introduced in 1905 by Albert Einstein, starts with the assumption that the Lorentz transformations do not only apply for electromagnetic phenomena, but for all physics phenomena" - which would be correct for Poincare's version of SRT, but Poincare isn't mentioned! In contrast, Einstein's 1905 version of the PoR has that:

"the same laws of electrodynamics and optics will be valid for all frames of reference for which the equations of mechanics hold good".

This difference is relevant for the discussion about its compatibility with QM.

Harald88 12:26, 11 September 2006 (UTC)

The statement, while poorly worded, is correct. Look again at the Einstein quote: In his 1905 article, Einstein was seeking "equations of mechanics" in which the "same laws of electrodynamics and optics" would "hold good". The equations of mechanics that he found are those which are now called the Lorentz transformations. So it very much was Einstein's conclusion that the Lorentz transformations apply to all physical phenomena, not just "electrodynamics and optics".

Einstein's 1905 article "On the Electrodynamics of Moving Bodies" is sufficient to document this statement. However, I consider that whole section to be poorly constructed, and the last paragraph's discussion of time dilation to be somewhat confused. Indeed, I would like to see the entire "History of the idea" section just plain removed. --EMS | Talk 14:19, 11 September 2006 (UTC)

The statement as worded implies that Einstein knew the Lorentz transformations before he started, and quoted them as his starting point, rather than derived them. It is of course possible that he knew the Lorentz transformations, since they been published in 1897, 1899 and 1904 so it is important not to suggest that Einstein merely quoted them. E4mmacro 21:05, 11 September 2006 (UTC)

I have deleted the last paragraph which showed Lorentz's understanding in 1899 that time dilation is required if a modified MM experiment was to return a null result, as a side issue. Is there any discussion anywhere on all the wiki relativity pages about light propagating through a medium? E4mmacro 21:19, 11 September 2006 (UTC)

I think a lot of the history section can be deleted. In particular, it seems like the whole Special Rel subsection can go. I haven't touched much of the synchronization or history sections, but I think it would probably be best to shorten them. As they are, I don't think many people will read much of them, or be able to glean much useful information from them.

On a slightly different note, I don't particularly think the quantum section in the leader should be in this article. Maybe it could be moved to Causality (physics). MOBle 22:47, 11 September 2006 (UTC)

As the relativity subsection is now mostly a repetition of the better expressed part higher up in the article, I agree with simply removing it - thus eliminating that point of discussion altogether.

Since this article is about relativity of simultaneity and not about causality, the implications of QM for it are certainly pertinent. If causality should be mentioned in this article is a different question. IMO there is no need for it, it would merely distract, and you point out that there is an article on it. Harald88 09:31, 12 September 2006 (UTC)

Removal of material

Once again I am trying to cut this article down to size. The topic of the article is the relativity of simultaneity, NOT the history of the Lorentz transformations. I strongly recommend that the bulk of the excised material be moved to such an article. I have also removed the section on clock synchronization. It just does not work. It is poorly worded and fails to communicate anything new.

What I have left makes for a good article on this difficult topic. There are times when less is better, and this is one of those times. --EMS | Talk 02:02, 12 September 2006 (UTC)

Before any knee-jerk reactions to EMS's edit happen, I'd like to encourage people to think about the situation. Personally, I had thought that those sections could be edited down and left in, until I saw the article without them. I think it's a much better article now. The content that was removed could be used as inspiration for making a clock synchronization (physics) page (disambiguated from this page), and making/improving a page on the history of the Lorentz contraction. The article as it is currently is entirely about this topic, rather than having extra information on closely related topics. Since we are able to make new pages, the quality of Wikipedia can be improved by creating those pages and making sure that individual articles don't veer off into other topics. Less is, indeed, more in this article. MOBle 04:10, 12 September 2006 (UTC)

I think that removing the section on synchronization was a mistake. It was the only thing in the article which described the PHYSICAL facts which make it impossible to use absolute simultaneity. The way the article is now, it just sounds like Einstein (and others) arbitrarily decided to discard absolute simultaneity so that they could make the two postulates true. We need something which shows why it is IMPOSSIBLE to stick to absolute simultaneity. As far as I am concerned that should be the main thrust of this article. JRSpriggs 06:04, 12 September 2006 (UTC)

I agree on that point, now the article is deteriorating. The history of synchronization issues is more than just a history, it served as a logical line of thought towards the LT. That was completely messed up and is now even gone. I'll do a repair attempt later. Harald88 09:35, 12 September 2006 (UTC)

I make no claim to being an historian, and I don't know what was going through Einstein's mind when he discarded absolute simultaneity. However, there are two very important physical facts:

• the equivalence of all inertial observers;
• the constancy of the speed of light.

Einstein put these in his very first article on special relativity. Without these, you have no Lorentz contraction, no time dilation, and no relativity of simultaneity -- no special relativity. I don't see any other argument that was being made or physical insight that was being brought out in that section, and these two physical facts are described in the article.

