General Physics

Thanks for calling me clueless Tom.  I really appreciate that.   

I won't argue the specifics with you, but there are ways to eliminate and account for those forces, which the author of the paper does not mention or reflects very inaccurately.

None of the Cavendish Experiments make an attempt to account for macro forces. This is thoroughly described in the article I linked.

You didn't answer my questions.  Do you accept gravitation by mass?

Seeing as the stars don't have very much mass, no I don't.

Strong and Weak forces decrease exponentially, not by inverse square.  Strong and Weak forces could not conceivably have affected this experiment.  The electro-static forces can be neutralized simply by touching the metal balls together prior to performing the experiment, and a closed box rules out buildup of unequal static charge by convection.

Wrong. At the levels of sensitivity we're talking about strong and weak forces absolutely will have an effect. Towards the end of the paper I linked the author even gives detailed equations which demonstrates this.

The one detail which gives very great credence to the existence of gravity is the consistency by which the gravitational constant has been found to be.  If gravity didn't exist at all, and all of the disruptive forces he lists did exist, every measurement would be wildly inaccurate.  Again, remember that each experiment was performed with a different apparatus, and experiments have even been performed (recently) which use a totally different technique.

Consistency in prediction and observation indicates that a theory is effective.  Key word: prediction.

Unfortunately for your argument the Cavendish Experiment is neither consistent or predictable. You would have known that if you had read the entirety of the article.

In an instance of a Cavendish Experiment shared on this forum the moderator TheEngineer, a mechanical engineer, had this to say about the "Bending Space-Time in the Basement" Cavendish experiment:--

There seems to me, to be some unexplainable things going on in the experiments.  The second video shows a large return of the balance after it contacts the weights.  Just from looking at the video and using an estimate of the angle and time using the stamp on the video, I've made a liberal estimate of the velocity when it makes contact with the weights.  This will result in a certain kinetic energy at the moment of impact.  Assuming a perfectly elastic collision (again, very liberal), the total energy must be conserved, so that the potential energy gained by the masses must equal the kinetic energy.  Using a simple equation, I've found the gravitational attraction of the weights and masses.  Using the kinetic energy as the maximum potential energy and solving for the distance that the mass can travel, I've found the rebound angle to be 0.126 degrees, not the nearly 30 that is shown in the video.  However, there is also a water brake which should damp this small movement, making the video highly suspect.

Now, as I have said, I've made assumptions and simplifications (as it's late, I'm tired and I'm not getting paid for this), and those have been on the larger side of things.

Perhaps I will do an in depth analysis of this if I get bored.

Using his analysis, anyone can look at the videos and see that "gravity" does not act in the way predicted.

Quote from: Euclid on April 22, 2009, 08:14:45 PM

Thanks for calling me clueless Tom.  I really appreciate that.   

I won't argue the specifics with you, but there are ways to eliminate and account for those forces, which the author of the paper does not mention or reflects very inaccurately.

None of the Cavendish Experiments make an attempt to account for macro forces. This is thoroughly described in the article I linked.

Quote from: Euclid on April 22, 2009, 08:14:45 PM

You didn't answer my questions.  Do you accept gravitation by mass?

Seeing as the stars don't have very much mass, no I don't.

How do you know the stars don't have very much mass?  If the stars' gravitation affects massive objects on the Earth, surely gravitation by mass/energy exists?

Quote from: Euclid on April 22, 2009, 08:14:45 PM

You didn't answer my questions.  Do you accept gravitation by mass?

Seeing as the stars don't have very much mass, no I don't.

Then how can the the whirling pattern of stars affect Foucault pendulums all over the world?  Or is there some other source for the star's gravitation?

Quote from: munkirench on April 22, 2009, 10:20:12 PM

Strong and Weak forces decrease exponentially, not by inverse square.  Strong and Weak forces could not conceivably have affected this experiment.  The electro-static forces can be neutralized simply by touching the metal balls together prior to performing the experiment, and a closed box rules out buildup of unequal static charge by convection.

