OWLS is not equal to c
in [Relativity]
|
Prev: What keeps electrons spinning around their nucleus?
Next: Ballistic Theory, Progress report...Suitable for 5yo Kids
From: msadkins04 on 30 Aug 2005 12:27 George Dishman wrote: > <msadkins04(a)yahoo.com> wrote in message > news:1123451473.180380.122870(a)g47g2000cwa.googlegroups.com... > > > > Have *you* looked up the original paper? Because your > > link is to a highly simplified abstract from what appears to be an > > undergraduate text. Since you've made some rather definite assertions > > about what *isn't* in the experiment, the question isn't rhetorical. > > You are right, my information comes from undergraduate > books (an old Halliday & Resnick) and web sites but I > have checked several sources to make sure they all > describe the same method. I haven't seen the whole > original paper, however there are scanned parts here > > http://spiff.rit.edu/classes/phys314/lectures/doppler/doppler.html > > > > > Now, I've tried to explain to you the meaning of the word "transverse" > > in the context of the term "transverse doppler shift", but you just > > keep ignoring me. > > That's only because you are saying something with which > I am entirely familiar. I have no disagreement with what > you said on that point at all, but Ives and Stilwell didn't > do the experiment the way you suggest. George, you could scarcely be "entirely familiar" with the experiment when you admit above that you haven't even read the original paper. Undergraduate texts and Web abstracts are notorious for playing fast and loose with the details of complicated historical experiments, simplifying (and in some cases completely transforming) those details in order to make a point about general principles clear to non-specialists. Frequently, these secondary sources borrow from one another, and errors present in some old abstract or undergrad text version propagate; then unsuspecting individuals mistake these for independent corroborating sources. In the present case, second-hand reports are particularly suspect, because as previously mentioned the original experimentors concluded that their results WERE NOT evidence for relativistic time dilation, whereas the sources you cite are all by relativists "deconstructing" the conclusions of Ives and Stilwell! I just don't think it is prudent for either of us to argue about the technical specifics of this experiment in the absence of the complete original papers (two of them, for two experiments -- another possible pitfall since details from both may be conflated into a single "experiment" by secondary sources). > The Ives-Stilwell experiment didn't measure the > times of any events, they only measured the second > order component of the Doppler shift using a > spectrograph. That's why I raised it because it > illustrates that time dilation can be measured > without clocks or syncronisation. How do you know what the Ives-Stillwell experiments measured? And again, the title of both papers (differentiated in the second case only by the addition of a roman-numeral II) was "An Experimental Study of the Rate of a Moving Atomic Clock". I won't go into the synchronization aspect because we don't have the details of the original experiment available. I will, however, repeat that, if the principle of the relativity of simultaneity is applied universally, including to the synchronization of clocks (instead of being the only exception of this principle, as in special relativity), the *mathematical basis* for time dilation disappears. That is, there is no longer any theoretical component to support any *interpretation* of experimental events as evidence of "time dilation". You don't seem to realize that there are many possible ways to interpret experimental data. Empirical data isn't evidence without a framework of interpretation. Witness the fundamental disagreement between, on the one hand, Ives and Stilwell, and on the other hand, the relativist community. Mark Adkins msadkins04(a)yahoo.com
From: George Dishman on 31 Aug 2005 10:40 <msadkins04(a)yahoo.com> wrote in message news:1125419247.424399.18310(a)g47g2000cwa.googlegroups.com... > George Dishman wrote: >> <msadkins04(a)yahoo.com> wrote in message >> news:1123451473.180380.122870(a)g47g2000cwa.googlegroups.com... >> > >> > Have *you* looked up the original paper? Because your >> > link is to a highly simplified abstract from what appears to be an >> > undergraduate text. Since you've made some rather definite assertions >> > about what *isn't* in the experiment, the question isn't rhetorical. >> >> You are right, my information comes from undergraduate >> books (an old Halliday & Resnick) and web sites but I >> have checked several sources to make sure they all >> describe the same method. I haven't seen the whole >> original paper, however there are scanned parts here >> >> http://spiff.rit.edu/classes/phys314/lectures/doppler/doppler.html >> >> > >> > Now, I've tried to explain to you the meaning of the word "transverse" >> > in the context of the term "transverse doppler shift", but you