Thread: Is the speed of light really a constant?

1. The speed of light was fixed in 1972, and the length of a metre was pegged to the speed of light. Therefore the speed of light (in metric) cannot vary, even though it does in fact fluctuate over time.

Is this true? I believe it is.

2.

3. Originally Posted by ultimatesceptic
The speed of light was fixed in 1972, and the length of a metre was pegged to the speed of light. Therefore the speed of light (in metric) cannot vary, even though it does in fact fluctuate over time.

Is this true? I believe it is.
Believing it doesn't make it so. Do you have any evidence that it varies over time?

4. The speed of light in a vacuum is a constant. It does not vary over time.

5. Originally Posted by ultimatesceptic
The speed of light was fixed in 1972, and the length of a metre was pegged to the speed of light. Therefore the speed of light (in metric) cannot vary, even though it does in fact fluctuate over time.

Is this true? I believe it is.
This is true except that measured fluctulactions in the speed of light can not be identified as true fluctulations or as experimental error. The length of a second is also pegged to the speed of light which makes the difficulty complete.
The speed of light is defined as c = 299,792,458 m/sec, *exactly.* It is therefore exact and hence technically unmeasurable.
The defined value of c is used as a constant to establish our standard units for length and time. And, since our units for length, time, and c are all mutually defined, the value of c can not be anything other than what it is defined to be. Given any two of the values for length, time , or c, we can calculate the value of the third. This makes the speed of light unmeasurable. The problem is not obvious when ordinary scales are involved but consider the absurdity of trying to measure the speed of light over the distance of a light year.

6. One of the contributors in another forum thread here said; "All science is wrong.."
and this subject is of the same ilk... and also wrong.
~ The answer is NO. Light speed, or the measured velocity of light photons through a void in a vacuum is c. No variation has been found or is found.. Tested and challenged and as yet no change is detected.. c. is a constant.

7. There are 2 choices, one is the speed of light is a constant, the other one is the speed of light is not a constant. If we accept the second one, then we have to accept the truth that we know nothing about Universe. If the speed of light can change before the light get to planet Earth, then there is no way to know how far the stars in the sky are away from us. So the Unit light year is meaningless.

You have to accept the first choice, no matter how wrong it is. Now you maybe understand the reason why the theory of relativity is so popular in Astronomy.

8. Tomjin's second incarnation was no better than the first.

9. There are 2 choices, one is the speed of light is a constant, the other one is the speed of light is not a constant.
The speed of light in vacuum is neither constant nor varying, it is a relativistic invariant. The distinction is very subtle and is irrelevant in flat space-time, but it is very important in curved space-times, such as extended regions around massive bodies, because then we have to account for the differences in coordinate speed and proper speed. Coordinate speed of light actually does vary ( in fact it drops to zero at the event horizon of a black hole ), but the speed of light being relativistically invariant means that proper speed - which is what is physically measured in a local rest frame - is the same everywhere. Only the latter is physically meaningful since in curved space-times there is no global notion of time. Proper speed to be everywhere locally exactly c just means that all observers experience the same laws of physics, regardless of where they are and how they move. This leads to the very counter-intuitive result that no stationary far-away observer ever sees anything reach the event horizon of a black hole, while at the same time someone in free fall will cross it in finite and well-defined proper time as measured on his own watch; this is not a contradiction, but a simple manifestation of the nature of gravity.

10. Damn I wish that like button was working Markus :-))

11. Originally Posted by Implicate Order
Damn I wish that like button was working Markus :-))
Ibidem

12. Originally Posted by bangstrom
Originally Posted by ultimatesceptic
The speed of light was fixed in 1972, and the length of a metre was pegged to the speed of light. Therefore the speed of light (in metric) cannot vary, even though it does in fact fluctuate over time.

Is this true? I believe it is.
This is true except that measured fluctulactions in the speed of light can not be identified as true fluctulations or as experimental error. The length of a second is also pegged to the speed of light which makes the difficulty complete.
The speed of light is defined as c = 299,792,458 m/sec, *exactly.* It is therefore exact and hence technically unmeasurable.
The defined value of c is used as a constant to establish our standard units for length and time. And, since our units for length, time, and c are all mutually defined, the value of c can not be anything other than what it is defined to be. Given any two of the values for length, time , or c, we can calculate the value of the third. This makes the speed of light unmeasurable. The problem is not obvious when ordinary scales are involved but consider the absurdity of trying to measure the speed of light over the distance of a light year.
Surely there is something wrong with this logic isn't there? The arbitrary act of defining a standard unit of length in terms of c does not preclude our ability to measure variations in c, if there were any. If c did vary, we would see it and this would call into question the validity of the standard unit definition.

13. Originally Posted by bangstrom
The length of a second is also pegged to the speed of light which makes the difficulty complete.
The length of a second is defined as:
"The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom. This definition refers to a caesium atom at rest at a temperature of 0 K."
I am not sure where the speed of light comes into it.

14. Originally Posted by RedPanda
Originally Posted by bangstrom
The length of a second is also pegged to the speed of light which makes the difficulty complete.
The length of a second is defined as:
"The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom. This definition refers to a caesium atom at rest at a temperature of 0 K."
I am not sure where the speed of light comes into it.
Ask Farsight. It is something he has been saying recently on The Physics Forum.

15. Originally Posted by KJW
Ask Farsight. It is something he has been saying recently on The Physics Forum.
I've not been reading ThePhysicsForum, and I couldn't find their comments about light.
I'm not really familiar with Farsight either.

16. Originally Posted by RedPanda
Originally Posted by KJW
Ask Farsight. It is something he has been saying recently on The Physics Forum.
I've not been reading ThePhysicsForum, and I couldn't find their comments about light.
I'm not really familiar with Farsight either.
Farsight is a notorious eccentric and self-taught amateur physicist. I would not go hunting out what he has to say about anything, if I were you.

But I take your point: given that the source is at rest relative to the observer, the frequency of this microwave radiation would not appear to have any scope to vary, even if c were not constant under all circumstances, and thus the duration of the standard second would seem to be fixed.

17. Thank you for the answers. I particularly like those from Tomjin and Markus Hanke.

Mathman, I am not claiming to be an expert, but apparently values of c in physics textbooks have not remained constant over time.

18. This is due to improvements in the accuracy and precision with which c can be measured not changes in c. The values given for the radius of the Earth have changed over the years/centuries as improved measurements have been made, do you think this reflects real changes in the size of the planet?

19. And to add to what PhDemon said, each successively more accurate measurement always fell within the margin of error for the previous measurement.

20. Originally Posted by RedPanda
Originally Posted by bangstrom
The length of a second is also pegged to the speed of light which makes the difficulty complete.
The length of a second is defined as:
"The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom. This definition refers to a caesium atom at rest at a temperature of 0 K."
I am not sure where the speed of light comes into it.
The speed of light is still defined as exactly 299,792,458 m/s and the length of a meter is defined as the distance traveled by light 1/299,792,458 second. This gives us a standard for both length and time. The duration of a second can best be measured with a cesium clock and it has been agreed by convention that the duration of a second be defined in cesium transitions but the original definition for the length of a second and the speed of light remain unchanged. The standard second is still pegged to the speed of light even though it is timed with a cesium clock.

21. Originally Posted by bangstrom
The speed of light is still defined as exactly 299,792,458 m/s and the length of a meter is defined as the distance traveled by light 1/299,792,458 second. This gives us a standard for both length and time.
It doesn't give the standard for time.
The standard definition for 1 second is: "The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom."
Certainly, you can calculate backwards from the speed of light to produce a number for how long a second is, but that is not how a second is defined.
As you said yourself: "it has been agreed by convention that the duration of a second be defined in cesium transitions" - it is not based on the speed of light.

Originally Posted by bangstrom
but the original definition for the length of a second ... remain[s] unchanged.
Original definition?
Do you mean this: "Between AD 1000 and 1960 the second was defined as 1/86,400 of a mean solar day"
Or this: "Between 1960 and 1967, it was defined in terms of the period of the Earth's orbit around the Sun in 1900"
Neither of which are based on the speed of light.

And what do archaic definitions of the length of a second have to do with the current definition?

Originally Posted by bangstrom
The standard second is still pegged to the speed of light even though it is timed with a cesium clock.
How is it 'pegged' to the speed of light if it is defined as "The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom."?
(I can see how the speed of light is 'pegged' to the length of a second - but that it not the same.)

22. Originally Posted by RedPanda
Originally Posted by bangstrom
The speed of light is still defined as exactly 299,792,458 m/s and the length of a meter is defined as the distance traveled by light 1/299,792,458 second. This gives us a standard for both length and time.
It doesn't give the standard for time.
The standard definition for 1 second is: "The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom."
Certainly, you can calculate backwards from the speed of light to produce a number for how long a second is, but that is not how a second is defined.
As you said yourself: "it has been agreed by convention that the duration of a second be defined in cesium transitions" - it is not based on the speed of light.

Originally Posted by bangstrom
but the original definition for the length of a second ... remain[s] unchanged.
Original definition?
Do you mean this: "Between AD 1000 and 1960 the second was defined as 1/86,400 of a mean solar day"
Or this: "Between 1960 and 1967, it was defined in terms of the period of the Earth's orbit around the Sun in 1900"
Neither of which are based on the speed of light.

And what do archaic definitions of the length of a second have to do with the current definition?

Originally Posted by bangstrom
The standard second is still pegged to the speed of light even though it is timed with a cesium clock.
How is it 'pegged' to the speed of light if it is defined as "The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom."?
(I can see how the speed of light is 'pegged' to the length of a second - but that it not the same.)
The one second length of time it takes light to travel some measurable fraction of 299,792,458 m is defined as equal to 9,192,631,770 periods of transition on a cesium clock so the two are exactly the same by definition.
By "original" definition for a second I meant the original definition based on the conventional speed of light defined as 299,792,458 m/s *exactly.*
Changing the defined units of a second does not change the length of a second or that it represents the period of time it takes light to travel 299,792,458 meters.

23. Originally Posted by bangstrom
The one second length of time it takes light to travel some measurable fraction of 299,792,458 m is defined as equal to 9,192,631,770 periods of transition on a cesium clock so the two are exactly the same by definition.
No it isn't.
It is based on the Ephemeris second.

Originally Posted by bangstrom
By "original" definition for a second I meant the original definition based on the conventional speed of light defined as 299,792,458 m/s *exactly.*
But that is not how the second was ever defined.

The definition of a second isn't - and hasn't ever been - based on (or pegged to) the speed of light.

24. Originally Posted by RedPanda
Originally Posted by bangstrom
The one second length of time it takes light to travel some measurable fraction of 299,792,458 m is defined as equal to 9,192,631,770 periods of transition on a cesium clock so the two are exactly the same by definition.
No it isn't.
It is based on the Ephemeris second.

Originally Posted by bangstrom
By "original" definition for a second I meant the original definition based on the conventional speed of light defined as 299,792,458 m/s *exactly.*
But that is not how the second was ever defined.

The definition of a second isn't - and hasn't ever been - based on (or pegged to) the speed of light.
The Ephemeris second is not easily determined, it is inexact, and it varies with time. To avoid confusion, its value is standardized by comparison with the SI second as measured with a cesium clock and the date on which the Ephemeris second was measured is also a part of the standard definition.
The speed of light would be a variable if the second in m/s were an Ephemeris second because the Ephemeris second varies randomly over short periods of time and it is growing longer over long periods of time. The last big earthquake and tsunami in Japan slowed the rate of Ephemeris time but the SI second and the standard speed of light remain unchanged.

25. Originally Posted by bangstrom
The Ephemeris second is not easily determined, it is inexact, and it varies with time. To avoid confusion, its value is standardized by comparison with the SI second as measured with a cesium clock and the date on which the Ephemeris second was measured is also a part of the standard definition.

