🇳🇱
**Interesting fact of the day**: The speed of light has only been proven to be a constant speed **round trip**. Not the instantaneous speed of light nor the speed of light while heading in a single direction has ever been able to be proven as constant.
In other words, it has never been experimentally disproved that light doesn't, for example, preferentially travel at half the speed of light in one orientation, but instantaneous in another. In fact it may very well be impossible to test the one-way speed of light due to the very consequences of relativity itself.
@freemo @Science This is not correct according to my understanding of physics. Specifically addressing the claim that it is not testable: you could have distance separated sensors send light to each other and see how long it takes. If it takes longer in one direction than the other than the theory that the speed of light is constant would be falsified. Maybe I'm not understanding something, but I believe your claim is false. If I am missing something I'd love to know what.
Yes you are misunderstanding the fundemental problem.. if you do this experiment as you propose it would require both sensors to have synchronized clocks with very high accuracy correct? Synchronizing the clocks would be impossible if the speed of light is asymmetrical without knowing what that asymmetry is ahead of time. Worse yet, the error in synchronizing those clocks due to the asymmetry would be exactly enough to cause the experiment to incorrectly show that the speed of light is the same in both directions.
Ergo, your experiment would not be able to detect asymmetry in the speed of light if it existed.
@freemo @Science No, the clocks are synchronized first then they are separated. For example one stays on Earth and one is sent to Mars. I understand that relativity will result in time dilation for whichever sensor accelerates to create the distance, but that can be accounted for with a Lorentz transformation. You still end up with two distance separated sensors that know what the other sensor's clock reads. Is there something I'm missing? Thanks for the response!
Yup you are still missing something.
When you synchronize the clocks and then separate them the act of separating them causes time dilation to occur on the clocks and thus desynchronizes them. Time dilation itself is dependent on the speed of light so if the speed of light is different in different directions then so too will be the effects of time dilation.
So if you synchronize the clocks first and separate them they will be out of sync and you will be unable to compensate for how much out sync they are at any one moment without knowing the extent of the speed of light asymmetry int he first place..
So with your proposed solution of synching the clocks first you haven't, as you can see, avoided the problem at all since the act of moving them breaks synchronization. And your lorentz transformation depends and requires symmetric values of C or to know the specific asymmetry up front to do the synchronization. So you are still shit out of luck :)
@freemo @Science From that quote, it sounds like this has been experimented and it's at least very very close to the same. Also the Lorentz transformation results in minuscule time dilation on Martian rovers so even if you say the clocks are off by the full factor of the Lorentz transform, the uncertainty that introduces is not very much. Thanks for responding!
See my other link where it explicitly states one-way speed of light has not been tested:
