Analysis of gas mode MM interferometer operation using standard SR formulae.
Surfer
using standard SR formulae. The result is consistent with a formula
given in:
"The Michelson and Morley 1887 Experiment and the Discovery of
Absolute Motion"
http://arxiv.org/abs/physics/0508174
The assumptions herein are:
An MM type interferometer with perpendicular arms AB and AC each of
length L. (Eg as in Fig. 1 in the above paper.)
The interferometer is moving relative to 3-space with velocity v in
direction AB.
The interferometer contains a co-moving gas.
The effective refractive index of the gas is assumed to vary slightly
as follows:
In the 3-space frame, the symmetry of paths AC and CA ensure that the
speed of light relative to the gas is the same in both paths. Take the
effective refractive index for this path as being a reference value of
(n).
In the 3-space frame, light travelling along path AB is _catching_up_
with the gas. Suppose this slightly changes the effective
refractive index for that light to a value of (n-x).
In the 3-space frame, light travelling along path BA is _heading_into_
the gas. Suppose this slightly changes the effective refractive index
for that light to a value of (n+x+d).
Choose units so that c==1,
Define variables for light speeds relative to the interferometer frame
as follows.
Vo = 1/(n-x) -- the outward speed along AB
Vr = 1/(n+x+d) -- the return speed along BA
Vt = 1/n -- the transverse speed along AC and CA
Using the SR formulae for composition of velocities, define variables
for the equivalent speeds relative to the 3-space frame as follows.
Vout = (Vo + v)/(1 + v Vo/c^2)
Vret = (Vr - v)/(1 - v Vr/c^2)
Vtran = Sqrt[v^2 + (1-v^2) Vt^2]
Consider path ABA
In the 3-space frame, Lorentz contraction causes the length of the AB
arm to be:
Lc = L Sqrt[1-v^2]
Hence the times taken by light to traverse the arm is:
tAB = Lc/(Vout - v)
tBA = Lc/(Vret + v)
So
tABA = tAB + tBA = L(Vo + Vr)/(Vo Vr Sqrt[1-v^2])
Consider path ACA and traversal times tAC and tCA
In the 3-space frame, the length of the AC arm remains equal to L
Pythagoras theorem gives
(Vtran tAC)^2 = L^2 + (v tAC)^2
which gives
tAC = L/Sqrt[Vtran^2-v^2]
By symmetry tCA = tAC, so
tACA = 2 tAC = 2 L / Sqrt[Vtran^2 - v^2]
= 2 L /(Vt Sqrt[1-v^2])
Consider both arms
The difference between return travel times in the two arms is,
Delta_t = tABA - tACA
= L (Vt (Vo+Vr) - 2 Vo Vr) / (Vt Vo Vr Sqrt[1-v^2])
If Vo, Vr and Vt were all equal, as would ordinarily be
assumed, then Delta_t would be zero. However, if we substitute the
refractive index dependent values defined above, then after
simplification we get:
Delta_t = d L/Sqrt[1-v^2]
If v<<c then we can use the approximation:
Delta_t = d L
This result is consistent with equation 5 in the above mentioned
paper. For units in which c==1, and with the directions assumed
herein, equation 5 reduces to:
Delta_t = n(n^2 - 1) v^2 L
Hence we can deduce that:
d = n(n^2 - 1) v^2
How does this affect the postulate of relativity, "The laws of
physical phenomena are the same in all inertial references frames"?
It seems to me that if the formula for d holds equally well in ALL
inertial reference frames, then it does not directly violate the
postulate.
Androcles
http://www.androcles01.pwp.blueyonder.co.uk/
"Surfer" <n...@spam.net> wrote in message
news:ksm514942d6jel2pu...@4ax.com...
Picking out one postulate when there are three is rather silly.
Treating MMX in isolation is rather silly too.
Einstein admits to two postulates and also states as a third:
"the ``time'' required by light to travel from A to B equals the
``time'' it requires to travel from B to A."
Ref: http://www.fourmilab.ch/etexts/einstein/specrel/www/
Whilst it is true for MMX, it is false for Sagnac which DOES
show a fringe shift.
http://www.androcles01.pwp.blueyonder.co.uk/Sagnac/Sagnac.htm
This leads to such ridiculously contradictory remarks as
"the counter-rotating pulse arrives at the "end" point slightly earlier than
the co-rotating pulse at exactly the same time. "
http://www.androcles01.pwp.blueyonder.co.uk/Sagnac/SagnacIdiocy.htm
Dono
Bzzt, no.
It is remarkable that you did some calculations for a change instead
of simply sucking up to Cahill.
Your "Vtran" (and everything that follows is wrong. Here are the
correct calculations:
Tom Roberts
> The following is an analysis of gas mode MM interferometer operation
> using standard SR formulae. The result is consistent with a formula
> given in:
> "The Michelson and Morley 1887 Experiment and the Discovery of
> Absolute Motion"
> http://arxiv.org/abs/physics/0508174
But this does not take the errorbars into account! Not a single figure
has errorbars. In every one the actual errorbars greatly exceed the
variation shown, indicating that the results are statistically
INSIGNIFICANT.
See Appendix I of http://arxiv.org/abs/physics/0608238 for an error
analysis of Michelson and Morley's measurements.
> The assumptions herein are:
... completely useless when one does not understand the basics of
experimental physics, and does not take errorbars into account.
They are also completely useless because you do not actually use "SR
formulae" -- it is simple and easy to show that SR predicts a null
result for such experiments, whether performed in vacuum or in optical
media, be they gas, liquid, or solid, including the case when the two
arms contain different optical media. And this prediction is confirmed
in literally zillions of experiments.
[In particular, in SR there is no such thing as a "3-space
frame", and Vo=Vr=Vt=c/n relative to the instantaneous
inertial frame in which the interferometer AND THE GAS are
at rest (n=1 for vacuum, which needs no "rest").]
> How does this affect the postulate of relativity, "The laws of
> physical phenomena are the same in all inertial references frames"?
There's no effect at all, because a) your analysis is wrong, and b) the
errorbars show that EVERY ONE of these plots and experiments is
consistent with the prediction of SR.
When you base arguments on a complete ignorance of modern experimental
technique, they are useless. You need to educate yourself (and Cahill)
about basic, undergraduate experimental physics.
And you also need to learn what SR actually says and how to apply it.
Tom Roberts
Dono
>
> And you also need to learn what SR actually says and how to apply it.
>
> Tom Roberts
..this will take another 10 years ....
Surfer
The second postulate is, "Light is always propagated in empty space
with a definite velocity c that is independent of the state of motion
of the emitting body"
In vacuum n is 1 and so d will be zero. So the formula predicts that
no fringe shifts will be seen if the gas is replaced with a vacuum.
This is consistent with the second postulate, so the second postulate
isn't affected by this analysis.
The following paper deals with light speed anisotropy in empty space,
but I think it is important to note that the anisotropy is measured
in an indirect way.
Resolving Spacecraft Earth-Flyby Anomalies with Measured Light Speed
Anisotropy
Cahill R.T.
http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Cahill_flyby.pdf
>
>Einstein admits to two postulates and also states as a third:
>"the ``time'' required by light to travel from A to B equals the
>``time'' it requires to travel from B to A."
>Ref: http://www.fourmilab.ch/etexts/einstein/specrel/www/
>
I see that as equivalent to the second postuate, because if direct
measurement of the speed of light always gives a value of c, then
direct measurement of the time for light to travel a fixed distance
will always give the same value in both directions.
>
>Whilst it is true for MMX, it is false for Sagnac which DOES
>show a fringe shift.
> http://www.androcles01.pwp.blueyonder.co.uk/Sagnac/Sagnac.htm
>
The postulates of special relativity (which is what these are) only
apply to inertial frames of reference. The Sagnac effect involves a
rotating frame, so they do not apply there.
Dono
>
> The postulates of special relativity (which is what these are) only
> apply to inertial frames of reference. The Sagnac effect involves a
> rotating frame, so they do not apply there.
While the postulates apply to inertial frames only, there are
extensions to SR for accelerated motion. You need to take a class in
relativity. The Sagnac experiment is fully explained within the SR
framework.
Maybe Cahill can teach you that?
Surfer
<tjrobe...@sbcglobal.net> wrote:
>Surfer wrote:
>> The following is an analysis of gas mode MM interferometer operation
>> using standard SR formulae. The result is consistent with a formula
>> given in:
>> "The Michelson and Morley 1887 Experiment and the Discovery of
>> Absolute Motion"
>> http://arxiv.org/abs/physics/0508174
>
>But this does not take the errorbars into account!
>Not a single figure has errorbars.
>
They are not necessary in the above paper, because it only concerns
the initial discovery.
Confirmation came from the Miller experiments. Eg.
Ether drift experiments at Mt. Wilson
http://www.pnas.org/cgi/reprint/11/6/306
The Ether-Drift Experiment and the Determination of the Absolute
Motion of the Earth.
http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
Androcles
http://www.androcles01.pwp.blueyonder.co.uk/
"Surfer" <n...@spam.net> wrote in message
news:m4i614ltg9no3nnik...@4ax.com...
Yes I know. That's a rather stupid postulate, unsupported by observation.
Unless this pencil is really broken, of course.
http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/optpic/brokpen.jpg
| In vacuum n is 1 and so d will be zero. So the formula predicts that
| no fringe shifts will be seen if the gas is replaced with a vacuum.
How odd. Plenty of shift is seen here:
http://www.flitemedia.com/images/album-art/red-shift.jpg
(and in ring laser gyroscopes).
Are you telling me they don't work in a vacuum?
| This is consistent with the second postulate, so the second postulate
| isn't affected by this analysis.
| The following paper deals with light speed anisotropy in empty space,
| but I think it is important to note that the anisotropy is measured
| in an indirect way.
Oh yeah?
|
| Resolving Spacecraft Earth-Flyby Anomalies with Measured Light Speed
| Anisotropy
| Cahill R.T.
| http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Cahill_flyby.pdf
|
So what? Any fool can write pages of drool, first he needs to understand
a simple equation, c = lambda * nu.
Since nothing changes the emitted frequency, red and blue shift are
proof enough of measuring light's speed to be c+v with respect to the
observer. Perhaps Cahill never heard of Doppler.
Even the crank Einstein said
"But the ray moves relatively to the initial point of k, when measured in
the stationary system, with the velocity c-v",
contradicting his own third postulate.
| >Einstein admits to two postulates and also states as a third:
| >"the ``time'' required by light to travel from A to B equals the
| >``time'' it requires to travel from B to A."
| >Ref: http://www.fourmilab.ch/etexts/einstein/specrel/www/
| >
| I see that as equivalent to the second postuate, because if direct
| measurement of the speed of light always gives a value of c, then
| direct measurement of the time for light to travel a fixed distance
| will always give the same value in both directions.
else everything you said is garbage.
Try IF <statement> THEN <statement> ELSE <statement>
| >
| >Whilst it is true for MMX, it is false for Sagnac which DOES
| >show a fringe shift.
| > http://www.androcles01.pwp.blueyonder.co.uk/Sagnac/Sagnac.htm
| >
|
| The postulates of special relativity (which is what these are) only
| apply to inertial frames of reference.
Wrong. Nowhere does the crank Einstein use ther word "inertial"
and he specifically states:
If we assume that the result proved for a polygonal line is also valid for a
continuously curved line, we arrive at this result: If one of two
synchronous clocks at A is moved in a closed curve with constant velocity
until it returns to A, the journey lasting t seconds, then by the clock
which has remained at rest the travelled clock on its arrival at A will be
1/2 tv^2/c^2 second slow. Thence we conclude that a balance-clock at the
equator must go more slowly, by a very small amount, than a precisely
similar clock situated at one of the poles under otherwise identical
conditions.
Maybe you are talking about relativity by Surfer, but that doesn't interest
anyone.
The Sagnac effect involves a
| rotating frame, so they do not apply there.
Oh yes it does.
"If we assume that the result proved for a polygonal line is also valid for
a
continuously curved line, ..."
The LYING ROGUE Einstein's third postulate is:
" the ``time'' required by light to travel from A to B equals the ``time''
it requires to travel from B to A."
What I want to know is why Einstein says
the speed of light is c-v from A to B,
the speed of light is c+v from B to A,
the time each way is the same.
He must have been a right kook...
Are you a kook too?
Surfer
wrote:
The expression for Vtran omits c.
But units are assumed in which c==1, so it doesn't affect the result.
Surfer
<Headm...@Hogwarts.physics> wrote:
>|
>| The second postulate is, "Light is always propagated in empty space
>| with a definite velocity c that is independent of the state of motion
>| of the emitting body"
>
>Yes I know. That's a rather stupid postulate, unsupported by observation.
>
What the postulate really means is that when the speed of light is
_measured_ in an inertial frame of reference, the result is always
found to be c.
That is supported by observation.
Androcles
--
This message is brought to you by Androcles
http://www.androcles01.pwp.blueyonder.co.uk/
"Surfer" <n...@spam.net> wrote in message
news:l0q614t0c97gpg2o2...@4ax.com...
| On Sat, 26 Apr 2008 17:27:41 +0100, "Androcles"
| <Headm...@Hogwarts.physics> wrote:
|
| >|
| >| The second postulate is, "Light is always propagated in empty space
| >| with a definite velocity c that is independent of the state of motion
| >| of the emitting body"
| >
| >Yes I know. That's a rather stupid postulate, unsupported by observation.
| >
| What the postulate
Did you snip something I wrote because you are an ignorant cretin
who cannot face the truth?
*plonk*
Dirk Van de moortel
"Surfer" <n...@spam.net> wrote in message news:crl614djptl55m15e...@4ax.com...
> On Sat, 26 Apr 2008 15:12:03 GMT, Tom Roberts
> <tjrobe...@sbcglobal.net> wrote:
>
>>Surfer wrote:
>>> The following is an analysis of gas mode MM interferometer operation
>>> using standard SR formulae. The result is consistent with a formula
>>> given in:
>>> "The Michelson and Morley 1887 Experiment and the Discovery of
>>> Absolute Motion"
>>> http://arxiv.org/abs/physics/0508174
>>
>>But this does not take the errorbars into account!
>>Not a single figure has errorbars.
>>
> They are not necessary in the above paper, because it only concerns
> the initial discovery.
"Errorbars not necessary for initial discovery":
http://users.telenet.be/vdmoortel/dirk/Physics/Fumbles/DiscoveryBars.html
Good one, Reginald.
Dirk Vdm
OG
But c cannot = = 1, as c is not a dimensionless number.
YBM
> But c cannot = = 1, as c is not a dimensionless number.
So YOU cannot drive at v = 1 mile per hour since v in
not a dimensionless number ? What so special about "1" in
your "system" ?
Tom Roberts
> On Sat, 26 Apr 2008 15:12:03 GMT, Tom Roberts
> <tjrobe...@sbcglobal.net> wrote:
>> Surfer wrote:
>>> "The Michelson and Morley 1887 Experiment and the Discovery of
>>> Absolute Motion"
>>> http://arxiv.org/abs/physics/0508174
>> But this does not take the errorbars into account!
>> Not a single figure has errorbars.
>>
> They are not necessary in the above paper, because it only concerns
> the initial discovery.
That's complete and utter nonsense! To claim a "discovery" you need a
SIGNIFICANT result, and Michelson and Morley did not have one.
I flipped this coin 8 times, and it came up heads 6 times.
Have I "discovered" an anomaly in the laws of physics, or in
the laws of statistics, or in the balance of this coin? --
OF COURSE NOT! Because the deviation from 50% heads is not
significant! You are claiming this would be a "discovery".
You REALLY need to learn the basics.
Had I flipped it 800 times and it came up heads 600 times,
then that would be significantly different from 50%. There
is a well known and elementary statistical analysis of such
measurements that tells you whether or not such a deviation
is statistically significant. You need to STUDY this.
Every modern experimental physicist knows this, and uses it in EVERY
analysis. You and Cahill merely display your ignorance by not doing so.
> Confirmation came from the Miller experiments.
That, too, is nonsense, as Miller's result is not significant, either.
You will just repeat your mistakes until and unless you get an
EDUCATION. You REALLY need to learn the basics of experimental physics
-- topics which are taught to undergraduates around the world.
Tom Roberts
Dono
>
> The expression for Vtran omits c.
> But units are assumed in which c==1, so it doesn't affect the result.
No, idiot.
Your expression for Vtran is plain wrong, I felt sorry for you and for
Rag and I gave you the complete, correct derivation. Download it and
learn. Feel free to share it with "mentor".
OG
I can drive at v = 1 mile per hour
I cannot drive at v = 1
Can you see the difference?
In the first case 1 is, in effect, multipled by the unit (miles per
hour) Determining a value of '1' for v demands that the multiplier
is "1 mile / 1 hour".
Dimensions are real things - you cannot subtract v from 1,
any more than you can subtract 1 cheese from a dollar.
Surfer wants to use Vtran = Sqrt[v^2 + (1-v^2) Vt^2] which includes a
component (1-v^2). How do you think we can subtract the square of my
veolcity from 1? It's like subtracting apples from oranges.
The only way that (1-v^2) makes any sense is to be explicit and say that
v and c must be in the same units - otherwise they don't cancel out.
If you choose to express 'v' not as m/h, but as 'm/h per 679,616,629
m/h' you can technically get away with it, but it makes your
calaulations very complicated
YBM
> YBM wrote:
>> OG a écrit :
>>
>>> But c cannot = = 1, as c is not a dimensionless number.
>>
>>
>> So YOU cannot drive at v = 1 mile per hour since v in
>> not a dimensionless number ? What so special about "1" in
>> your "system" ?
>>
>
> I can drive at v = 1 mile per hour
> I cannot drive at v = 1
>
> Can you see the difference?
> In the first case 1 is, in effect, multipled by the unit (miles per
> hour) Determining a value of '1' for v demands that the multiplier
> is "1 mile / 1 hour".
>
> Dimensions are real things - you cannot subtract v from 1,
> any more than you can subtract 1 cheese from a dollar.
>
> Surfer wants to use Vtran = Sqrt[v^2 + (1-v^2) Vt^2] which includes a
> component (1-v^2). How do you think we can subtract the square of my
> veolcity from 1? It's like subtracting apples from oranges.
My point is not about Surfer's claims, it's about your argument.
> If you choose to express 'v' not as m/h, but as 'm/h per 679,616,629
> m/h' you can technically get away with it, but it makes your
> calaulations very complicated
Quite the contrary, using a unit system where c=1 makes calculation
far more easier.
OG
"YBM" <ybm...@nooos.fr> wrote in message
news:4813e0ea$0$32543$426a...@news.free.fr...
Before we continue - can I ask to what level you have had formal education
in physics?
If you've not had formal education in dimensional analysis, I can give you
that - and hope you can understand it; but if you've already been taught
(and failed to understand) it I'll have to take longer.
Apologies if it sounds patronising, but it really is worth knowing where you
are starting from.
You would like to express 'c' as '1', but that means that you are changing
the definition of 'velocity' and I really don't think you know what it
implies.
YBM
Is graduate in math enough for you ? And by math I mean a cursus where
physics is taught as well.
> You would like to express 'c' as '1', but that means that you are changing
> the definition of 'velocity' and I really don't think you know what it
> implies.
Don't make a fool of yourself. Did you ever heard of Planck units ?
You goofed, admit it, there is nothing worse than using a cranky
argument against a crank as you did with surfer.
Eric Gisse
[snip all, unread]
Why? Your - and by extension, Cahill's - spew has been debunked and
discredited every goddamn time. We're tired of it and don't you have
better things to do?
Eric Gisse
> >Surfer wrote:
> >> The following is an analysis of gas mode MM interferometer operation
> >> using standard SR formulae. The result is consistent with a formula
> >> given in:
> >> "The Michelson and Morley 1887 Experiment and the Discovery of
> >> Absolute Motion"
> >>http://arxiv.org/abs/physics/0508174
>
> >But this does not take the errorbars into account!
> >Not a single figure has errorbars.
>
> They are not necessary in the above paper, because it only concerns
> the initial discovery.
...the fuck?
Did you just argue that error bars are "not necessary" ?!
>
> Confirmation came from the Miller experiments. Eg.
Well suddenly the insane and stupid position /makes sense/ when one
realizes that Cahill's conclusions are null and void when error bars
otherwise negate the "signal" found in Miller's results.
It isn't stupidity, it is just plain ol' intellectual dishonesty!
>
> Ether drift experiments at Mt. Wilsonhttp://www.pnas.org/cgi/reprint/11/6/306
>
> The Ether-Drift Experiment and the Determination of the Absolute
> Motion of the Earth.http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
I can't help but wonder what Cahill's coworkers think of him when he
writes stuff like this even after ignoring every refutation of his
claims.
OG
Hey - it's not my needs that we're talking about :-)
It's your level of understanding that we're looking at.
What do you know about Dimensional Analysis?
doug
> On Sat, 26 Apr 2008 15:12:03 GMT, Tom Roberts
> <tjrobe...@sbcglobal.net> wrote:
>
>> Surfer wrote:
>>> The following is an analysis of gas mode MM interferometer operation
>>> using standard SR formulae. The result is consistent with a formula
>>> given in:
>>> "The Michelson and Morley 1887 Experiment and the Discovery of
>>> Absolute Motion"
>>> http://arxiv.org/abs/physics/0508174
>> But this does not take the errorbars into account!
>> Not a single figure has errorbars.
>>
> They are not necessary in the above paper, because it only concerns
> the initial discovery.
>
I love this. If it can be used to support your hope for
something spectacular, it is ok to ignore the error bars
and thus misrepresent the facts. You should consider
politics.
> Confirmation came from the Miller experiments. Eg.
>
I guess that ignoring facts is a consistent thread
in your hopes.
YBM
> It's your level of understanding that we're looking at.
As a matter of fact it is yours, as you've shown by evading most
of my comments.
> What do you know about Dimensional Analysis?
Far enough to know that choosing a unit system where c=1
is not a problem, no more that choosing a system where c=2
or 299 792 458. It is even routine to choose such a system
as long as relativistic effects are considered.
You seem to like authority arguments (without any kind of
reference). How many references from academic physics papers
or book where such a unit system is in used would you want
to consider you goofed as a usual crank ?
If you pretend that any numerical value, like 1, will mean something
special on the basis of dimensional analysis, then you are a joke.
I would be especially amused to see you explain dimensional
analysis on such a ridiculous idea.
At first I took you seriously, and thought you expressed yourself
poorly even if educated in physics. Now I ask you the question you
had the impudence to ask : where (and how) in hell did you learn
anything about dimensional analysis and physics ?
Tom Roberts
>>> But c cannot = = 1, as c is not a dimensionless number.
It is merely a choice of units. You act as if an artifact of human
history were a law of nature -- it isn't. When one measures time in
meters, then all speeds are unitless (being a ratio of meters/meters).
Essentially all theoretical physicists use such units (and
experimentalists translate from such units to the ones they use in their
measurements).
When you make a claim that implies that >99% of all physicists are doing
something wrong, it is time for YOU to check YOUR premises.
There's nothing strange or wrong about this, because after all, units
are CONVENTIONAL, meaning they are a free choice of humans. Remember
that a unit of length is merely a standard length to which all others
are referenced by a ratio; ditto for a unit of time, and ditto for a
unit of ANY OTHER QUANTITY. In units with c=1, the speed of light in
vacuum is the standard speed to which all other speeds are referenced by
a simple ratio. In such units, time is measured in meters (or cm). To
relate that to the conventional unit (seconds), 1 meter of time is
1/299,792,458 second.
Generally we use units with c=G=hbar=1 (speed of light,
gravitational constant, Planck's constant divided by 2 pi).
This determines the units for time, mass, and charge, plus
all subsidiary units related to them. This permits one to
study relativity, electrodynamics, quantum mechanics, and
quantum field theory, all with just one unit: length
(usually cm, for historical reasons). Sometimes 8*pi*G
is set to 1 rather than G itself (in general relativity).
Tom Roberts
Surfer
<tjrobe...@sbcglobal.net> wrote:
>Surfer wrote:
>> On Sat, 26 Apr 2008 15:12:03 GMT, Tom Roberts
>> <tjrobe...@sbcglobal.net> wrote:
>>> Surfer wrote:
>>>> "The Michelson and Morley 1887 Experiment and the Discovery of
>>>> Absolute Motion"
>>>> http://arxiv.org/abs/physics/0508174
>>> But this does not take the errorbars into account!
>>> Not a single figure has errorbars.
>>>
>> They are not necessary in the above paper, because it only concerns
>> the initial discovery.
>
>That's complete and utter nonsense! To claim a "discovery" you need a
>SIGNIFICANT result, and Michelson and Morley did not have one.
>
If the result had not been SIGNIFICANT, there would have been no need
to try to explain the result by introducing the Lorentz-FitzGerald
contraction hypothesis.
http://en.wikipedia.org/wiki/Lorentz-FitzGerald_contraction_hypothesis#Lorentz.E2.80.93FitzGerald_contraction_hypothesis
In fact the result was sufficiently significant to bring about a
revolution in physics.
That being the case, the experiment is worth re-examining. When that
was done it was found that:
"Their published data revealed to them the expected fringe shifts, but
that data gave a speed of some 8km/s using a Newtonian theory for the
calibration of the interferometer, and so was rejected by them solely
because it was less than the 30km/s orbital speed of the earth. A 2002
post relativistic-effects analysis for the operation of the device
however gives a different calibration leading to a speed > 300km/s. So
this experiment detected both absolute motion and the breakdown of
Newtonian physics. So far another six experiments have confirmed this
first detection of absolute motion in 1887."
>
> I flipped this coin 8 times, and it came up heads 6 times.
> Have I "discovered" an anomaly in the laws of physics, or in
> the laws of statistics, or in the balance of this coin? --
> OF COURSE NOT! Because the deviation from 50% heads is not
> significant! You are claiming this would be a "discovery".
>
If the MM experiment was like that, people would have ignored the
result and stuck with ether theory :-)
Surfer
wrote:
>On Apr 26, 10:10 am, Surfer <n...@spam.net> wrote:
I had a look.
In my analysis:
Vtran is the speed of light relative to 3-space (Sigma in your paper)
Vt is the speed of the same light relative to the lab.
In your analysis:
cs is the speed of light relative to Sigma
c0 is the speed of the same light relative to the lab
Expression 2.11 in your analysis is
cs^2 -v^2 = c0^2/gamma^2
Re-arranging:
cs = Sqrt[v^2 + (1/gamma)^2 c0^2]
Substituting Vtran for cs, Vt for c0 and 1/Sqrt[1-v^2] for gamma
gives:
Vtran = Sqrt[v^2 + (1-v^2) Vt^2]
which is the expression I provided.
