Leap seconds are what keep the passage of time - intervals of fixed duration - in sync with the earth's rotation. A day is measured as 24 hours or 86,400 seconds, the earth currently takes 86,400.001 seconds to rotate; in 50,000 years it is expected to take about 86,401 seconds. The extra 0.001 seconds per day are made up for by leap seconds, occasionally adding 1 second to the end of a month. Since 1972, we've had 24 leap seconds. They, a United Nations affiliated organization known as the Radiocommunication Sector of the International Telecommunications Union (ITU-R), are considering breaking the link between the passage of time and the earth's motion.
The July/August 2011 issue of
American Scientist has an interesting article,
The Future of Time:UTC and the Leap Second, on this issue. The text of the article is available
online. The actual article has some informative illustrations that aren't available.
A short except:
Before atomic timekeeping, clocks were set to the skies. But starting in 1972, radio signals began
broadcasting atomic seconds. Since then, leap seconds have occasionally been added to that stream of
atomic seconds to keep the signals synchronized with the actual rotation of Earth. Adjustments are needed
because Earth’s rotation is slightly less regular and a bit slower on average than cesium-133’s quantumscale
rhythms.
Leap seconds have remained a means to an end—that is, to reconcile and combine two entirely different
yet useful notions of timekeeping. But the ITU-R proposal would cease issuing leap seconds entirely.
Clocks everywhere—on your wall, wrist, phone and computer—would begin to diverge from the heavens.
The change would bring scientific, technological, legal, philosophical and social implications too. These
range from abandoning the requirement to preserve knowledge of Earth’s rotation through timekeeping, to
whether the word day will mean “one turn of the Earth” versus “794,243,384,928,000 cycles of cesium-133
radiation.” As scientists and engineers engaged in astronautics, space navigation and astronomy, we seek to
draw attention to this unprecedented situation.
...
The chief advantage for repealing leap seconds today is that the logistical overhead needed to keep
track of and display them will no longer be required by systems that do not need astronomical time to
function. This would provide a notable simplification to telecommunications and some electronic
navigation systems, but those who deal with historically time-stamped information will be handicapped.
Reprogramming of operational software that already presumes UT1 and UTC are always within a second of
each other would be required, and some space operations and astronomical applications would need to
distinguish between the UTC without leap seconds and UT1. There would also be a need for statutory and
regulatory changes to national legal systems for which astronomical time is the (explicit or implicit)
standard. And a review, update and republication of documents and standards discussing timekeeping and
UTC would be necessary. Because the difference between apparent time of day and clock time would grow
without leap seconds, a longer-duration adjustment would be necessary in the future. That would be less
easy to overlook compared to today’s leap seconds. Hardware would still need to represent such an adjustment
in an untraditional way. Without any clock adjustments, regional governments in a distant time would
eventually be compelled to redefine the time zones, and the International Date Line would have to be
moved to avoid having the calendar date change during normal daylight hours.
There is no consensus in industry, academia or among governments regarding the next best step for
civil timekeeping, Coordinated Universal Time and leap seconds. The confounding aspects of UTC, when
they exist, are more often seen during its implementation than in establishing its definition. And even if
leap seconds ceased, the underlying issue of how to reconcile the steady advance of SI seconds with the
astronomical day will never go away. Will deliberations at the ITU-R Radiocommunication Assembly in
January 2012 resolve or cloud these issues?