Wrong. At the levels of sensitivity we're talking about strong and weak forces absolutely will have an effect. Towards the end of the paper I linked the author even gives detailed equations which demonstrates this.

Tom, please explain how strong and/or weak nuclear forces are significant beyond the nucleus of an atom.

Quote from: munkirench on April 22, 2009, 10:20:12 PM

The one detail which gives very great credence to the existence of gravity is the consistency by which the gravitational constant has been found to be.  If gravity didn't exist at all, and all of the disruptive forces he lists did exist, every measurement would be wildly inaccurate.  Again, remember that each experiment was performed with a different apparatus, and experiments have even been performed (recently) which use a totally different technique.

Consistency in prediction and observation indicates that a theory is effective.  Key word: prediction.

Unfortunately for your argument the Cavendish Experiment is neither consistent or predictable. You would have known that if you had read the entirety of the article.

In an instance of a Cavendish Experiment shared on this forum the moderator TheEngineer, a mechanical engineer, had this to say about the "Bending Space-Time in the Basement" Cavendish experiment:--

There seems to me, to be some unexplainable things going on in the experiments.  The second video shows a large return of the balance after it contacts the weights.  Just from looking at the video and using an estimate of the angle and time using the stamp on the video, I've made a liberal estimate of the velocity when it makes contact with the weights.  This will result in a certain kinetic energy at the moment of impact.  Assuming a perfectly elastic collision (again, very liberal), the total energy must be conserved, so that the potential energy gained by the masses must equal the kinetic energy.  Using a simple equation, I've found the gravitational attraction of the weights and masses.  Using the kinetic energy as the maximum potential energy and solving for the distance that the mass can travel, I've found the rebound angle to be 0.126 degrees, not the nearly 30 that is shown in the video.  However, there is also a water brake which should damp this small movement, making the video highly suspect.

Now, as I have said, I've made assumptions and simplifications (as it's late, I'm tired and I'm not getting paid for this), and those have been on the larger side of things.

Perhaps I will do an in depth analysis of this if I get bored.

Using his analysis, anyone can look at the videos and see that "gravity" does not act in the way predicted.

I take it that Engy never got bored enough to visit that guy's basement to do an in depth analysis of his experiment, did he?  I notice that Engy never did flat out refute the experiment, he only pointed out a few potential issues that he never followed up on to see if they were indeed problems.

How do you know the stars don't have very much mass?

Because by plane trigonometry, in special connection with carefully measured base lines, Samuel Birley Rowbotham has demonstrated and placed beyond all power of doubt that the sun, moon, stars, and planets are all bodies within a distance of a few thousand miles from the surface of the earth. Therefore they are very small objects. Therefore not worlds. Therefore not hundreds of thousands of miles across or from each other.

If the stars' gravitation affects massive objects on the Earth, surely gravitation by mass/energy exists?

If the stars do affect bodies on earth, it's certainly not through "gravitation by mass."

Tom, please explain how strong and/or weak nuclear forces are significant beyond the nucleus of an atom.

Read the article. The author does an in-depth explanation and provides equations demonstrating that macro forces are significant enough to overcome the experiment.

I take it that Engy never got bored enough to visit that guy's basement to do an in depth analysis of his experiment, did he?  I notice that Engy never did flat out refute the experiment, he only pointed out a few potential issues that he never followed up on to see if they were indeed problems.

TheEngineer refutes the experiment just fine.

It's also stated on the Wiki:

"Bending Spacetime in the Basement (do-it-yourself Cavendish apparatus - appears to be seriously flawed[1])"

Quote

I'd say that article does nothing to refute the Cavendish experiment or other experiments that have measured G with different methods.

Actually the paper I linked demolishes the validity of the Cavendish Experiment pretty badly. It demonstrates that the experiment is entirely invalid.

I got suspicious when I read this:

An apple weighs, what, 2 or 3 ounces?

Then I stopped reading when I read this:

Those are two separate margins of error, so they have to multiply. Ten percent times 3 percent. That?s a thirty percent error.

It's all a load of idiotic babble.