So its correct :-)
Surfer
>Surfer wrote:
>> On Sat, 26 Apr 2008 15:12:03 GMT, Tom Roberts
>> <tjrobe...@sbcglobal.net> wrote:
>>
>>> Surfer wrote:
>>>> The following is an analysis of gas mode MM interferometer operation
>>>> using standard SR formulae. The result is consistent with a formula
>>>> given in:
>>>> "The Michelson and Morley 1887 Experiment and the Discovery of
>>>> Absolute Motion"
>>>> http://arxiv.org/abs/physics/0508174
>>> But this does not take the errorbars into account!
>>> Not a single figure has errorbars.
>>>
>> They are not necessary in the above paper, because it only concerns
>> the initial discovery.
>>
>
>I love this. If it can be used to support your hope for
>something spectacular, it is ok to ignore the error bars
>and thus misrepresent the facts.
Miller who could directly investigate sources of measurement error in
his equipment, estimated probable errors as follows.
From Page 218 of Miller's paper:
http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
"...A study of numerical results as plotted in Fig. 26 shows that the
probable error of the observed velocity, which has a magnitude of from
ten to eleven kilometers per second, is +/- 0.33 kilometer per second,
while the probable error in the determination of the azimuth is +/-
2.5%. The probably error in the right ascensions and declinations of
the polar chart, Fig. 28, is +/- 0.5%"
I take those estimates seriously (this is from a physics paper
published in a peer reviewed journal), so I am not ignoring error bars
or misrepresenting the facts.
Jerry
> On Sat, 26 Apr 2008 19:01:14 -0500, Tom Roberts
> >That's complete and utter nonsense! To claim a "discovery" you need a
> >SIGNIFICANT result, and Michelson and Morley did not have one.
>
> If the result had not been SIGNIFICANT, there would have been no need
> to try to explain the result by introducing the Lorentz-FitzGerald
> contraction hypothesis.
> http://en.wikipedia.org/wiki/Lorentz-FitzGerald_contraction_hypothesis
As Tom Roberts might say, you have just made an atrocious PUN on
the word "significant."
A statistical hypothesis test is a method for making a statistical
decision about a data set. In "null hypothesis testing", the
question to be answered is, "what is the probability that the
observed results might be due simply to chance factors?"
For the MMX experiment, the null hypothesis would be, "there is
no anisotropy in the speed of light measured in different
directions."
One uses statistical testing to determine whether the available
data is sufficient to reject the null hypothesis. In other words,
is the distribution of data such that it is highly unlikely that
the observed deviations from the the null hypothesis could have
arisen by chance?
If the null hypothesis cannot be rejected, the results are not
significant in sense 1 of the word "significant."
The insignificant sense 1 result of the MMX experiment was a
highly significant sense 2 result, because it was completely
unexpected.
Jerry
Surfer
Statistical testing has its place, but the physics of an MM
interferometer is deterministic. Measurement errors can be determined
by testing, as was done by Miller.
From Page 218 of Miller's paper:
http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
"...A study of numerical results as plotted in Fig. 26 shows that the
probable error of the observed velocity, which has a magnitude of from
ten to eleven kilometers per second, is +/- 0.33 kilometer per second,
while the probable error in the determination of the azimuth is +/-
2.5%. The probably error in the right ascensions and declinations of
the polar chart, Fig. 28, is +/- 0.5%"
Cahill's success in using Miller's results to resolve the spacecraft
earth flyby anomalies, suggest to me that these are the true estimates
of the probable errors.
I showed elsewhere that Tom Robert's statistical analysis was based on
false premises.
Surfer
In the interest of self consistency:
Are you going to reject Boyles law because it was discovered without
errorbars?
http://en.wikipedia.org/wiki/Boyle's_Law
"The relationship between pressure and volume was brought to the
attention of Boyle by two friends and amateur scientists, Richard
Towneley and Henry Power, who discovered it."
Surfer
<Headm...@Hogwarts.physics> wrote:
>
>Did you snip something I wrote because you are an ignorant cretin
>who cannot face the truth?
I just wanted to focus on one point at a time.
If we live in a universe in which measuring equipment experiences time
dilation and length contraction, then the "measured speed of light"
will not be the same as the speed that would be measured if those
effects didn't happen.
[Surfer]
>|
>| The second postulate is, "Light is always propagated in empty space
>| with a definite velocity c that is independent of the state of motion
>| of the emitting body"
>
[Androcles]
>Yes I know. That's a rather stupid postulate, unsupported by observation.
>
[Surfer]
What the postulate really means is that when the speed of light is
_measured_ in an inertial frame of reference, the result is always
found to be c.
That is supported by observation.
I believe that is what observations reveal.
Considering that measuring equipment is subject to relativistic
effects, do you agree or disagree?
Dono
> I had a look.
>
> In my analysis:
> Vtran is the speed of light relative to 3-space (Sigma in your paper)
> Vt is the speed of the same light relative to the lab.
>
> In your analysis:
> cs is the speed of light relative to Sigma
> c0 is the speed of the same light relative to the lab
>
> Expression 2.11 in your analysis is
>
> cs^2 -v^2 = c0^2/gamma^2
>
> Re-arranging:
> cs = Sqrt[v^2 + (1/gamma)^2 c0^2]
>
> Substituting Vtran for cs, Vt for c0 and 1/Sqrt[1-v^2] for gamma
> gives:
>
> Vtran = Sqrt[v^2 + (1-v^2) Vt^2]
>
> which is the expression I provided.
>
> So its correct :-)
No shithead, it is not correct
If it were, you would have obtained the correct prediction , NULL.
Dono
>
> I had a look.
>
> In my analysis:
> Vtran is the speed of light relative to 3-space (Sigma in your paper)
> Vt is the speed of the same light relative to the lab.
>
> In your analysis:
> cs is the speed of light relative to Sigma
> c0 is the speed of the same light relative to the lab
>
> Expression 2.11 in your analysis is
>
> cs^2 -v^2 = c0^2/gamma^2
>
> Re-arranging:
> cs = Sqrt[v^2 + (1/gamma)^2 c0^2]
>
> Substituting Vtran for cs, Vt for c0 and 1/Sqrt[1-v^2] for gamma
> gives:
>
> Vtran = Sqrt[v^2 + (1-v^2) Vt^2]
>
> which is the expression I provided.
>
> So its correct :-)
But, shithead
If you got all the way to 2.11, you could see that applying it
correctly (2.13, two lines below) shows a NULL result. This is NOT
what you got.
Dirk Van de moortel
"Surfer" <n...@spam.net> wrote in message news:j0r81410aqodt9dhj...@4ax.com...
> On Sat, 26 Apr 2008 23:03:52 +0200, "Dirk Van de moortel"
> <dirkvand...@ThankS-NO-SperM.hotmail.com> wrote:
>
>>
>>"Surfer" <n...@spam.net> wrote in message news:crl614djptl55m15e...@4ax.com...
>>> On Sat, 26 Apr 2008 15:12:03 GMT, Tom Roberts
>>> <tjrobe...@sbcglobal.net> wrote:
>>>
>>>>Surfer wrote:
>>>>> The following is an analysis of gas mode MM interferometer operation
>>>>> using standard SR formulae. The result is consistent with a formula
>>>>> given in:
>>>>> "The Michelson and Morley 1887 Experiment and the Discovery of
>>>>> Absolute Motion"
>>>>> http://arxiv.org/abs/physics/0508174
>>>>
>>>>But this does not take the errorbars into account!
>>>>Not a single figure has errorbars.
>>>>
>>> They are not necessary in the above paper, because it only concerns
>>> the initial discovery.
>>
>>"Errorbars not necessary for initial discovery":
>> http://users.telenet.be/vdmoortel/dirk/Physics/Fumbles/DiscoveryBars.html
>>Good one, Reginald.
>>
> In the interest of self consistency:
>
> Are you going to reject Boyles law because it was discovered without
> errorbars?
I am going keep having a big laugh over you, Reginald,
Deputy Head of School
School Courses and Curricula (Chairman)
School Space Committee (Chairman
Janitor's Advisor
Toilet cleaner
Lawn mower
Dirk Vdm
Tom Roberts
> On Sat, 26 Apr 2008 19:01:14 -0500, Tom Roberts
> <tjrobe...@sbcglobal.net> wrote:
>> Surfer wrote:
>>> On Sat, 26 Apr 2008 15:12:03 GMT, Tom Roberts
>>> <tjrobe...@sbcglobal.net> wrote:
>>>> Surfer wrote:
>>>>> "The Michelson and Morley 1887 Experiment and the Discovery of
>>>>> Absolute Motion"
>>>>> http://arxiv.org/abs/physics/0508174
>> SIGNIFICANT result, and Michelson and Morley did not have one.
>>
> If the result had not been SIGNIFICANT, there would have been no need
> to try to explain the result by introducing the Lorentz-FitzGerald
> contraction hypothesis.
You're not paying attention. THEY did not know this, because the
knowledge and techniques I applied were developed after all these
physicists died. Quantitative error analysis did not become commonplace
until the 1950's, and you cannot expect Michelson, Morley, Miller,
Fitzgerald, or Lorentz to be clairvoyant.
But a glance at the calendar shows that YOU have no such excuse. And
there is no reason TODAY to believe any of these experiments have a
non-null result, because we KNOW their results are not significantly
different from null.
Just look at the difference in the errorbars between Fig. 5 (Miller's
original analysis) and Fig. 11 (my modern analysis of the same data)
from my paper http://arxiv.org/abs/physics/0608238 . This 50-fold
reduction of errorbars in the analysis of Miller's data is due to MODERN
KNOWLEDGE AND TECHNIQUES.
> In fact the result was sufficiently significant to bring about a
> revolution in physics.
Yes, because the results did not agree with the predictions of the
then-current theory. They were luckier than Miller, primarily due to a
better experimental design for temperature stability, and their
systematic drift was small enough that they chose to just claim an upper
limit of 7.5 km/s, much smaller than the minimum expected value of 30 km/s.
I contented myself with an analysis of their data showing they were
consistent with a null result. Had I carried it through to an upper
bound on "absolute motion", I'm pretty sure it would have been less than
30 km/s, so their basic conclusion would hold up.
> That being the case, the experiment is worth re-examining.
Yes. BUT YOU MUST KNOW WHAT YOU ARE DOING! You and Cahill clearly do
not. Your "conclusion" is inconsistent with modern knowledge of basic
experimental physics.
>> I flipped this coin 8 times, and it came up heads 6 times.
>> Have I "discovered" an anomaly in the laws of physics, or in
>> the laws of statistics, or in the balance of this coin? --
>> OF COURSE NOT! Because the deviation from 50% heads is not
>> significant! You are claiming this would be a "discovery".
>
> If the MM experiment was like that, people would have ignored the
> result and stuck with ether theory :-)
It _WAS_ at least in part like that, but THEY did not know it. And they
were fortunate to get the answer that led the way to modern physics.
Their conclusion does hold up to modern scrutiny.
This example does show that your claims about not needing errorbars for
"discovery" are ridiculous. Indeed, essentially NONE of your claims hold
up to modern scrutiny. Michelson, Morley, and Miller were all competent
experimental physicists in their day, and I have no doubt that were they
alive today they would fit right in and LEARN the necessary techniques.
You and Cahill, on the other hand, have demonstrated a complete
inability to do so. Sad.
Tom Roberts
Tom Roberts
> Miller who could directly investigate sources of measurement error in
> his equipment, estimated probable errors as follows.
>
> From Page 218 of Miller's paper:
>
> I take those estimates seriously (this is from a physics paper
> published in a peer reviewed journal), so I am not ignoring error bars
> or misrepresenting the facts.
Neither you nor Miller take into account the errorbar inherent in
AVERAGING THE DATA. It is ENORMOUS, and completely overwhelms the other
contributions to errors. Miller did not know to include it, but you and
Cahill have no such excuse. Had you any knowledge of basic experimental
physics, you would know this, and cease repeating your mistakes. Any
undergraduate physics major knows this technique.
Tom Roberts
Tom Roberts
> Statistical testing has its place, but the physics of an MM
> interferometer is deterministic.
More nonsense. Just LOOK at their data: it varies turn-by-turn VASTLY
more than the variation in the average. The STATISTICS of their multiple
measurements at a given orientation show CONCLUSIVELY that there is no
significant orientation dependence in their data. This is true for
Michelson-Morley, for Miller, and for Illingworth, as shown in the
appendices of my paper.
As I keep saying: you and Cahill REALLY need to learn the basics.
Tom Roberts
Surfer
wrote:
>
>No shithead, it is not correct
>If it were, you would have obtained the correct prediction , NULL.
>
If you read my post carefully you would have seen.
<Start extract>
The effective refractive index of the gas is assumed to vary slightly
as follows:
In the 3-space frame, the symmetry of paths AC and CA ensure that the
speed of light relative to the gas is the same in both paths. Take the
effective refractive index for this path as being a reference value of
(n).
In the 3-space frame, light travelling along path AB is _catching_up_
with the gas. Suppose this slightly changes the effective
refractive index for that light to a value of (n-x).
In the 3-space frame, light travelling along path BA is _heading_into_
the gas. Suppose this slightly changes the effective refractive index
for that light to a value of (n+x+d).
<End extract>
After doing the maths it turns out that
Delta_t = d L/Sqrt[1-v^2]
In other words fringe shifts are predicted, IF and ONLY IF the
effective refractive index VARIES in such a way that 'd' is NON-ZERO.