just >> > keep ignoring me. >> >> That's only because you are saying something with which >> I am entirely familiar. I have no disagreement with what >> you said on that point at all, but Ives and Stilwell didn't >> do the experiment the way you suggest. > > George, you could scarcely be "entirely familiar" with the experiment > when you admit above that you haven't even read the original paper. Mark, why did you have to twist my reply? I said I am entirely familiar with the meaning of "transverse Doppler shift" and I didn't comment because I don't disagree with what you said about that meaning. The phrase "teaching your granny to suck eggs" is appropriate, though I'm not sure how well that is known outside the UK. > Undergraduate texts and Web abstracts are notorious for playing fast > and loose with the details of complicated historical experiments, > simplifying (and in some cases completely transforming) those details > in order to make a point about general principles clear to > non-specialists. Frequently, these secondary sources borrow from one > another, and errors present in some old abstract or undergrad text > version propagate; then unsuspecting individuals mistake these for > independent corroborating sources. In the present case, second-hand > reports are particularly suspect, because as previously mentioned the > original experimentors concluded that their results WERE NOT evidence > for relativistic time dilation, whereas the sources you cite are all by > relativists "deconstructing" the conclusions of Ives and Stilwell! Results can be taken as evidence for many things. That has little bearing on what was actually done during the experiment which is what we were discussing. > I just don't think it is prudent for either of us to argue about the > technical specifics of this experiment in the absence of the complete > original papers (two of them, for two experiments -- another possible > pitfall since details from both may be conflated into a single > "experiment" by secondary sources). The method is quite clear and the equipment is shown in the diagram scanned from (one of) the original papers which I cited http://spiff.rit.edu/classes/phys314/lectures/doppler/doppler.html The method they used was to take the light emitted parallel to the beam direction and measure its wavelength using a spectroscope, the resulting lines being recorded on film. To see how they processed the resulting values you would certainly need to look at the details of the papers but the technique is clear. >> The Ives-Stilwell experiment didn't measure the >> times of any events, they only measured the second >> order component of the Doppler shift using a >> spectrograph. That's why I raised it because it >> illustrates that time dilation can be measured >> without clocks or syncronisation. > > How do you know what the Ives-Stillwell experiments measured? Because in those days they didn't have the ability to measure frequency directly (e.g. by heterodyning with lasers), the only precise instruments avalailable were spectroscopes which measure wavelength. They used a precision spectroscope and recorded the spectra on Eastman plates and then used a measuring microscope to find the line displacements for the direct beam and that reflected in the mirror. I believe they used the second line of the Balmer series at 4861.3A. I know the two lines, direct and via the mirror, were measured parallel to the beam motion because the mirror was perpendicular the beam as shown in their own diagram scanned on the page above. > And > again, the title of both papers (differentiated in the second case only > by the addition of a roman-numeral II) was "An Experimental Study of > the Rate of a Moving Atomic Clock". And we agreed the spectral lines of ions are in effect small clocks. > I won't go into the > synchronization aspect because we don't have the details of the > original experiment available. Well I'll wait for you to find them, in the meantime if you want to suggest how you "synchronise" a spectrum on a photographic plate, I'm very curious to hear how you think that might be done. > I will, however, repeat that, if the > principle of the relativity of simultaneity is applied universally, > including to the synchronization of clocks (instead of being the only > exception of this principle, as in special relativity), the > *mathematical basis* for time dilation disappears. I don't intend to get into discussing that at this stage. SR is quite simple and time dilation is a fundamental result but that and relativity of simultaneity are very closely linked. My point is simply that it is possible to confirm what is called time dilation by measurements of frequency only without any requirement for synchronisation which makes me dubious of your statement. It may be that your way of looking at it means you could substitute one for the other though. > That is, there is > no longer any theoretical component to support any *interpretation* of > experimental events as evidence of "time dilation". You don't seem to > realize that there