The speed of light would be a variable if the second in m/s were an Ephemeris second because the Ephemeris second varies randomly over short periods of time and it is growing longer over long periods of time. The last big earthquake and tsunami in Japan slowed the rate of Ephemeris time but the SI second and the standard speed of light remain unchanged.
No - wrong again.
The second is defined using a specific value based on the average length of Ephemeris second.
That specific value is fixed; it doesn't change.

In summary:
The second is based on the Ephemeris second.
The Ephemeris second is not based on the speed of light.
Therefore the definition of a second is not based on (or pegged to) the speed of light.

26. [QUOTE=RedPanda;584080]
Originally Posted by bangstrom
No - wrong again.
The second is defined using a specific value based on the average length of Ephemeris second.
That specific value is fixed; it doesn't change.

In summary:
The second is based on the Ephemeris second.
The Ephemeris second is not based on the speed of light.
Therefore the definition of a second is not based on (or pegged to) the speed of light.
The SI second is the 'second' that is now based on the speed of light. The Ephemeris second and the SI second have both evolved over time by definition. The two seconds were both originally based on the rotation of the Earth but the SI second is now pegged to the value of c and the Ephemeris second is pegged to the value of the SI second since they can both be standardized by a caesium clock. This gives the Ephemeris second a precisely defined value when previously it was inexact and variable. I would think that since, the Ephemeris second is pegged to the value of the SI second and the SI second is pegged to the value of c, this would mean they are both pegged to the value of c by extention.

The SI units for meters, seconds, and the value of c are all mutually defined by convention so the speed of light in a vacuum can not be measured as anything other than c.

A quote from Wikipedia:

The hyperfine structure transition can be used to make a microwave notch filter with very high stability, repeatability and Q factor, which can thus be used as a basis for very precise atomic clocks. Typically, the hyperfine structure transition frequency of a particular isotope of caesium or rubidium atoms is used as a basis for these clocks.
Due to the accuracy of hyperfine structure transition-based atomic clocks, they are now used as the basis for the definition of the second. One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms.
Since 1983, the meter is defined by declaring the speed of light in a vacuum to be exactly 299,792,458 metres per second. Dividing the number of cycles with the speed of light yields:
The metre is the length of the path travelled by light in vacuum during a time interval of 30.6633189884984 caesium-133 hyperfine transition cycles.

Hyperfine structure - Wikipedia, the free encyclopedia

27. Originally Posted by bangstrom
The SI second is the 'second' that is now based on the speed of light.
No, it isn't.
I'll post it again, just in case you missed the first 3 times I posted it:
"The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom."

Originally Posted by bangstrom
The Ephemeris second and the SI second have both evolved over time by definition.
That doesn't really make sense - but regardless, what ever the previous definitions were, they are irrelevant when describing what the current definition is.

Originally Posted by bangstrom
The two seconds were both originally based on the rotation of the Earth but the SI second is now pegged to the value of c and the Ephemeris second is pegged to the value of the SI second since they can both be standardized by a caesium clock.
This gives the Ephemeris second a precisely defined value when previously it was inexact and variable. I would think that since, the Ephemeris second is pegged to the value of the SI second and the SI second is pegged to the value of c, this would mean they are both pegged to the value of c by extention.

The SI units for meters, seconds, and the value of c are all mutually defined by convention so the speed of light in a vacuum can not be measured as anything other than c.
All of that is wrong.
I am not sure where you are getting your information from, but it is not correct.

Originally Posted by bangstrom
A quote from Wikipedia:

The hyperfine structure transition can be used to make a microwave notch filter with very high stability, repeatability and Q factor, which can thus be used as a basis for very precise atomic clocks. Typically, the hyperfine structure transition frequency of a particular isotope of caesium or rubidium atoms is used as a basis for these clocks.
Due to the accuracy of hyperfine structure transition-based atomic clocks, they are now used as the basis for the definition of the second. One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms.
Since 1983, the meter is defined by declaring the speed of light in a vacuum to be exactly 299,792,458 metres per second. Dividing the number of cycles with the speed of light yields:
The metre is the length of the path travelled by light in vacuum during a time interval of 30.6633189884984 caesium-133 hyperfine transition cycles.
Nowhere in that quote does it say that the second is based on the speed of light.

Are you not able to find anything supporting your claim?

28. [QUOTE=bangstrom;584132]
Originally Posted by RedPanda
Originally Posted by bangstrom
No - wrong again.
The second is defined using a specific value based on the average length of Ephemeris second.
That specific value is fixed; it doesn't change.

In summary:
The second is based on the Ephemeris second.
The Ephemeris second is not based on the speed of light.
Therefore the definition of a second is not based on (or pegged to) the speed of light.
The SI second is the 'second' that is now based on the speed of light. The Ephemeris second and the SI second have both evolved over time by definition. The two seconds were both originally based on the rotation of the Earth but the SI second is now pegged to the value of c and the Ephemeris second is pegged to the value of the SI second since they can both be standardized by a caesium clock. This gives the Ephemeris second a precisely defined value when previously it was inexact and variable. I would think that since, the Ephemeris second is pegged to the value of the SI second and the SI second is pegged to the value of c, this would mean they are both pegged to the value of c by extention.

The SI units for meters, seconds, and the value of c are all mutually defined by convention so the speed of light in a vacuum can not be measured as anything other than c.

A quote from Wikipedia:

The hyperfine structure transition can be used to make a microwave notch filter with very high stability, repeatability and Q factor, which can thus be used as a basis for very precise atomic clocks. Typically, the hyperfine structure transition frequency of a particular isotope of caesium or rubidium atoms is used as a basis for these clocks.
Due to the accuracy of hyperfine structure transition-based atomic clocks, they are now used as the basis for the definition of the second. One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms.
Since 1983, the meter is defined by declaring the speed of light in a vacuum to be exactly 299,792,458 metres per second. Dividing the number of cycles with the speed of light yields:
The metre is the length of the path travelled by light in vacuum during a time interval of 30.6633189884984 caesium-133 hyperfine transition cycles.

Hyperfine structure - Wikipedia, the free encyclopedia
As Red Panda is saying, this definition of the second is based on the time required to count a number of cycles of microwave radiation, assuming the radiating source and the counting device are not in relative motion. That being so, the speed of the radiation does not seem to come into it.

The definition of the metre does depends on c, certainly, but surely that is irrelevant, as the definition of the second does not rely on measuring any length?

29. Originally Posted by RedPanda
I'll post it again, just in case you missed the first 3 times I posted it:
"The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom."
The
9,192,631,770 periods is the length of time it takes light to travel 299,792,458 meters. That is the connection between the SI second and the speed of light.

Also, the Wikipedia article includes the connection between the SI second and the speed of light but you have to do the math to find it.

The article says, "One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms.
Since 1983, the meter is defined by declaring the speed of light in a vacuum to be exactly 299,792,458 metres per second. Dividing the number of cycles with the speed of light yields:
The metre is the length of the path travelled by light in vacuum during a time interval of
30.6633189884984 caesium-133 hyperfine transition cycles."

So, if light travels at c (299,792,458 m/s) and one second is defined as 9,192,631,770 transitions on a caesium clock, then the time it takes light to travel one meter equals 9,192,631,770 transitions divided by 299,792,458 meters which gives you the value of meter expressed in units of time as 30.6633189884984 caesium transitions as stated in the article.
You can continue with the same numbers to solve for the value of one second (
9,192,631,770 cycles) the length of a meter, or the value of c. Given any two of the values for meters, seconds, or c, you can solve for the value of the third since they are all mutually defined in the SI system. Everything is based on c as a universal absolute.

30. Originally Posted by exchemist
As Red Panda is saying, this definition of the second is based on the time required to count a number of cycles of microwave radiation, assuming the radiating source and the counting device are not in relative motion. That being so, the speed of the radiation does not seem to come into it.

The definition of the metre does depends on c, certainly, but surely that is irrelevant, as the definition of the second does not rely on measuring any length?
The definition of a second is length related since the number of cycles you mentioned is the number of cycles of the microwave radiation in 299,792,458 meters. This is directly related to the wavelength of the radiating source. One second is the time it takes light to travel 299,792,458 meters.

31. Originally Posted by bangstrom
Originally Posted by RedPanda
I'll post it again, just in case you missed the first 3 times I posted it:
"The duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom."
The 9,192,631,770 periods is the length of time it takes light to travel 299,792,458 meters. That is the connection between the SI second and the speed of light.
The 9,192,631,770 periods is also the length of time it takes Usain Bolt to travel 10.44 meters. That is the connection between the SI second and the speed of Usain Bolt.

Any speed is "connected" to seconds - of course it is! It is meters per second.
But seconds are not defined by the speed of anything.

Originally Posted by bangstrom
So, if light travels at c (299,792,458 m/s) and one second is defined as 9,192,631,770 transitions on a caesium clock, then the time it takes light to travel one meter equals 9,192,631,770 transitions divided by 299,792,458 meters which gives you the value of meter expressed in units of time as 30.6633189884984 caesium transitions as stated in the article.
But that has nothing to do with the definition of a second.
Because, as you have agreed, the definition of a second is:
"One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms."
Nowhere in that definition is the speed of light even mentioned.

I find it weird that you continue to claim that the definition of a second is based on the speed of light but then are unable to show exactly where in the definition the speed of light is used.
Do you not also find that strange?

Originally Posted by bangstrom
You can continue with the same numbers to solve for the value of one second (9,192,631,770 cycles) the length of a meter, or the value of c.
Given any two of the values for meters, seconds, or c, you can solve for the value of the third since they are all mutually defined in the SI system.
That is simply how equations work.

V = I x R
Given any two of the values for voltage, current, or resistance, you can solve for the value of the third since they are all mutually defined in the SI system?

F = m x a
Given any two of the values for force, mass, or acceleration, you can solve for the value of the third since they are all mutually defined in the SI system?

So - according to you - all definitions are circular?

Originally Posted by bangstrom
Everything is based on c as a universal absolute.
Nope.
I'll post it again:
"One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms."
As you can see, that definition has nothing to do with the speed of light.

32. Originally Posted by bangstrom
Originally Posted by exchemist
As Red Panda is saying, this definition of the second is based on the time required to count a number of cycles of microwave radiation, assuming the radiating source and the counting device are not in relative motion. That being so, the speed of the radiation does not seem to come into it.

The definition of the metre does depends on c, certainly, but surely that is irrelevant, as the definition of the second does not rely on measuring any length?
The definition of a second is length related since the number of cycles you mentioned is the number of cycles of the microwave radiation in 299,792,458 meters. This is directly related to the wavelength of the radiating source. One second is the time it takes light to travel 299,792,458 meters.
No, the number of cycles in a second is purely a count of the arrival, at your detector, of a number of wavecrests. You do not need length to do that.

33. Besides, both arguments ignore the fact that one "Day" is defined as 86400 seconds, and one "Day" is one complete spin of the Earth (or well... slightly more so that it ends up facing the Sun.)

Then there is the matter of a "Year" being 365 days except on "Leap Years" when it is 366 days.

So if the speed of light were to change by more than a miniscule fraction of its current accepted value, and we were using the speed of light to measure seconds, I think people would notice when they went to take their lunch break and saw that it was dark outside.

Originally Posted by Markus Hanke
This leads to the very counter-intuitive result that no stationary far-away observer ever sees anything reach the event horizon of a black hole, while at the same time someone in free fall will cross it in finite and well-defined proper time as measured on his own watch; this is not a contradiction, but a simple manifestation of the nature of gravity.

This in turn, gets more confusing when stationary, far away observers, start to ask ourselves whether the infalling observer has crossed the event horizon yet.

