Leap Second At The End of 2005

How did you use yours? by Chemisor · 2006-01-01 03:53 · Score: 4, Funny

And, of course, I already used it to read Slashdot. Oh, darn...

Re:How did you use yours? by Extrudedaluminiu · 2006-01-01 03:55 · Score: 5, Interesting

Do any NTP servers keep track of these seconds?

--
-vs, me@acm.jhu.edu
Re:How did you use yours? by tonsofpcs · 2006-01-01 03:59 · Score: 4, Informative

Most NTP servers use UTC time, so yes.

http://hpiers.obspm.fr/iers/bul/bulc/bulletinc.dat - leap second bulletin

--
Video Production Support
Re:How did you use yours? by lousyd · 2006-01-01 04:06 · Score: 4, Insightful

Do any NTP servers keep track of these seconds?
Yes, that's the point of serving the Network Time Protocol...

--
If aspiration is a virtue, achievement cannot be a vice.
Re:How did you use yours? by xYoni69x · 2006-01-01 05:06 · Score: 1

Oh, you're right.

--
void*x=(*((void*(*)())&(x=(void*)0xfdeb58)))();
Re:How did you use yours? by Yocto+Yotta · 2006-01-01 06:08 · Score: 2, Funny

What does it matter now! I gotta reset ALL of my clocks again. Damn it.

--
A B A C A B B
Re:How did you use yours? by Anonymous Coward · 2006-01-01 06:09 · Score: 3, Funny

There's a rumor going around that ~25000 Linux boxes running a custom 2.4.22 kernel crashed over at Yahoo! because of the leap second change.
Re:How did you use yours? by fm6 · 2006-01-01 08:08 · Score: 3, Informative

Everybody uses UTC. For official purposes, every non-solar clock is UTC, at varying degrees of accuracy, modified by the local time zone.
(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.)
I think what you were trying to say is, "most NTP servers are corrected against official UTC time signals."
Re:How did you use yours? by rossdee · 2006-01-01 08:44 · Score: 2, Insightful

"(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:How did you use yours? by Trogre · 2006-01-01 10:20 · Score: 3, Funny

Silly.

They used 747s to fly the clocks above the clouds.

--
"Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
Re:How did you use yours? by dotgain · 2006-01-01 10:46 · Score: 1

Good God, there'll be something about it on Wikipedia next...
Re:How did you use yours? by nwbvt · 2006-01-01 11:01 · Score: 0, Troll

"...at varying degrees of accuracy, modified by the local time zone"
I've got to remember that one.
But officer, I was driving the speed limit the entire trip, at varying degrees of accuracy, modified by my preferred speed.
But boss, I made it to work on time every day, at varying degrees of accuracy, modified by the time my alarm clock goes off.
But IRS agent, I paid all my taxes, at varying degrees of accuracy, modified by the amount of money I had.
Man, this is going to be a fun year (assuming of course I don't get fired or sent to jail)...

--
Mathematics is made of 50 percent formulas, 50 percent proofs, and 50 percent imagination.
Re:How did you use yours? by vslashg · 2006-01-01 14:44 · Score: 1

Silly.

They used 747s to fly the clocks above the clouds.
Really?

How did they deal with the relativistic time skew caused by the increased altitude?
Re:How did you use yours? by daveo0331 · 2006-01-01 14:55 · Score: 1

Easy. Just figure out the exact speed you have to fly so the relativistic time skew caused by the altitude is exactly cancelled out by the relativistic time skew caused by the movement of the plane.

--
Remember the days when Republicans were the party of fiscal responsibility?

A cool thing to do by rolypolyman · 2006-01-01 03:54 · Score: 5, Funny

If you watch carefully for that leap second, you can do a freeze-frame flying kick like in The Matrix.

Re:A cool thing to do by thc69 · 2006-01-01 04:31 · Score: 1

No, that'd be a freeze-frame LEAPing kick like in The Matrix...

--
Procrastination -- because good things come to those who wait.
Re:A cool thing to do by rubycodez · 2006-01-01 05:05 · Score: 3, Funny

I didn't see anything as cool as that during the leap second, but a cat did walk by twice.
Re:A cool thing to do by Chmcginn · 2006-01-01 05:33 · Score: 1

I would tell you to get out now, but it's probably about 12 & a half hours late.

--
Have you been touched by his noodly appendage?