If you want to get the standard SR result, all you need to do is
ASSUME that the effective refractive index DOESN'T VARY. If you do
that, then x and d will be zero and the formula will give NULL.
So the occurance or otherwise of fringe shifts doesn't depend on SR
theory in this case, but rather on how motion through 3-space affects
the refractive index of gases.
From Cahill's formula for Delta_t, it is trivial to deduce that:
d = n(n^2 - 1) v^2
So another way to get the standard SR result is to replace the gas
with vacuum. Then n==1, causing d to be zero, thus again ensuring no
fringe shifts.
Note that if the formula for d holds equally well in ALL
inertial reference frames, then the Principle of Relativity:
"The laws of physical phenomena are the same in all inertial
references frames".
still holds.
So this result is compatible with SR.
Surfer
wrote:
>On Apr 27, 2:12 am, Surfer <n...@spam.net> wrote:
I explained why in my answer to your post that was dated
Surfer
<tjrobe...@sbcglobal.net> wrote:
>Surfer wrote:
>> Statistical testing has its place, but the physics of an MM
>> interferometer is deterministic.
>
>More nonsense. Just LOOK at their data: it varies turn-by-turn VASTLY
>more than the variation in the average.
>
That is sometimes true. But at other times, variation with direction
is clearly visible.
Fig. 4 in Cahill's paper is almost perfect.
http://arxiv.org/abs/physics/0508174
>
> The STATISTICS of their multiple
>measurements at a given orientation show CONCLUSIVELY that there is no
>significant orientation dependence in their data.
>
Since I identified several false premises in your paper, I think we
are better off to believe Miller and Cahill.
Furthermore, Maurice Allias, a French physicist and Nobel Laureate in
economics (so he had excellent knowledge of statistics) demonstrated
that Miller's results were NOT null.
See:
"The Experiments of Dayton C. Miller (1925-1926) and the Theory of
Relativity" 21st century Science & Technology, Spring 1998, p. 26
Maurice Allais
http://allais.maurice.free.fr/English/media12-1.htm
and
"The origin of the very significant regularities displayed in the
interferometric observations of Dayton C. Miller 1925-1926:
temperatures effects or space anisotropy ?"
Note dated December 2000, sent by Professor Allais to the French
Academy of Sciences In this paper, Maurice Allais refutes all the
criticisms opened about the Miller's works, from Shankland (1955) up
to now.
http://allais.maurice.free.fr/English/media16-1.htm
Surfer
<tjrobe...@sbcglobal.net> wrote:
>Surfer wrote:
>> On Sat, 26 Apr 2008 19:01:14 -0500, Tom Roberts
>> <tjrobe...@sbcglobal.net> wrote:
>>> Surfer wrote:
>>>> On Sat, 26 Apr 2008 15:12:03 GMT, Tom Roberts
>>>> <tjrobe...@sbcglobal.net> wrote:
>>>>> Surfer wrote:
>>>>>> "The Michelson and Morley 1887 Experiment and the Discovery of
>>>>>> Absolute Motion"
>>>>>> http://arxiv.org/abs/physics/0508174
>>> To claim a "discovery" you need a
>>> SIGNIFICANT result, and Michelson and Morley did not have one.
>>>
>> If the result had not been SIGNIFICANT, there would have been no need
>> to try to explain the result by introducing the Lorentz-FitzGerald
>> contraction hypothesis.
>
>You're not paying attention. THEY did not know this, because the
>knowledge and techniques I applied were developed after all these
>physicists died.
>
Your comments would be relevant if your error analysis was correct.
But as I showed earlier, your paper contained several false premises.
No matter how modern your technique, if your premises are wrong your
conclusions will be wrong.
Surfer
<tjrobe...@sbcglobal.net> wrote:
>Surfer wrote:
>> Miller who could directly investigate sources of measurement error in
>> his equipment, estimated probable errors as follows.
>>
>> From Page 218 of Miller's paper:
>> [...]
>>
>> I take those estimates seriously (this is from a physics paper
>> published in a peer reviewed journal), so I am not ignoring error bars
>> or misrepresenting the facts.
>
>Neither you nor Miller take into account the errorbar inherent in
>AVERAGING THE DATA. It is ENORMOUS,
>
I don't find that credible because:
1) I spotted false premises in your paper, and,
2) There is no way to confirm if your paper analysed the data
that Miller actually used in his final results.
The run that you analysed could have been one of the very
many that Miller discarded. When Miller used a run, he used all the
data in it, but he only used a small proportional of all the runs he
conducted.
Dono
>
> Define variables for light speeds relative to the interferometer frame
> as follows.
>
> Vo = 1/(n-x) -- the outward speed along AB
> Vr = 1/(n+x+d) -- the return speed along BA
> Vt = 1/n -- the transverse speed along AC and CA
>
If your grandmother had wheels she would have been a ...trolleybus.
You are pulling the anisotropy from your ass and then , when searching
for it in your ass, you...find it!.
>
> Consider path ABA
> In the 3-space frame, Lorentz contraction causes the length of the AB
> arm to be:
>
> Lc = L Sqrt[1-v^2]
>
> Hence the times taken by light to traverse the arm is:
>
> tAC = L/Sqrt[Vtran^2-v^2]
>
> Delta_t = tABA - tACA
>
> = L (Vt (Vo+Vr) - 2 Vo Vr) / (Vt Vo Vr Sqrt[1-v^2])
>
> If Vo, Vr and Vt were all equal, as would ordinarily be
> assumed, then Delta_t would be zero.
Good, you can stop here, since this is what numerous non-crackpot
experiments have already had confirmed.
> However, if we substitute the
> refractive index dependent values defined above, then after
> simplification we get:
>
> Delta_t = d L/Sqrt[1-v^2]
>
However, if your grandmother was born with wheels, she would be
running on the no.4 trolleybus line.
It doesn't work this way, if yoy assume an absurdity, you will get
an ...absurdity.
Come on Rag, give it up, this is getting really lame.
<rest of the nonsense mercifully snipped>
Dono
>
> No matter how modern your technique, if your premises are wrong your
> conclusions will be wrong.
Yes, you are right, this applies perfectly to the absurd premises of
your writeup. Do you know the "GiGo principle" ?
Surfer
wrote:
>On Apr 26, 12:53 am, Surfer <n...@spam.net> wrote:
>
>>
>> Define variables for light speeds relative to the interferometer frame
>> as follows.
>>
>> Vo = 1/(n-x) -- the outward speed along AB
>> Vr = 1/(n+x+d) -- the return speed along BA
>> Vt = 1/n -- the transverse speed along AC and CA
>>
>
>If your grandmother had wheels she would have been a ...trolleybus.
>You are pulling the anisotropy from your ass and then , when searching
>for it in your ass, you...find it!.
>
If you can't prove something wrong, why challange it?
By attempting to insult, you just make a spectacle of yourself.
Tom Roberts
> Since I identified several false premises in your paper, I think we
> are better off to believe Miller and Cahill.
There is no "premise" in the FACT that when one averages data, the
variance of the average is related to the variance of the data that were
averaged. You cannot possibly escape this FACT. And it shows that for
Michelson and Morley, for Miller, and for Illingworth, the "orientation
dependence" you (and others) claim to see in the AVERAGES of their data
are significantly exceeded by the variance in those AVERAGES.
Get an education. Then you won't have to "believe"....
This is going nowhere. Don't expect me to respond until you actually
LEARN something about the subject.
Tom Roberts
Dono
> On Sun, 27 Apr 2008 15:21:35 -0700 (PDT), Dono <sa...@comcast.net>
> wrote:
>
> >On Apr 26, 12:53 am, Surfer <n...@spam.net> wrote:
>
> >> Define variables for light speeds relative to the interferometer frame
> >> as follows.
>
> >> Vo = 1/(n-x) -- the outward speed along AB
> >> Vr = 1/(n+x+d) -- the return speed along BA
> >> Vt = 1/n -- the transverse speed along AC and CA
>
> >If your grandmother had wheels she would have been a ...trolleybus.
> >You are pulling the anisotropy from your ass and then , when searching
> >for it in your ass, you...find it!.
>
> If you can't prove something wrong, why challange it?
> By attempting to insult, you just make a spectacle of yourself.
Because you are both an idiot and a cheat:
1. There is no reason to assume light speed anisotropy in the lab
frame (which you clearly pulled out from your ass)
2. Once that you assume light speed anisotropy in the lab frame, why
are you surprised that your shit experiment predicts....light speed
anisotropy!
You share the same cheating style with "Rag" Cahill :-)
Dono
> wrote:
>
> >On Apr 26, 12:53 am, Surfer <n...@spam.net> wrote:
>
> >> Define variables for light speeds relative to the interferometer frame
> >> as follows.
>
> >> Vo = 1/(n-x) -- the outward speed along AB
> >> Vr = 1/(n+x+d) -- the return speed along BA
> >> Vt = 1/n -- the transverse speed along AC and CA
>
> >If your grandmother had wheels she would have been a ...trolleybus.
> >You are pulling the anisotropy from your ass and then , when searching
> >for it in your ass, you...find it!.
>
> If you can't prove something wrong, why challange it?
> By attempting to insult, you just make a spectacle of yourself.
I can understand that you are getting frustrated by your cheap cheats
getting exposed :-)
Surfer
wrote:
>On Apr 27, 11:25 am, Surfer <n...@spam.net> wrote:
>
>>
>> No matter how modern your technique, if your premises are wrong your
>> conclusions will be wrong.
>
>Yes, you are right, this applies perfectly to the absurd premises of
>your writeup.
>
If you don't point them out, people won't believe you.
Dono
Cranko,
You are assuming light speed to be anisotropic in the lab frame. I
pointed this out to you several times, now you are pulling the
standard crackpot "You haven't pointed out any flaws in my paper" :-)
Surfer
<tjrobe...@sbcglobal.net> wrote:
>Surfer wrote:
>> Since I identified several false premises in your paper, I think we
>> are better off to believe Miller and Cahill.
>
>There is no "premise" in the FACT that when one averages data, the
>variance of the average is related to the variance of the data that were
>averaged.
>
True. But this is an attempt at a red herring.
>
>You cannot possibly escape this FACT.
>
True. But this is a continuation of the attempt at a red herring.
>
>And it shows that for
>Michelson and Morley, for Miller, and for Illingworth, the "orientation
>dependence" you (and others) claim to see in the AVERAGES of their data
>are significantly exceeded by the variance in those AVERAGES.
>
FALSE. That is what your analysis CLAIMs to show. But owing to TWO
false premises in your analysis it fails. The false premises are as
follows:
In:
http://www.arxiv.org/abs/physics/060823
FALSE PREMISE 1
=====================
The caption under Fig 3. says:
"The assumed-linear systematic drift from the data of Fig. 1.
The lines are between successive Marker 1 values and the points are
Marker 9. These markers are 180 degrees apart, so any real signal has
the same value for every corner and every point--the variations are
purely an instrumentation effect."
This statement is FALSE, because measurements at Marker 1 and Marker 9
were not made simultaneously. So any real FLUCTUATING signal would
have different values at the two markers.
Consequently the analysis that relies on the statement is FALSE.
======================
FALSE PREMISE 2
At the top of page 6, Tom Roberts wrote:
data = signal(orientation) + systematic(time)
The key point is that signal(orientation) is independent of time,
and for each orientation (marker) it has the same value for every
turn of the interferometer within a given data run Therefore if the
data from the first turn is subtracted marker-by-marker from the
data of every turn, the result is completely independent of any
orientation dependence, and contains only systematic(time).
The above claims are false, because any real FLUCTUATING signal would
vary with orientation AND TIME.
So in particular, the claim that:
"Therefore if the data from the first turn is subtracted marker-by-marker
from the data of every turn, the result is completely independent of any
orientation dependence, and contains only systematic(time)."
is FALSE.
Consequently all analysis that relies on the claim is FALSE.
=======================
Surfer
wrote:
>On Apr 27, 8:27 pm, Surfer <n...@spam.net> wrote:
>> On Sun, 27 Apr 2008 15:21:35 -0700 (PDT), Dono <sa...@comcast.net>
>> wrote:
>>
>> >On Apr 26, 12:53 am, Surfer <n...@spam.net> wrote:
>>
>> >> Define variables for light speeds relative to the interferometer frame
>> >> as follows.
>>
>> >> Vo = 1/(n-x) -- the outward speed along AB
>> >> Vr = 1/(n+x+d) -- the return speed along BA
>> >> Vt = 1/n -- the transverse speed along AC and CA
>>
>> >If your grandmother had wheels she would have been a ...trolleybus.
>> >You are pulling the anisotropy from your ass and then , when searching
>> >for it in your ass, you...find it!.
>>
>> If you can't prove something wrong, why challange it?
>> By attempting to insult, you just make a spectacle of yourself.
>
>
>Because you are both an idiot and a cheat:
>
>1. There is no reason to assume light speed anisotropy in the lab
>frame (which you clearly pulled out from your ass)
>
What voices are you listening to?
We have peer reviewed papers that tell us otherwise.
Ether drift experiments at Mt. Wilson
Proc. of the National Academy of Sciences of the USA
http://www.pnas.org/cgi/reprint/11/6/306
The Ether-Drift Experiment and the Determination of the Absolute
Motion of the Earth.