Quote from: Tom Bishop on April 22, 2009, 07:36:42 PM

Quote

I'd say that article does nothing to refute the Cavendish experiment or other experiments that have measured G with different methods.

Actually the paper I linked demolishes the validity of the Cavendish Experiment pretty badly. It demonstrates that the experiment is entirely invalid.

I got suspicious when I read this:

Quote

An apple weighs, what, 2 or 3 ounces?

Then I stopped reading when I read this:

Quote

Those are two separate margins of error, so they have to multiply. Ten percent times 3 percent. That?s a thirty percent error.

It's all a load of idiotic babble.

Read more. Post less.

Wrong. Read more. Post less.

The guy can't even work out proper margins of error! What a failure!

The guy can't even work out proper margins of error! What a failure!

Incorrect. The author describes the total multi-variable margin of error just fine. In the experiment the ten percent margin of error recurs with discrepancies of five to fifteen seconds. That means the total margin of error is recurring. Over extended trial runs the margin of error accumulates and therefore the error margin must be multiplied by the total number of recursions.

Please educate yourself.

Incorrect. The author describes the total multi-variable margin of error over multiple trial runs just fine. In the experiment the ten percent margin of error recurs with discrepancies of five to fifteen seconds in seven minutes. That means the total margin of error is recurring. Over extended trial runs the margin of error accumulates and therefore the variables can be multiplied.

Please educate yourself.

Do you understand that a ten percent error, multiplied by a 3 percent error does not equal a thirty percent error?

Do you understand that a ten percent error, multiplied by a 3 percent error does not equal a thirty percent error?

That ten percent error is happening over and over again. Ergo, recurring. Ergo, that margin of error must be multiplied by the number of times it recurs in a trial.

Quote from: liedetector on April 23, 2009, 12:21:22 PM

Do you understand that a ten percent error, multiplied by a 3 percent error does not equal a thirty percent error?

That ten percent error is happening over and over again. Ergo, recurring. The 3 percent variable is a multiplier variable since the three percent is the rate at which the ten percent error recurs.

Oh dear. I'll take that as a "no".

If the period varied between 5 and 15, then the average would be 10 seconds, this would in fact be an average error of 2 percent. (Another failure). However, even then I'm confused. An error is usually expressed as a +/- value. 5 to 15 doesn't make much sense.


Anyway, how this error would feed into the final error of the result is more complicated. See the equations here: http://en.wikipedia.org/wiki/Cavendish_experiment

However, it still stands that a ten percent error multiplied by a three percent error is not a thirty percent error. The fact that neither you nor Miles Mathis understand this is irrellevent.

However, it still stands that a ten percent error multiplied by a three percent error is not a thirty percent error. The fact that neither you nor Miles Mathis understand this is irrellevent.

Nope. You're not understanding the fundamentals of the scenerio.

We have an experiment with one hundred data points. We know that ten percent of those data points are wrong. Now, there is also a three percent recursion on top of that ten percent, meaning that we know that the ten percent error rate recurs three times across the hundred data points. Therefore, we multiply ten by three and get a thirty percent error margin.

Pick any one of those data points and there's a 30% chance of it being wrong.

I'd have to see TheEngineer's estimates of the velocity of the balls.  Just judging by the angles involved, energy is indeed conserved.

You can't disregard the fact that, using a well-designed apparatus, the value of G is consistent.  Just pointing out that people have gotten wildly different values doesn't mean anything because these outlier values were obtained using faulty apparatus that probably had the very problems that you are describing... Explain to me, how could hundreds of measurements, using non-identical apparatus and parameters, produce the same answer, within a 1% error, stipulating that the apparatus was built by a competent engineer?  If you still insist that the values obtained are so different as to preclude the mere possibility of correlation, please point out to me one of the experiments performed which indicate an inconsistent value, so that I can evaluate it for myself?  (although I must point out once more that the existence of a correlation much looser than the slim errors found in the value would still constitute valid evidence toward the existence of a gravitational constant.  the errors would simply be showing the inconsistency of the apparatus)

Miles Mathis' paper is just ridiculous. I skimmed over the whole thing... first off, not once does he mention the strong or weak force.  Just do a search for "strong force"... nothing.  He does mention some completely incorrect ideas about E/M fields affecting the attractions, but again, it is obvious that he just doesn't understand the underlying theories of E/M.  His idea that EM actually is part of the "pseudo-field" is completely wrong.  Seriously... how could you believe even a word of that paper?

