It still seems to be common among groups outside the computing and
scientific worlds to talk about “GMT” rather than “UTC”. The trouble
with “GMT” is that it doesn’t seem to be clearly defined; it might or
might not differ from UTC by up to 0.9 seconds.
The (unadjusted) time reference defined by an official collection of
atomic clocks worldwide is called “TAI”. This pays no heed to the
slowing rotation of the Earth. Astronomers prefer a time reference
called “UT1”, which is based on the actual rotation of the Earth
(subject to some time-averaging, I think). The definition of “UTC” comes
from TAI, adjusted by applying leap seconds. These accumulate as needed
so as to keep UTC within 0.9 seconds of UT1, and we are currently up to
36 of them. That is, at 0:00:00 UTC today, TAI was actually 0:00:36.
The trouble with GMT is that it is not officially clear whether it is
the same as UT1, or the same as UTC. Given that it was originally
defined at the Greenwich Observatory, which was (is?) run by
astronomers, one would think they would make it the same as UT1. But
the practical fact is, our usual everyday timekeeping is based on
standard time zones, which are all defined as whole-second
(indeed, whole multiples of a half-hour) offsets from UTC.