are many possible ways to interpret experimental > data. Empirical data isn't evidence without a framework of > interpretation. Witness the fundamental disagreement between, on the > one hand, Ives and Stilwell, and on the other hand, the relativist > community. Of course there are many ways to interpret anything and I never suggested otherwise. What I did say is that it is perfectly possible to measure time dilation without requiring any explicit synchronisation because it affects frequencies, and that is how Ives and Stilwell first confirmed it. Mark, you sit down and ask yourself, could time dilation be measured the way I suggest, just taking the midpoint between two spectral lines seen parallel to the beam and comparing with a static source? You should realise there is no theoretical reson why not, whether that was the method used or not. George
From: George Dishman on 1 Sep 2005 10:11 "Henri Wilson" <H@..> wrote in message news:pl85h1hu9p8datrrc9rgb0lnsluant81pi(a)4ax.com... .... > That's for a rotating star. I have calculated the distribution for an > EXPANDING > star. Of course, my mistake. In general of course both would apply. It would be interesting to see which predominates. >>> I hope to include that today. >> >>Cool. > > haven't had time but will get there soon. Any luck yet? George
From: George Dishman on 2 Sep 2005 16:04 "Henri Wilson" <H@..> wrote in message news:8cd4h1tf86k5crkn8iu5beteempeact9an(a)4ax.com... > On Sat, 27 Aug 2005 12:35:07 +0100, "George Dishman" > <george(a)briar.demon.co.uk> > wrote: > >> >>"Henri Wilson" <H@..> wrote in message >>news:nk8ng11b87o6bvuhs6uimn1uasko6p0qrp(a)4ax.com... > >>> My use of the term 'short shove' is just a concise way of saying, "a >>> force >>> is >>> applied to one end of the rod for a short period of time so as to >>> accelerate it >>> to a different speed relative to the initial one. The force is then >>> released." >>> The question is, after the force is released, has the rod's length >>> 'physically' >>> increased or decreased? >> >>The answer to that is of course NO for SR, taking >>'physical' length to mean proper length. There is >>no aether in SR so there is nothing with which the >>rod could interact to cause a change. > > well how is it that the GPS clock correction assumes that the clocks DO > physically change with velocity. The GPS correction assumes the clocks suffer NO 'physical' change and the only the geometric GR term needs to be corrected. > You can't have it both ways George. > >> >>> The point of the question is that the 'v' in SR's gamma term will have >>> both >>> increased and decreased depending on observer velocity. Therefore the >>> rod >>> will >>> have both increased and decreased its length, according to SR. >> >>No, according to SR, the rod is unchanged. The change >>of motion rotates the axes so that_measurements_ of >>the length by observers not at rest relative to the >>rod will be changed. You have to learn that SR and >>GR are about geometric effects on measurements. > > this 'rotation ' business is just more nonsense. I am aware you choose to remain ignorant of the theory. > A single observer will SEE a rotation but there is no rotation in an > observer's > frame. > >> >>> Can you see now why the proof is trivial? The rod's length cannot >>> PHYSICALLY >>> increase and decrease simultaneously. >> >>I agree, that's what SR says. > > > .....but it isn't what SR DOES.... > SR requires that the changes are REAL PHYSICAL changes. Nope, it predicts differences between measured and proper values due to geometry and that's all. > When you get down to the physics behind the maths, it is no different from > aether theory. I am aware you choose to remain ignorant of the physics behind the maths, preferring to stick with the more comfortable aether-based philosophy. >>> George I cannot see any point in what you are doing. >> >>As you say later, we have very different ways >>of looking at things. Firstly, I am trying to >>create some common foundations that we can >>agree that can form the basis of a conversation. >>I had hoped you could agree that what I said >>about the stick and rulers was valid and >>uncontentious. >> >>Secondly I am trying to educate you a little in >>SR and GR so that the converation is relevant. >>Both are geometric theories but you seem to be >>having trouble grasping that and continually >>talk about "physical" changes that are relevant >>only to aether theories. > > George, I don't wish to understand a theory ... Indeed, I am aware of that. However, you cannot then expect me to answer questions in a way you will follow based on the theory if you don't understand (and don't want to understand) a word of what I am saying. As you said, we have very different ways of looking at this. > Tell me why my 'constant density theory' is any worse that relativity. Does it predict the correct value for the MMX, stellar aberration, Fizeau's drag experiments, the Ives-Stlwell experiment, cosmic muon lifetime, the Sagnac experiment, the precession of Mercury, the Pound-Rebka experiment and the