# What is one way speed of light is about? Why is it said undefined?

Tags:
1. Jul 13, 2014

### Sreenath Skr

I've gone though a lot of google searches and couldn't find the reason why one way speed of light is undefined.
All i know is that it is about synchronizing clocks.
So i made my own thought experiment

Imagine two clocks A & B. Both are not active
I keep A with me and keep B at 1 light second away.
First i send a signal to B to activate its clock

When A sends a pulse, it's clock starts running from 0 seconds and when it reaches B, it reads 1 second on it. So that we have two clocks that runs at same rate.

Now i send a pulse to B when A reads 10s.
Since B is 1Ls away, the pulse will reach there in 11s of B time. So now i can subtract 11-10 and that is 1s. d/t=s ie c

So what's the problem?

2. Jul 13, 2014

### TumblingDice

How do you know what distance one light second is without already knowing the one-way speed of light? Your intention is to measure speed, but you can't assuredly position the clocks at one light second apart without knowing the speed first.

The experiment setup presupposes what the one-way speed of light is.

3. Jul 13, 2014

### Simon Bridge

I think that's just an artifact of the chosen wording.
The underlying question is, why not just time light over a known distance like we do for any other 1-way speed we measure?

We can get the distance, say, in multiples of a particular rod of platinum-iridium alloy kept in a vault in France, or something, without presupposing some speed for light ... can't we?

There are other troubles with the description ...
If the distance between the clocks is d, then the time for light get from A to B is d/c... here's what was described:

You send a pulse to B to start it ... then, an unmeasured time later, you send another pulse at the same time as you start A.

The second pulse arrives a bit after the first one, so when the second pulse arrives, B reads the difference in time between the start pulse and the pulse from A. i.e. not d/c

You send a third pulse when A reads 10s ... this is 10s after the second pulse.
Now what?

4. Jul 13, 2014

### TumblingDice

I've tried to recognize anything ambiguous and read the OP several times. It seems to state very clearly that we place the clocks one light second apart, and if a pulse is sent from A to B when clock A reads zero, clock B is assumed to read one second when the pulse reaches it. EDIT: Or is it clock A?

We know there are a million and one ways to try get around the problem, and each one has it's own flaw.

5. Jul 13, 2014

### Sreenath Skr

So the problem is that i already defined a distance for calculation?

6. Jul 13, 2014

### TumblingDice

If I understand your experiment setup, the problem isn't that you already have a distance. The problem is that the distance (one light second) for your setup is based on the speed of light. It requires knowing the speed of light before doing the experiment.

As Simon replied, there are other troubles with your description. This is the first one.

7. Jul 13, 2014

### Simon Bridge

Yeah - you defined the distance in terms of the speed of light.
Try doing the same thing with an arbitrary distance d between the clocks... then you'll probably see the other problems.

8. Jul 13, 2014

### ghwellsjr

You've got it backwards. The one way speed of light cannot be measured, instead, under Einstein's Special Relativity, it is defined to be the same as the measured two way speed of light. If we want to measure the speed of light, it will always be a two way measurement.

Yes, that is exactly what it is about. We synchronize clocks in different locations so that the one way speed of light comes out to be c.

As long as the clocks are at rest with respect to each other, that is, neither one is accelerating and the distance remains the same, then they will both run at the same rate. What you mean is that we have two clocks that are synchronized according to Einstein's definition of the one way speed of light.

As long as you started clock A one second after you sent the signal to B to activate its clock, then you have properly synchronized the two clocks. But your process will require a third clock because otherwise, how will you know that you have started clock A one second after you sent the signal to B?

No problem. Once you define the one way speed of light to be c and synchronize the clocks so that this is true, then when you later "measure" the one way speed of light, you will, of course, get c as your answer.

9. Jul 13, 2014

### TumblingDice

If this were basketball, you just stepped back from the crowd and made a three-point shot hitting "nothing but net" on my scoreboard. That's not only the beginning - as you mentioned, it's the only end game, too.