I have to inquire, Tom... what is the highest level of mathematics and classical physics that you have actually learned?  Based on your acceptance of Mathis' paper, I would estimate that you have only the most basic introduction to classical physics, probably only learned up to Calculus I, and have obviously never heard of a vector.  If I'm wrong here, let me know, so I know what level of technicality I can include in my responses.

I'm not going to refute everything in Mathis' article, because literally 95% is just meaningless drivel.  Mathis doesn't even have a technical degree... while I admit that degrees in themselves don't mean much, his actual formal education has been limited to Art, Politics, and Management.  Hardly a trustworthy source... and have you looked at his other papers?  In one of them, he claims that the constant pi is actually not dimensionless, has the units of velocity, and is actually closer to 4 than 3.14.  He doesn't even know the difference between velocity and speed.

Since I know you won't be pleased about my utter disregard for anything Mathis has written, I'll respond to the one part of the paper which makes physical sense, but is mathematically incorrect in its interpretation:  the idea that the walls of the apparatus would have affected the calculation.  It is true that under the accepted theories of gravity, the walls would indeed exert a gravitational force on the balls.  However, even if the walls were 1000 times the mass of the balls, negating any inverse-squared-distance argument I could make regarding the attenuation of the fields produced, the walls would STILL not have affected the experiment.  If you examine the apparatus described by Cavendish, you see that the line between the centers of mass of the two balls is parallel to the nearest wall.  As such, the force does no actual work on the balls, because the balls are constrained to move in a single dimension - or at least, they are constrained to a single degree of freedom which is orthogonal to the wall's field.

So, here's my conclusion:  The Cavendish Experiment is valid.  The wall's affect is negligible, both by design of the experiment, and because of the significant difference between the distances involved (the distance of the non-parallel walls negates any affect they would have).  Mathis' analysis is just wrong... he has the foundation of a high schooler, and his axioms are just plain old false.  Furthermore, Mathis obviously doesn't have any type of formal training in these types of analysis... he isn't knowledgeable enough of the theories of classical physics to properly refute them.

The idea that either EM or strong/weak forces could affect the apparatus is also false.  Yes, even at the 10^-10N levels that Mathis constantly brays about.  Also, strong/weak forces are NOT macro forces... they would NEVER be observed at greater than the molecular level, and Mathis doesn't even address them.  Strong/Weak forces are called Nuclear forces because that is where they operate... on the nucleus of atoms.

You keep referencing sources which are just plain old unreliable... While I admit that I don't have all of the data I would need to evaluate TheEngineer's analysis, it appears at the surface to be invalid.

And just for the record, Mathis believes in gravity.  He just doesn't accept the validity of experiments which measure the Gravitational Constant.

Tom... that is just wrong.  .03*.10 does NOT equal .30.  The proper way to measure the error is (1 - x1)(1 - x2) = (1 - y), where x are the margins of error and y is the final... the final error is .9*.97 = .873, so the actual margin of error throughout is 12.7%.

I have no idea how you concluded that 3% = 3x.

I think your error comes from not understanding that 10% actually means .10... the value of 10 is totally irrelevant in real analysis unless you include the factor of 100 which you use to convert.  It is also necessary to recognize that the value of 100 is totally arbitrary based on whichever number base you want to use.  This is just more indication to me that you don't understand mathematics beyond basic arithmetic.

Tom... that is just wrong.  .03*.10 does NOT equal .30.  The proper way to measure the error is (1 - x1)(1 - x2) = (1 - y), where x are the margins of error and y is the final... the final error is .9*.97 = .873, so the actual margin of error throughout is 12.7%.

I have no idea how you concluded that 3% = 3x.