Rev. Mod. Phys. 5, 203 - 242 (1933)
http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
Eric Gisse
> On Sun, 27 Apr 2008 20:49:17 -0700 (PDT), Dono <sa...@comcast.net>
> wrote:
>
>
>
> >On Apr 27, 8:27 pm, Surfer <n...@spam.net> wrote:
> >> On Sun, 27 Apr 2008 15:21:35 -0700 (PDT), Dono <sa...@comcast.net>
> >> wrote:
>
> >> >On Apr 26, 12:53 am, Surfer <n...@spam.net> wrote:
>
> >> >> Define variables for light speeds relative to the interferometer frame
> >> >> as follows.
>
> >> >> Vo = 1/(n-x) -- the outward speed along AB
> >> >> Vr = 1/(n+x+d) -- the return speed along BA
> >> >> Vt = 1/n -- the transverse speed along AC and CA
>
> >> >If your grandmother had wheels she would have been a ...trolleybus.
> >> >You are pulling the anisotropy from your ass and then , when searching
> >> >for it in your ass, you...find it!.
>
> >> If you can't prove something wrong, why challange it?
> >> By attempting to insult, you just make a spectacle of yourself.
>
> >Because you are both an idiot and a cheat:
>
> >1. There is no reason to assume light speed anisotropy in the lab
> >frame (which you clearly pulled out from your ass)
>
> What voices are you listening to?
>
> We have peer reviewed papers that tell us otherwise.
>
> Ether drift experiments at Mt. Wilson
> Proc. of the National Academy of Sciences of the USAhttp://www.pnas.org/cgi/reprint/11/6/306
Not actually peer reviewed.
>
> The Ether-Drift Experiment and the Determination of the Absolute
> Motion of the Earth.
> Rev. Mod. Phys. 5, 203 - 242 (1933)http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
The only reason you continue to cite something from 1933 is that every
modern analysis debunks Miller's conclusions.
Surfer
No. I cite Miller because competent analysis supports his conclusions.
Eg:
Dono
> >> wrote:
>
> >> >On Apr 26, 12:53 am, Surfer <n...@spam.net> wrote:
>
> >> >> Define variables for light speeds relative to the interferometer frame
> >> >> as follows.
>
> >> >> Vo = 1/(n-x) -- the outward speed along AB
> >> >> Vr = 1/(n+x+d) -- the return speed along BA
> >> >> Vt = 1/n -- the transverse speed along AC and CA
>
>
> The Ether-Drift Experiment and the Determination of the Absolute
> Motion of the Earth.
> Rev. Mod. Phys. 5, 203 - 242 (1933)http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
Exactly as I said, "both" you and Rag Cahill pull out the anisotropy
from your respective butts:
>On Apr 26, 12:53 am, Surfer <n...@spam.net> wrote:
> Define variables for light speeds relative to the interferometer frame
> as follows.
> Vo = 1/(n-x) -- the outward speed along AB
> Vr = 1/(n+x+d) -- the return speed along BA
> Vt = 1/n -- the transverse speed along AC and CA
Then, surprise(!), when you look up your own butts you
discover....anisotropy!
Interesting , you two cheat the same exact way :-)
Eric Gisse
A nice statement that falls flat on its' face when you are confronted
with competent analysis.
> Eg:
>
> "The origin of the very significant regularities displayed
> in the interferometric observations of Dayton C. Miller
> 1925-1926: temperatures effects or space anisotropy ?"
> Note dated December 2000, sent by Professor Allais to the French
> Academy of Sciences In this paper, Maurice Allais refutes all the
> criticisms opened about the Miller's works, from Shankland (1955) up
> to now.http://allais.maurice.free.fr/English/media16-1.htm
Allias? Cite Podkletnov and you get the trifecta.
Tom Roberts
> [referring to Miller's 1933 paper]
> The only reason you continue to cite something from 1933 is that every
> modern analysis debunks Miller's conclusions.
Do you know of any, other than Shankland's 1955 paper, and my 2006 preprint?
Tom Roberts
Tom Roberts
> On Mon, 28 Apr 2008 03:45:28 GMT, Tom Roberts
> <tjrobe...@sbcglobal.net> wrote:
>> Surfer wrote:
>>> Since I identified several false premises in your paper, I think we
>>> are better off to believe Miller and Cahill.
>> There is no "premise" in the FACT that when one averages data, the
>> variance of the average is related to the variance of the data that were
>> averaged.
>>
> True. But this is an attempt at a red herring.
No red herring, this is directly applicable to Miller's analysis.
Just go read Miller's description of his analysis (near Fig 8 of his
1933 RMP article, p213). You'll find that AVERAGING is the primary
component of the analysis.
>> You cannot possibly escape this FACT.
>>
> True. But this is a continuation of the attempt at a red herring.
No red herring, this is directly applicable to Miller's analysis.
Just go read Miller's description of his analysis (near Fig 8 of his
1933 RMP article, p213). You'll find that AVERAGING is the primary
component of the analysis.
> FALSE PREMISE 1
> =====================
> The caption under Fig 3. says:
>
> "The assumed-linear systematic drift from the data of Fig. 1.
> The lines are between successive Marker 1 values and the points are
> Marker 9. These markers are 180 degrees apart, so any real signal has
> the same value for every corner and every point--the variations are
> purely an instrumentation effect."
>
> This statement is FALSE, because measurements at Marker 1 and Marker 9
> were not made simultaneously. So any real FLUCTUATING signal would
> have different values at the two markers.
You're not paying attention:
1) This has NOTHING WHATEVER to do with the errorbars on the averages of
Miller's data.
2) Even if you assume that the "fluctuations" are a true "signal", this
does not affect the ORIENTATION DEPENDENCE of the result. It is the
ORIENTATION DEPENDENCE OF THE AVERAGES that you and Cahill use, and THAT
is subject to the errorbars I keep mentioning and you keep ignoring.
> FALSE PREMISE 2
> At the top of page 6, Tom Roberts wrote:
> data = signal(orientation) + systematic(time)
> The key point is that signal(orientation) is independent of time,
> and for each orientation (marker) it has the same value for every
> turn of the interferometer within a given data run Therefore if the
> data from the first turn is subtracted marker-by-marker from the
> data of every turn, the result is completely independent of any
> orientation dependence, and contains only systematic(time).
>
> The above claims are false, because any real FLUCTUATING signal would
> vary with orientation AND TIME.
This is just a repeat of your error. For the ORIENTATION DEPENDENCE OF
THE AVERAGES, this is likewise not relevant when computing the ERRORBAR
DUE TO THE AVERAGING.
Just go read Miller's description of his analysis (near Fig 8 of his
1933 RMP article, p213). You'll find that AVERAGING is the primary
component of the analysis. And it ought to be easily within your
abilities to COMPUTE the corresponding errorbar on just one of those
averages. Do it! Don't keep avoiding the real issue, COMPUTE THE
ERRORBAR YOURSELF for one orientation of Miller's data from his Fig 8.
Tom Roberts
Tom Roberts
> "The origin of the very significant regularities displayed
> in the interferometric observations of Dayton C. Miller
> 1925-1926: temperatures effects or space anisotropy ?"
> Note dated December 2000, sent by Professor Allais to the French
> Academy of Sciences In this paper, Maurice Allais refutes all the
> criticisms opened about the Miller's works, from Shankland (1955) up
> to now.
> http://allais.maurice.free.fr/English/media16-1.htm
Allais did not check that the individual runs had significant
orientation dependence. He makes from the same error that Miller, you,
and Cahill make -- lack of knowledge of modern experimental physics.
Miller can be excused, but the rest of you are just plain silly.
Tom Roberts
Surfer
<tjrobe...@sbcglobal.net> wrote:
If we were to measure the values of a fluctuating signal, with perfect
accuracy, the measurement error would be zero.
If we were to average those values, then the error in the average
would still be zero.
So there is something seriously wrong with your argument.
Miller calculated the probable error in his results.
He was an experienced physicist and I see no reason to doubt him.
The established way to calculate propagation of errors is
described here:
http://science.widener.edu/svb/stats/error.html
Miller's results enabled accurate calculation of a dynamical 3-space
velocity. Here is an updated paper on how that was used to resolve the
spacecraft earth-flyby anomalies.
Resolving Spacecraft Earth-Flyby Anomalies with Measured
Light Speed Anisotropy
http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Cahill_flyby.pdf
The anomalies could not have been resolved in this way if Miller's
results had been wildly wrong.
Surfer
<tjrobe...@sbcglobal.net> wrote:
>Surfer wrote:
>> "The origin of the very significant regularities displayed
>> in the interferometric observations of Dayton C. Miller
>> 1925-1926: temperatures effects or space anisotropy ?"
>> Note dated December 2000, sent by Professor Allais to the French
>> Academy of Sciences In this paper, Maurice Allais refutes all the
>> criticisms opened about the Miller's works, from Shankland (1955) up
>> to now.
>> http://allais.maurice.free.fr/English/media16-1.htm
>
>Allais did not check that the individual runs had significant
>orientation dependence.
>
What evidence do you have that individual runs used by Miller to
derive his FINAL RESULTS didn't have significant orientation
dependence?
You have only analysed one run. It included adjustments for
temperature drift. But Miller says in one of his papers that his best
runs didn't require such adjustments.
Also, if we have a formula for absolute motion effects that holds
equally well in all inertial frames of reference, then it seems to me
the Principle of Relativity continues to hold.
So why not adopt a relaxed attitude to the possiblity of such effects?
Jerry
> On Tue, 29 Apr 2008 01:30:25 GMT, Tom Roberts
> >Allais did not check that the individual runs had significant
> >orientation dependence.
>
> What evidence do you have that individual runs used by Miller to
> derive his FINAL RESULTS didn't have significant orientation
> dependence?
>
> You have only analysed one run. It included adjustments for
> temperature drift. But Miller says in one of his papers that his best
> runs didn't require such adjustments.
A false assumption employed by Miller was the assumption of
linearity in the temperature correction needed, i.e. that he
could subtract 0, 1/16, 2/16, 3/16 etc of the difference between
the first and last readings to adjust to a constant baseline.
A careful reading of Miller's papers shows this to have been a
false assumption. A displacement of up to two fringes in a single
turn might be followed by a displacement far less than two fringes
in the next turn, or even one of different sign. With such a
wildly swingling baseline, the artificial adjustment to a constant
baseline by the subtraction of 0, 1/16, 2/16, 3/16 etc. imposes
an imaginary periodic signal on the readings.
Let's try it on the digits of pi. EXCEPT WHERE INDICATED, each
of the following steps roughly corresponds with a step taken in
Miller's analysis:
pi
3.1415926535897932384626433832795028
Take the first 17 digits after the decimal point
14159265358979323
Space each digit and add an extra zero
10 40 10 50 90 20 60 50 30 50 80 90 70 90 30 20 30
'
The difference between first and last is 20. Subtract 0, 1/16,
2/16 etc. of the difference
10 39 8 46 85 14 52 42 20 39 68 76 55 74 12 1 10
Combine the first half-turn with the second half-turn:
10 39 8 46 85 14 52 42 20
20 39 68 76 55 74 12 1 10
----------------------------------
30 78 76 122 140 88 64 43 30
14 . . . . x . . . .
13 . . . . . . . . .
12 . . . x . . . . .
11 . . . . . . . . .
10 . . . . . . . . .
09 . . . . . x . . .
08 . x x . . . . . .
07 . . . . . . . . .
06 . . . . . . x . .
05 . . . . . . . . .
04 . . . . . . . x .
03 x . . . . . . . x
02 . . . . . . . . .
01 . . . . . . . . .
00 . . . . . . . . .
Draw a smooth curve through the points:
14 . . . . __x . . . .
13 . . . ./ .\ . . . .
12 . . . x . | . . . .
11 . . . _/. . \ . . . .
10 . . _/ . . \. . . .
09 . . _/. . . x . . .
08 . x/ x . . .\_ . . .
07 . /. . . . . \. . .
06 . | . . . . . x . .
05 . | . . . . . .\__. .
04 ./ . . . . . . x__ .
03 x . . . . . . . \x
02 . . . . . . . . .
01 . . . . . . . . .
00 . . . . . . . . .
(Not a step shown in Miller's papers)
A pronounced sine wave modulation is obvious with a period of
half a revolution:
. . . . __x . . . . . . . __x . . . .
. . . ./ .\ . . . . . . ./ .\ . . . .
. . . x . | . . . . . . x . | . . . .
. . . _/. . \ . . . . . . _/. . \ . . . .
. . _/ . . \. . . . . _/ . . \. . . .
. . _/. . . x . . . . _/. . . x . . .
. x/ x . . .\_ . . . x/ x . . .\_ . . .
. /. . . . . \. . . /. . . . . \. . .
. | . . . . . x . . | . . . . . x . .
. | . . . . . .\__. . | . . . . . .\__. .
./ . . . . . . x__ ./ . . . . . . x__ .
x . . . . . . . \x . . . . . . . \x
Jerry
Eric Gisse
[...]
OK, that's cool. I didn't quite see how the signal was self-imposed
until that.
Dono
This is VERY nice!
Surfer
Nice simulation. But to duplicate Miller's results it will not be
sufficient enough to merely generate a constant sinewave.
As the earth rotates, the angle of the interferometer with respect to
the dynamical 3-space velocity vector changes.
To duplicate that you must make the phase of your sinewave pattern
vary appropriately.
Tom Roberts
> What evidence do you have that individual runs used by Miller to
> derive his FINAL RESULTS didn't have significant orientation
> dependence?
Fig 11 of my paper.
> You have only analysed one run.
READ MY PAPER.
> But Miller says in one of his papers that his best
> runs didn't require such adjustments.
His "best runs" also display MUCH less orientation dependence in the
averages, and that is STILL significantly smaller than their errorbars.
