Leap Second At The End of 2005

Ruff_ilb writes "Because of the discrepency between an ephemeris second (the fraction 1/31,556,925.9747 of the tropical year for 1900 January 0 at 12 hours ephemeris time) and the second of atomic time (the duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the cesium 133 atom), we're left with more than leap years. In order to ensure that the the atomic time and civil stay coordinated, "Civil time is occasionally adjusted by one second increments to ensure that the difference between a uniform time scale defined by atomic clocks does not differ from the Earth's rotational time by more than 0.9 seconds."" And Happy New Years everyone ;)

Comments

And those with computers...

[jZnat]

Those with updated (don't know from when) libc or equivalent automatically encountered the leap second at midnight. Even if you didn't, you'll still sync back up with an NTP server eventually (I'd hope).

Re:How did you use yours?

[lousyd]

Do any NTP servers keep track of these seconds?

Yes, that's the point of serving the Network Time Protocol...

Re:Highlights problem with ntp...

[thehickcoder]

Actually, if you where doing high precision scientific applications, it seems this type of behavior would be preferred. Because of the leap second there was not 2 seconds between 11:59:59 and 12:00:01 last night. So, using the NTP behavior if I want a timestamp that was exactly 10000000 seconds ago, I get one that represents 10000000 actually elapsed seconds.

Just because everybody agrees to change their clocks doesn't mean time actually slows down or speeds up.

Re:It's not the 'ephemeris second' that's the prob

[Jonathan McDowell]

Right. If you want a timescale that guarantees that it's dark at midnight, you use UTC, locked to the Earth's rotation. If you want a timescale which guarantees a simple calculation of the elapsed time between two time stamps, use TAI or TT or GPS time or another timescale that's linked to atomic time or proper time in some rest frame. The scientific community provides all the different timescales that you might want, all of them within a couple of minutes of being GMT. The Earth-locked one makes sense for civil time. There's a problem that many technical applications - for instance a lot of stuff at NASA - use UTC in places where it would really make more sense to use TT or TAI, causing needless software grief.

Re:How did you use yours?

[rossdee]

"(By "solar" I don't just mean sundials. I mean clocks that are set by pointing the hour hand straight up at noon, which is how all clocks were set before time zones were invented.)"

How do they know its noon then?

(Its overcast here)

Re:The hard way

[jc42]

I think if we level all mountain ranges and melt both ice caps it should be enough to make Earth spin faster enough to compensate for this leap second

Um, if you melt the ice caps, won't the water spread itself out through the oceans? This will, on the average, make the molecules move away from the axis, thus slowing the planet's rotation.

What you want to do is pile mass up close to the axis, i.e., at the poles. With water, you'd need to precipitate it out at the poles. This isn't what we're doing, though.

For mountains, you don't want to just level them; you want to move the mass closer to the axis. So you want to pile the rubble on the poleward side of where the mountain used to be (north in the northern hemisphere, etc.).