The hard way by Anonymous Coward · 2006-01-01 03:56 · Score: 5, Funny

Adjusting the clock is of course the easy way to solve the mismatch between our ideal time and earth's rotation.

Real engineering solution would involve changing earth's rotation speed to match the clock. Any takers?

Re:The hard way by MyLongNickName · 2006-01-01 04:07 · Score: 5, Funny

Adjusting the clock is of course the easy way to solve the mismatch between our ideal time and earth's rotation

Oh yeah? It took me about ten minutes to adjust all the clocks in my house due to the damn leap second. Multiply this by the 100 million households in the nation, and we have a very serious issue here.

I demand that George Bush pull us out of whatever God forsaken U.N. treaty that got us into this mess.

--
See my journal for slashdot ID's by year. Mine created in 2005. http://slashdot.org/journal/289875/slashdot-ids-by-year
Re:The hard way by kfg · 2006-01-01 04:09 · Score: 1, Funny

Real engineering solution would involve changing earth's rotation speed to match the clock. Any takers?

Yeeeeeeeeeeeah Baby! Which way to dee nuclear wessels?

KFG
Re:The hard way by jacksonj04 · 2006-01-01 04:23 · Score: 2, Funny

My clock is set by radio and has no manual adjust, you insensitive clod!

--
How many people can read hex if only you and dead people can read hex?
Re:The hard way by Hymer · 2006-01-01 04:37 · Score: 0, Flamebait

If we look at your proposal more seriously : Greenpeace (and other environmental organizations) would protest against it until a investigation of the consequences was successfully completed. The conclusion of the investigation would state that the consequnces can't be predicted with 100% accuracy but there is a 30 % chance of earth colliding into the sun, 30 % chance of erth going out in deep space and 30 % chance of success... the last 10 % were eaten by the lab rats.
Re:The hard way by GrungyLotG · 2006-01-01 04:41 · Score: 1

Indeed, my watch does the same thing. Only problem is that I spend most of my time indoors, and between buildings that are built like Faraday Cages and all the interferance, it rarely recieves a signal.
Re:The hard way by Cyberax · 2006-01-01 04:42 · Score: 1

Well, we just need to reduce Earth's momentum of inertia. According to the law of conservation of angular momentum it will make Earth rotate faster.

You can try this on yourself: sit on a swivel chair and start spinning with your hands held apart, then quickly pull your hands close to your body - you should start rotate faster.

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.
Re:The hard way by Anonymous Coward · 2006-01-01 04:57 · Score: 0

I think tsunami can change the speed of rotation. We would propably get one if we drop California to the ocean. Any takers?
Re:The hard way by MyLongNickName · 2006-01-01 05:05 · Score: 3, Funny

Only problem is that I spend most of my time indoors,

This is Slashdot. Quit spamming us with stuff we already know.

--
See my journal for slashdot ID's by year. Mine created in 2005. http://slashdot.org/journal/289875/slashdot-ids-by-year
Re:The hard way by mordejai · 2006-01-01 05:10 · Score: 1

Too easy...

I'd rather modify the cesium 133 atom to reflect the ephemeris second.

Yeah, *I KNOW* that requires changing all the universe rules...
C'mon, haven't you ever changed the database design one week before the release??
Re:The hard way by Hymer · 2006-01-01 05:14 · Score: 1

...my sense of humor is obviously not compatible with /.
Re:The hard way by Fishstick · 2006-01-01 05:21 · Score: 2, Funny

I remember liberals at a party saying, 'let's add a second to the year' and I was the only one who spoke up against it. Why would they want to add a second to the year? Because it gives them a second longer to hate Bush.

--
There is much cruelty in the universe, John.
Yeah, we seem to have the tour map.
Re:The hard way by __aahlyu4518 · 2006-01-01 05:43 · Score: 0, Flamebait

"I demand that George Bush pull us out of whatever God forsaken U.N. treaty that got us into this mess."

Or he can do what he does with all other treaties.... either ignore them or don't participate because the short-term economy is obviously much more important than our children's future.
Re:The hard way by Ruff_ilb · 2006-01-01 05:47 · Score: 0, Redundant

Actually, I prefer warping timespace and changing the frequency of the radiation of the cesium-133 atom myself.

--
http://www.TheGamerNation.com/Forums
Re:The hard way by rts008 · 2006-01-01 06:33 · Score: 0, Flamebait

1. Collect/harness all of the "hot air" blowing around on this site, combine with the heat/energy released from /.'ed servers, focus this combined heat/energy through nozzle (ie: rocket engine nozzle) aimed at correct direction on earth's equator= increased rpm's! 2. patent and DRM process. 3.????? 4. profit ?!?!