I think your error comes from not understanding that 10% actually means .10... the value of 10 is totally irrelevant in real analysis unless you include the factor of 100 which you use to convert.  It is also necessary to recognize that the value of 100 is totally arbitrary based on whichever number base you want to use.  This is just more indication to me that you don't understand mathematics beyond basic arithmetic.

You're talking nonsense. In the experiment we know that ten percent of the results are already wrong. That ten percent recurs three times out of a hundred (3%), meaning that in 100 data points the ten percent error recurs three times across the data set.

And the 100 value is not arbitrary. 3% is 3 out of 100. We can even expand the data set to 1000 data points and the 3% recursion is now 30 out of 1000. We're just increasing our ratios. Under 1000 data points, the ten percent total error now recurs 30 times across the set, meaning that 30% (300/1000) of the points are wrong again.

Go back to school.

Honestly, I'm a bit stunned that you so strongly believe in this little bit of clearly incorrect mathematics.

Here's the origin of the error:

In the original experiment of Cavendish there seems to have been an irregularity in the position of rest of one-tenth of the deflection obtained, while the period showed discrepancies of five to fifteen seconds in seven minutes.

Let's analyze. 
The relationship between period and deflection:  period determines the torsional spring constant, which is used to determine the force as related to deflection.  The relationship isn't really linear, but let's assume it is for the sake of your argument.
error in deflection: 10% = .10.
error in period: 10s / (7*60s) (that is, the error of 10 seconds in 7 minutes) = .0238.
While it doesn't say so, let's assume that the error is actual x +/- 10% (the actual value is probably x +/- 5%), in order to maximize the error.
Minimum deflection: .9x (where x is the average deflection)
maximum deflection: 1.1x
minimum period: 6:45 = 405s (a conservative estimate of 15 seconds difference from the average)
maximum period: 7:15 = 435s

minimum value: .9x * 405 = 364.5x s
maximum value: 1.1x * 434 = 478.5x s
average value, including error:  421.5
421.5 - 364.5 = 57
57/421.5 = .135 = 13.5%

Final value:
421.5 +/- 57 = 421.5 +/- 13.5%

Before you tell people to go back to school, maybe you should get your high school diploma.

where in the world do you get the arbitrary number of 3 to multiply 10% by?

Honestly, I'm a bit stunned that you so strongly believe in this little bit of clearly incorrect mathematics.

You're new here, aren't you?  FE'ers (Tom especially) will defend to the death their misinterpretations of the facts, no matter how obvious.

Slightly off-topic, but I think it would be awesome to be called General Physics.

Carry on.

Honestly, I'm a bit stunned that you so strongly believe in this little bit of clearly incorrect mathematics.

Here's the origin of the error:

Quote

In the original experiment of Cavendish there seems to have been an irregularity in the position of rest of one-tenth of the deflection obtained, while the period showed discrepancies of five to fifteen seconds in seven minutes.

Let's analyze. 
The relationship between period and deflection:  period determines the torsional spring constant, which is used to determine the force as related to deflection.  The relationship isn't really linear, but let's assume it is for the sake of your argument.
error in deflection: 10% = .10.
error in period: 10s / (7*60s) (that is, the error of 10 seconds in 7 minutes) = .0238.
While it doesn't say so, let's assume that the error is actual x +/- 10% (the actual value is probably x +/- 5%), in order to maximize the error.
Minimum deflection: .9x (where x is the average deflection)
maximum deflection: 1.1x
minimum period: 6:45 = 405s (a conservative estimate of 15 seconds difference from the average)
maximum period: 7:15 = 435s

minimum value: .9x * 405 = 364.5x s
maximum value: 1.1x * 434 = 478.5x s
average value, including error:  421.5
421.5 - 364.5 = 57
57/421.5 = .135 = 13.5%

Final value:
421.5 +/- 57 = 421.5 +/- 13.5%

Before you tell people to go back to school, maybe you should get your high school diploma.

Nope.

1. Cavandish says that 1 out of 10 of the deflections are wrong. Therefore we know that 10% of the total data set is  wrong.

