From our other sites
The story
The “leap second,” which has adjusted the one-second discrepancy between Earth’s rotation and atomic clocks for half a century, has been officially abolished. Introduced in 1972, it has been inserted 27 times since, but growing concerns over computer glitches and trouble in financial transactions had fueled calls for a review. In its place comes the “leap hour,” which will tolerate discrepancies of up to one hour, meaning no adjustment should be needed for at least several centuries. On 5ch, the thread lit up with trivia-flavored chatter — sarcastic remarks that daylight saving time should be scrapped first, speculation on why Earth’s rotation has been speeding up lately, and simple bafflement over why leap years and leap seconds are named so differently.
Published 9/22 (Tue) 5:00 AM
Asahi Shimbun
The “leap second” is being abolished. Introduced in 1972 to reconcile the gap between Earth’s rotation and atomic clock time, it has been inserted 27 times over the past half-century. In recent years, however, concerns over computer malfunctions and trouble in financial transactions have fueled growing calls for a review. In its place, a “leap hour” will be introduced, tolerating discrepancies of up to one hour — meaning time adjustments likely won’t be needed again for at least several centuries.
Source: news.yahoo.co.jp / Original article here
What people said
> But once the internet spread, leap seconds started causing communication outages and financial trading glitches.
If that's the case, they should've scrapped daylight saving time first.
Daylight saving time works in one-hour increments, so it's handled without issues.
That's exactly why they went with the "leap hour" — they're leaning on DST's know-how and track record.
In a few thousand years, a day might end up being shorter.
Read the article.
Rotation is slowing down, which means the day is getting longer.
Re: #27
It also changes with the amount of glacial ice, so there are fluctuations on cycles of hundreds of thousands or even tens of millions of years, too.
Seconds and hours are adjustments for Earth's rotation
Days and months are adjustments for Earth's orbit
Both "leap month" and "leap second" add or subtract a month/second, but a "leap year" doesn't actually add a year.
Strictly speaking, a leap year is "a year in which a leap day is inserted into February."
I was just explaining what Re: #52 meant — I never said the leap second was some new concept.
No need to pick a pointless fight with me.
and in fact, up until now it had been gradually slowing, with some fluctuation along the way.
So why is it spinning faster now?
Oil drilling, ore mining, tunnel excavation reshaping the terrain,
plus shifts in internal structures — probably something in the mantle —
that seem to have moved quite abruptly. I think that's the cause.
Maybe it's the other way around?
What if the atomic clock's own "tick speed" has slowed down for some reason,
and the clock's rate is what's off?
To put it in extreme terms:
suppose something used to take 10 seconds by the atomic clock.
If the clock's ticking rate becomes half as fast as before,
the second hand only advances 5 seconds' worth,
so the "measured elapsed time" comes out smaller —
making the object's motion look like it sped up,
leading to the conclusion "the motion got faster."
But we're talking about such incredibly tiny fractions of time
that comparison with the past is basically impossible,
and since the very "ruler for measuring the passage of time"
is itself stretching and shrinking,
an accurate comparison with before just isn't possible.
They also say (giant) dams shift the Earth's center of gravity.
Why does it even need adjusting?
It'll keep drifting until summer becomes winter, lol.
Apparently they're scrapping it because it became "what's even the point of correcting things down to the second? Isn't this just a pain?"
A year is 365 days, but Earth's actual orbital period is about 365.25 days.
Leave that alone and it drifts a full day every 4 years,
so once every 4 years they make it 366 days to fix the gap.
and Obon becomes the tail end of summer — "still hot, but mornings and evenings
are getting more comfortable" by then.
Even if you say it'd be fine,
years/months/days run on the orbital cycle
while hours/minutes/seconds run on rotation.
The "365 days" thing doesn't line up with that at all —
you're just listing two completely separate things side by side.
If the goal is to cut down how often you have to adjust, I'd think minute-level increments would've made more sense.
No need to worry about something that'll only matter after you're dead.
Yeah, it's actually pretty huge.
Computerization ended up making this less rational, not more —
they could've just let some central authority handle clocks however it wanted.
For now they've just decided to tolerate up to a 1-hour gap,
but how it'll actually be operated hasn't been decided yet.
For example, they might end up inserting a "leap minute" once things drift by a minute,
but that's roughly a century away, so there's plenty of time to hash out the details.
The day currently designated as New Year's isn't the winter solstice, isn't perihelion, and has no particular astronomical significance — so why was this the day chosen?
The day mochi tastes best, probably.
Is this story going to be a long one?
Probably because the leap second is a concept that only emerged in theory once precise scientific calculation became possible.
If you don't phrase it in intuitive terms, it's hard to grasp.
That's not what I meant. It's:
"correcting with a second" → leap second
"correcting with a month" → leap month
but then "correcting with a day" becomes "leap year" — that mismatch is what feels off.
No, that's exactly what we're talking about.
Keep up.
Even under the old lunisolar calendar, a year with an inserted leap month was also called a "leap year."
Conversely, even today, February 29th itself is called a "leap day"
(though people rarely bother with that term since it's quicker to just say "February 29th").
If you center it on the winter solstice you land around February,
and Christ's birthday is a bit over a month before that,
so somewhere around there felt right,
and they probably just settled on the tail end of the Christmas holidays.
New Year's Day is set so that the spring equinox lands around March 21st —
that's how it was arranged when the Gregorian calendar was adopted in the 16th century.
In that sense it's not totally meaningless astronomically.
The spring equinox also ties into Easter on top of farming, so there's a strong religious dimension to it too.
Mostly it's the Roman emperors' fault.
Under the Julian calendar that preceded today's Gregorian calendar, February was the end of the year and March was the start — because that's how the emperor decided it.
That's why February got stuck with the leftover 28 days, with one extra day tacked on in leap years to balance things out.
The Roman emperors also did plenty of other self-serving things, like adding months or shifting which months got 31 days, pretty much as they pleased.
The Gregorian calendar was later codified by the Pope based on astronomical observations and the like, but since it built on the existing Julian calendar, it inherited that habit of balancing things out in February.
There's no real benefit to keeping time perfectly synced with Earth's motion.
Scientifically speaking, time that flows uniformly without exception is far more useful.
My guess is that by the time the gap hits an hour, they'll just decide to "let it keep drifting as-is."
It'll stop matching up with longitude —
the International Date Line would have to shift, right?
Maybe it'll split into "everyday clock time" versus "measured/scientific time." Though that does seem like it'd need a pretty major technical breakthrough.
Background and key points of this discussion
The leap second corrects drift in Earth’s rotation (the length of a day), while the leap year corrects drift in Earth’s orbit (the length of a year) — different targets, different mechanisms entirely. But in the thread, the shared character “leap” (閏) in both terms made them easy to mix up, which became a talking point in itself. What’s being abolished this time is the former; its replacement, the “leap hour,” allows discrepancies to build up to a full hour before correcting, meaning no adjustment work will be needed at all for the foreseeable future. As for why Earth’s rotation has been speeding up slightly in recent years, commenters floated various theories — melting glaciers, crustal shifts, and more — but neither the article nor the thread pinned it on any single definitive cause, treating each as merely a contributing factor. The operational details of exactly when a “leap hour” would actually be inserted are also still an open question, but since there’s no need to rush an answer for decades at least, it’s been left for future discussion.
*This article is excerpted and summarized from the 5ch (Newsflash+) thread “Leap second abolition officially decided in October — Earth’s rotation speeding up, leap hour to be introduced.”
Leave a Reply