299,792,458

The figure was chosen to match the best measurements then available rather than rounded to 300,000,000, so that the new metre would be indistinguishable in practice from the krypton-86 wavelength standard it replaced. Rounding would have been tidier and would have silently changed the length of every existing metre.

Before 1983 the metre was defined by the orange-red emission line of krypton-86, and before 1960 by a platinum-iridium bar kept near Paris. Each redefinition kept the length the same and changed what it was anchored to — from an object, to an atomic property, to a constant of nature.

Because the value is exact by definition, distances derived from timing — the basis of satellite navigation and of lunar laser ranging — carry no uncertainty from the constant itself. The remaining error budget belongs to the clocks and the atmosphere, which is a meaningfully different engineering problem from one where the constant is also in doubt.

The 15th Conference generale des poids et mesures recommended 299,792,458 metres per second as the best value in 1975, eight years before the metre was redefined from it. The 1983 decision therefore ratified a figure already in use rather than introducing a new one.

Light covers very close to 30 centimetres in a nanosecond — the fact Grace Hopper made concrete by handing out lengths of wire that long. It is the most useful way to hold the number: signal delay in a cable is not an abstraction but roughly a nanosecond per foot.

One light-second is 299,792,458 metres by definition. The Moon averages about 1.28 light-seconds from Earth, which is why a two-and-a-half-second round trip was audible in the Apollo transmissions — a delay imposed by this constant and by nothing that better equipment could remove.

The 17th Conference generale des poids et mesures fixed the speed of light in vacuum at exactly 299,792,458 metres per second in 1983, and defined the metre from it: the distance light travels in 1/299,792,458 of a second. The constant stopped being something experiments measure and became something they calibrate against.

The speed of light in vacuum is the same for every observer regardless of their motion, which is what makes it usable as a definition: an anchor that changed with the laboratory's velocity would define a different metre in every laboratory. That invariance is the substantive reason for the choice, not the speed's magnitude.