The Future of Leap Seconds
@10u8 writes "Since 1972 precision clocks around the world have ticked using atomic seconds, but earth rotation is slowing down. Leap seconds have been inserted in order to keep noon happening at noon, but they upset some timekeepers. Recent discussions have considered
discontinuing leap seconds in UTC, and a colloquium in Torino next month will present results. It is a matter of international significance."
And why do we care?
Read the article!
It's important for systems programmers, and lots of folks here are at least systems programming fanboys.
It's important for navigation. Yeah, that includes your GPS toys.
It's important for a number of scientific disciplines, including a number of subdisciplines of radio astronomy.
It's also really interesting that the change in the Earth's rotation can't yet be predicted with enough accuracy to set a schedule in advance for adding leap seconds, but must be measured. This is relatively prosaic stuff that's nonetheless at the limits of our current understanding. Doesn't anyone get excited or curious about science anymore?
And just think, if no leap seconds were added since 1972, you'd be having your Noon Lunch at 11:59:38!
:)
Oh the horror...
Accuracy isn't everything...
"Can of worms? The can is open... the worms are everywhere."
The concept of leap years is because the ratio of the length of a year divided by the length of an earth day is not an iteger. No calendar can get around that fact. You either add intercalation days whenever the remainders of your divisions exceed 1, or you keep track of huge numbers and cycles that greatly complicate your timekeeping.
The Julian roman calendar did suck because they didn't get the ratios quite right and it drifted. (The Gregorian calendar fixed this for all practical purposes.) However, prior to Julius Caesar, it sucked even more because there was no mathematical formula. Instead, priests were supposed to observe the sun each year and decide when leap days were needed.
The priests were also involved in politics, so they chose to shorten political terms more often than not by omitting leap days. IIRC, by the time the Julian calendar was instituted, the Romans were off by several months due to these partisan shenanigans.
The stated problem with leap seconds is that some software gets confused by them. Guess what? That same software probably gets confused if the time zone changes, or when it moves into daylight savings time.
The Right Way to solve this problem is for computers to work with TAI internally, and treat the difference introduced by leap seconds as part of the time zone, for human consumption only. Instead of defining PST to be UTC - 08:00, define PST = TAI - 08:00:22.
Computers can keep their straightforward time system, humans can keep our astronomically synchronized system. No need to lose either of those qualities.
Tarsnap: Online backups for the truly paranoid
With a manned mission to Mars possibly less than 20 years away, shouldn't we start looking at timekeeping systems that aren't tied to this rock?
I personally hate daylight savings. Why do we need to change our clocks just to keep people awake during dayligt? Why can't we just get up/start work one hour earlier if it's really that important?
Assuming you said that because you hate Bush, bad plan. Time is your friend. If Nov 2004 was, say, tommorow, Bush would win by a landslide. Time is your friend. If anything, you need to push it back.
First of all, whether or not an observer can tell how many matches are in a matchbox is immaterial--nay, completely unrelated to the fact that there is a particular number of matches in the box, a precise number that can be determined.
Secondly, time needs not be exact for most people most of the time, but perhaps you can recognize that there are certain applications, especially scientific and technological, for which a measure of exactitude is quite necessary.
Thirdly, I would venture to say that the society depicted in 1984 would rather that people be unable to tell what time it was.
Lastly, all these fitful worries are meaningless, because my man Flavor Flav always knows what time it is. Word.
I watched C-beams glitter in the dark near the Tannhauser gate.
leap seconds are only evil when you try to commingle UTC and TAI. If your system operates on a straight TAI, then leap seconds become a presentation issue right along with time zones and daylight savings.
-russ
Don't piss off The Angry Economist
I think that the removal of the leap second is a big mistake. Sure any noticeable changes will be extremely gradual, keeping time has more purposes than just knowing what time to leave for lunch. If we read about cowboys fighting at high noon, we know what it is. If we read about Paul Revere's midnight ride, we know that it did indeed happen at night.
Removing the leap second makes most history recorded with reference to time of day pretty useless. Noon is defined by most people as the time that the sun is in the middle of the sky. Let's keep it that way. If method of keeping time based on exact seconds from one point in time to another (which is actually pretty useless for most things that happen within timeframes longer than a couple of minutes) then let a separate system be designed for it. Start reading off an atomic clock and never account for leap seconds, but don't screw up the rest of the world to please a few.
There are occasional problems when they add the leap seconds (programs that don't expect 61 seconds in a minute, for example), or programs that don't realize that there are X number of seconds (15 or so?) that simply didn't exist since 1970. (sometimes this stuff matters).
I think that the present UTC compromise is quite reasonable. In almost all civilian systems, including non-real-time computers (like the one you are using right now), you really don't need perfectly constant real time and they are probably usually off from the correct time by a few seconds to a few minutes anyway. The leap second is handled seamlessly as just regular clock skew. I've never seen a PC that didn't gain or lose about 30 seconds a day anyway. (That's really pathetic when you consider what a $5 watch can do.)
If you have some kind of real-time system, then just use TAI. It's about 35 seconds off from UTC. I'd like to have a civil time that is closely synchronized to the real world (UT-1).
AFAIK, the loss of time is fairly predictable since a rotation day is about 86400.002 seconds long so a leap second accumulates about every 500 days. They have a leap second about every 1.5 years (on June 30 and December 31).
All this techno mumbo-jumbo is just an overly verbose representation of an old saw:
A man with a watch always knows what time it is. A man with two watches is never quite sure...
I have no problem with your religion until you decide it's reason to deprive others of the truth.