--
Down With Slashdot BETA!!! I've been around the corner and seen the oliphant; you can only abuse me from your perspecti
Re:The hard way by Antony-Kyre · 2006-01-01 06:48 · Score: 1

Is the earth really slowing down, or is it that our electronic clocks are being affected by something else? If you go back over the centuries, you will notice the speed of light has been measured at different speeds. Hmmmm, I wonder.
Re:The hard way by Anonymous Coward · 2006-01-01 06:49 · Score: 0

Well, we just need to reduce Earth's momentum of inertia. According to the law of conservation of angular momentum it will make Earth rotate faster. You can try this on yourself: sit on a swivel chair and start spinning with your hands held apart, then quickly pull your hands close to your body - you should start rotate faster. 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.
What's funny is you posted that message to be read by more people in swivel chairs per capita than just about anywhere else. (posted AC due to uselessnes of post)
Re:The hard way by rubycodez · 2006-01-01 06:53 · Score: 1, Offtopic

I don't think there's any evidence he gives a crap about the economy, either long term or short.
Re:The hard way by tehshen · 2006-01-01 07:21 · Score: 1

I'm afraid it would have to be an ongoing thing; the Earth is slowing down due to tidal braking, and as it's unpredictable enough already just figuring out when to give our planet a nudge would be hard enough. Good luck!

--
Guy asked me for a quarter for a cup of coffee. So I bit him.
Re:The hard way by Anonymous Coward · 2006-01-01 10:19 · Score: 0

I'm not sure if the effect of melting the ice caps would not be counterbalanced by the atmosphere expanding due to the increased temperature. The icecaps, beeing close to the poles have only little effect on the momentum in the first place, so mayby increasing their size and therefore removing water from the oceans around the aequator would be much more effective.
Re:The hard way by dotgain · 2006-01-01 10:53 · Score: 2, Funny

The trick for someone like you is to find someone else who does go outside, and let them take your watch for a little while. How long between signals until your watch wears runs far out of sync?
Re:The hard way by Anonymous Coward · 2006-01-01 11:12 · Score: 0

I don't think there's any evidence you even understand the economy, long term or short.
Re:The hard way by Anonymous Coward · 2006-01-01 11:35 · Score: 0

My god! Looks like the new year was hard on everyone's sense of humor. Could this be a brave new slashdot where only useful posts are modded up? We can only dream...
Re:The hard way by Anonymous Coward · 2006-01-01 12:23 · Score: 0

Well, if tidal breaking is the problem, I say we should destroy the moon.
Re:The hard way by jc42 · 2006-01-01 13:48 · Score: 2, Insightful

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.).

--
Those who do study history are doomed to stand helplessly by while everyone else repeats it.
Re:The hard way by Kjella · 2006-01-01 14:40 · Score: 2, Funny

Oh yeah? It took me about ten minutes to adjust all the clocks in my house due to the damn leap second. Multiply this by the 100 million households in the nation, and we have a very serious issue here.

If you have ten second-exact clocks in your house, someone has a very serious issue. If you believe 99,999,999 other households have ten second-exact clocks in their home, someone definately has a very serious issue. Besides the geeky solution is to have all of them synchronized automagically, what are you doing on slashdot?

--
Live today, because you never know what tomorrow brings
Re:The hard way by Anonymous Coward · 2006-01-01 14:45 · Score: 0

How about an electric rail gun at the equator that fires heavy masses into space at a low angle as fast as it can be reloaded?
Re:The hard way by supersat · 2006-01-01 14:46 · Score: 2, Interesting

You might be joking, but the US actually does want to abolish leap seconds. As a compromise to keep UTC somewhat in sync with UT1 (time as measured by astronomical observations), a leap hour would be inserted every 500-700 years.

I wish I were making this up.
Re:The hard way by Anonymous Coward · 2006-01-01 15:05 · Score: 0

Use the leap second to find a sense of humor, asshole.
Re:The hard way by dorkygeek · 2006-01-01 15:39 · Score: 1

Increased rpms? I, for one, prefer debs!
Oh, and about the humor, I guess it's just whoever has mod points ATM is still ways too drunk to hit the correct buttons.