The relativity of simultaneity is not a separate postulate and can be derived from the two postulates. I wouldn't be surprised if Einstein did in fact arbitrarily discard absolute simultaneity to "make the postulates true" (to avoid contradicting the postulates he had just introduced). MOBle 07:05, 12 September 2006 (UTC)

Your incorrect statement that without the two postulates "you have no relativity of simultaneity" demonstrates the need to reinsert some of the deleted facts. Harald88 22:44, 12 September 2006 (UTC)

How is that statement incorrect? Historically, people came up with relativity of simultaneity without the postulates, but their understanding of the physics was flawed. I think it will only be confusing to introduce flawed and antiquated notions in an attempt to explain something that can easily be understood without them. MOBle 23:31, 12 September 2006 (UTC)

Just curious about this quote from MOBle above. "Without [A: equivalence of all inertial observers and B: constancy of the speed of light] , you have no (a) Lorentz contraction, (b) no time dilation, and (c) no relativity of simultaneity -- ". Since Lorentz and Larmor had (a) and (b) and a form of (c) does this mean they must have had A and B? E4mmacro 08:04, 12 September 2006 (UTC)

The discussion is about physics as it is currently understood. Lorentz and Larmor had (a), (b), and some form of (c), but they misunderstood those things. In modern terms, all you need are A and B to derive (a), (b), and (c). There is no additional physics needed, or even desired. MOBle 18:48, 12 September 2006 (UTC)

To MOBle: You are ignoring my point. Why is it physically impossible to define time in a way that maintains absolute simultaneity? The article does not currently answer that question. But it is THE VITAL question. Certainly people would use absolute simultaneity, if there were anyway that they could. JRSpriggs 08:43, 12 September 2006 (UTC)

There is no problem in defining time in a way that maintains - conventionally! - "absolute" simultaneity. A first paragmatic approach already exists in the form of "universal time" which is probably OK for a lifetime: if I'm not mistaken, its time rate would only change (relative to a clock in the centre of the universe) due to a change of speed or position of the solar system. Harald88 22:18, 12 September 2006 (UTC)

...unless, of course, you want to take to heart the postulate that all inertial observers are equivalent. By choosing universal time, you have chosen a reference frame (indeed, a non-inertial reference frame). By choosing a clock in the center of the universe, you have chosen a reference frame. Einstein says that such a choice has no absolute meaning. MOBle 23:31, 12 September 2006 (UTC)

This encyclopdia does not prefer Einstein's opinions; we could put "according to Einstein". However, I doubt that he said that, since he said effectively the opposite: in physics only statements should be made that can be empirically verified. A statement that "such a choice has no absolute meaning" cannot be empirically verified. Harald88 12:09, 1 October 2006 (UTC)

My reason for the removal of the section on clock synchronization is that it failed to answer the questions of what synchonization is and how it shows that absolute simultaneity is impossible in relativity. As written, that section metioned sea travel and blabbered incoherently about time dilation and length contraction, and then at the end suddenly declares a lack of absolute simultaneity. The goal is laudable, but the text failed to establish a coherent logical path leading to the stated conclusion. My suggestion is to describe the Einstein synchronization procedure, and use the first two illustrations from the now-deleted history section to show why observers in relative motion must have incompatible volumes of simultaneity given the postulates of relativity. Most importantly, leave out time dilation and the Lorentz contraction. In this 1905 article, Einstein described the relativity of simultaneity first, before deriving the Lorentz transformation and showing that they call for the other effects. So time dilation and length contraction, while important parts of relativity theory, are not required here. --EMS | Talk 14:39, 12 September 2006 (UTC)

In a for me mysterious way, EMS and E4mmacro appear to think that one can get away with describing relativity of simultaneity without mention of the other effects, but on grounds on which they disagree heavily.

Contrary to their views, to me it makes most sense to introduce it in the pragmatic historical context, as the last version by Cleonis had it; and outside that context the relativity postulates wouldn't have emerged, making it a pertinent development of thought for many readers. Probably for the same reasons some textbooks use a similar introduction.

In the truncated version the PoR just comes falling out of the air.

On top of that, EMS had as criticism that the article wasn't fair to Einstein. Now the opposite happened, with his version it sounds as if Einstein wrote down "his" postulates by devine inspiration. Also the fact that it already existed in practice before SRT is lacking now, leading to the wrong idea. Thus we must reinsert some history with proper acknowledgements - and preferably in the first part to allow readers to follow the logical development that led to it.

Insofar as the old version wasn't sufficiently clear (according to EMS, "that section metioned sea travel and blabbered incoherently about time dilation and length contraction, and then at the end suddenly declares a lack of absolute simultaneity"):

obviously it must be improved if even an editor couldn't follow it.

IOW, the discussion is far from over - I'd say it's just starting.

Harald88 22:38, 12 September 2006 (UTC)

To answer JRSpriggs's question: if you take Einstein's two postulates to be true, then it is physically impossible to define time in a way that maintains absolute simultaneity. I think the article makes that clear. That is the physics behind abandoning absolute simultaneity. That is the beginning and end of the story, unless you want to talk about history -- which I now think would be best placed in a different article. I really can't see what insight was delivered by the clock synchronization section.