There are seven runs in my Fig 11 without any adjustments, and every one
of them has zero for this amplitude.
> Also, if we have a formula for absolute motion effects that holds
> equally well in all inertial frames of reference, then it seems to me
> the Principle of Relativity continues to hold.
>
> So why not adopt a relaxed attitude to the possiblity of such effects?
Huh???? YOU NEED TO READ MY PAPER!!!!! My analysis in section IV
specifically included the possibility of such effects (evidenced by an
orientation dependence in the data). This is not "relaxed", this is a
SERIOUS "attitude".
I do not have a "relaxed attitude" because I do science, not whatever it
is that you are trying to do.
Miller's data show no evidence of such effects.
> If we were to measure the values of a fluctuating signal, with perfect
> accuracy, the measurement error would be zero.
Yes.
> If we were to average those values, then the error in the average
> would still be zero.
NONSENSE! You _REALLY_ need to learn the basics of experimental physics.
Example: imagine the true signal is a sinewave [#], and you
measure its value at random times [@] with perfect accuracy
[$]. Obviously the true average is zero, and you obtain a
series of apparently-random values that would average to
zero if you took an infinite number of measurements. But
for a finite number of measurements the average is unlikely
to be exactly zero [%]. Standard statistical analysis will
give an errorbar from the values of your measurements, and
will almost certainly show that the average of your
measurements is not SIGNIFICANTLY different from zero (the
true average), even though the actual value of your average
is different from zero.
[#] This is what Miller's "true signal" would be.
[@] This is not what Miller did -- I'm giving an example.
[$] Nor did he do this.
[%] Clearly refuting your claim "the error in the average
would still be zero".
Grow up. Get an education. LEARN about the subject before attempting to
discuss it.
Tom Roberts
Surfer
I am referring to how errors propagate via calculations. If there are
no measurement errors to start with, there will be no error propagated
into the average.
> Example: imagine the true signal is a sinewave [#], and you
> measure its value at random times [@] with perfect accuracy
> [$]. Obviously the true average is zero, and you obtain a
> series of apparently-random values that would average to
> zero if you took an infinite number of measurements. But
> for a finite number of measurements the average is unlikely
> to be exactly zero [%]. Standard statistical analysis will
> give an errorbar from the values of your measurements, and
> will almost certainly show that the average of your
> measurements is not SIGNIFICANTLY different from zero (the
> true average), even though the actual value of your average
> is different from zero.
>
Agreed
> [#] This is what Miller's "true signal" would be.
>
If the velocity vector didn't fluctuate.
Fig 4 in
http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Cahill_flyby.pdf
shows large variations from 150 km/s up to 580 km/s in the component
of the vector in the plane of the interferometer.
>
> [@] This is not what Miller did -- I'm giving an example.
>
Agreed.
> [$] Nor did he do this.
>
Agreed. He calculated a probable error in his final results. To do
that I expect he would have determined the probable errors in raw data
values, and then determined how they would propagate through his
calculations.
>
> [%] Clearly refuting your claim "the error in the average
> would still be zero".
>
I was not referring to error in the sense that you are using it here,
but to propagated error, which would be zero.
It seems to me that if one is dealing with a fluctuating velocity
vector, then the best way to apply the method you have just
illustrated, would be to apply it to samples of that vector. That is,
calculate a velocity vector for each run and then regard that velocity
as a sample of what we are dealing with. The fact that velocity
changes while a run is in progress is awkward, but I think it is the
best that could be done in Miller's situation.
Then after collecting many samples of the velocity vector, they can be
averaged and statistics calculated in the way you suggest.
If the samples point in random directions, then the average would be
close to zero. That would be true proof of a null result.
However the result has not been null.
In 2003, this paper
http://arxiv.org/abs/physics/0306196
obtained an average speed of 415 km/s from Miller's results.
Nearly five years later, it turns out that spacecraft earth flyby
anomalies can be resolved by allowing for light speed anisotropy due
to 3-space velocities in the range 420-450 km/s.
This is in remarkable concordance with the earlier calculated average
velocity.
Surfer
<tjrobe...@sbcglobal.net> wrote:
>Surfer wrote:
>> What evidence do you have that individual runs used by Miller to
>> derive his FINAL RESULTS didn't have significant orientation
>> dependence?
>
>Fig 11 of my paper.
>
>> You have only analysed one run.
>
>READ MY PAPER.
>
My apologies. I stopped reading after I got to what I saw as the
second false premise, but you do later claim to analyse 67 runs.
Our differences depend on what we believe about the 3-space velocity
vector. You believe that any real signal would be due to a constant
velocity. Therefore fluctuations in your view are evidence of
systematic error that need to be subtracted to leave the real signal.
I believe the 3-space velocity fluctuates. Therefore any real signal
must fluctuate. Therefore subtraction of fluctuations will subtract
important components of the signal and possibly destroy it altogether.
To sample the effects of a fluctuating velocity one should ideally
measure the fringe shifts for all markers simultaneously. However this
is not possible with interferometers, because the fringe shifts at
each marker must be measured in sequence. Hence by the time one gets
to the last marker, the correlation with the first marker no longer
exists. Hence a single rotation will not be capable of showing any
orientation dependence.
However, there is a way to improvise. If you conduct say 20 turns and
then average the results for each marker, then for each marker you
will get an indication of the average effect of the velocity vector on
that marker, during the time that it took to conduct the twenty turns.
If a correlation exists between the average effects on each marker
during that period, then it will show up in the final results.
As it happened it is the procedure that Miller came to follow.
It appears to have been effective, because looking at the plots of the
processed data (eg as in your Fig. 1 taken from Miller's paper) an
orientation dependence can very clearly be seen.
Of course in order to be genuine, such signals must vary appropriately
with the rotation of the earth and its orbital motion. This has been
confirmed to be the case.
We also require that independent experiments give consistent values
(within error) for average speed and direction. This has also been
confirmed to be the case.
Jerry
> Nice simulation. But to duplicate Miller's results it will not be
> sufficient enough to merely generate a constant sinewave.
> As the earth rotates, the angle of the interferometer with respect to
> the dynamical 3-space velocity vector changes.
>
> To duplicate that you must make the phase of your sinewave pattern
> vary appropriately.
You asked for it, you got it.
I found that I could greatly improve the appearance of the periodic
waves in my simulation immensely by using, not the raw digits of pi,
but rather using the digits to generate a Markov chain. Another
improvement was to average twenty turns of data rather than merely
a single turn as I did in my initial simulation.
Four runs were performed at the latitude/longitude of North Dallas
on 5/1/2008. Remarkable internal consistencies were apparent in the
collected data, which clearly indicate that I was measuring a true
signal resulting from the movement of the Earth through pi-aether.
The runs conducted at 6:00 AM and 6:00 PM showed much larger signals
than the runs conducted at midnight and noon, and the magnitudes
were consistent with each other.
From my preliminary data, it should not be too difficult to calculate
the absolute direction in which Earth is moving through pi-aether.
====================================================================
Digits of pi after the decimal point
1 4 1 5 9 2 6 5 3 5 8 9 7 9 3 2 3
Markov chain generated using digits of pi
S[0] = 0
S[n] = S[n-1] - pi[n] + 5
4 5 9 9 5 8 7 7 9 9 6 2 0 -4 -2 1 3
====================================================================
Run 1 using digits 1 through 361
The value in col 17 is the same as the value in col 1 on the next row
4 5 9 9 5 8 7 7 9 9 6 2 0 -4 -2 1 3
3 0 1 0 3 2 3 5 7 4 6 9 7 3 3 8 11
11 8 5 6 10 6 4 8 7 3 5 1 -3 -1 -3 -3 1
1 6 6 3 6 11 7 5 6 2 3 4 4 0 3 5 10
10 8 5 9 8 9 14 13 16 13 12 15 20 17 13 9 6
6 5 8 5 10 12 13 10 13 13 15 16 19 23 27 25 30
30 29 27 23 20 23 27 28 25 30 27 26 26 30 32 35 32
32 35 37 42 41 40 41 39 44 40 42 39 40 41 40 45 41
41 41 41 46 46 43 46 49 51 55 53 56 56 58 58 54 55
55 60 57 61 64 61 62 59 63 67 71 69 70 70 75 78 75
75 76 80 85 88 86 91 95 91 93 90 90 93 97 101 106 106
106 106 106 102 101 102 103 102 105 108 104 105 102 98 98 99 95
95 97 102 104 101 105 101 100 101 102 105 102 99 103 108 104 102
102 102 101 100 100 96 98 100 101 102 101 105 108 105 106 104 104
104 103 104 101 104 106 108 106 103 102 100 97 99 103 102 102 105
105 103 107 110 115 119 115 120 116 120 121 121 120 121 118 118 117
117 116 112 115 117 118 117 122 124 125 122 121 125 130 131 131 132
132 134 137 136 135 136 133 136 140 142 144 140 142 141 146 144 147
147 146 151 154 155 151 155 156 160 163 161 163 161 164 165 165 162
162 160 165 170 169 168 173 172 174 178 178 178 175 172 176 174 175
Take the mean of each column. The math is being performed using full
double-precision, even though space limitations only allow me to
display to the nearest integer.
67 67 68 69 70 70 71 72 73 74 73 73 73 74 75 75 75
The difference between the last column (75.45) and the first column
(66.90) is 8.55
Subtract 0, 1/16, 2/16, 3/16 etc. of 8.55
67 66 67 67 68 67 68 68 69 69 68 67 67 67 67 67 67
Combine the first half-turn with the second half-turn
135.43 135.21 134.94 134.52 134.50 134.03 135.06 135.04 135.43
Normalize the summed values to zero
0.70 0.48 0.21 -0.21 -0.23 -0.70 0.33 0.32 0.70
Plot the normalized values
====================================================================
Run 1 using digits 1 through 361 conducted 5/1/2008 00:00 (Midnight)
x . . . . . . . -x . . . . . . . -x
\ / \ /
. x . . . . . . . x . . . . . . .
\ / \ /
. . x . . . x x . . x . . . x x .
\ / /
. . . \ . . . . . . . . \ . . . . . .
\ / /
. . . \ . . . . . . . \ . . . . .
\ / /
. . . x \ x . . . . . . x \ x . . . .
\ / /
. . . . \ . . . . . . . \ . . . .
\ / \ /
. . . . . --x . . . . . . --x . . .
|North |West |South |East |N
====================================================================
Run 2 using digits 401 through 761 conducted 5/1/2008 06:00
. . . . . x . x . . . . . x . x .
___ ___
. . . . . . . . . . . . . . . . .
/ \ / \
. . . . . . . . . . . . . . . . .
/ \ / \
. . . . . . x . . . . . . . x . .
/ \ / \
. . x . x . . . . . x . x . . . .
/ \ / \
. . . ./ . . . . . . . ./ . . . . .
/ \ / \
. . . /x . . . . . . . x . . . . .
/ \ / \
. . . . . . . . . . . . . . . . .
/ \ / \
. . . . . . . . . . . . . . . . .
/ /
x . . . . . . . x . . . . . . . x
/ /
. . . . . . . . . . . . . . . . .
\ / \ /
. . . . . . . . . . . . . . . . .
\ / \ /
. . . . . . . . . . . . . . . . .
\ / \ /
. . . . . . . . . . . . . . . . .
\ / \ /
. . . . . . . . . . . . . . . . .
. x . . . . . . . x . . . . . . .
|North |West |South |East |N
====================================================================
Run 3 using digits 801 through 1161 conducted 5/1/2008 12:00 (noon)
. . . . _ x . . . . . . . _ x . . . .
/ \ / \
. . . . . . . . . . . . . . . . .
/ \ / \
. . . . . . . . . . . . . . . . .
/ \ / \
. . . . . . x . . . . . . . x . .
/ \ / \
. . x . . x . . . . x . . x . . .
/ \ / \
. . . . . . . . . . . . . . . . .
/ \ / \
. . . x . . . . . . . x . . . . .
/ \ / \
. . . . . . . . . . . . . . . . .
/ \ / \
. . . . . . . x . . . . . . . x .
/ \ / \
x __ x . . . . . . x __ x . . . . . . x
|North |West |South |East |N
====================================================================
Run 4 using digits 1200 through 1561 conducted 5/1/2008 18:00
x . . . . . . . x . . . . . . . x
/ /
\ . . . . . . / \ . . . . . . / \
.\ . . . . . . /. .\ . . . . . . /. .
.\ . . . . . . / x .\ . . . . . . / x .
. \ . . . . . x/ . . \ . . . . . x/ . .
. \. . . . . / . . \. . . . . / . .
. \. . . . . /. . . \. . . . . /. . .
. \. . . . . / . . . \. . . . . / . . .
. \ . . . ./ . . . \ . . . ./ . . .
. x . . . x . . . x . . . x . . .
. .\ . . . / . . . .\ . . . / . . .
. . \ . . . /. . . . . \ . . . /. . . .
. . x . . /. . . . . x . . /. . . .
. . .\ x . /. . . . . .\ x . /. . . .
. . . \ . . / . . . . . . \ . . / . . . .
. . . \. . / . . . . . . \. . / . . . .
. . . \ ./ . . . . . . \ ./ . . . .
. . . .\ ./ . . . . . . .\ ./ . . . .
\ \
. . . . x . . . . . . . x . . . .
|North |West |South |East |N
====================================================================
Jerry
Jerry
> Run 1 using digits 1 through 361
Correction:
As should be obvious, the runs used digits 1 through 321,
401 through 721, 801 through 1121, and 1200 through 1521.