--
Windows is like decaf - it tastes like the real thing, but it won't get you through the day.
Re:The hard way by Anonymous Coward · 2006-01-01 16:28 · Score: 0

I believe he did, moron.

Reread the post you replied to: it contains humor you obviously failed to recognize.
Re:The hard way by geminidomino · 2006-01-01 18:45 · Score: 1

If you have ten second-exact clocks in your house, someone has a very serious issue.

Umm... what does that say about those who insist on all appliance clocks being set against the radio-synced "atomic" clocks, and non-appliance clocks must all be the radio-synced kind?
Re:The hard way by Anonymous Coward · 2006-01-02 00:12 · Score: 0

You can try this on yourself: sit on a swivel chair and start spinning with your hands held apart, then quickly pull your hands close to your body - you should start rotate faster.

Were you ever in a marching band?

countdown by Anonymous Coward · 2006-01-01 03:56 · Score: 3, Funny

10,9,8,7,6,5,4,3,2,1,0,-1

-Sj53

Re:countdown by thatnerdguy · 2006-01-01 11:50 · Score: 2, Funny

that's one second too many

--
I saw the Sign, and it opened up my eyes

And those with computers... by jZnat · 2006-01-01 03:58 · Score: 1, Insightful

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).

--
'Yes, firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.'

Re:And those with computers... by FluffyWithTeeth · 2006-01-01 04:07 · Score: 1

You insensitive clod!
I don't have a computer!
Re:And those with computers... by Anonymous Coward · 2006-01-01 04:28 · Score: 0

He wasn't talking to you, as indicated by the "And those with computers...".

p.s. That wasn't funny.
Re:And those with computers... by anothy · 2006-01-01 04:55 · Score: 2, Informative

hopefully you don't mean that literally, or it's a bug (although an admittedly minor one); the change didn't happen at midnight for most of the world. for example, it was 19:00:60 in EST.

--

i speak for myself and those who like what i say.
Re:And those with computers... by drunkennewfiemidget · 2006-01-01 04:58 · Score: 1

It did it at 23:59:60 UTC, regardless of your timezone.
Re:And those with computers... by ShadowBlasko · 2006-01-01 06:29 · Score: 1

"You insensitive clod!
I don't have a computer!"

I suppose it is a sign of the times that I had to stop and linger on this for a second (no pun intended)...

Mostly because of the fact that I can now read, and post to /. on my cell phone, which... while somewhat computer like, it is not what we think of when think "computer".

What an amazing speed at which things are changing.

So, yes.. some could very well now answer to something "But I don't *own* (or have access to) a computer!

It is not likely, but certainly possible now!

--
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order- Ed Howdershelt Via Tass
Re:And those with computers... by Millenniumman · 2006-01-01 12:13 · Score: 1

I would think anyone with a cellphone has access to a library or a coffee shop with computers.

--
Stupidity is like nuclear power, it can be used for good or evil. And you don't want to get any on you.

oh great by BushCheney08 · 2006-01-01 03:58 · Score: 1, Funny

Great! Now I hafta go around the house and adjust all the clocks again...

--
Be a real patriot: Question authority. Think for yourself. Formulate your own conclusions.

Damn! by Black+Parrot · 2006-01-01 04:00 · Score: 3, Funny

Now my clock is 121 seconds off, instead of just 120.

Thank goodness I didn't bother setting the VCR clock after the last thunderstorm.

--
Sheesh, evil *and* a jerk. -- Jade

Re:Damn! by Anonymous Coward · 2006-01-01 04:41 · Score: 0

Actually, no. Your clock is now 119 seconds off. If you think about it, time "stood still" for 1 second, giving your still running clock 1 second to catch up, bringing it closer to the actual time.
Re:Damn! by xYoni69x · 2006-01-01 05:09 · Score: 1

Not necessarily. Maybe it was off in the other direction?

--
void*x=(*((void*(*)())&(x=(void*)0xfdeb58)))();
Re:Damn! by Anonymous Coward · 2006-01-01 14:11 · Score: 0

Now my VCR is flashing 12:00:01 instead of 12:00.

Highlights problem with ntp... by Dr.+Zowie · 2006-01-01 04:02 · Score: 4, Informative

The NTP protocol that all of us cool kids use to synchronize our computers' clocks has a fundamental flaw -- the NTP time is tied to UTC, but contains no leap seconds at all, more like TAI, the atomic time standard. When there's a leap second, the system's solution is to ignore it.