I think it's really important to note that we don't need to talk about the Lorentz transformations first. That's math, which will scare some people off before they realize that this is basically a simple geometric idea. We also don't need to mention time dilation or length contraction at all. In short, I agree with EMS's response directly above. (Except that if the first two illustrations from the deleted history section are used, the axis labels need to be changed.) MOBle 18:48, 12 September 2006 (UTC)

Okay. There seem to be a couple fundamental disconnects between the views of the editors. The questions are: should we present the idea pedagogically or historically? Should this article be a sub-topic of special relativity (the currently accepted standard), or should it also present other (rejected) notions?

The modern physicist -- when thinking about relativity -- is not served by recalling the mistaken theories of the ether any more than an astronomer is served by recalling the flat earth idea or the geocentric universe model. If we want to reflect the modern understanding of the relativity of simultaneity, the history should either come last, or in a separate article. Similarly, we might wonder about the value in presenting the historic treatment. I'll draw the analogy of explaining the solar system. Does the geocentric model need to be presented there? I don't think so, as it will just lead to confusion or distraction. If you look at the solar system article, the geocentric model isn't mentioned until the very last section, which is on the history of the idea.

A separate argument also seems to be coming out of the discussion seems to be that this article shouldn't focus on the status of simultaneity of relativity in the context of special relativity, but on any idea which has ever fallen under the rubric of simultaneity of relativity. I really hope that's not what people actually want, but if it is, we should discuss it.

Basically, I really don't understand why the suggested changes would be an improvement to any reader's understanding of the article or its topic. MOBle 23:31, 12 September 2006 (UTC)

Actually, you have the question wrong! The question is: what is pedagogically best.

Moreover, your claim that the subject is by "currently accepted standard" a sub-topic of relativity is simply your claim. I don't know such standards to exist, and insofar they do, it's a matter of taste and POV.

I already explained to deaf man's ears above; of course, the same did some who partly agree with me, as well as some who partly agree with you. Still, one more try:

Contrary to a model of the solar system, the concept of relativity of simultaneity isn't obvious and its early use was not slightly incorrect but (in hindsight) fully correct. From discussions I discovered that many people who can operate the Lorentz transformations still don't understand relativity of simultaneity; personally I only came to a good understanding of how everything connects and why after reading up on the historical development.

IMO for most Wikipedia readers the historical development of the concept will be much easier to understand than a dry mathematical derivation of it. Harald88 23:59, 12 September 2006 (UTC)

Harald88 23:59, 12 September 2006 (UTC)

Harald88 wrote above

The question is: what is pedagogically best. ... IMO for most Wikipedia readers the historical development of the concept will be much easier to understand than a dry mathematical derivation of it.

To be quite blunt about it, I find the current treatment to be much, much preferable padagogically. The first job of the article is to answer the question "what is the relativity of simultaneity"? As the article currently stands, the first section with its diagrams answers that question very well. So the stage is then set for showing how the effect is represented in the Lorentz transformations. OTOH, the historical essay that existed previously took the reader through a whole series of maths without ever explaining what this is about. You are right that we need to avoid using math in the initial explanation, but that is currently being done.

As for your concerns named above, I can go along with a section on the history of the concept, but it must be the last section, not the first one. Also be advised that Einstein was genuinely inspired when the came up with his postulates. Even though they were based on known observations and came to a known place (the Lorentz transformations, although Einstein appears to have been unaware of them and independently re-derived them himself), noone else had come up with such a scheme before.

Your wanting to associate the relativity of simultaneity with the other effects I also find worthwhile, but that has to be done right. The relativity of simultaneity is the thing that keeps SR from being self-contradictory, and than should be noted. However, I would save such a discussion for a new penultimate section (meaning before the history section) of this article. --EMS | Talk 03:25, 13 September 2006 (UTC)

I disagree with your claim about Einstein's inspiration - it has been pointed out that his success was due to his approach of problem solving based on a few accepted assumptions - but I'm again optimistic that we can find a way of presentation of this article that will be acceptable to most editors. Indeed it may not be optimal to explain how the concept arose before describing the concept itself. Harald88 08:36, 13 September 2006 (UTC)

Keep in mind that my biggest objections to the old history section was that it was way too long and failed to tell the reader what the relativity of simultaneity is. I also objected to the lack on any mention of Einstein and how he put this concept on a sound theoretical foundation in 1905. (Please realize that before 1905 spacetime was assumed to be governed by the Galilean transformations and that Lorentz equations were thought to describe effects created by the aether). On the other hand, Lamor does deserve priority for time dilation, and maybe too for the relativity of simultaneity. Certainly a short, concise description of the history of this concept is appropriate for an encyclopedia article. However, I reiterate that it must be at the end, and must leave the details of the hisotory of the Lorentz transformations for another article. (I don't object to the removed material per se, just its presense in this article.)

I am glad to see that you now agree that the concept must be explained before its history is reviewed. I also think that the issue with Einstein is partially one of presentation, but as noted in the previous paragraph the publication of the 1905 SR article was a seminal event for this concept, and that cannot be ignored. --EMS | Talk 15:18, 13 September 2006 (UTC)