Present theory asserts that he will eventually fall through, but it can't say when. If the infalling observer is very near the event horizon, then the infalling observer's answer is "about a billionth of a second from now".

And what was his answer yesterday? "About a billionth of a second from now." What will his answer probably be tomorrow? "About a billionth of a second from now." He's always just about to fall through.

34. Originally Posted by RedPanda
I'll post it again:
"One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms."
As you can see, that definition has nothing to do with the speed of light.
The number 9,192,631,770 is in Hz or cycles per second of EM radiation. There is no need to mention the speed of light or the wavelength of a 9 GHz signal because these values redundant when all you need is the number of cycles. But if one needs to know the precise wavelength of a 9 GHz signal it can be calculated given the value of c. When you use EM waves to measure distance or time, you are using a c related measure.

35. Originally Posted by exchemist
No, the number of cycles in a second is purely a count of the arrival, at your detector, of a number of wavecrests. You do not need length to do that.
Counting cps in an EM radiation is meaningless unless you know the precise c related frequency of the signal. Someone had to calculate the values of the frequency and wavelength of a caesium clock to determine the number of cycles you need to count.

36. Originally Posted by kojax
Besides, both arguments ignore the fact that one "Day" is defined as 86400 seconds, and one "Day" is one complete spin of the Earth (or well... slightly more so that it ends up facing the Sun.)

Then there is the matter of a "Year" being 365 days except on "Leap Years" when it is 366 days.

So if the speed of light were to change by more than a miniscule fraction of its current accepted value, and we were using the speed of light to measure seconds, I think people would notice when they went to take their lunch break and saw that it was dark outside.
The speed of light is independent of the Earth's rotation. A change in the speed of light would not be apparent locally.

We have two seconds, the SI second is based on the speed of light and the Ephemeris second is based on the rotation of the Earth. The Ephemeris second is used for astronomical work where the rotation of the Earth is critical and the SI second is used for nearly everything else. The two seconds are made equal by definition so they are essentially the same. The Ephemeris second is less exact than the SI second so if the two are not precisely the same the difference is below detection. The discrepancies that do arise are handled every few years by adding or subtracting a "leap second" to the international clock.
Originally Posted by kojax
This in turn, gets more confusing when stationary, far away observers, start to ask ourselves whether the infalling observer has crossed the event horizon yet.

Present theory asserts that he will eventually fall through, but it can't say when. If the infalling observer is very near the event horizon, then the infalling observer's answer is "about a billionth of a second from now".

And what was his answer yesterday? "About a billionth of a second from now." What will his answer probably be tomorrow? "About a billionth of a second from now." He's always just about to fall through.
Our yesterday was not his yesterday. If he reports back every billionth of a second, then his "billionth of a second" would be his last report but from our perspective, his last report may take days or weeks to complete as he sloooooooly reports back.

37. Originally Posted by bangstrom
The number 9,192,631,770 is in Hz or cycles per second of EM radiation.
So, the definition of a second is 9,192,631,770 cycles per second of EM radiation?
Really?
Do you really think that is the definition of a second?
Can you not see a problem with that?

The number 9,192,631,770 is not measured in Hz or cycles per second.

Originally Posted by bangstrom
But if one needs to know the precise wavelength of a 9 GHz signal it can be calculated given the value of c. When you use EM waves to measure distance or time, you are using a c related measure.
As I said:
That is simply how equations work.

V = I x R
Given any two of the values for voltage, current, or resistance, you can solve for the value of the third since they are all mutually defined in the SI system?

F = m x a
Given any two of the values for force, mass, or acceleration, you can solve for the value of the third since they are all mutually defined in the SI system?

So - according to you - all definitions are circular?

Originally Posted by bangstrom
the SI second is based on the speed of light
Nope.
I'll post it again:
"One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms."
As you can see, that definition has nothing to do with the speed of light.

38. Originally Posted by bangstrom
Originally Posted by exchemist
No, the number of cycles in a second is purely a count of the arrival, at your detector, of a number of wavecrests. You do not need length to do that.
Counting cps in an EM radiation is meaningless unless you know the precise c related frequency of the signal. Someone had to calculate the values of the frequency and wavelength of a caesium clock to determine the number of cycles you need to count.
No you just count the cycles and after the requisite number you know one second has gone by.

39. Originally Posted by RedPanda
So, the definition of a second is 9,192,631,770 cycles per second of EM radiation?
Really?
Do you really think that is the definition of a second?
Can you not see a problem with that?
I know my explanation is not the standard definition of a second but I am trying to explain what a SI second is and not just repeat the same old definition. The standard definition involves transition cycles in a caesium atom but we detect those transitions as a microwave radio frequency emission. So the number 9,192,631,770 is the radio frequency emitted by electrons in a caesium atom when they drop from a high energy level to a lower energy level.

The definition of a second was written by people who know what they are doing and, for simplicity, they only include the essential information that can be followed by even by those who do not know what they are doing.

Originally Posted by RedPanda
The number 9,192,631,770 is not measured in Hz or cycles per second.
I figure that, if a second is defined as 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133, atoms, then the number 9,192,631,770 must be in units of Hz or cycles per second. Where did I go wrong?

40. Originally Posted by exchemist
Originally Posted by bangstrom
Originally Posted by exchemist
No, the number of cycles in a second is purely a count of the arrival, at your detector, of a number of wavecrests. You do not need length to do that.
Counting cps in an EM radiation is meaningless unless you know the precise c related frequency of the signal. Someone had to calculate the values of the frequency and wavelength of a caesium clock to determine the number of cycles you need to count.
No you just count the cycles and after the requisite number you know one second has gone by.
Do you understand that the "cycles" you are counting are cycles of a microwave radio signal traveling at c with a specific wavelength so you are not just measuring an interval of time but also a c related interval of distance even though time is the only part of the measurement you are interested in? And, the wavelength of the signal matters because, the longer the wavelength, the fewer cycles you need to count to determine the duration of a second.

41. Originally Posted by bangstrom
Originally Posted by RedPanda
So, the definition of a second is 9,192,631,770 cycles per second of EM radiation?
Really?
Do you really think that is the definition of a second?
Can you not see a problem with that?
I know my explanation is not the standard definition of a second but I am trying to explain what a SI second is and not just repeat the same old definition. The standard definition involves transition cycles in a caesium atom but we detect those transitions as a microwave radio frequency emission. So the number 9,192,631,770 is the radio frequency emitted by electrons in a caesium atom when they drop from a high energy level to a lower energy level.

The definition of a second was written by people who know what they are doing and, for simplicity, they only include the essential information that can be followed by even by those who do not know what they are doing.

Originally Posted by RedPanda
The number 9,192,631,770 is not measured in Hz or cycles per second.
I figure that, if a second is defined as 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133, atoms, then the number 9,192,631,770 must be in units of Hz or cycles per second. Where did I go wrong?
It's a pure number. It has no units.

42. It's been explained why bangstrom is wrong numerous times, if he still can't grasp it I guess he's just one of those people who are dumb...

43. Originally Posted by exchemist
It's a pure number. It has no units.
The number 9,192,631,770 is not just a number. It has a unitary value of cycles per second (Hz). There are 9,192,631,770 cycles in a second so the number is in cycles per second. The cycles you are counting are cycles of a 9 GHz radio signal. Which part of that does not make sense? Do you not understand that 9,192,631,770 is in units of cycles per second or that the cycles you are counting are cycles of a radio signal?

Just to toss out some big numbers, a radio signal of 9,192,631,770 Hz has a wavelength of 0.03261225571749406 m and 9,192,631,770 cycles of that length amounts to a total distance of 299,792,458 m. That is the distance a radio signal travels in one second. And, 299,792,458 m/s is the value of c. This is no coincidence because our units of length, time, and c are all mutually defined by convention and using units of meters and seconds to measure the speed of light can only return the same value for c that was used in the calibration of meters and seconds. This is no different from trying to measure the speed of light over the distance of a light year. It makes no sense to even try. The system is rigged so we can only measure the speed of light in a vacuum as c.

44. Originally Posted by bangstrom
The number 9,192,631,770 is not just a number. It has a unitary value of cycles per second (Hz). There are 9,192,631,770 cycles in a second so the number is in cycles per second.
Yes, you're really having problems of comprehension here, aren't you.
We picked that NUMBER and THEN defined it as second.
Hence it is "cycles per second" post hoc ONLY.
The post hoc part is what seems to be eluding you.
If we'd decided that 3 cycles was the count required to define a second then the current value would actually be 3064210590 Hz.

45. Originally Posted by bangstrom
Originally Posted by RedPanda
Can you not see a problem with that?
I know my explanation is not the standard definition of a second but I am trying to explain what a SI second is and not just repeat the same old definition. The standard definition involves transition cycles in a caesium atom but we detect those transitions as a microwave radio frequency emission. So the number 9,192,631,770 is the radio frequency emitted by electrons in a caesium atom when they drop from a high energy level to a lower energy level.
The problem you don't seem to be able to see is that (according to you) the definition includes what it is defining.
You say that a second is 9,192,631,770 caesium cycles per second.
But that still leaves you not knowing how long a second is.

But, as pointed out by others, it is just a number; a simple count of a particular object.
It doesn't have a unit.

Originally Posted by bangstrom
The definition of a second was written by people who know what they are doing and, for simplicity, they only include the essential information that can be followed by even by those who do not know what they are doing.
But you are also claiming that they left out key information.

I'll post it again:
"One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms."

So, despite the definition being written by "people who know what they are doing" who "include[d] the essential information", they forgot to mention they were counting cycles per second.
(It also assumes they weren't clever enough to notice that their definition relied on the very thing they are defining.)

Originally Posted by bangstrom
Originally Posted by RedPanda
The number 9,192,631,770 is not measured in Hz or cycles per second.
I figure that, if a second is defined as 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133, atoms, then the number 9,192,631,770 must be in units of Hz or cycles per second. Where did I go wrong?
Ok - let's look at this part of your definition: "Cycles per second"
Clearly, 'cycles' are not 'seconds', correct?
And if you wanted to measure the number of cycles per second, you would need to count both the 'cycles' and the 'seconds', yes?
If you counted 20 cycles in 10 seconds then you would say there were 20 cycles per 10 seconds (or 2 cycles per second), yes?

Now, when you were counting the cycles, what unit were they?
Was each cycle you counted "1 cycle per second"?
Would your final figure actually be "20 cycles per second every 10 seconds"?

Does this help show you where you went wrong?

46. cycles per second per second?
wouldn't that mean an accelerating frequency?

47. Originally Posted by Dywyddyr
Yes, you're really having problems of comprehension here, aren't you.
We picked that NUMBER and THEN defined it as second.
Hence it is "cycles per second" post hoc ONLY.
The post hoc part is what seems to be eluding you.
You are exactly right but the post hoc did not escape my attention. If you look carefully at the numbers in my Wikipedia quote it should be apparent that the ENTIRE SI system is post hoc beginning with the premise that c is an absolute. They start with the value of c and work backwards from there to define our units of length and time.
The confusion began when I defined the "second" as the time it takes to light to travel the distance of 299,792,458 m. I don't see how this should be controversial because the speed of light is 299,792,458 m/s but everyone complained because that was NOT the definition of a second.

The second is defined as, " Exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms." OK, but if you crunch the numbers considering both distance and time, the standard definition tells us that light travels 299,792,458 m in one second. And the number 9,192,631,770 is in units of cycles per second. The cycles have a specific wavelength so counting cycles gives us values of both time and distance.
Originally Posted by Dywyddyr
If we'd decided that 3 cycles was the count required to define a second then the current value would actually be 3064210590 Hz.
Yes, we could define one second as 3 cycles but that would require a signal of 3 Hz. Not 3064210590 Hz. Do you agree?