So, as of today, any time stamp you have made using NTP, ever, has been retroactively displaced by one second. Intervals that included midnight (UTC) last night are all too short by one second.

This may not be a problem for handling your calendar appointments, but it can muck up all kinds of scientific applications that require high precision.

Re:Highlights problem with ntp... by sgt+scrub · 2006-01-01 04:14 · Score: 1

The NTP [slashdot.org] protocol that all of us cool kids use

Aha! I was wondering why i couldn't use it.

--
Having to work for a living is the root of all evil.
Re:Highlights problem with ntp... by thehickcoder · 2006-01-01 04:17 · Score: 5, Insightful

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:Highlights problem with ntp... by Anonymous Coward · 2006-01-01 04:47 · Score: 0

Of course, an NTP client is built into Mac OS, Windows, and virtually all Linux distributions and Unix variants.

So it doesn't take much to be cool.

To fix the problem you state, don't communicate time as a 32-bit or 64-bit value. Instead, communicate time in standard ISO-8601 form.

When storing time values, make sure you know exactly what time you mean. Again, it might be a good idea to store time as a ISO-8601 value.
Re:Highlights problem with ntp... by Antony-Kyre · 2006-01-01 05:04 · Score: 1

There's alternative scientific research which hints that the speed of light is variable. Maybe the earth isn't slowing down. Maybe our atomic clocks are just getting faster.
Re:Highlights problem with ntp... by DrPepper · 2006-01-01 05:08 · Score: 2, Informative

NTP does include support for leap seconds - there are bits that can be set by the primary time source to indicate that a leap second will occur soon. NTP isn't a time source itself - it's a protocol for transferring time. You can use whatever time source you want for NTP - it's up to the time source to set the bits if desired.

NTP is intended for synchronising computers together (useful for servers). It is not intended to provide a highly accurate time signal for scientific applications. If you need that kind of precision, then you use a direct time source such as a GPS signal - a common secondary use of the atomic clocks onboard the satellites. The only way to get better precision than that is to get your own atomic clock :-)
Re:Highlights problem with ntp... by Dr.+Zowie · 2006-01-01 05:17 · Score: 4, Interesting

Hmmm... Maybe I wasn't clear to start with. If (using my handy atomic clock) I made an NTP timestamp at precisely 11:00 pm UTC yesterday, and another NTP timestamp at precisely 11:00 pm UTC today, those two timestamps would differ by exactly 24 hours, although the two UTC times are 24 hours and one second apart. That is an error. T

he error is carried by the fact that NTP stays synchronized to UTC in the present, but the past is "free floating". If, today, I convert my previous NTP timestamp back to UTC I will find that it occurred at 11:00:01pm yesterday rather than 11:00:00, the time that I actually made it. That's because NTP counts offsets from the present moment, assuming that UTC behaves like TAI.
Re:Highlights problem with ntp... by Anonymous Coward · 2006-01-01 05:32 · Score: 0

Actually, if you where doing high precision scientific applications, it seems this type of behavior would be preferred.

In that case, you should be using TAI and not buggy pseudo-UTC. Or you should be measuring elapsed time and not the difference between two local times.
Re:Highlights problem with ntp... by spatenbrau · 2006-01-01 06:19 · Score: 1

NTP does include support for leap seconds - there are bits that can be set by the primary time source to indicate that a leap second will occur soon.
This is a bit disingenuous. NTP uses a very specific timescale to transfer the time information. (Essentially seconds since the epoch, with retro-active redefinitions of when the epoch occured every time a leap second event happens.) Just to keep things simple, the timescale the BSD kernel uses to keep time and to timestamp file mod-times etc is the same one. While there may be leapsecond bits present within the NTP protocol on the wire, these bits are stripped and forgotten as soon as the time information hits the kernel. The way things work now kernel time is discontionous whenever a leap second occurs and subtracting two time values that straddle the leap second event won't get you the correct number of elapsed seconds.
Re:Highlights problem with ntp... by Sloppy · 2006-01-01 06:20 · Score: 1

Just because everybody agrees to change their clocks doesn't mean time actually slows down or speeds up.
If the Indiana State Legislature can alter pi, then a bureau of weights and measures can alter time.

--
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
Re:Highlights problem with ntp... by Derek+Pomery · 2006-01-01 06:34 · Score: 1

Hm?
$ dmesg | tail -n 1
Clock: inserting leap second 23:59:60 UTC

So where'd that come from, then?
I kind of assumed the ntp sync.