orbital variation of the PSR 1913+16 system? If so, and your working stands up to peer review, then it might be "as good". >>> Running its ends past a couple of accurately spaced photocells .... >> >>Only tells you the time when the each end passes >>a cell. Unless you deal with synchronisation of >>the clocks measuring those times, you are no >>further forward. > > It is easy to synch the clocks at rest. At rest relative to the cells of course in order to get a measurement valid in that frame. > Just make tAB=tBA. Einstein was right on that one...even if for the wrong > reasons. Yes, that works, but then they aren't in sync relative to the rod so you get a different result from the length measured in that frame. The measured length is still not single valued. >>>>.. Now think about this observer compared to >>>>one he hurtles past. Let's say they align the x >>>>axis with the direction of movement and they use >>>>the point and time at which they pass as the origin >>>>of their x and t axes. A vector drawn from (0,0) >>>>to (0,1), one second later is a unit vector lying >>>>along the time axis. Each of them says that the >>>>(0,1) point is located with them of course since >>>>the x coordinate is zero. Can you see that these >>>>unit time vectors have an angle between them on >>>>the x-t plane which depends on the speed? .... >>> If two dolphins performed your experiment under water, using sonar to >>> line up >>> the axes, the angle would be very different from your visually >>> determined one. >>> What does that prove? >> >>The angle is exactly 180 degrees as a given. I think >>you must have misread the paragraph. > > Plotting events in x-t doesn't change 3D physical space. True. That still bears no discernable relation to what I wrote though. > I use grids of synched clocks to measure everything. That way, all frames > are equal and I don't have to correct for light travel time. You still have to define your method for synchronising the clocks and all physical methods turn out to be the same as sending light pulses so no you haven't avoided correcting for the propagation time. >>>>If >>>>the time axes have that angle between them and >>>>each considers the x axis perpendicular to the >>>>time axis, then the x axes also have that angle >>>>between them. In other words one observer is using >>>>axes that are rotated relative to the other's. That >>>>is the answer to your "Why would you want to do >>>>that?" earlier, you don't want to but it is >>>>unavoidable. .... > You plot different angle on and x-t diagram. So what? So if the t axes used by the two observers are not parallel, they won't agree on the duration measured along that axis. You already agreed that some time ago but couldn't understand "why anyone would want to do that". They don't want to but it is inevitable. >>>>> Interestingly, even though rotation direction is not absolute, >>>>> rotation itself IS absolutely detectable. >>>> >>>>What is really interesting is that acceleration >>>>is also absolute and acceleration is a rotation >>>>in the x-t plane. >>> >>> That is a mathematical concept only. It has no physical meaning. >>> Obviously, a constant speed is represented on an x-t diagram as a >>> straight line. An acceleration has a constantly changing gradient. >>> I don't see how this kind of graphical demonstration can change the >>> nature of space and time. >> >>That's why it is interesting to find that it does. > > ....and when was that ever FOUND? About 1907, by Minkowski. >>>>>>Constant, but not single valued. You can trade >>>>>>space and time by rotating the x-t axes just as >>>>>>you can trade space-space by rotating the axes >>>>>>in the x-y plane. >>>>> >>>>> It IS single valued....magnitude ONE. >>>> >>>>No, it has two values, you just want to call them >>>>both "one" but I won't let you :-) >>> >>> According to you, it has an infininte number of values....because you >>> can have an infinite number of observers. >>> I agree, it can have an infinite number of MEASURED values. >> >>Thank you, that has been my point all the weeks. > > It can only have an infinite number if light is used directly. No, it has as many values as there are orientations of the axes. As you say, an infinite number. > It will have only one value in all frames if my 'grid of clocks' is used. Each grid has to be synchronised. That means setting the clocks so that all the clocks along a line perpendicular to the t axis (i.e. a suface of constant t) show the same value. If the t axes of the observers are not parallel, it is impossible to achieve that with one set of clocks. Each obssrver has his own criteria for determining if the clocks are in sync, and they can never agree. >>> However, the fact >>> remains none of those values will change during the course of the >>> experiment... >> >>Yes, and I agree that. We are now in complete >>agreement :-) > > No we are not. > You have lost me here. I have been saying that the measured duration would be "Constant, but not single valued." You agreed with "not single valued" when you said "I agree, it can