We could also define one second as 3,064,210,590 Hz for a more reasonable value and that would not change the fact that, "One second is the time it takes for light to travel 299,792,458 m." So that definition of a second would still apply but it would change the standard definition that everyone keeps quoting. It would require that we either change our signal source to 3,064,210,590 Hz or count three cycles on a quartz clock as one. I think that is the point you are trying to make. It is all post hoc beginning with the speed of light as an absolute and working backwards from there.

48. Originally Posted by bangstrom
You are exactly right but the post hoc did not escape my attention.
And yet you persist in using it as an a priori argument.

If you look carefully at the numbers in my Wikipedia quote it should be apparent that the ENTIRE SI system is post hoc beginning with the premise that c is an absolute. They start with the value of c and work backwards from there to define our units of length and time.
You have been shown, and told, time and time again that you are wrong.

The confusion began when I defined the "second" as the time it takes to light to travel the distance of 299,792,458 m. I don't see how this should be controversial because the speed of light is 299,792,458 m/s but everyone complained because that was NOT the definition of a second.
The "confusion" appears because it is NOT the definition of a second while you persist in claiming that it is.

The second is defined as, " Exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms." OK, but if you crunch the numbers considering both distance and time, the standard definition tells us that light travels 299,792,458 m in one second. And the number 9,192,631,770 is in units of cycles per second. The cycles have a specific wavelength so counting cycles gives us values of both time and distance.
Still wrong.
Ignore, completely the "cycles per second": as I have pointed out, and - as you have claimed you weren't ignoring (but apparently still are) - the "per second" is entirely post hoc.
We count a number that occurs naturally - a cycle. When we have counted a particular quantity of them we THEN define the period taken to be one second.
There is NO a priori "second" in that count.

]Yes, we could define one second as 3 cycles but that would require a signal of 3 Hz. Not 3064210590 Hz. Do you agree?
No, you're back to bollocks again.
There is no requirement whatsoever for Hz. The Hz is reliant on the definition of a second and that definition is a count of a specified NUMBER. Not a frequency.

We could also define one second as 3,064,210,590 Hz for a more reasonable value and that would not change the fact that, "One second is the time it takes for light to travel 299,792,458 m."
Of course it would.
If we redefine a second then light will travel a different distance in that second.

So that definition of a second would still apply but it would change the standard definition that everyone keeps quoting. It would require that we either change our signal source to 3,064,210,590 Hz or count three cycles on a quartz clock as one. I think that is the point you are trying to make. It is all post hoc beginning with the speed of light as an absolute and working backwards from there.
You're utterly clueless.

49. Originally Posted by RedPanda
The problem you don't seem to be able to see is that (according to you) the definition includes what it is defining.
You say that a second is 9,192,631,770 caesium cycles per second.
But that still leaves you not knowing how long a second is.
Obviously my definition includes the word it is defining but it does tell you how long a second is. A second is 9,192,631,770 caesium cycles per second.

I know there is a catch 22 to this explanation because you need to know how long a second is to determine how long a second is but the duration of a second has already been determined by international convention and the problem is how to measure a second in your laboratory that is consistent with the standard. The 9,192,631,770 Hz is a reference value for converting cycles of EM radiation to units of time.
If you have measured an interval of time on a caesium clock, the conversion is easy because because you only need to divide by 9,192,631,770 Hz to convert your measurement to seconds. If your measurement is in cycles of a light laser, you need to know the Hz value of your laser to make the same conversion.

If you multiply the Hz of your laser by the wavelength of your laser, the result is 299,792,458m/s. The wavelength of EM radiation is c/Hz. So Hz times the wavelength (c/Hz) equals c. This should tell you that the 9,192,631,770 in the definition is in units of cycles per second and it is not just a number. The whole SI system is based on value of c as a starting point.
Originally Posted by RedPanda
I'll post it again:
"One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms."

So, despite the definition being written by "people who know what they are doing" who "include[d] the essential information", they forgot to mention they were counting cycles per second. The time interval is included in
(It also assumes they weren't clever enough to notice that their definition relied on the very thing they are defining.)
The catch 22 never escaped anyone's notice. The purpose of the definition is to make it possible any laboratory in the world to measure the interval of a second that is consistent with the international standard.
And they did include the information that the number of 9,192,631,770 is in cps (Hz). Look at the definition, "One second is now defined to be exactly 9,192,631,770 cycles..." If there are 9,192,631,770 cycles in a second, then 9,192,631,770 must be in cycles per second. There is no need to state it twice by putting a Hz behind the number.

Originally Posted by RedPanda
And if you wanted to measure the number of cycles per second, you would need to count both the 'cycles' and the 'seconds', yes?
No, this is where we disagree. If you measure a number of cycles and you have a reverence value where 9,192,631,770 cycles equals one second, then you can divide your number of cycles by 9,192,631,770 and convert the number of cycles into seconds. The 9,192,631,770 is in units of Hz which allows you to convert cycles to seconds without a separate measurement of seconds. The time interval is included in the measurement of cycles alone. You just need to know the conversion factor.

50. A second is 9,192,631,770 caesium cycles per second.
This is nonsensical.

51. Originally Posted by bangstrom
Obviously my definition includes the word it is defining but it does tell you how long a second is. A second is 9,192,631,770 caesium cycles per second.
Your "definition" is incorrect.

I know there is a catch 22 to this explanation because you need to know how long a second is to determine how long a second is
No you don't.
A second is however long it takes for caesium to go through that set number of cycles.

This should tell you that the 9,192,631,770 in the definition is in units of cycles per second and it is not just a number.
No.

The whole SI system is based on value of c as a starting point.
And no.

If there are 9,192,631,770 cycles in a second, then 9,192,631,770 must be in cycles per second.
No.

There is no need to state it twice by putting a Hz behind the number.
You REALLY should get an education.

No, this is where we disagree. If you measure a number of cycles and you have a reverence value where 9,192,631,770 cycles equals one second, then you can divide your number of cycles by 9,192,631,770 and convert the number of cycles into seconds. The 9,192,631,770 is in units of Hz
You start off as if you're beginning to understand and then persist in adding Hz.
(Oh, and the word is "reference" not "reverence").
A second is the length of time it takes for that number of cycles.
The second is DEFINED by however long it takes for that number of cycles to occur.
Thus, if we had no other method of determining time (i.e. no other comparative measures), and if those cycles slowed down then the length of a (defined) second would increase. And we would never know.
The definition is not recursive and is not reliant on Hz.
Nor the speed of light.

52. Originally Posted by dan hunter
cycles per second per second?
wouldn't that mean an accelerating frequency?
I don't know what this is in reference to but cycles per second per second would be an accelerating frequency.

There was a time when I wrote cps (HZ) meaning that the units could be written either as cps or Hz. Was that it?

Or my comment that, "One second is now defined to be exactly 9,192,631,770 cycles..." If there are 9,192,631,770 cycles in a second, then 9,192,631,770 must be in cycles per second. There is no need to state it twice by putting a Hz behind the number."

The number 9,192,631,770 is in units of cycles per second as per the definition and to put the units directly behind the number would be redundant. I reject the opinion that 9,192,631,770 is a unitless number.

53. Originally Posted by bangstrom
I reject the opinion that 9,192,631,770 is a unitless number.
It's not a f*cking opinion.

54. Originally Posted by bangstrom
Originally Posted by RedPanda
And if you wanted to measure the number of cycles per second, you would need to count both the 'cycles' and the 'seconds', yes?
No, this is where we disagree. If you measure a number of cycles and you have a reverence value where 9,192,631,770 cycles equals one second, then you can divide your number of cycles by 9,192,631,770 and convert the number of cycles into seconds.
So, are you saying it is impossible to count the number of cycles whilst counting the number of seconds?
I presume you don't think it is impossible, so please answer my questions.

These are not just random questions.
I posted them in response to your request to show you where you went wrong.
(I have rephrased one of the questions to deter you from going down the rabbit hole of alternative calculations.)

So - I repeat:

Ok - let's look at this part of your definition: "Cycles per second"
Clearly, 'cycles' are not 'seconds', correct?
And if you wanted to measure the number of cycles per second, you could count both the 'cycles' and the 'seconds', yes?
If you counted 20 cycles in 10 seconds then you would say there were 20 cycles per 10 seconds (or 2 cycles per second), yes?

Now, when you were counting the cycles, what unit were they?
Was each cycle you counted "1 cycle per second"?
Would your final figure actually be "20 cycles per second every 10 seconds"?

55. Originally Posted by bangstrom
Originally Posted by kojax
Besides, both arguments ignore the fact that one "Day" is defined as 86400 seconds, and one "Day" is one complete spin of the Earth (or well... slightly more so that it ends up facing the Sun.)

Then there is the matter of a "Year" being 365 days except on "Leap Years" when it is 366 days.

So if the speed of light were to change by more than a miniscule fraction of its current accepted value, and we were using the speed of light to measure seconds, I think people would notice when they went to take their lunch break and saw that it was dark outside.
The speed of light is independent of the Earth's rotation. A change in the speed of light would not be apparent locally.

We have two seconds, the SI second is based on the speed of light and the Ephemeris second is based on the rotation of the Earth. The Ephemeris second is used for astronomical work where the rotation of the Earth is critical and the SI second is used for nearly everything else. The two seconds are made equal by definition so they are essentially the same. The Ephemeris second is less exact than the SI second so if the two are not precisely the same the difference is below detection. The discrepancies that do arise are handled every few years by adding or subtracting a "leap second" to the international clock.
If the difference between an Ephemeris second and a SI second is that small, then why do you care? The point is that if the speed of light were changing, the SI second would begin to deviate further from the Ephemeris second and someone would notice.

Personally I never bother using 299,792,458 m as my value for C. I just use 300,000,000 and don't worry much about it, because I think a difference as small as 0.0692% isn't probably going to affect my calculations very much, unless I'm doing something that requires extremely high precision.

Originally Posted by kojax
This in turn, gets more confusing when stationary, far away observers, start to ask ourselves whether the infalling observer has crossed the event horizon yet.

Present theory asserts that he will eventually fall through, but it can't say when. If the infalling observer is very near the event horizon, then the infalling observer's answer is "about a billionth of a second from now".

And what was his answer yesterday? "About a billionth of a second from now." What will his answer probably be tomorrow? "About a billionth of a second from now." He's always just about to fall through.
Our yesterday was not his yesterday. If he reports back every billionth of a second, then his "billionth of a second" would be his last report but from our perspective, his last report may take days or weeks to complete as he sloooooooly reports back.
Or his last report could take millions or billions of years to complete, as he gets closer and closer to the event horizon. If he's ten billionths of a second away in free fall, and reporting in increments of one billionth of a second, then a lot of time might pass out here before he gets from 10 to 9.

Of course, by that point we also won't be able to pick up his signals, because any EM signals he is emitting will be badly redshifted. Like if he's sending them at 100 mhz, we might receive the signal at a rate of one cycle per 10 centuries. We'd need an awfully big antenna.

Originally Posted by bangstrom
The system is rigged so we can only measure the speed of light in a vacuum as c.
You've admitted that there are units available which can be used to confirm the validity of our measurements, though. If you think the SI second is messed up, you are free to use Ephemeral seconds.

If you don't like using a "light year" as a unit of measurement for distance, feel free to use Astronomical Units.

Astronomical unit - Wikipedia, the free encyclopedia

Or just use a meter stick. Or if you really don't like meters, you can always do like we do in the USA and go back to using feet, yards, and miles.

56. Originally Posted by mathman
Originally Posted by ultimatesceptic
The speed of light was fixed in 1972, and the length of a metre was pegged to the speed of light. Therefore the speed of light (in metric) cannot vary, even though it does in fact fluctuate over time.

