4 ms·
FWIW, those are still in use. Here's the (rather interesting) relationship: * UT1 is really not a time, but an angular measurement - it measures the rotation o
by FabHK 7y ago
FWIW, those are still in use. Here's the (rather interesting) relationship:
* UT1 is really not a time, but an angular measurement - it measures the rotation of the earth. In UT1, a day (= 1 rotation) has 86400 seconds (= 1 rotation), but those seconds are not SI seconds.
* TAI is time since some fixed epoch, in SI seconds. In TAI, a day doesn't really exist as a primitive concept, but a day (=1 rotation) has a variable number of seconds, and those seconds are SI seconds. TAI and UT1 diverge over time, continuously (due to earth's rotational speed variations).
* UTC is sort of an amalgamation of them: it uses SI seconds (like TAI), but occasionally jumps a second to keep close to UT1. In UTC, a day has 86400 seconds, or 86399 or 86401, and those seconds are SI seconds. UTC and TAI diverge, and their difference is piecewise constant with occasional jumps. UTC and UT1 diverge, but at most a second continuously, and then with occasional jumps come together again.
- gumby 7y agoThis nicely recapitulates phylogeny: in the origin of Babylonian timekeeping (we all now use a system based on that), hours were not of a fixed length; there were twelve of them between dawn and dusk and twelve more between dusk and dawn. Minutes (as can be see from their clearly Latin name, "tiny") and the second division of them (seconds) are brand new and AFAIK have always been of fixed length, a consequence of mechanical timekeeping.