have an infinite number of MEASURED values." and I agree with you when you say ".. none of those values will change during the course of the experiment..." which is why I said "Constant, but ...". I cannot see any difference now. >>> The rotation period has an inherent constancy. >> >>Correct. Now see if you can reconstruct your argument >>so that it only uses the constancy of each measured >>value without assuming there is only a single value. > > I don't see the connection between 'constancy' and 'single valued'. Good, there isn't one. That has been my point throughout. You set up the experiment so that the duration of the orbit would be constant, I agree with that. However your proof relies on the measured duration being single-valued so that you can use it as a reference measurement, also called "a unit". >>> You (and Einstein) are claiming that if its duration is measured using >>> manmade >>> clocks and if the readings are different for different frames or >>> circumstances, >>> then this is an indication that TIME itself must vary according to those >>> circumstances. >> >>Not quite, what SR and GR say is that a duration >>measured by a clock is really the length of the >>worldline of the clock between those two events. > > it says that time itself is affecterd by the spatial path taken by a > moving object. It is more explicit, proper (or 'physical' if you prefer) time is the nothing other than the length of the spatio-temporal path. > That is totally meaningless. I am aware you choose to remain ignorant of the subject. It's your choice to work hard to avoid understanding the meaning. >>> You are not even prepared to consider the possibility that the clocks >>> are not perfect and might have malfunctioned under the different >>> conditions. >> >>Not at all, if the clock didn't show exactly the >>value corresponding to the length of the worldline >>then the clock has malfunctioned or is imperfect. > > You make a wrong logical assumption here. > What is a worldline? A straightline between any two points on a space/time > graph. No, it is seldom straight. It is the path through spacetime or the locus of events at which the object is present between two. > If you draw any other path on that graph between the two end points, how > does that change the time coordinates there? Different path, different length. If you move the clock along _any_ path and it doesn't indicate the length of the path, that is the criterion for saying the clock is not perfect. There is only one correct value, it just isn't the number you would expect. >>> ......Very strange logic if you ask me.... >> >>It is the same logic as your except that the >>expected "correct" value is not the same for all >>clocks, it depends on the path. It is no >>different from car odometers showing different >>distances between locations depending on the >>route taken, there is still only one correct >>value for each route and an odometer that shows >>a different value isn't working properly. > > ....but as I have pointed out many times before, the car clocks still show > exactly the same times no matter what routes the cars take. But not the odometers. In the analogy it is the car odometers that explain how real world clocks are expected to behave in GR. > There no x-t analogy for this kind of x-y effect... > If you accept that there are three time dimensions then you can talk about > 'taking different routes in time'. Nope, it is "different routes in spacetime" which has four dimensions. Just remember to use 'c' to fix your units and that time is distinguished from space by having the opposite sign. >>>>>>Of course, but you are still trying to define >>>>>>two different durations as the same 1, or 100 >>>>>>or whatever number of units. You can't do that >>>>>>and your attempt at a proof fails as a result. >>>>> >>>>> George, an orbit occupies a duration of time. >>>> >>>>No, it occupies as many durations as there are >>>>angles for the time axis for it to be measured >>>>against. >>> >>> You are introducing measurement errors, mainly due to EM's finite travel >>> time. >>> You have to eliminate that. >> >>No, I am acknowledging that different routes >>have different lengths. This assumes finite >>communication time for remote measurements >>has been eliminated. In your example I did >>that by proposing a marker in space so there >>was no separation between the satelite and >>the marker at the ends of the orbit. The >>finite communication time from there to >>the ground is the same for all orbits so >>doesn't affect the duration measurement. > > You seem to have difficulty understanding the difference between path > length > and the actual distance between two points. > > If two cars take different routes to go from A to B, they travel different > distances... > > BUT THE DISTANCE BETWEEN POINTS A AND B HAS NOT CHANGED. Right, but clocks measure the DISTANCE TRAVELLED THROUGH SPACETIME, not the DISTANCE BETWEEN THE POINTS. Get it yet? Empirically that's what we find all clocks do whether we like it or not. > The same applies in the x-t