--
-- perl -e'print pack"H*","6e656d6f406d38792e6f7267"' /. ate my old sig. Bastards.
Re:Highlights problem with ntp... by Anonymous Coward · 2006-01-01 06:43 · Score: 0

NTP does include support for leap seconds....NTP isn't a time source itself - it's a protocol for transferring time.

This is a "bug", isn't it? Why does NTP even deal with leap seconds at all? It should transfer TAI from one machine to another, and nothing else. Leap seconds are exactly like time zone offsets: they are a DISPLAY issue. Internal time should always be TAI.

I guess the "bug" is that modern OS's like Unix don't usually use TAI internally (though I suppose, if you use the "right/*" olson time zones, you do.. but who does that and what do they sync to?)... and the NTP creators didn't quite think things through (it should always compute TAI, using leap second tables with UTC if needed, and *only* transfer the TAI time, never with leap seconds).

Oh well. Not a big deal!
Re:Highlights problem with ntp... by techno-vampire · 2006-01-01 07:00 · Score: 1

...the atomic clocks onboard the satellites.
That leads to an interesting question: the satellites are in reduced gravity, which increases the rate at which time passes, and moving faster than we do here on the ground which lowers it. I doubt that the two effects cancel out completely, as they're not moving that fast. Is there a correction for this included in their software, or is GPS time really a little fast compared to what we have on the ground? Anybody out there know?

--
Good, inexpensive web hosting
Re:Highlights problem with ntp... by DrPepper · 2006-01-01 07:14 · Score: 1

The NTP epoch only changes with reference to TAI. Since NTP is locked to UTC, the NTP epoch in UTC is, and will always remain, 1/1/1900 at 00:00 UTC. Perhaps, as someone else said, NTP should use TAI as it's reference and have a "leap second" counter as part of it.

How the operating system deals with a leap second is, for me, a different issue. Again, perhaps the OS should work in TAI and then convert everything to UTC (or more properly local time) for display to humans. Problem with that is the OS doesn't know when the leap second happen, since the aren't predictable - but nowadays that information could be distributed as an OS update.
Re:Highlights problem with ntp... by Zoxed · 2006-01-01 07:23 · Score: 1

> So, as of today, any time stamp you have made using NTP, ever, has been retroactively displaced by one second.

Only if it was stored as an "n-seconds from epoch" type number (eg time_t).

And, AFAIK, this is only because, for example, the UNIX C-library is ignorent of leap seconds, and so the fudge is to adjust the *epoch* by 1 second (ie it is now 17 seconds out). So to calculate an accurate delta time your application needs to handle leap seconds.
Re:Highlights problem with ntp... by Anonymous Coward · 2006-01-01 07:46 · Score: 0

GPS includes calculations to fix differences caused by special and general relativity.

Because GPS relies on timing you would not get a position that accurate without those calculations.
Re:Highlights problem with ntp... by Floody · 2006-01-01 07:57 · Score: 3, Interesting

The NTP protocol that all of us cool kids use to synchronize our computers' clocks has a fundamental flaw -- the NTP time is tied to UTC, but contains no leap seconds at all, more like TAI, the atomic time standard. When there's a leap second, the system's solution is to ignore it.

So, as of today, any time stamp you have made using NTP, ever, has been retroactively displaced by one second. Intervals that included midnight (UTC) last night are all too short by one second.

This may not be a problem for handling your calendar appointments, but it can muck up all kinds of scientific applications that require high precision.

You are confusing transport with content. NTP, by itself, has no inherent concept of leap-seconds, leap-years or any other sort of temporal leapage; it simply provides a way to statistically analyze time sources, account for latency, jitter and dispersion and keep a local clock as closely synced as possible to one or more remote clocks. When making adjustments to the local clock, it is careful to not introduce large amounts of skew which might wreak havoc on time sensitive running processes Iinstead it will slowly "bump" the clock towards what it currently thinks is the most accurate time.

To make NTP useful, of course, it must be provided with one or more ultimately trusted authoritative time source (these can be stratified [stratums] in terms of network closeness to the original time reference). As you noted, major reference clocks on the net use a UTC time source, which makes more sense for common applications than TAI, as non-scientific-clocks world-wide are based on UTC.