2. Cavandish also says that the error rate recurs at 3%, meaning that in a data set of 100 data points the 10% total error recurs three times across the set.

where in the world do you get the arbitrary number of 3 to multiply 10% by?

The three percent figure is recurring, meaning that a total error of 10% recurs 3 out of 100 times in a 100 point data set.

I'm done explaining middle school math to you. If you need any further help please ask a parent or guardian.

the error rate recurs at 3%, meaning that in a data set of 100 data points the 10% total error recurs three times.

explain the definition of the word "recurs" here, and explain how 3% = 10% * 3?  You're just plain wrong... you don't understand statistical analysis of error.  You accept Mathis' explanation, but Mathis is wildly wrong on many claims, including this one.  I just showed you a REAL error analysis... one that doesn't involve statistical axioms which you don't understand... I showed that using maximum error in both inputs, the output error is only 13.5%.  Isn't this exactly what you FEers always whine about?  What you SEE and what ACTUALLY happens?  This is a prime example... your calculations MUST be wrong, because they don't show what happens in REALITY.  My calculations don't use statistical theory, they just show the ACTUAL error through combining both minimum values and both maximum values.  I PROVED what the error is.  Beyond doubt.  Your math is fundamentally flawed.

Now answer my question... what is the highest formal (or informal, for that matter) training in mathematics?

Quote

the error rate recurs at 3%, meaning that in a data set of 100 data points the 10% total error recurs three times.

explain the definition of the word "recurs" here, and explain how 3% = 10% * 3?  You're just plain wrong... you don't understand statistical analysis of error.  You accept Mathis' explanation, but Mathis is wildly wrong on many claims, including this one.  I just showed you a REAL error analysis... one that doesn't involve statistical axioms which you don't understand... I showed that using maximum error in both inputs, the output error is only 13.5%.  Isn't this exactly what you FEers always whine about?  What you SEE and what ACTUALLY happens?  This is a prime example... your calculations MUST be wrong, because they don't show what happens in REALITY.  My calculations don't use statistical theory, they just show the ACTUAL error through combining both minimum values and both maximum values.  I PROVED what the error is.  Beyond doubt.  Your math is fundamentally flawed.

Now answer my question... what is the highest formal (or informal, for that matter) training in mathematics?

tl;dr, please refer to your parent or guardian for further assistance. If they are at work please find a friendly neighbor to help you out.

Now answer my question... what is the highest formal (or informal, for that matter) training in mathematics?

So what you're saying is that if he's not formally educated but correct, you can dismiss him since he "clearly doesn't know what he's talking about", but if he's got a Ph.D. in Maths but is utterly wrong then you can go "Ahh bullshit, you don't have a Ph.D.!"?

What's the point in even asking this question?

Quote from: munkirench on April 23, 2009, 02:19:53 PM

Quote

the error rate recurs at 3%, meaning that in a data set of 100 data points the 10% total error recurs three times.

explain the definition of the word "recurs" here, and explain how 3% = 10% * 3?  You're just plain wrong... you don't understand statistical analysis of error.  You accept Mathis' explanation, but Mathis is wildly wrong on many claims, including this one.  I just showed you a REAL error analysis... one that doesn't involve statistical axioms which you don't understand... I showed that using maximum error in both inputs, the output error is only 13.5%.  Isn't this exactly what you FEers always whine about?  What you SEE and what ACTUALLY happens?  This is a prime example... your calculations MUST be wrong, because they don't show what happens in REALITY.  My calculations don't use statistical theory, they just show the ACTUAL error through combining both minimum values and both maximum values.  I PROVED what the error is.  Beyond doubt.  Your math is fundamentally flawed.

Now answer my question... what is the highest formal (or informal, for that matter) training in mathematics?

tl;dr, please refer to your parent or guardian for further assistance. If they are at work please find a friendly neighbor to help you out.

So when I PROVED that you are wrong, the only option left was to cover your ears and say "la la la la"?

