Best of all, none of those natural reference values are constant. They drift gradually, and lunar months won’t be 30 days forever just like a day won’t be 24 hours in the future.
Hmm, I wonder… our current standard of time might end up being the standard for a long time, primarily because of GPS. Before we had global data networks it wasn’t really possible to syncronize clocks all around the world. There used to be a telephone service that you could dial which would tell you “The time is now eight fifty-five PM” or w/e because that was the most effective way to distribute a coordinated time signal, and then you could manually set your local clock/watch to match.
But GPS depends heavily on accurate time information, and keeping it accurate is very complicated. Relativitistic time dilation applies because the satellites are:
far enough away from Earth (~20000km) that they experience different gravity than devices on Earth’s surface, causing local time for the satellite to be be faster, and
moving so fast in their orbit that they experience a measurable slowing of local time.
(that’s right, using GPS on your phone is a real-world demonstration of the theory of relativity in practical effect)
…and all those satellites are constantly checking in with each other and ground stations to make sure they’re in agreement.
As a result there is now a de facto standard time reference for the entire world, and all networked devices depend on it for their own timing, and it is accurate to microseconds at worst.
100 years ago people were still winding mechanical clocks every day, and setting them by the local churchbell.
Yep. This stuff is surprisingly complicated, and that’s why we need to measure the day using a standardized unit instead of defining the unit with the day.
Incidentally, Wikipedia has a nice graph about the variation of the length of day. It’s surprisingly messy and pretty far from the ideals of antiquity.
The lunar month is currently ~29.5306 days (synodic, on average). Some day in the future (probably in a few million years), the average synodic moon cycle will reach 30 days (it is slowing down) - if, and that is a big if - we keep the current definition of “second” and “day”, because both the length of the day as well as the lenght of the year will have changed…
I suggest we switch to Planck based units. Just slap on a suitable multiplier to make the final unit practical in normal life and engineering. Basically like the mol unit but with a more precise and natural foundation.
If the fundamental constants of the universe aren’t exactly constant, we’re screwed and there are no good units.
Best of all, none of those natural reference values are constant. They drift gradually, and lunar months won’t be 30 days forever just like a day won’t be 24 hours in the future.
Hmm, I wonder… our current standard of time might end up being the standard for a long time, primarily because of GPS. Before we had global data networks it wasn’t really possible to syncronize clocks all around the world. There used to be a telephone service that you could dial which would tell you “The time is now eight fifty-five PM” or w/e because that was the most effective way to distribute a coordinated time signal, and then you could manually set your local clock/watch to match.
But GPS depends heavily on accurate time information, and keeping it accurate is very complicated. Relativitistic time dilation applies because the satellites are:
(that’s right, using GPS on your phone is a real-world demonstration of the theory of relativity in practical effect)
…and all those satellites are constantly checking in with each other and ground stations to make sure they’re in agreement.
As a result there is now a de facto standard time reference for the entire world, and all networked devices depend on it for their own timing, and it is accurate to microseconds at worst.
100 years ago people were still winding mechanical clocks every day, and setting them by the local churchbell.
Yep. This stuff is surprisingly complicated, and that’s why we need to measure the day using a standardized unit instead of defining the unit with the day.
Incidentally, Wikipedia has a nice graph about the variation of the length of day. It’s surprisingly messy and pretty far from the ideals of antiquity.
Planck second is constant.
The lunar month is currently ~29.5306 days (synodic, on average). Some day in the future (probably in a few million years), the average synodic moon cycle will reach 30 days (it is slowing down) - if, and that is a big if - we keep the current definition of “second” and “day”, because both the length of the day as well as the lenght of the year will have changed…
I suggest we switch to Planck based units. Just slap on a suitable multiplier to make the final unit practical in normal life and engineering. Basically like the mol unit but with a more precise and natural foundation.
If the fundamental constants of the universe aren’t exactly constant, we’re screwed and there are no good units.