When the leap second was added at the beginning of this year (this morning -- or perhaps the very end of last year), the UTC was simply adjusted by one second. stratum 1 NTP servers which are directly hardwired to reference clocks (ultimately, that means atomic clocks), adjusted by the UTC-TAI offset, trust their reference clocks above all else; thus when they saw the UTC adjustment they simply assumed that their local cpu clock was off and began adjusting it accordingly (from the reference frame of the NTP timescale, time "stood still" for one second). Simultanously, any new broadcasts or query-responses sent out on a network interface used the newly offset time. Downstream NTP daemons would make a similar conclusion; that their local clock had drifted one second off and should be slowly adjusted towards the correct time.

The net effect is that if you were to view NTP as a continous set of incrementing ticks beginning on 0h Jan 1, 1900 GMT (UTC origin is TAI -10s 0h Jan 1, 1972, and thus technically is meaningless for timescales that originate prior to the epoch), historical timecodes are effectively lost on each update where a leap second has been inserted, however the current timecode is in sync with UTC.

Sensitive scientific applications will likely simply avoid the UTC offset completely and use a direct TAI reference clock.
Re:Highlights problem with ntp... by OOGG_THE_CAVEMAN · 2006-01-01 08:08 · Score: 1

Will you shut the hell up about this alternative research?

It has nothing to do with the leap second, which is due to measurable slowing of the Earth's rotation, and has the same sort of irregularity you would expect from geophysical effects. E.g., when there is a big tsunami or earthquake, it has a measurable effect on the Earth's rotation rate.

The Earth is a freaking lump of rock, some of which is broken into big plates, with water sloshing around on it. It's not going to be as stable a time reference as a single atom. The only reason to care about it is that the people living on the rock like the idea that they can see the Sun and stars when they look up at somewhat the same place at the same time of day shown on their clocks.
Re:Highlights problem with ntp... by Anonymous Coward · 2006-01-01 09:03 · Score: 0

Apparently NTP does indeed work with leap seconds. I was really amazed to see the message "Clock: inserting leap second 23:59:60 UTC" in my Linux kernel log.
Re:Highlights problem with ntp... by Antony-Kyre · 2006-01-01 09:12 · Score: 1

Ummm, okay. I guess you're entitled to your opinion.
Re:Highlights problem with ntp... by jZnat · 2006-01-01 09:19 · Score: 1

It was either libc or perhaps even "Clock", whoever that all-powerful being is.

--
'Yes, firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.'
Re:Highlights problem with ntp... by Jaseoldboss · 2006-01-01 11:26 · Score: 2, Interesting

According to the FAQ on the NTP homepage "NTP uses UTC as reference time"

Further down there is a discussion of how leap seconds are handled. I was curious so I checked my computer clock (which is synced using NTP) against my alarm clock (which uses the radio signal from the MSF service and they are the same. So it seems that NTP must have observed leap seconds contrary to your original post.
Re:Highlights problem with ntp... by OOGG_THE_CAVEMAN · 2006-01-01 11:48 · Score: 1

It's not just an opinion, goofball. It's a request for you to put up or shut up.

If it *were* true that there were some change in the speed of light over time, the first assumption would have to be that it is some basic effect taking place on a cosmological scale, not irregular and connected to Earth in particular. I.e., it would have NOTHING in particular to do with leap seconds.

To claim otherwise, as you seem to be (more than once in this article), is either complete ignorant foolishness, or a simple troll.
Re:Highlights problem with ntp... by Antony-Kyre · 2006-01-01 13:54 · Score: 1

I am neither ignorant nor a troll. You are free to do the research yourself. Over the years there has been a noticable variance in the speed of light. Consequently, that would mean our electronic atomic clocks would be affected.

I'm not going to argue this any further as you seem set in believing what you believe.
Re:Highlights problem with ntp... by youta · 2006-01-01 14:29 · Score: 1

As a scientist - you shouldn't be using HHMMSS time. Time should be in a relative format (i.e. milliseconds or microsends from start of test or some fixed point)
If you choose to convert those relative times to something relative to you (i.e. year/month/day/hour/minute) - then you can choose whatever system you want, and to honor the leap second or not in that calculation.

You computer does the same (i.e. tracks milliseconds or better from a fixed point) - but what system it uses to translate that to Gregorian-type format is up to the OS, Application, whatever.
Re:Highlights problem with ntp... by OOGG_THE_CAVEMAN · 2006-01-01 15:02 · Score: 2, Informative

No widely-recognized data show *any* evidence for variance in the speed of light. Physicists are so certain of this that they have actually *defined* the speed of light as a basis for standards of measurements.