The purpose of trying to find out just what math he understands is to see exactly what level of expertise I am addressing... note that I said "or informal"... I don't care what degrees he has, I've stated that.  But the fact remains that someone whose highest understanding of mathematics is basic algebra is not worth debating with, because they will just reject mathematical proof that they do not understand, even though the proof is still logically equivalent to saying "1 + 1 = 2".

So Tom, now that we've established that you have the mathematical background of a 12 year old, are you going to address any of my statements regarding the absurdity of Mathis' paper?  I understand that refuting the statement that "your ENTIRE argument is BS" is pretty difficult... so how about this... just name for me 2 or 3 parts of Mathis' paper which you think most relevant and important to debunking the Cavendish Experiment, and we'll go from there.

Nope.

1. Cavandish says that 1 out of 10 of the deflections are wrong. Therefore we know that 10% of the total data set is  wrong.

2. Cavandish also says that the error rate recurs at 3%, meaning that in a data set of 100 data points the 10% total error recurs three times.

Quote

where in the world do you get the arbitrary number of 3 to multiply 10% by?

The three percent figure is recurring, meaning that a total error of 10% recurs 3 out of 100 times in a 100 point data set.

I'm done explaining middle school math to you. If you need any further help please ask a parent or guardian.

What is your profession? I am certain that it isn't in any math-related field because this is all orders-of-magnitude wrong, and would be thrown out by any actual scientists or mathematicians. Hell, Munkirench is doing his best to explain this to you, yet you persist in your error.

3% is 3 times out of 100, or 3/100. A "recurring" (which I'm not sure you understand) figure simply means its compounded continually. That is to say, you would have to take 3/100, or 0.03, as your base, not 3. 3% is not 3. % literally is a symbol representing 1/100.

The "error rate" of 10% recurs three times, yes, out of one hundred. For every three times it occurs, ninety-seven times it doesn't occur. That is probability and statistics, my friend. This is indeed middle-school math, this is something you should have gone over in Algebra.

Let me see if I have you right, however.

10% = 10/100. This much we can agree on.

3% = 3/100. Surely, you can agree that, since it logically follows from the previous example.

To take 10/100 and multiply it by 3 would assume an incredible recurrence rate of 300%, which is impossible, because 300% is 3/1, or simply 3.

To take 10/100 and multiply it by 3/100 would follow from 3%, which is the case. After following the arithmetic Munkirench outlined you will find the margin of error closer to 12%.

You also seem to place a lot of faith ("Go back to school") in an institution controlled by the conspiracy, anyway. If I didn't know any better I'd say you lot could get in bed with that Time Cube lunatic.

So when I PROVED that you are wrong, the only option left was to cover your ears and say "la la la la"?

I've explained it several times now. Please ask a parent or guardian for assistance. You are still treating the 3% variable as some other error margin separate from the 10% error. You're wrong. The 3% represents a multiplier of which 10% affects the data set. The 3% is at a higher order than that 10%, and represents a recurrence in any given set.

So Tom, now that we've established that you have the mathematical background of a 12 year old, are you going to address any of my statements regarding the absurdity of Mathis' paper?  I understand that refuting the statement that "your ENTIRE argument is BS" is pretty difficult... so how about this... just name for me 2 or 3 parts of Mathis' paper which you think most relevant and important to debunking the Cavendish Experiment, and we'll go from there.

The most relevant points are:

- The results of the Cavendish Experiment are not consistent

- The results of the Cavendish Experiment are not predictive

- The macro forces such as E/M are more powerful than "gravity" at the level of measurement in the experiment.
 
- The experimenters make no attempt to account for such forces.

- Even if gravitation by mass did exist, many of the experimenters are operating in basements surrounded by earth. The experimenters think that merely moving to the center of the room will overcome any outside influences.

It's really a mess all around.

Quote from: Tom Bishop on April 23, 2009, 02:14:26 PM

Nope.

1. Cavandish says that 1 out of 10 of the deflections are wrong. Therefore we know that 10% of the total data set is  wrong.

2. Cavandish also says that the error rate recurs at 3%, meaning that in a data set of 100 data points the 10% total error recurs three times.