Is this true? I believe it is.
Believing it doesn't make it so. Do you have any evidence that it varies over time?
If something is travelling at 300,000 meters per second squared or at one meter per second we still see it. And the speed of light isn't the same travelling at 300,000 meters per second as it is travelling at one meter per second. Things don't look the same travelling at different speeds, therefore, it follows that the speed of light varies!*

57. Originally Posted by Dywyddyr
You start off as if you're beginning to understand and then persist in adding Hz.
(Oh, and the word is "reference" not "reverence").
A second is the length of time it takes for that number of cycles.
The second is DEFINED by however long it takes for that number of cycles to occur.
Thus, if we had no other method of determining time (i.e. no other comparative measures), and if those cycles slowed down then the length of a (defined) second would increase. And we would never know.
The definition is not recursive and is not reliant on Hz.
Nor the speed of light.
I get the point that I am wrong, wrong, wrong but if that statement is not followed by an explanation of why I am wrong then I have no idea what you are talking about.

Are there points we can agree upon? Here are some of your statements in quotes and my comments below.

"The 9,192,631,770 is in units of Hz You start off as if you're beginning to understand and then persist in adding Hz."

Can you explain why this number has no units? Or far better, demonstrate that this number is independent of c.

A radio signal is said to travel the distance 299,792,458 meters in one second as do all EM signals. So, while we are counting 9,192,631,770 cycles to determine the duration of one second, the same signal can propagate to a distance of 299,792,458 m. There are 9,192,631,770 cycles (or wavelengths) in the signal's 299,792,458 m and, since the signal is propagating at c, we are counting cycles at the rate of 9,192,631,770 cycles per second. I find it strange to explain how, if there are 9,192,631,770 cycles in a one second interval of a radio signal, then the signal has a frequency of 9,192,631,770 Hz. It should be kind of obvious.

Also, by this same explanation, a "second" is the time it takes a EM signal to travel 299,792,458 m. This follows directly from the definition of c as 299,792,458 m/s.

"The second is DEFINED by however long it takes for that number of cycles to occur."

Agreed.

"Thus, if we had no other method of determining time (i.e. no other comparative measures), and if those cycles slowed down then the length of a (defined) second would increase. And we would never know."

Agreed.

"The definition is not recursive and is not reliant on Hz.
Nor the speed of light."

I disagree with all three as I explained in the beginning.

58. Originally Posted by bangstrom
I get the point that I am wrong, wrong, wrong but if that statement is not followed by an explanation of why I am wrong then I have no idea what you are talking about.
It HAS been explained. Numerous times.

Can you explain why this number has no units? Or far better, demonstrate that this number is independent of c.
It requires no units because it's a simple count.

A radio signal is said to travel the distance 299,792,458 meters in one second as do all EM signals.
Utterly and totally irrelevant.

So, while we are counting 9,192,631,770 cycles to determine the duration of one second, the same signal can propagate to a distance of 299,792,458 m.
So what?
The distance that signal travels is irrelevant to the actual counting.
Even if (for some reason) the signal propagated at less than c (for example if we used some other "device" to get that count) the second is defined purely and only in terms of the number of counts.

"The second is DEFINED by however long it takes for that number of cycles to occur."
Agreed.
Then you agree that it's merely a count.

"Thus, if we had no other method of determining time (i.e. no other comparative measures), and if those cycles slowed down then the length of a (defined) second would increase. And we would never know."
Agreed.
Then you agree that it's merely a count.

I disagree with all three as I explained in the beginning.
And this is why you are wrong.
You have already agreed that we simply count the number of cycles: i.e. how many times it has occurred.
After we get to the required number we define the time taken to reach that number as one second. (For at least the second time of writing this out).
It is not, and cannot be recursive, nor is it, and cannot be, Hz because the "second" is defined ONLY afterwards as the time that has elapsed from start to finish of that count.

59. Originally Posted by RedPanda
Originally Posted by bangstrom
Originally Posted by RedPanda
And if you wanted to measure the number of cycles per second, you would need to count both the 'cycles' and the 'seconds', yes?
No, this is where we disagree. If you measure a number of cycles and you have a reverence value where 9,192,631,770 cycles equals one second, then you can divide your number of cycles by 9,192,631,770 and convert the number of cycles into seconds.
So, are you saying it is impossible to count the number of cycles whilst counting the number of seconds?
I presume you don't think it is impossible, so please answer my questions.

These are not just random questions.
I posted them in response to your request to show you where you went wrong.
(I have rephrased one of the questions to deter you from going down the rabbit hole of alternative calculations.)

So - I repeat:

Ok - let's look at this part of your definition: "Cycles per second"
Clearly, 'cycles' are not 'seconds', correct?
And if you wanted to measure the number of cycles per second, you could count both the 'cycles' and the 'seconds', yes?
If you counted 20 cycles in 10 seconds then you would say there were 20 cycles per 10 seconds (or 2 cycles per second), yes?

Now, when you were counting the cycles, what unit were they?
Was each cycle you counted "1 cycle per second"?
Would your final figure actually be "20 cycles per second every 10 seconds"?
Here are your questions in quotes and my answers below:

"And if you wanted to measure the number of cycles per second, you could count both the 'cycles' and the 'seconds', yes?"

Correct. By counting cycles you are simultaneously counting seconds- if you have a conversion factor.

"If you counted 20 cycles in 10 seconds then you would say there were 20 cycles per 10 seconds (or 2 cycles per second), yes?"

Yes.

"Now, when you were counting the cycles, what unit were they?"

They were in units of 1/2 cps.

"Was each cycle you counted "1 cycle per second"?"

No, it was 1/2 cps.

"Would your final figure actually be "20 cycles per second every 10 seconds"?"

Yes, or more simply, 2 cps.

60. Bangstrom, even I wouldn't write something as daft and senseless as you just did. If you are trolling that is one thing, but if you really think you are making sense I have to wonder if you are insane or just have access to some extreme drugs.

61. [QUOTE=Dywyddyr;584928]
Originally Posted by bangstrom
And this is why you are wrong.
You have already agreed that we simply count the number of cycles: i.e. how many times it has occurred.
After we get to the required number we define the time taken to reach that number as one second. (For at least the second time of writing this out).
It is not, and cannot be recursive, nor is it, and cannot be, Hz because the "second" is defined ONLY afterwards as the time that has elapsed from start to finish of that count.
No, the definition of a second has been defined by international convention BEFORE you begin your count. The definition of a second provides a conversion factor of 9,192,631,770 Hz so any laboratory can convert their counts of cycles into units of seconds. If they know the value of their counts in Hz, they can convert that value to seconds.

62. Originally Posted by bangstrom
No, the definition of a second has been defined by international convention BEFORE you begin your count.
Perhaps I worded that ambiguously.
The definition is stated: the second itself isn't established.
I.e. the value of a second isn't there until the count has occurred.

The definition of a second provides a conversion factor of 9,192,631,770 Hz
NO!
It is NOT Hz, it's a count ONLY.

If they know the value of their counts in Hz, they can convert that value to seconds.
This is physically and mathematically meaningless.
The count, number, (or rather the time elapsed to make the count) is what a second is.
It cannot be Hz.
It only becomes Hz AFTER we have set the value of a second - and that value is set by making the count.

As you agreed earlier, if caesium somehow "slowed down" then it would take longer than a second for that count to occur.
Therefore the second, as we define it, would be longer. An "actual second" would not change.
But we'd notice because light would travel further in that time than it had previously (all our previous measurements would be "wrong"), there'd be no agreement with an ephemeris second etc etc.

Put it this way - if you have no (other) method whatsoever of establishing the time it's still possible to do a count.
That count - in and of itself - tells you how long a second is.
But, because you do not, prior to making that count, have any method of telling time then it cannot be claimed to be a frequency until AFTER the count is made and a second is established (defined).
Once we've done that then we can say it's in Hz (which is, I suspect, where your confusion comes in: we've already established the length of a second so, in your mind, it's inextricably subsumed in the count itself).

Oh, PS, nicely avoided on the rest of my points BTW.

63. Originally Posted by kojax
If the difference between an Ephemeris second and a SI second is that small, then why do you care? The point is that if the speed of light were changing, the SI second would begin to deviate further from the Ephemeris second and someone would notice.
The only difference is that the SI second is based on the premise that c is a constant while the Ephemeris second needs occasional tweaking in the form of "leap seconds" to keep it in line with observation. A change in the speed of light would change the Ephemeris second proportionally so no one would notice the difference. An acceleration of c would cause the Earth to rotate faster but it would also cause the Earth to orbit the sun faster so there would be no day and night discrepancy. An acceleration in c would cause the entire 'movie' to run faster so no one would notice the difference.

64. Firstly, you missed answering this question:
Clearly, 'cycles' are not 'seconds', correct?

Originally Posted by bangstrom
"And if you wanted to measure the number of cycles per second, you could count both the 'cycles' and the 'seconds', yes?"
Correct. By counting cycles you are simultaneously counting seconds- if you have a conversion factor.
No - please read what I wrote again.

"you could count both the 'cycles' and the 'seconds', yes?"
Do you agree that you could count the cycles and the seconds separately?
(If it helps, imagine that there are two people: one counting the cycles and one counting the seconds.)

65. I never expected when I wrote the OP that I would be provoking a lengthy debate about the duration of a second. Anyway, going back to an earlier post, I have a question.

Originally Posted by Janus
And to add to what PhDemon said, each successively more accurate measurement always fell within the margin of error for the previous measurement.
Do you know what that margin for error was? And more importantly, do you know how it was decided upon?

66. Have you ever done an experiment or made a scientific measurement? If so did you consider the accuracy an precision of your results? Quantification of experimental uncertainties is a major part of the work and the experiment is designed to minimize them with the equipment available (if you did science at school think back to your write ups, a lot of the marks are given for this), it is the same in professional labs, all sources of uncertainty will have been considered, propagated through the calculations and reflected in the uncertainty quoted in the final result (you will need to read the original papers for info on how they were minimized and assessed in each measurement of c).

67. Originally Posted by PhDemon
Have you ever done an experiment or made a scientific measurement? If so did you consider the accuracy an precision of your results? Quantification of experimental uncertainties is a major part of the work and the experiment is designed to minimize them with the equipment available (if you did science at school think back to your write ups, a lot of the marks are given for this), it is the same in professional labs, all sources of uncertainty will have been considered, propagated through the calculations and reflected in the uncertainty quoted in the final result (you will need to read the original papers for info on how they were minimized and assessed in each measurement of c).
Does that answer my question? Yes of course there are margins for error, but they can be big or small depending on how accurate you think your research is.

68. Originally Posted by RedPanda
Firstly, you missed answering this question:
Clearly, 'cycles' are not 'seconds', correct?.
Absolutely, no doubt. Cycles are not seconds.

Originally Posted by RedPanda
"you could count both the 'cycles' and the 'seconds', yes?"
Do you agree that you could count the cycles and the seconds separately?
(If it helps, imagine that there are two people: one counting the cycles and one counting the seconds.)
Yes, one person could count cycles and another could count seconds separately.

Here are some questions based on your example with 20 cycles in ten seconds.

I hope we agree that 20 cycles in 10 seconds can be reduced to 2 cycles per second. If the number of seconds is an unknown, we can determine the number of seconds by dividing 20 cycles by 2 cycles per second.
N seconds= N cycles/ N cycles per second. The cycles cancel leaving our value for N in units of one second. (20 cycles/2cps = 10 s) OK?

We have the same problem in the previous discussion where 9,192,631,770 cycles from the definition of a second is a simple number having specific units. The definition tells us that there are 9,192,631,770 cycles in one second. So we can convert cycles to seconds by dividing by 9,192,631,770 and, as before, N seconds= N cycles/ N cycles per second. In the case of one second we have 9,192,631,770 cycles/ 9,192,631,770 cycles per second = one second.