Jerry
Dono
>
> If there are no measurement errors to start with, there will be no error propagated
> into the average.
>
Came on, Peter, give it up, you are getting lame in your denials. How
can there be "no measurement errors".
No, don't answer that, you will only embarass yourself only further
with your permanent weaseling :-)
Dono
>
> Four runs were performed at the latitude/longitude of North Dallas
> on 5/1/2008. Remarkable internal consistencies were apparent in the
> collected data, which clearly indicate that I was measuring a true
> signal resulting from the movement of the Earth through pi-aether.
> The runs conducted at 6:00 AM and 6:00 PM showed much larger signals
> than the runs conducted at midnight and noon, and the magnitudes
> were consistent with each other.
>
> From my preliminary data, it should not be too difficult to calculate
> the absolute direction in which Earth is moving through pi-aether.
>
You should publish this :-)
It is brilliant!
Surfer
wrote:
>....
You guessed wrong again. Does this mean anything to you?
"Great minds discuss ideas. Average minds discuss events. Small minds
discuss people." -- Eleanor Roosevelt
Dono
No, this time I know.
Dono
> wrote:
>
> >On Apr 30, 1:28 am, Peter R.F. Brown <n...@spam.net> wrote:
>
> ....
>
> You guessed wrong again. Does this mean anything to you?
Not really, see here:
http://www.mountainman.com.au/cv_in_it.htm
Jerry
The truly wonderful thing about analysis of the pi-aether data is
that the digits of pi are known EXACTLY. Error analysis is hence
totally unnecessary, because every summation, every average, every
plotted point is known to infinite precision. Even Miller's data
is not that accurate!
Since the error bars are infinitesimal, the apparent fluctuations
in the data that lead to deviations of the plotted points from
perfect smoothness represent REAL signals and can not be dismissed
as measurement error. Rather, they represent fluctuations in the
dynamical vacuum pi-background. As Surfer has so aptly stated,
"If we were to measure the values of a fluctuating signal, with
perfect accuracy, the measurement error would be zero."
Discovery of the dynamical vacuum pi-background is a major, major
event! I will definitely thank Cahill and Surfer in my forthcoming
acceptance speech in Stockholm for their inspired guidance.
Jerry
Dono
I think "mountainMan" will take this straight to Cahill. Dang! You
scooped Reginald!
Dirk Van de moortel
b342d03d-9d0f-46cf...@27g2000hsf.googlegroups.com
You have opened Pandora's box here.
Do you realize that when you replace pi with e, you can
materialize Ken Seto's E-matrix ?
He will be thrilled - provided someone manages to explain
to him first what numbers are.
Dirk Vdm
Jerry
I am happy to report my preliminary determination of the direction
of Earth's absolute motion through the pi-aether, made with the
help of the Fourmilab Your Sky interactive planetarium
http://www.fourmilab.ch/cgi-bin/Yoursky
Later on I plan to perform a quantitative determination of the
direction and magnitude of absolute motion using the formulas on
pp. 225-226 of Miller's 1933 review, but I'm preparing for final
exams right now, so my rough determinations using the planetarium
software will have to do.
The existing data is sufficient to narrow down the motions to
one of two opposing vectors:
1) A quadrangle centered around the constellation of Draco,
roughly delimited by Ursa Minor, Bootes, and Cygnus, the most
probable aim point being RA 18h 30m DEC +63 degrees within the
body of Draco, or
2) The opposite vector RA 6h 30m DEC -63 degrees within the
constellation of Volans.
By an absolutely marvelous coincidence, and I AM NOT KIDDING,
THIS IS REAL, I DID NOT FUDGE MY FIT TO MATCH MILLER'S DATA,
Miller reported in Science 63:436(1926) a probable apex of solar
motion of RA 17h DEC +65 degrees towards the head of Draco,
alternatively by 1933 he had decided that his data were more
consistent with the southern apex of RA 4h 54m DEC -70d 33m
within the constellation of Dorado, which is right next to
Volans!
Unlike Miller, I did not collect sufficient data to distinguish
between the effects of solar motion through the pi-aether versus
Earth's movements about the Sun, but the preliminary matchup
between my results and Miller's results is truly remarkable.
Jerry
Dono
> On May 2, 10:51 pm, Dono <sa...@comcast.net> wrote:
>
>
>
>
>
> > On May 2, 8:22 pm, Jerry <Cephalobus_alie...@comcast.net> wrote:
> > > The truly wonderful thing about analysis of the pi-aether data is
> > > that the digits of pi are known EXACTLY. Error analysis is hence
> > > totally unnecessary, because every summation, every average, every
> > > plotted point is known to infinite precision. Even Miller's data
> > > is not that accurate!
>
> > > Since the error bars are infinitesimal, the apparent fluctuations
> > > in the data that lead to deviations of the plotted points from
> > > perfect smoothness represent REAL signals and can not be dismissed
> > > as measurement error. Rather, they represent fluctuations in the
> > > dynamical vacuum pi-background. As Surfer has so aptly stated,
> > > "If we were to measure the values of a fluctuating signal, with
> > > perfect accuracy, the measurement error would be zero."
>
> > > Discovery of the dynamical vacuum pi-background is a major, major
> > > event! I will definitely thank Cahill and Surfer in my forthcoming
> > > acceptance speech in Stockholm for their inspired guidance.
>
> > > Jerry
>
> > I think "mountainMan" will take this straight to Cahill. Dang! You
> > scooped Reginald!
>
> I am happy to report my preliminary determination of the direction
> of Earth's absolute motion through the pi-aether, made with the
>
> - Show quoted text -
You absolutely have to publish this ! It is absolutely hilarious. If
this will not stop the idiots that keep arguing about Dayton Miller
and Allais, nothing will.
Dirk Van de moortel
b8744c1f-7ae3-4525...@f24g2000prh.googlegroups.com
I agree 100%.
> If
> this will not stop the idiots that keep arguing about Dayton Miller
> and Allais, nothing will.
Nothing will.
Dirk Vdm
Tom Roberts
> Dono <sa...@comcast.net> wrote in message
> b8744c1f-7ae3-4525...@f24g2000prh.googlegroups.com
>> On May 4, 1:57 am, Jerry <Cephalobus_alie...@comcast.net> wrote:
>>> I am happy to report my preliminary determination of the direction
>> You absolutely have to publish this ! It is absolutely hilarious.
> I agree 100%.
The difficulty is where to publish it. I doubt any serious physics
journal will be interested, as it does not really present any new
physics. An interesting possibility is Apieron (!), and I bet you can
rather easily meet their criteria. This has the advantage that the
target audience might actually see it (I doubt any of them read Phys.
Rev. :-). I have not read your article in detail, but to make it most
convincing, re-read Miller's 1933 RMP article and reproduce exactly his
analysis algorithm.
I'm wondering if there is some additional mathematical quirk that
essentially forces such an analysis to give an "answer" that is normal
to the ecliptic, and thus reasonably close to one of those
constellations. One way to explore that is to use different regions of
pi as your data -- if 10 analyses using 10 different regions all give
the same "answer" than that would be quite interesting; once you have
code set up it might not take much effort to do that.
>> If
>> this will not stop the idiots that keep arguing about Dayton Miller
>> and Allais, nothing will.
>
> Nothing will.
I agree, the idiots will keep being idiots -- "Ignorance can be cured
but stupidity is forever" [attribution lost].
Tom Roberts
Jerry
> Dirk Van de moortel wrote:
>
> > Dono <sa...@comcast.net> wrote in message
> >> You absolutely have to publish this ! It is absolutely hilarious.
> > I agree 100%.
>
> The difficulty is where to publish it. I doubt any serious physics
> journal will be interested, as it does not really present any new
> physics. An interesting possibility is Apieron (!), and I bet you can
> rather easily meet their criteria. This has the advantage that the
> target audience might actually see it (I doubt any of them read Phys.
> Rev. :-). I have not read your article in detail, but to make it most
> convincing, re-read Miller's 1933 RMP article and reproduce exactly his
> analysis algorithm.
>
> I'm wondering if there is some additional mathematical quirk that
> essentially forces such an analysis to give an "answer" that is normal
> to the ecliptic, and thus reasonably close to one of those
> constellations.
At this point, I shall have to confess a slight dodge of mine, so
that my statement written in bold caps "I AM NOT KIDDING, THIS IS
REAL, I DID NOT FUDGE MY FIT TO MATCH MILLER'S DATA" is only 99%
true. When I wrote that I performed my "runs" at 0:00, 6:00, 12:00
and 18:00, I intentionally omitted stating whether I meant "local
time" or "GMT" to double my chances of maybe having a chance
alignment with either Miller's results, the CMBR anisotropy, the
direction of the galactic center (Sagitarrius), or the direction
of the "Great Attractor" (Centaurus).
Also, I was wondering whether I could -possibly- get away with
using Standard Time if neither Daylight Saving Time nor GMT
offered any interesting correlations.
I initially set the planetarium software to GMT - 5h, i.e. to
Central Standard Time. Nothing. Then I tried GMT.
BINGO!
Jerry
Dono
Jerry
After a lot of thought, I think Apeiron is the best place.
Surfer
You need to take into account the fact that as the earth moves in its
orbit, the velocity projected onto the plane of the interferometer
will be modified by how the earths orbital velocity adds to the
velocity of 3-space in the frame of the solar system.
So you ultimately need to get an output that varies appropriately as:
1) The earth rotates on its axis, thus changing the angle with which
the 3-space velocity relative to earth is projected onto the plane of
the interferometer.
2) The earth moves in its orbit thus causing an aberration of 3-space
velocity relative to earth.
To test for 2) you would need to simulate runs for a period of a year.
That could be a good exercise.
If you got an average velocity that didn't vary during the year, that
would suggest the earth maintains a constant velocity relative to
pi-ether, irrespective of how its velocity changes relative to the
sun. That would suggest strange properties (probably unphysical) for
pi-ether.
-- Surfer
Dono
> To test for 2) you would need to simulate runs for a period of a year.
>
Yes, this is called the Kennedy-Thorndike experiment. Came back with a
null result as well.
Give it up, Peter.
Jerry
Surfer, the sad fact of the matter is that Miller performed
hundreds upon hundreds of data runs through the years and
processed the data using an algorithm that -guaranteed- that he
would get imaginary half-period modulations, even as I extracted
imaginary half-period modulations from purely random data.
Unaware of modern statistical methods, much less today's digital
signal processing techniques, and utterly convinced that the
modulations that he saw were real, he wasted decades of his life
trying to make sense of the senseless, ultimately grouping and
fitting together contradictory data into a pattern that he
imagined meaningful, and communicating his results to the world.
Even a cursory modern analysis shows the modulations to lack
statistical significance. Nevertheless the modulations were
-always- present. We have Tom Roberts to thank for explaining
the origin of these seeming regularities.
Sorry, Surfer. Only a person such as yourself, who has invested
too many years of his life chasing an utter fantasy to admit that
it was all wasted, could possibly still believe that what Miller
saw was anything real.
I understand your unwillingness to drop your illusions. But that
is all that they are.
Jerry
Surfer
If you think your half-period modulations are as good as Miller's
half-period modulations why don't you do as I suggest and show that
your pi-ether ends up with sensible properties?
If you can do that, then I would have a reason to take your claims
seriously.
>
>Unaware of modern statistical methods, much less today's digital
>signal processing techniques, and utterly convinced that the
>modulations that he saw were real, he wasted decades of his life
>trying to make sense of the senseless, ultimately grouping and
>fitting together contradictory data into a pattern that he
>imagined meaningful, and communicating his results to the world.
>Even a cursory modern analysis shows the modulations to lack
>statistical significance.
>
You might believe a "cursory" analysis. I prefer to believe the
analyses by Professor Allais who is experienced in statistics and who
has done a proper job.
http://allais.maurice.free.fr/English/Science.htm
>
>Nevertheless the modulations were always- present.
>We have Tom Roberts to thank for explaining
>the origin of these seeming regularities.
>
Tom Robert's account is based on FALSE PREMISES.
See his paper at:
http://www.arxiv.org/abs/physics/060823
FALSE PREMISE 1
=====================
The caption under Fig 3. says:
"The assumed-linear systematic drift from the data of Fig. 1.
The lines are between successive Marker 1 values and the points are
Marker 9. These markers are 180 degrees apart, so any real signal has
the same value for every corner and every point--the variations are
purely an instrumentation effect."
This statement is FALSE, because measurements at Marker 1 and Marker 9
were not made simultaneously. If we are to consider ANY real signal,
then we must be prepared to consider real signals that FLUCTUATE. The
value of any real signal that fluctuates would change between readings
at Marker 1 and Marker 9.
======================
FALSE PREMISE 2
At the top of page 6, Tom Roberts wrote:
data = signal(orientation) + systematic(time)
The key point is that signal(orientation) is independent of time,
and for each orientation (marker) it has the same value for every
turn of the interferometer within a given data run Therefore if the
data from the first turn is subtracted marker-by-marker from the
data of every turn, the result is completely independent of any
orientation dependence, and contains only systematic(time).
The above claims are false, because any real FLUCTUATING signal would
vary with orientation AND TIME.
So in particular, the claim that:
"Therefore if the data from the first turn is subtracted marker-by-marker
from the data of every turn, the result is completely independent of any
orientation dependence, and contains only systematic(time)."
is FALSE.
=======================
If you read Tom Robert's paper you will see that nearly all of his
analysis is based on one or the other of the above false premises. So
the only safe thing to do is ignore his paper.