Quote

where in the world do you get the arbitrary number of 3 to multiply 10% by?

The three percent figure is recurring, meaning that a total error of 10% recurs 3 out of 100 times in a 100 point data set.

I'm done explaining middle school math to you. If you need any further help please ask a parent or guardian.

What is your profession? I am certain that it isn't in any math-related field because this is all orders-of-magnitude wrong, and would be thrown out by any actual scientists or mathematicians. Hell, Munkirench is doing his best to explain this to you, yet you persist in your error.

3% is 3 times out of 100, or 3/100. A "recurring" (which I'm not sure you understand) figure simply means its compounded continually. That is to say, you would have to take 3/100, or 0.03, as your base, not 3. 3% is not 3. % literally is a symbol representing 1/100.

The "error rate" of 10% recurs three times, yes, out of one hundred. For every three times it occurs, ninety-seven times it doesn't occur. That is probability and statistics, my friend. This is indeed middle-school math, this is something you should have gone over in Algebra.

Let me see if I have you right, however.

10% = 10/100. This much we can agree on.

3% = 3/100. Surely, you can agree that, since it logically follows from the previous example.

To take 10/100 and multiply it by 3 would assume an incredible recurrence rate of 300%, which is impossible, because 300% is 3/1, or simply 3.

To take 10/100 and multiply it by 3/100 would follow from 3%, which is the case. After following the arithmetic Munkirench outlined you will find the margin of error closer to 12%.

You also seem to place a lot of faith ("Go back to school") in an institution controlled by the conspiracy, anyway. If I didn't know any better I'd say you lot could get in bed with that Time Cube lunatic.

Nope. Still treating the 3% as a sibling rather than a value of a higher order. In the experiment there's a divergence error and a periodic error. The divergence error is 10% and the periodic error is 3%. Since with every periodic error of one percent, the total accuracy of the experiment decreases, the 10% error is multiplied across this new error variable.

The 3% is affecting the experiment at a higher order because this experiment is in free flow. The experiment is predicting very specific values and positions. Each new variable of error added to the experiment is throwing off the first error variable.

An analogy is a swinging pendulum. Newtonian physics says that the pendulum will swing in perfect arcs. We set the pendulum swinging and it turns out that ten out of one hundred swings the pendulum does not match what Newtonian Theory predicts. Now, we add in another variable: a periodic error which is directly affecting the entire system.

Remember, the Newtonian Theory is predicting very specific positions for the pendulum at each point, and the added periodic error, however small, is multiplying the inherent 10% error, resulting in completely incorrect positions of the pendulum as the effects add up. Each percent of the new periodic error is throwing off the device into a deeper and deeper rate of error. The 10% divergence causes the predicted position to diverge from the predicted value and the 3% periodic causes the divergence to diverge even more.

For arguments sake say that the divergence error is caused by air movements and the periodic error is caused by the E/M force. The air movement affects the horizontal movements of the swing and the E/M force affects the vertical movements of the swing. Under the Newtonian Theory the pendulum is supposed to match specific positions at every swing over 100 swings. The air movements are decreasing the accuracy by ten percent and the E/M force is affecting not only the experiment, but the inherent ten percent error rate from the air movements. With these two conflicting variables occurring at every swing its of no surprise that the experiment quickly becomes unreliable and inaccurate to those precise vertical and horizontal coordinates Newtonian Theory predicts. This is why any new variable of error must be treated on a higher order.

Quote

So when I PROVED that you are wrong, the only option left was to cover your ears and say "la la la la"?

I've explained it several times now. Please ask a parent or guardian for assistance. You're still treating the 3% variable as some other error margin separate from the 10% error. You're wrong. The 3% represents a multiplier of which 10% affects the data set. The 3% is at a higher order than that 10%, and represents a recurrence in any given set.

And you are treating 3% and 10% as integers.  They are not.  3%=0.03 and 10%=0.10.  Therefore 3% x 10% = 0.03 x 0.10 = 0.003 = 0.3%.

I lol at the general incompetence in error analysis showed by both sides and the author of this paper.  Tom's arguments are quite epic.