case. It matters not how the path is drawn, > the two end points have the same coordinates. No, as always the coordinates depend on the orientation, origin and scaling of the axes. You know that, it's trivial. I think you meant the interval between them doesn't matter how the path is drawn, you can think of that as 'absolute' (we had this conversation before). >>>>> The time axis is made imaginary to keep it orthogonal. >>>> >>>>Yes, empirically we find it has to have the opposite >>>>sign from spatial dimensions. >>> >>> empirically? >>> Since when was anything in relativity backed up by things 'empirical'? >> >>SR was derived from Maxwell's equations which in >>turn came from 19th century empirical relationships >>between the voltages and currents in coils and >>Layden jars. > > SR was derived from the mistake that a vertical light beam moving at c in > one frame becomes a diagonal light beam moving at c in another. Nope, it comes from the fact that the speed can be derived from the permittivity and permeability of a vacuum. > Einstein obviously spent too much time watching falling raindrops through > the windows of moving trains. That merely explains the consequence to those who might like to understand. You ahve already said you prefer not to. >>>>You can't combine furlongs and Fahrenheit, but as >>>>long as you correct the units using 299792458m=1s, >>>>you can combine time and space exactly as we can >>>>combine distance north and distance east using >>>>Pythagoras, while the negative sign distinguishes >>>>space from time. That has to be telling us >>>>something significant about the nature of our >>>>universe. >>> >>> But you are not combining space and time at all. >>> You are subtracting space, as specified in terms of 'light's time to get >>> there', from space Euclidean. >> >>I specifically chose the word "combining" because >>it involves not only both addition and subtraction >>but also taking the square root of the combined >>squares. >> >>> What is the purpose of doing that? >> >>Only one, we find that doing so gives a value >>which is independent of rotation of the axes >>just as Pythagoras gives the distance between >>two points regardless of rotation of the x-y >>axes. > > But in that case, tau would be constant. You got it :-) It is constant regardless of the orientation of the axes. We call that "invariant". > CMIIW, but 'tau' represents the distance derived from subtracting 'ct' > from the > Euclidean length between two points. > > What is its physical significance? Does it have any? The fact that it _is_ constant in all known real world experiments tells me that 4D spacetime is what is 'physical' and that since space and time separately depend on the orientation of the axes along which they are measured, they are only components. > I doubt it. > >>> Relativity says, "the length AB does not appear to be AB because it >>> takes >>> different times for light to go from A to O and from B to O. This >>> implies >>> that >>> AB is not a fixed length of space". >> >>No it doesn't. What is says is that there is >>some absolute separation in spacetime between >>two events (I like to use two supernovae) and >>that when we measure that, the measured values >>for the distance and time between the bangs >>will depend on how we choose to rotate the >>space and time axes, which we do by changing >>the speed of the observing instruments. > > The mistake is that it assumes spacetime is a physical entity rather than > a mathematical convenience. It is no more a mistake than assuming space is physical in Newtonian mechanics, with the benefit that it matches experimental results. >>To really understand SR, you need to look at c >>as only a way of converting between different >>units of spacetime measurements, exactly like >>using 25.4 to convert between inches and >>millimetres.] > > .....but the space or time interval between any two points on a space/time > diagram does not depend on the path taken. > Why should it? The spatial and temporal components vary with the axes but the straight-line separation never does as long as they are orthogonal. If I knew why ... Intriguing, isn't it. George
From: Henri Wilson on 2 Sep 2005 18:56
On Thu, 1 Sep 2005 15:11:13 +0100, "George Dishman" <george(a)briar.demon.co.uk> wrote: > >"Henri Wilson" <H@..> wrote in message >news:pl85h1hu9p8datrrc9rgb0lnsluant81pi(a)4ax.com... >... >> That's for a rotating star. I have calculated the distribution for an >> EXPANDING >> star. > >Of course, my mistake. In general of course both >would apply. It would be interesting to see which >predominates. > >>>> I hope to include that today. >>> >>>Cool. >> >> haven't had time but will get there soon. > >Any luck yet? my program now includes figures for magnitude variation. It can be used as a research tool although there are still many aspects I want to diagnose and improve. www.users.bigpond.com/hewn/variablestars.exe > >George > HW. www.users.bigpond.com/hewn/index.htm Sometimes I feel like a complete failure. The most useful thing I have ever done is prove Einstein wrong. |