>
>Sorry, Surfer. Only a person such as yourself, who has invested
>too many years of his life chasing an utter fantasy to admit that
>it was all wasted, could possibly still believe that what Miller
>saw was anything real.
>
Actually, until very recently I assumed SR was true. I only looked at
Miller's paper after reading Tom Robert's paper. I was favorably
impressed by the effort Miller made to exclude temperature effects
from measurement error. I was also impressed by the smallness of his
estimated probable error.
>
>I understand your unwillingness to drop your illusions.
>
By and large I have not regarded my belief in SR as an illusion. I
still believe SR is generally true. However I think it would be a
mistake to regard it as dogmatically true. In other words I think one
should consider the possibility that motion relative to 3-space could
cause detectable effects--such as observed by Miller.
-- Surfer
Jerry
> On Sun, 4 May 2008 22:27:43 -0700 (PDT), Jerry
> >Nevertheless the modulations were always- present.
> >We have Tom Roberts to thank for explaining
> >the origin of these seeming regularities.
>
> Tom Robert's account is based on FALSE PREMISES.
> See his paper at:http://www.arxiv.org/abs/physics/060823
Work it out for yourself on any data set, random or non-
random, it doesn't matter.
Miller's data reduction procedure FORCES a "return to zero"
every half turn.
Jerry
Surfer
That is incorrect.
Have a look at "Reduction of the interferometer observations" on Pages
213/214 of Miller's paper.
http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
To compensate for the effects of a linear temperature drift Miller
explains:
"A compensation for the shift is made by adding to the SUM of the
seventeenth column such a number that will make it equal to the SUM of
the first column and adding one-sixteenth, two-sixteenths, etc. of
this compensating number to the second, third, etc. columns; this
renders the axis of reference horizontal.
The word SUM lets us know that he is referring to the sum of 20 turns.
So the "forcing to zero" is only done ONCE every TWENTY FULL turns.
This process is not going to produce an output with a period of half a
turn, unless a signal with a period of half a turn is present in the
raw data that was summed.
Dono
>
> - Show quoted text -
I think that we are in violent agreement that Peter will never admit
that his phantasms are just...ghosts.
Jerry
> On Mon, 5 May 2008 10:28:10 -0700 (PDT), Jerry
>
>
>
>
>
> <Cephalobus_alie...@comcast.net> wrote:
> >On May 5, 8:24 am, Surfer <n...@spam.net> wrote:
> >> On Sun, 4 May 2008 22:27:43 -0700 (PDT), Jerry
>
> >> >Nevertheless the modulations were always- present.
> >> >We have Tom Roberts to thank for explaining
> >> >the origin of these seeming regularities.
>
> >> Tom Robert's account is based on FALSE PREMISES.
> >> See his paper at:http://www.arxiv.org/abs/physics/060823
>
> >Work it out for yourself on any data set, random or non-
> >random, it doesn't matter.
>
> >Miller's data reduction procedure FORCES a "return to zero"
> >every half turn.
>
> That is incorrect.
>
> Have a look at "Reduction of the interferometer observations" on Pages
>
> To compensate for the effects of a linear temperature drift Miller
> explains:
>
> "A compensation for the shift is made by adding to the SUM of the
> seventeenth column such a number that will make it equal to the SUM of
> the first column and adding one-sixteenth, two-sixteenths, etc. of
> this compensating number to the second, third, etc. columns; this
> renders the axis of reference horizontal.
>
> The word SUM lets us know that he is referring to the sum of 20 turns.
> So the "forcing to zero" is only done ONCE every TWENTY FULL turns.
>
> This process is not going to produce an output with a period of half a
> turn, unless a signal with a period of half a turn is present in the
> raw data that was summed.
"For the purpose of a preliminary study of the observations...the
second half of the line of sixteen average readings is placed under
the first half and the mean of the two numbers in each column is
obtained."
This procedure combined with the first forces ANY sequence of numbers
back to baseline when combined with the first procedure.
You may object that in his "definitive" study, he made use of a
harmonic analyzer, and did not use this approximate second step.
However, the FIRST step, by forcing the endpoints to equality,
"primed" the data so that any apparent second order effects were
greatly exaggerated. Raw data with no second harmonic component
at all can show a second harmonic when doctored in this fashion.
Jerry
Tom Roberts
> I prefer to believe the
> analyses by Professor Allais who is experienced in statistics and who
> has done a proper job.
> http://allais.maurice.free.fr/English/Science.htm
Allais never checked that the "signal" from each run is significant.
This negates his entire approach.
Children can find faces in clouds. But they are not significant, either.
> Tom Robert's account is based on FALSE PREMISES. [...]
[Your "premise 2" is just a re-statement of "premise 1".]
You are being SELF-INCONSISTENT!
That "premise" is at the base of Miller's analysis. So if that "premise"
is false, than NO USE WHATSOEVER can be made of Miller's own analysis,
either.
Note that this "premise" is NOT needed to compute the errorbars on
Miller's analysis, which shows his "signal" to be INSIGNIFICANT. Those
errorbars are INHERENT in the algorithm he used to reduce the data of
each run to 8 points. The fact that his method ASSUMES that "premise"
does not affect the FACT that those errorbars apply to the analysis that
he did perform.
I put "premise" in quotes, because it is YOUR word, and does not really
apply. All that is needed is that the "fluctuations" of the real signal
be slow compared to the rotation rate of the interferometer. A glance at
Figure 9 of my paper shows this is CLEARLY valid -- the time variations
of the different orientations track each other quite well.
In MILLER'S MODEL, the real signal does not "fluctuate" at
all. He ascribed the large-scale drift to the instrument,
not to the real signal or any cosmic effect; he did not
realize that this instrument drift actually generated 100%
of his "signal" -- his assumption that it is linear is
WOEFULLY inadequate.
Your claims of "fluctuations" do not hold up at the level of
invalidating my analysis. There could be such "fluctuations" at a small
level, but they are obscured by the much larger SYSTEMATIC drift (which
does not "fluctuate" very much between orientations -- remember that
adjacent points of each orientation in Fig 9 are 1/2 turn apart.
You need to step back and check YOUR premises -- you clearly have too
much emotional commitment in Miller's result to be objective. And you
need to educate yourself on modern experimental techniques, specifically
quantitative error analysis.
Tom Roberts
Tom Roberts
> Have a look at "Reduction of the interferometer observations" on Pages
> 213/214 of Miller's paper.
> http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
>
> To compensate for the effects of a linear temperature drift Miller
> explains:
>
> "A compensation for the shift is made by adding to the SUM of the
> seventeenth column such a number that will make it equal to the SUM of
> the first column and adding one-sixteenth, two-sixteenths, etc. of
> this compensating number to the second, third, etc. columns; this
> renders the axis of reference horizontal.
>
> The word SUM lets us know that he is referring to the sum of 20 turns.
> So the "forcing to zero" is only done ONCE every TWENTY FULL turns.
This is just plain wrong. Miller AVERAGED the values for each marker for
the 20 turns, and then applied that paragraph to the AVERAGES. Those
averages represent ONE turn.
Jerry misspoke slightly -- the result is forced to be NEAR zero at
markers 1 and 16 (it is near zero, not exactly zero, because the 20
values for marker "17" are not quite the same as the 20 values for
marker 1 [19 of them are identical]). But Miller then averaged the two
1/2 turns, and COPIED the first value to the 9th position. THAT makes
the resulting graph be periodic with period 1/2 turn, and the zero is
constrained to be near the endpoints, but not constrained to be
precisely at an endpoint. And this is EXACTLY what his Fig 8 (my fig 1)
shows.
You have spent more time grasping at straws than you would have spent
getting an education in basic error analysis. And in your ignorance, all
your straws prove to be inadequate to sustain your argument. There is no
substitute for knowledge -- it's time you get some, and there is only
one way to do that....
Tom Roberts
Surfer
<tjrobe...@sbcglobal.net> wrote:
>Surfer wrote:
>> Have a look at "Reduction of the interferometer observations" on Pages
>> 213/214 of Miller's paper.
>> http://www.scieng.flinders.edu.au/cpes/people/cahill_r/Miller1933.pdf
>>
>> To compensate for the effects of a linear temperature drift Miller
>> explains:
>>
>> "A compensation for the shift is made by adding to the SUM of the
>> seventeenth column such a number that will make it equal to the SUM of
>> the first column and adding one-sixteenth, two-sixteenths, etc. of
>> this compensating number to the second, third, etc. columns; this
>> renders the axis of reference horizontal.
>>
>> The word SUM lets us know that he is referring to the sum of 20 turns.
>> So the "forcing to zero" is only done ONCE every TWENTY FULL turns.
>
>This is just plain wrong. Miller AVERAGED the values for each marker for
>the 20 turns, and then applied that paragraph to the AVERAGES. Those
>averages represent ONE turn.
>
It still remains the case that the "forcing to zero" is only done ONCE
every TWENTY FULL turns. That is to say, the process of averaging does
not do any such forcing.
>
>Jerry misspoke slightly -- the result is forced to be NEAR zero at
>markers 1 and 16 (it is near zero, not exactly zero, because the 20
>values for marker "17" are not quite the same as the 20 values for
>marker 1 [19 of them are identical]). But Miller then averaged the two
>1/2 turns, and COPIED the first value to the 9th position. THAT makes
>the resulting graph be periodic with period 1/2 turn,
>
Here are some counter examples:
Averaged raw data =
{10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10}
Output =
{10., 90., 10., 90., 10., 90., 10., 90., 10.}
Averaged raw data=
{10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150,
160, 170}
Output=
{10., 10., 10., 10., 10., 10., 10., 10., 10.}
Big swings in raw data values, but they do not cause outputs periodic
in half a turn.
>
>and the zero is
>constrained to be near the endpoints, but not constrained to be
>precisely at an endpoint. And this is EXACTLY what his Fig 8 (my fig 1)
>shows.
>
Here is a counter example:
Averaged raw data =
{30, 20, 10, 0, 10, 20, 30, 40, 30, 20, 10, 0, 10, 10, 20, 30, 30}
Output =
{30., 20., 10., 0., 10., 15., 25., 35., 30.}
>
>You have spent more time grasping at straws than you would have spent
>getting an education in basic error analysis.
>
Whatever my education, it doesn't alter the fact that your analysis of
Miller's experiment in:
http://www.arxiv.org/abs/physics/060823
is based on false premises.
FALSE PREMISE 1
=====================
The caption under Fig 3. says:
"The assumed-linear systematic drift from the data of Fig. 1.
The lines are between successive Marker 1 values and the points are
Marker 9. These markers are 180 degrees apart, so any real signal has
the same value for every corner and every point--the variations are
purely an instrumentation effect."
This statement is FALSE, because measurements at Marker 1 and Marker 9
were not made simultaneously. If we are to consider ANY real signal,
then we must be prepared to consider real signals that FLUCTUATE. The
value of any real signal that fluctuates would change between readings
at Marker 1 and Marker 9.
======================
FALSE PREMISE 2
At the top of page 6, you wrote:
data = signal(orientation) + systematic(time)
The key point is that signal(orientation) is independent of time,
and for each orientation (marker) it has the same value for every
turn of the interferometer within a given data run Therefore if the
data from the first turn is subtracted marker-by-marker from the
data of every turn, the result is completely independent of any
orientation dependence, and contains only systematic(time).
The above claims are false, because any real FLUCTUATING signal would
vary with orientation AND TIME.
So in particular, the claim that:
"Therefore if the data from the first turn is subtracted marker-by-marker
from the data of every turn, the result is completely independent of any
orientation dependence, and contains only systematic(time)."
is FALSE.
=======================
Consequently the main products of your analysis, including Figure 11,
have no validity.
-- Surfer
Jerry
> On Mon, 05 May 2008 22:08:42 -0500, Tom Roberts
> >Jerry misspoke slightly -- the result is forced to be NEAR zero at
> >markers 1 and 16 (it is near zero, not exactly zero, because the 20
> >values for marker "17" are not quite the same as the 20 values for
> >marker 1 [19 of them are identical]). But Miller then averaged the two
> >1/2 turns, and COPIED the first value to the 9th position. THAT makes
> >the resulting graph be periodic with period 1/2 turn,
>
> Here are some counter examples:
>
> Averaged raw data =
> {10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10}
> Output =
> {10., 90., 10., 90., 10., 90., 10., 90., 10.}
Pathetic. This is simultaneously periodic over 1/2, 1/4,
and 1/8 turn.
> Averaged raw data=
> {10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150,
> 160, 170}
> Output=
> {10., 10., 10., 10., 10., 10., 10., 10., 10.}
>
> Big swings in raw data values, but they do not cause outputs periodic
> in half a turn.
Doubly pathetic.
> >and the zero is
> >constrained to be near the endpoints, but not constrained to be
> >precisely at an endpoint. And this is EXACTLY what his Fig 8 (my fig 1)
> >shows.
>
> Here is a counter example:
>
> Averaged raw data =
> {30, 20, 10, 0, 10, 20, 30, 40, 30, 20, 10, 0, 10, 10, 20, 30, 30}
> Output =
> {30., 20., 10., 0., 10., 15., 25., 35., 30.}
>
Despite your carefully engineered "raw data", note the beautiful
periodicities:
35 35
30--------------30--------------30--
25 25
20 20
15 15
10 10 10 10
00 00
Jerry
Dono
Peter,
You are by far the moost interesting crackpot in this forum. Congrats!