My conclusion from this is that the number 9,192,631,770 must be in units of cycles per second so the number has the value of 9,192,631,770 Hz.

How could it be a number without units and make any sense?

69. Your question showed ignorance of error analysis in an experiment. I gave you the basic, relevant information to help you answer your question, if you're too lazy to follow it up :shrug: (Edit this is a response to post #66)

70. Originally Posted by bangstrom
Originally Posted by RedPanda
"you could count both the 'cycles' and the 'seconds', yes?"
Do you agree that you could count the cycles and the seconds separately?
(If it helps, imagine that there are two people: one counting the cycles and one counting the seconds.)
Yes, one person could count cycles and another could count seconds separately.
Ok.
Now, when you were counting the cycles, what unit were they?
Was each cycle you counted "1 cycle per second" or were they simply 'cycles'?

Originally Posted by bangstrom
Here are some questions based on your example with 20 cycles in ten seconds.
I hope we agree that 20 cycles in 10 seconds can be reduced to 2 cycles per second. If the number of seconds is an unknown, we can determine the number of seconds by dividing 20 cycles by 2 cycles per second.
No. If the number of seconds is unknown, then all we know is the number of cycles.

To show this is true, try answering this question:
I have just counted 34 cycles over an unknown number of seconds.
How many seconds did it take?

Originally Posted by bangstrom
N seconds= N cycles/ N cycles per second.
The cycles cancel leaving our value for N in units of one second. (20 cycles/2cps = 10 s) OK?
I think this is what you are doing:
frequency = cycles per second
therefore
cycles per second = cycles per second
therefore
second = cycles / cycles per second
correct?
But you still have the unknown quantity of 'second' on both sides of the formula.
It is still unsolvable.

Sure the cycles cancel out, but that just leaves you with
second = second
Which is trivially true.

Originally Posted by bangstrom
How could it be a number without units and make any sense?
Much like if you counted how many toes you have you would have 10 toes.
'cycles' and 'toes' would be the units.
But it would not be cycles per second, the same way it would not be toes per second.
'Cycles' and 'seconds' are different and separate things.

71. Originally Posted by bangstrom
How could it be a number without units and make any sense?
You are making this much, much more complicated than necessary.

We have a source that we have agreed to call the defining reference (in this case, caesium's hyperfine transitions). It spits out cycles. After spitting out N of those cycles, you say "That's a second."

Done.

No meters.

No kilograms.

Just N cycles, and you define one second to be the interval of time corresponding to N of them.

72. Originally Posted by tk421
Originally Posted by bangstrom
How could it be a number without units and make any sense?
You are making this much, much more complicated than necessary.

We have a source that we have agreed to call the defining reference (in this case, caesium's hyperfine transitions). It spits out cycles. After spitting out N of those cycles, you say "That's a second."

Done.

No meters.

No kilograms.

Just N cycles, and you define one second to be the interval of time corresponding to N of them.
I agree we have a defining reference and it serves as on ordinary conversion factor. It is no more complicated than that.

The problem is that the definition of a second does not mention a few basics we need to understand its derivation. It is the end of the story and not the beginning. If we understand its derivation, we can understand how seconds, meters, and c are all related. What everyone appears to be doing is trying to reverse engineer everything starting with the definition of a second and that is a poor and difficult place to start. To consider seconds with the exclusion of other units amounts to looking at only one part of the elephant.

73. speed of mind = faster than speed of light

74. Originally Posted by bangstrom
What everyone appears to be doing is trying to reverse engineer everything starting with the definition of a second and that is a poor and difficult place to start.
No it's not. The definition of the second: "One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms." is self-contained and the perfect place to start.

75. Originally Posted by RedPanda
Ok.
Now, when you were counting the cycles, what unit were they?
Was each cycle you counted "1 cycle per second" or were they simply 'cycles'?

Each cycle was a single unit. They were not cycles per second but simply cycles.

Originally Posted by RedPanda
No. If the number of seconds is unknown, then all we know is the number of cycles.

To show this is true, try answering this question:
I have just counted 34 cycles over an unknown number of seconds.
How many seconds did it take?

The number of seconds is unknown.

Originally Posted by RedPanda
I think this is what you are doing:
frequency = cycles per second
therefore
cycles per second = cycles per second
therefore
second = cycles / cycles per second
correct?

Correcto.

Originally Posted by RedPanda
But you still have the unknown quantity of 'second' on both sides of the formula.
It is still unsolvable.

No, this is where our views part company. The value of a second is known and has been defined by international convention. We also have a conversion factor for converting cycles into cycles per second. So the conversion can be done and there is no need to reinvent the second.

76. Originally Posted by KJW
Originally Posted by bangstrom
What everyone appears to be doing is trying to reverse engineer everything starting with the definition of a second and that is a poor and difficult place to start.
No it's not. The definition of the second: "One second is now defined to be exactly 9,192,631,770 cycles of the hyperfine structure transition frequency of caesium-133 atoms." is self-contained and the perfect place to start.
This is what everyone appears to be trying to do so lets see how it goes.

77. Originally Posted by Bat-man
speed of mind = faster than speed of light
My mind departed one day and arrived on the previous night- I hope.

78. Originally Posted by bangstrom
]What everyone appears to be doing is trying to reverse engineer everything starting with the definition of a second and that is a poor and difficult place to start.
It takes active misunderstanding (as opposed to ordinary passive misunderstanding) to assert that reverse engineering is going on. Show me where, in the definition of the second given, there is any reverse cognitive flow. Diagram the sentence for me and identify where this occurs.

Of course you cannot, because there is no reverse engineering. We have a thing going "tick-tock" and when it's done that 9 gazillion times, we call that interval of time 1 second.

Period.

Done.

End of story.

No circularity here. No retrograde cognitive flow. Quit claiming that there is. I don't think you're cognitively impaired. I do think, though, that you have an unstated belief in how things work, and it's causing you to twist logic in ways that frustrate. I suggest you set aside whatever belief that is, and just look at the statements as they are constructed. They are simple and clear.

Again, as far as the definition of the second is concerned, there are no slugs, no furlongs, no drams. Just counting cycles.

79. Originally Posted by bangstrom
Originally Posted by RedPanda
Ok.
Now, when you were counting the cycles, what unit were they?
Was each cycle you counted "1 cycle per second" or were they simply 'cycles'?
Each cycle was a single unit. They were not cycles per second but simply cycles.
Yes - exactly like the caesium detector does.
The caesium cycle detector simply counts cycles.
It doesn't have a clock. It doesn't count seconds. It doesn't know how long it takes.
It is not measuring cycles per second - it is just counting cycles.

Or do you disagree?

Originally Posted by bangstrom
Originally Posted by RedPanda
I think this is what you are doing:
frequency = cycles per second
therefore
cycles per second = cycles per second
therefore
second = cycles / cycles per second
correct?
Correcto.
So why are you talking about that?
You've simply rearranged Hz=Hz.
What does that have to do with the definition of a second?

Originally Posted by bangstrom
The value of a second is known and has been defined by international convention. We also have a conversion factor for converting cycles into cycles per second. So the conversion can be done and there is no need to reinvent the second.
Ok, then I'll repeat this question:
I have just counted 34 cycles over an unknown length of time.
How many seconds did it take?
Do you still agree that "The number of seconds is unknown."?

Well, let me ask you a slightly different question:
I have just counted 9,192,631,770 cycles over an unknown length of time.
How many seconds did it take?
Do you agree that "The number of seconds is unknown."?

80. Originally Posted by RedPanda
Yes - exactly like the caesium detector does.
The caesium cycle detector simply counts cycles.
It doesn't have a clock. It doesn't count seconds. It doesn't know how long it takes.
It is not measuring cycles per second - it is just counting cycles.

Or do you disagree?
I mostly disagree and fortunately there is nothing esoteric about our points of disagreement. First, a minor point, a caesium clock is a clock. It is also called an "atomic clock" but it is not a "caesium detector' because it does not detect caesium. I have no desire to quibble about semantics so you can call it caesium whatever and I think everyone will understand what you mean.

A caesium clock counts cycles and it doesn't count seconds but that is a simple technical fix. My quartz clock on the wall counts nothing but cycles but it has a chip that steps the cycles down to the rate of one per second. A similar chip could step down the cycles of a caesium clock to one cycle per second so a caesium clock could count both seconds and cycles simultaneously. A caesium clock emits a RF signal of 9,192,631,770 Hz and, given that information alone, we have a well tried technology to step the signal down to one cycle per second.

We don't even need to consider the definition of a second. The technical problem is to convert a 9 GHz sigmal to 1 Hz.

Originally Posted by RedPanda

So why are you talking about that?
You've simply rearranged Hz=Hz.
What does that have to do with the definition of a second?.
I hope my comment above answers this question.

Originally Posted by RedPanda;585100
Ok, then I'll repeat this question:
I have just counted 34 cycles over an unknown length of time.
How many seconds did it take?
Do you still agree that [I
"The number of seconds is unknown."[/I]?.
The number of seconds is unknown but, if someone knows the Hz value for the signal, they can calculate the number of seconds.

Originally Posted by RedPanda
Well, let me ask you a slightly different question:
I have just counted 9,192,631,770 cycles over an unknown length of time.
How many seconds did it take?
Do you agree that "The number of seconds is unknown."?
The number of seconds is unknown but if you know the Hz of the signal you can calculate the seconds.

81. Originally Posted by bangstrom
We don't even need to consider the definition of a second. The technical problem is to convert a 9 GHz sigmal to 1 Hz.
That's not much of a problem. All you need is a simple piece of digital logic called...ahem...a counter. You configure it so that each cycle from the caesium physics package increments the counter by one. When it reaches 9billion (or whatever), it outputs a pulse and simultaneously resets the counter. Thus, you get one pulse each second. That is, the pulses provide the drumbeat that defines the second.

Why is this so hard for you to understand?

82. This guy reminds me of theorist.

83. Originally Posted by bangstrom
I mostly disagree and fortunately there is nothing esoteric about our points of disagreement. First, a minor point, a caesium clock is a clock. It is also called an "atomic clock" but it is not a "caesium detector' because it does not detect caesium. I have no desire to quibble about semantics so you can call it caesium whatever and I think everyone will understand what you mean.
I am not talking about a caesium clock.

Originally Posted by bangstrom
A caesium clock counts cycles and it doesn't count seconds but that is a simple technical fix.
I am not talking about a caesium clock.

Originally Posted by bangstrom
My quartz clock on the wall counts nothing but cycles but it has a chip that steps the cycles down to the rate of one per second. A similar chip could step down the cycles of a caesium clock to one cycle per second so a caesium clock could count both seconds and cycles simultaneously. A caesium clock emits a RF signal of 9,192,631,770 Hz and, given that information alone, we have a well tried technology to step the signal down to one cycle per second.
I am not talking about a caesium clock.

Just to be clear: I am not talking about a caesium clock.

I repeat:
The caesium cycle detector simply counts cycles.
It doesn't have a clock. It doesn't count seconds. It doesn't know how long it takes.
It is not measuring cycles per second - it is just counting cycles.
Or do you disagree?

Originally Posted by bangstrom
Originally Posted by RedPanda
So why are you talking about that?
You've simply rearranged Hz=Hz.
What does that have to do with the definition of a second?.
I hope my comment above answers this question.
No it doesn't.

Originally Posted by bangstrom
Originally Posted by RedPanda
Well, let me ask you a slightly different question:
I have just counted 9,192,631,770 cycles over an unknown length of time.
How many seconds did it take?
Do you agree that "The number of seconds is unknown."?
The number of seconds is unknown but if you know the Hz of the signal you can calculate the seconds.
So, we have the problem of:
9,192,631,770 cycles per ?? seconds.
As you agree, this problem is currently unsolvable.

