Register boffinry confab: Mass debate on the Perfect Kilogram For more than a century the world has relied upon a lump of metal protected under high security in a location outside Paris to accurately measure the kilogram. Problem is, the metal lump's mass might not actually amount to one kilogram - meaning the kilogram as you know it is wrong. Now, more than 30 years since they began …
How is self-reference avoided? For example the metre is defined as the distance light travels in (whatever fraction of) a second. But when the speed of light was calculated, surely there must have been both a reference distance and a reference time. Since the speed of light wasn't known at the time, the original reference distance needed to be calibrated against a physical metre.
Same with time, as far as I know the second is now calculated as the amount of time in which a particular radioactive element will have such-and-such a decay. But surely the element's radioactive decay had to be originally measured against something else.
The actual size of the base units doesn't matter; the fact that they can be consistently reproduced does.
So, you can define a second as being X number of decays of Y atom, as long as X and Y are constant you have a second. Measure how far light travels in that time and call some constant fraction of that a metre. And so on...
If you can then define a kilogram using only the definition of a metre and a second and other known constants, you're sorted. That's what they are trying to do...
I think you answered your own question (for the first part at least).
Originally the speed of light was measured against a reference metre and reference second (the former being a measure against electrical conductance or some such thing between 2 infinitely parallel conductors - can't remember). But now they have switched the references around so that the speed of light in a vaccumn is the reference against which the metre is obtained.
Obviously they chose the value to be the closest integer match with the original light/metre values.
Example: Imagine that a 'laptop' width is exactly the same as the distance from which a 5p piece blocks the moon when held 268.4 5p pieces away from your eye on 3rd-jan 2006 at 3:35.23 in the morning at a specific latitude/longitude (how contrived was that!). And my bedroom is exactly 104.0000000000000013 laptops wide.
Now assume that my bedroom width was a 'universal constant' , it makes sense to redefine the 'laptop' width as being 1/104 of a bedroom instead.
Make sense?
Ah, I get it - teh original reference metre was basically arbitrary (or based on some ancient measure), and the re-defining in terms of speed of light makes sure that ll metres are perfectly consistent with one another. It doesn't matter that the excat sizes / times / distances are basically arbitrarily created, as long as they can be reproduced with very high precision and infinite consistency
This post has been deleted by its author
And...
One 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 [cesium] 133 atom."
So they *haven't* properly defined the meter, because the definition parses to: "The length that light travels in a vacuum during 30.663318988498369762190615215544 transitions of the two hyperfine levels of the ground state of the caesium 133 atom."
They could measure the 30. , and they could estimate the fraction 0.66, but the 0.003318988498369762190615215544 cannot be measured directly.
The meter is thus a secondary constant. Speed of light is fundamental.
SI needs a reboot.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
