It won't bother you once you understand it. See the OP. See the definition of the second? That concerns the NIST caesium clock. It's like their optical clock, but it employs microwaves rather than visible light. Look carefully at the definition of the second: "Since 1967, the second has been defined to be 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". It's periods of radiation. Not frequency. It can't be frequency because frequency is cycles per second and we are defining the second isn't defined yet. So we're counting the microwaves coming past us, and when we get to 9,192,631,770 we say that's a second then we say the frequency of those microwaves is 9,192,631,770 Hertz by definition. Imagine we had a parallel-mirror light-clock and we counted 9,192,631,770 reflections then said that's a second/. Then the reflection frequency is 9,192,631,770 Hertz by definition. Oh for God's sake,read the OP. We define the second and the metre using the local motion of light. Then we use them to measure the local motion of light. So we are guaranteed to always measure 299,792,458 m/s. No. See above. Think of an ocean. Think of a wave traversing that ocean. Imagine you're above that ocean in a helicopter. Pacing that wave. It looks like a hump of water. A static hump of water. It doesn't have a frequency. It only appears to have a frequency to the guy in the boat. Because it passes him at some speed. And if it passes him slower, the frequency looks lower. Shrug. What they say is wrong. The NIST optical clocks will set the standard eventually.