So, do you therefore agree that the number of cycles provides no information about how long it took?
Simply counting the number of cycles provides no clue as to how long it takes, correct?

84. Originally Posted by tk421
Originally Posted by bangstrom
]What everyone appears to be doing is trying to reverse engineer everything starting with the definition of a second and that is a poor and difficult place to start.
It takes active misunderstanding (as opposed to ordinary passive misunderstanding) to assert that reverse engineering is going on. Show me where, in the definition of the second given, there is any reverse cognitive flow. Diagram the sentence for me and identify where this occurs.

Of course you cannot, because there is no reverse engineering. We have a thing going "tick-tock" and when it's done that 9 gazillion times, we call that interval of time 1 second.

Period.

Done.

End of story.
There is no reverse engineering in the definition itself. The definition is a simple conversion factor for converting a gazillion cycles to one cycle and it tells us to divide by a gazillion cycles per second to get the answer in seconds. End of story. This is exactly what I have been trying to explain.

My understanding (correct me if I am wrong) is that opinions counter to mine are saying you can't divide by cycles per second before you have determined what a second is so the conversion factor can't be in units of cycles per sec. It is a plane number with no value except for What? I have yet to understand.

The reverse engineering begins where there is a going back from the definition of a second to find the meaning of a second and the mysterious nine gazillion number with no units.
I thought all that work was done years ago. Now all we need is the conversion factor (which we have) so hang up the phone.

85. Originally Posted by RedPanda
Just to be clear: I am not talking about a caesium clock.
It has no hands and it doesn't hang on the wall so don't call it a clock but I call it a clock.

Originally Posted by RedPanda
I repeat:
The caesium cycle detector simply counts cycles.
It doesn't have a clock. It doesn't count seconds. It doesn't know how long it takes.
It is not measuring cycles per second - it is just counting cycles.
Or do you disagree?
I agree but this is a technical fix. Just install a counter fribbit and it can count cycles and seconds at the same time.

Originally Posted by RedPanda
Well, let me ask you a slightly different question:
I have just counted 9,192,631,770 cycles over an unknown length of time.
How many seconds did it take?
Do you agree that "The number of seconds is unknown."?
We have a conversion factor in units 9,192,631,770 Hz so the number of seconds= 1 if my calculator is right.

Originally Posted by RedPanda
So, we have the problem of:
9,192,631,770 cycles per ?? seconds.
As you agree, this problem is currently unsolvable.

So, do you therefore agree that the number of cycles provides no information about how long it took?
Simply counting the number of cycles provides no clue as to how long it takes, correct?
I have a conversion factor so I know I know long it took. If you have a conversion factor, then the number of cycles is all the more information you need.

86. Originally Posted by bangstrom
There is no reverse engineering in the definition itself.
Good. Then what's the problem?

The definition is a simple conversion factor for converting a gazillion cycles to one cycle and it tells us to divide by a gazillion cycles per second to get the answer in seconds. End of story. This is exactly what I have been trying to explain.
I'd word it differently. In fact, I have worded it differently. You have unnecessarily complicated the definition by insisting on this "conversion factor" business. It's as if you've become so fixated on dimensional analysis that you've overwhelmed your cognitive apparatus with it.

I'll repeat myself one more time: We, as humans, have chosen arbitrarily to define an interval of time -- called, also arbitrarily, the second -- in terms of a number of drumbeats. We count to about 9 billion caesium drumbeats. When that happens, we say that one second has elapsed.

My understanding (correct me if I am wrong) is that opinions counter to mine are saying you can't divide by cycles per second before you have determined what a second is so the conversion factor can't be in units of cycles per sec. It is a plane number with no value except for What? I have yet to understand.
I haven't been following this thread that closely, so I have no idea what the above is supposed to mean. I would therefore simply go back to the definition of the second in terms of the drumbeat count. If you truly understand that, then there's no need (or point) in wasting more time on other ways of thinking about it.

The reverse engineering begins where there is a going back from the definition of a second to find the meaning of a second and the mysterious nine gazillion number with no units.
I'm not sure what you mean by "the meaning" of a second. It's just an interval of time that humans have named.

I thought all that work was done years ago. Now all we need is the conversion factor (which we have) so hang up the phone.
Again, I have no idea what you are talking about. What "work" was done years ago? The agreement by standards bodies to define the second in terms of caesium's hyperfine transition? Then, yes, that happened years ago. But I have no idea why you insist on getting hung up on this "conversion factor" business. That's why I speculated earlier that you have some unstated contextual basis in your mind that is getting in the way of understanding that this is all extremely simple. Trivially simple.

87. Originally Posted by bangstrom
Originally Posted by RedPanda
Just to be clear: I am not talking about a caesium clock.
It has no hands and it doesn't hang on the wall so don't call it a clock but I call it a clock.
Simply saying that you would call it a clock does not make it a clock.
It is not a clock - it simply counts caesium cycles.
Maybe I wasn't clear: I am not talking about a caesium clock.

Originally Posted by bangstrom
Originally Posted by RedPanda
I repeat:
The caesium cycle detector simply counts cycles.
It doesn't have a clock. It doesn't count seconds. It doesn't know how long it takes.
It is not measuring cycles per second - it is just counting cycles.
Or do you disagree?
I agree ...
So, we both agree that it is a caesium cycle detector that just counts caesium cycles.
It is not measuring cycles per second - it is just counting cycles.
Good.

Originally Posted by bangstrom
...but this is a technical fix. Just install a counter fribbit and it can count cycles and seconds at the same time.
It doesn't need fixing.
It counts cycles.
That's all it is meant to do.

Originally Posted by bangstrom
Originally Posted by RedPanda
Well, let me ask you a slightly different question:
I have just counted 9,192,631,770 cycles over an unknown length of time.
How many seconds did it take?
Do you agree that "The number of seconds is unknown."?
We have a conversion factor in units 9,192,631,770 Hz so the number of seconds= 1 if my calculator is right.
Where did you get that conversion factor from?
I have re-read my question and it doesn't mention any conversion factor.

It is just a count of cycles.
As you said: "Each cycle was a single unit. They were not cycles per second but simply cycles."
Therefore, it is not 9,192,631,770 Hz nor 9,192,631,770 cycles per second.
It is simply 9,192,631,770 cycles.

Maybe you would be less distracted if I asked the previous question:
I have just counted 34 cycles over an unknown length of time.
How many seconds did it take?
Do you agree that the number of seconds is unknown?

Originally Posted by bangstrom
Originally Posted by RedPanda
So, we have the problem of:
9,192,631,770 cycles per ?? seconds.
As you agree, this problem is currently unsolvable.

So, do you therefore agree that the number of cycles provides no information about how long it took?
Simply counting the number of cycles provides no clue as to how long it takes, correct?
I have a conversion factor so I know I know long it took. If you have a conversion factor, then the number of cycles is all the more information you need.
But the question provides no extra information.
It only tells you the number of cycles.

So, do you agree that the number of cycles provides no information about how long it took?
Simply counting the number of cycles provides no clue as to how long it takes, correct?

88. Originally Posted by bangstrom
Originally Posted by kojax
If the difference between an Ephemeris second and a SI second is that small, then why do you care? The point is that if the speed of light were changing, the SI second would begin to deviate further from the Ephemeris second and someone would notice.
The only difference is that the SI second is based on the premise that c is a constant while the Ephemeris second needs occasional tweaking in the form of "leap seconds" to keep it in line with observation. A change in the speed of light would change the Ephemeris second proportionally so no one would notice the difference. An acceleration of c would cause the Earth to rotate faster but it would also cause the Earth to orbit the sun faster so there would be no day and night discrepancy. An acceleration in c would cause the entire 'movie' to run faster so no one would notice the difference.

This is the direction you were taking this? Yes. If you believe that an increase in C would simultaneously increase the rate at which all phenomena occur.... then we're just splitting hairs.

A slowing of "all phenomena" and a slowing of "time itself" are identical propositions in modern physics. For example, it is believed that "time slows down" for objects that are in a strong gravitational field (when compared with the speed of "time" in other frames). However nobody can say for sure whether "time" is slowing down, or "all phenomena are slowing down" or "c is slowing down", because there would never be any way to differentiate.

Any test that would reveal it to be "time itself", would also reveal it to be "all phenomena", or "c" that is slowing instead.

Originally Posted by bangstrom

"Thus, if we had no other method of determining time (i.e. no other comparative measures), and if those cycles slowed down then the length of a (defined) second would increase. And we would never know."

Agreed.

But we do have other methods of determining time. So why focus on just one thing.

If you are assuming that all clocks are attuned to the speed of light in such a manner so that all clocks would slow down or speed up if the speed of light were to increase or decrease, then you're really not in disagreement with physics at all.

The main point of contention is that, in relativity, you can't say something is going "faster" or "slower" unless you can say what you are comparing against. Since you are suggesting that there is nothing to compare C against (since all phenomena are timed by it), you are talking about the "absolute speed" of C. C has no absolute speed. Nothing has an absolute speed in relativity.

C only has fixed proportional rates of motion when compared with other phenomena. And it appears that you are agreeing that those proportional rates of motion are invariant.

89. Originally Posted by tk421
You have unnecessarily complicated the definition by insisting on this "conversion factor" business. It's as if you've become so fixated on dimensional analysis that you've overwhelmed your cognitive apparatus with it
I hope you are not "saying" what you are saying about this conversion factor business being an unnecessary complication. I see it as a stopping point since it provides all the information we need to convert cycles to seconds so any conjecturing about the conundrums we need to overcome to implement its use amounts to trying to reinvent the "second" in my opinion.

My view is that the much quoted definition of a second is nothing more than a bench-top guideline for measuring the duration of a second so any laboratory in the world with a caesium clock can determine the interval of a second and that interval will be consistent with every other laboratory in the world with the same equipment.

The essential part of the definition is that it provides us with a conversion factor of 9,192,631,770 in units of Hz and, once we have the conversion factor, we can then convert cycles into seconds without a problem- or so I thought. This appears to be a contentious point of view.

I admit to a obsession about the conversion factor. Every time someone says, 'Cycles can't be converted to seconds because, because...' I say it can be done. We have the conversion factor.
Originally Posted by tk421
We, as humans, have chosen arbitrarily to define an interval of time -- called, also arbitrarily, the second -- in terms of a number of drumbeats. We count to about 9 billion caesium drumbeats. When that happens, we say that one second has elapsed.
I agree and I don't think I have ever seen it beter stated.

Originally Posted by tk421
I'm not sure what you mean by "the meaning" of a second. It's just an interval of time that humans have named.
I am not the one pondering the problem. The definition gives us a conversion factor and that is all we need. The issue is, How do you determine what a second is from the definition of a second so you can determine the length of a second? I say it has already been done.

Originally Posted by tk421
Again, I have no idea what you are talking about. What "work" was done years ago? The agreement by standards bodies to define the second in terms of caesium's hyperfine transition? Then, yes, that happened years ago. But I have no idea why you insist on getting hung up on this "conversion factor" business. That's why I speculated earlier that you have some unstated contextual basis in your mind that is getting in the way of understanding that this is all extremely simple. Trivially simple.
Trivially simple? I thought so too but you haven't been following the thread that close.

Yes, the "work" I had in mind was the work that yielded the conversion factor for converting cycles into seconds. Not to say that it hasn't been done many times over the years.

My complaints are directed towards anyone who examines the instructions for measuring the duration of a second with the intent of figuring out how to determine what a second is so we can determine the length of a second. Or how the 9 GHz number can not be in units of Hz because we have not yet determined the duration of one second. The number was picked simply because it worked is the explanation as I have been told.

The debate over the conversion factor continues.

90. Originally Posted by RedPanda
Simply saying that you would call it a clock does not make it a clock.
It is not a clock - it simply counts caesium cycles.
Maybe I wasn't clear: I am not talking about a caesium clock.
If your caesium gismo counts cycles, then I would call it a clock. Some clocks count cycles and some count seconds.
But you can call it whatever you want.

Originally Posted by RedPanda
Where did you get that conversion factor from?
I have re-read my question and it doesn't mention any conversion factor.

It is just a count of cycles.
As you said: "Each cycle was a single unit. They were not cycles per second but simply cycles."
Therefore, it is not 9,192,631,770 Hz nor 9,192,631,770 cycles per second.
It is simply 9,192,631,770 cycles.

Maybe you would be less distracted if I asked the previous question:
I have just counted 34 cycles over an unknown length of time.
How many seconds did it take?
Do you agree that the number of seconds is unknown?
The conversion factor came from the definition of an SI second. (I think someone may have quoted that a time or two) And the definition states "almost" perfectly clearly that the units are in cycles per second.
9,192,631,770 Hz is also the radio frequency of a caesium gismo so to convert cycles to seconds you divide cycles by cycles per second.
In your second example with 34 cycles the number of seconds is unknown so we can't tell how long it took.
Originally Posted by RedPanda
So, do you therefore agree that the number of cycles provides no information about how long it took?
Simply counting the number of cycles provides no clue as to how long it takes, correct?
Each cycle in a uniform signal has a uniform interval of time so cycle counts tell us how long it took in cycle units of time. If we have a conversion factor, we can convert cycles to seconds.

91. Originally Posted by bangstrom
If your caesium gismo counts cycles, then I would call it a clock. Some clocks count cycles and some count seconds.
But you can call it whatever you want.
A clock measures time.
The detector only counts cycles - which you agreed is not measuring cycles per second - it is just counting cycles.
The detector doesn't measure time, it is therefore not a clock.

Originally Posted by bangstrom
Originally Posted by RedPanda
Where did you get that conversion factor from?
I have re-read my question and it doesn't mention any conversion factor.
The conversion factor came from the definition of an SI second. (I think someone may have quoted that a time or two) And the definition states "almost" perfectly clearly that the units are in cycles per second.
9,192,631,770 Hz is also the radio frequency of a caesium gismo so to convert cycles to seconds you divide cycles by cycles per second.
I have re-read my question and it doesn't mention caesium, Hz or conversion factors.
It is just a simple count of cycles.

I'll repeat the question:
I have just counted 9,192,631,770 cycles over an unknown length of time.
How many seconds did it take?
Do you agree that the number of seconds is unknown?

Originally Posted by bangstrom
Originally Posted by RedPanda
So, do you therefore agree that the number of cycles provides no information about how long it took?
Simply counting the number of cycles provides no clue as to how long it takes, correct?
Each cycle in a uniform signal has a uniform interval of time so cycle counts tell us how long it took in cycle units of time. If we have a conversion factor, we can convert cycles to seconds.
But the question does not provide a conversion factor.
It only tells you the number of cycles.

So, do you agree that the number of cycles provides no information about how long it took?
Simply counting the number of cycles provides no clue as to how long it takes, correct?

92. Originally Posted by bangstrom
I admit to a obsession about the conversion factor.
That is, unfortunately, all too evident. It's preventing you from learning.

Every time someone says, 'Cycles can't be converted to seconds because, because...' I say it can be done. We have the conversion factor.
Again, for the last time that I will say, we DEFINE the second as the INTERVAL OF TIME it takes for 9+ billion cycles of the hyperfine transition's blah blah blah.

That definition is self-contained, non self-referential, and complete.

My complaints are directed towards anyone who examines the instructions for measuring the duration of a second with the intent of figuring out how to determine what a second is so we can determine the length of a second.
That's not a sentence I can parse.

Or how the 9 GHz number can not be in units of Hz because we have not yet determined the duration of one second. The number was picked simply because it worked is the explanation as I have been told.
I don't see anyone saying that 9GHz is not Hz (neglecting the trivially pedantic observation that GHz and Hz are not, strictly speaking, the same). What people are saying is that the number of cycles of caesium's hyperfine transition that we have selected to define the second is arbitrary in the sense that a second is arbitrary. The number of cycles was selected somewhat less arbitrarily to correspond most closely to the previous standard second, so as to have minimal practical impact (the clocks of ordinary people would not have to be thrown away, etc.).

The debate over the conversion factor continues.
Only in your mind, apparently. I'm sorry that I cannot help you. You are stuck on a trivial point, apparently because you have some other agenda that somehow requires the fabrication of this complication. But it is solely of your own making.

The second is defined as it is. Its definition could hardly be more clear. That you refuse -- not merely fail -- to grasp it is not something that I or anyone else can fix, apparently. The problem is entirely in your head. You will have to make a choice to get unstuck. Until then, I'm done. I'm happy to educate anyone who is interested in it, but I have learned to avoid those who actively reject knowledge for one reason or another. That seems to be the case here, regrettably.

93. I know exactly what the OP is talking about and a simple explanation for it.

The speed of light in a vacuum IS always the same, so how can light arrive slower over increased distances of vacuum?

Well there answer is simple in a way, I have thought of a easy to visualize analogy for this concept. I will use a baseball being thrown to simulate a photon and the catcher to be the object receiving the light.

Lets say in this virtual world the speed of light (the ball) is 1M/s.

As the ball is thrown it stops for a VERY slim amount of time breaking apart and coming back together.

The speed of the broken ball is now 0M/s, but this is not LIGHT (the completed ball).

The ball comes back together almost instantly and continues to travel at 1M/s.

These "pauses" are what causes the ball to get to the catcher in more expected time relative to the distance, even though the ball is always travelling at 1M/s.

I hope that makes sense!

94. Just more total nonsense.

95. Oh great...

As the ball is thrown it stops for a VERY slim amount of time breaking apart and coming back together.
Does light "break apart"?
Does it stop at all?

If no then how is this an "explanation"?
(Oh, and you appear to be ignoring Emmy Noether, even IF the "ball" broke apart its constituents would continue in the same direction at the same speed. Hard luck).

96. Originally Posted by Dywyddyr
Oh great...

As the ball is thrown it stops for a VERY slim amount of time breaking apart and coming back together.
Does light "break apart"?
Does it stop at all?

If no then how is this an "explanation"?
(Oh, and you appear to be ignoring Emmy Noether, even IF the "ball" broke apart its constituents would continue in the same direction at the same speed. Hard luck).

To answer your question yes, photons split into positrons and electrons in a process called vacuum polarization. That would create a gravitational differential which would cause to two to have a small amount of impact energy. This energy is not lost simply transferred, when the positron and electron recombines the photon continues at the speed of light.

Many of these splits and recombines account for added up time.

I don't know the exact statement Emmy Noether made, because I am not him and don't want to speak for him. But if that is what he said then he is WRONG.

Science changes and old research is constantly being broken.

Please check up on some reputable articles it is explained quite well. Physicist suggests speed of light might be slower than thought

97. Originally Posted by AlexG
Just more total nonsense.
Do some research buddy Physicist suggests speed of light might be slower than thought

98. You are quick to challenge me even when I am not the publisher of these concepts, this time you aren't attacking me you are attacking University of Maryland.

99. Originally Posted by ScienceNoob
To answer your question yes, photons split into positrons and electrons in a process called vacuum polarization.
Ah yes.
You've called on vacuum polarisation before, in a different thread.
And ignored the fact that it doesn't "split into" positron-electron pairs. It creates virtual ones.
In other words: no, light does not "break apart".

I don't know the exact statement Emmy Noether made, because I am not him and don't want to speak for him. But if that is what he said then he is WRONG.
Yeah right.
A) You don't know Noether's Theorem.
B) You're not even aware that Emmy Noether was female.
C) She was wrong? Because some ill-educated half wit on the internet says so?

Please check up on some reputable articles it is explained quite well. Physicist suggests speed of light might be slower than thought
Yeah, ONE guy.
In an article 3 years old and only recently resurrected by newspapers (but not, apparently, taken up by any other scientist. I wonder why?).
An article that was "peer-reviewed" by reviewers who are James Franson's peers, i.e. other hardcore cranks.
"At any rate, Franson's paper is an uninterrupted stream of nonsense. It's not a paper with several bugs at several places. It's a paper whose almost every sentence shows that the author is completely ignorant about basic physics."
You didn't write it did you?

The thing that struck me was this "the arrival of the photons was later than expected, by 4.7 hours".
Wait, what?
Until the photons arrive (they being faster than anything else coming from the supernova) how the f*ck would we know what "time" it happened?
And therefore the only way we get a "time of the event" is from those photons.
So how can they arrive any earlier (or even later) than "expected"?

100. Originally Posted by Dywyddyr
Originally Posted by ScienceNoob
To answer your question yes, photons split into positrons and electrons in a process called vacuum polarization.
Ah yes.
You've called on vacuum polarisation before, in a different thread.
And ignored the fact that it doesn't "split into" positron-electron pairs. It creates virtual ones.
In other words: no, light does not "break apart".

I don't know the exact statement Emmy Noether made, because I am not him and don't want to speak for him. But if that is what he said then he is WRONG.
Yeah right.
A) You don't know Noether's Theorem.
B) You're not even aware that Emmy Noether was female.
C) She was wrong? Because some ill-educated half wit on the internet says so?

Please check up on some reputable articles it is explained quite well. Physicist suggests speed of light might be slower than thought
Yeah, ONE guy.
In an article 3 years old and only recently resurrected by newspapers (but not, apparently, taken up by any other scientist. I wonder why?).
An article that was "peer-reviewed" by reviewers who are James Franson's peers, i.e. other hardcore cranks.
"At any rate, Franson's paper is an uninterrupted stream of nonsense. It's not a paper with several bugs at several places. It's a paper whose almost every sentence shows that the author is completely ignorant about basic physics."
You didn't write it did you?

The thing that struck me was this "the arrival of the photons was later than expected, by 4.7 hours".
Wait, what?
Until the photons arrive (they being faster than anything else coming from the supernova) how the f*ck would we know what "time" it happened?
And therefore the only way we get a "time of the event" is from those photons.
So how can they arrive any earlier (or even later) than "expected"?
Positrons and Electrons are far form virtual, so I will just ignore that you even said that.

Yes I made a large mistake, in referring to Noether as a man, but like I said I don't know what she said, or want to speak for her so I didn't make any false claims.

Also we measure the distance to galaxies by brightness not with a measuring stick.

This is how we can calculate the difference in values between what used to be expected for that distance.

Hope this helped.

101. Originally Posted by ScienceNoob
Positrons and Electrons are far form virtual, so I will just ignore that you even said that.
You don't even know what a virtual particle is?
You're dumber than I thought.

Yes I made a large mistake, in referring to Noether as a man, but like I said I don't know what she said, or want to speak for her so I didn't make any false claims.
And you were so arrogant (and willing to hang on to your ignorance) that you didn't bother to look.
And you "didn't want to speak for her (or make any false claims)" but you had no hesitation at all in flatly declaring that "what [s]he said then he is WRONG".

Also we measure the distance to galaxies by brightness not with a measuring stick.
This is how we can calculate the difference in values between what used to be expected for that distance.
Hope this helped.
And we see that brightness because of...?
Oh wait. It wouldn't be the arrival of photons, would it?
So, we know how far away it is before it explodes.
How do we get a time of explosion?
By seeing the photons of that explosion arrive where we are.
So no, it didn't help. (Except to further illustrate your cluelessness).

Page 1 of 2 12 Last
 Bookmarks
Bookmarks
 Posting Permissions
 You may not post new threads You may not post replies You may not post attachments You may not edit your posts   BB code is On Smilies are On [IMG] code is On [VIDEO] code is On HTML code is Off Trackbacks are Off Pingbacks are Off Refbacks are On Terms of Use Agreement