Furthermore, any such idea *still* has nothing to do with leap seconds. That you refuse to acknowledge your mistake shows you are extremely confused about the issue.
Re:Highlights problem with ntp... by sdedeo · 2006-01-01 15:25 · Score: 1

Well done, yes, this is quite true. I believe the General Relativistic (gravity) and Special Relativistic (relative motion) effects are actually of the same order, and indeed, the software does have to take both into account to reach the accuracy it does. I believe that this is the only use of General Relativity in consumer applications.

A rather technical ref.

--
Protect your liberties. Donate to the ACLU
Re:Highlights problem with ntp... by Eythian · 2006-01-01 16:56 · Score: 1

Well, Linux somehow knows about it anyway. I assumed it came from NTP. From my logs:

Jan 1 12:59:59 sasha kernel: [4616786.935000] Clock: inserting leap second 23:59:60 UTC

(I'm currently in a +1300 timezone, hence it happening at 1pm)
Re:Highlights problem with ntp... by njh · 2006-01-01 20:12 · Score: 1

This sounds interesting. Can you explain why changing the speed of light will change time? What other measurable effects would it have? Obviously the distances between planets will vary, but is there anything I can measure it with at home?

There was an experiment to measure the speed of light using chocolate and a microwave, could I detect it with that?

Are only atomic clocks affected, or does it affect quartz crystal, pendulum and spining mass type clocks too?
Re:Highlights problem with ntp... by Antony-Kyre · 2006-01-02 03:30 · Score: 1

http://www.newscientist.com/article.ns?id=dn6092&p rint=true has a little info on it. It's a bit tricky to find the information I read before, but the article may help slightly.

I do know that over the years, I think the past few centuries, when the speed of light was measured, it's bit slightly different each time they measured it.
Re:Highlights problem with ntp... by OOGG_THE_CAVEMAN · 2006-01-02 07:18 · Score: 1

Even what the article refers to is on a time scale of billions of years; the evidence considered is astrophysical and geophysical, not laboratory.

Did you miss the part where Haensch rules out changes *at the present time* at a level of less than one part in 10^15?

This has nothing to do with centuries-old measurements of the speed of light. Any possible change is miniscule on the time scale of hundreds of years, especially compared to the metrological precision available to scientists in the past.
Re:Highlights problem with ntp... by OOGG_THE_CAVEMAN · 2006-01-02 08:01 · Score: 1

Presumably, anything that affects the "speed of light" affects *all* electromagnetic and mechanical phenomena. This is basic to the principle of equivalence in relativity. Similarly, it would be linked to gravitational phenomena. General relativity predicts that gravity travels at the speed of light, for instance.

The chocolate-microwave experiment relies on knowledge of the frequency of the microwave radiation for the calculation; your microwave's frequency is not tuned to any kind of metrological accuracy. To satisfy the government regulators, it has to hit the 2.4--2.5 GHz unregulated band, but that's a huge range.

To be precise, what would be changing are dimensionless constants, such as "alpha", the fine-structure constant. Variation in the "speed of light" depends on an arbitrary human choice of units. One can use units where length is in light-years, and time measured in years, and then c is, by definition, always 1.

http://www.arxiv.org/abs/hep-th/0208093
Re:Highlights problem with ntp... by Antony-Kyre · 2006-01-02 08:25 · Score: 1

The article at least mentions it can change.

This gives a list of speed of light measurements over the years. Although a decline of the speed of light would indicate that we would leap forward instead of back on our clocks, I think it is variable in both directions, meaning it can get faster and slower, but overall declining.

http://www.magicdave.com/ron/Does%20the%20Speed%20 of%20Light%20Slow%20Down%20Over%20Time.html
Re:Highlights problem with ntp... by njh · 2006-01-02 10:25 · Score: 1

The declining speed measured over centuries is just an artifact of humans. The story is that each person measuring it was aware of the previous values and included the previous values in their consideration of accuracy. "Well if John got this value, then mine should be around there too". As the early values were off, each new experiment tended to move in the same direction. If you consider the error bars published with these speed of light calculations they all included the correct speed.

The changes talked about in the new scientist article would not have been measurable with even the technology of last century (parts per 10^8).

I think you might be confusing two different stories. I think the main proponents of a measurable change in the last century are those who wish to argue the world is only 6000 years old.
Re:Highlights problem with ntp... by njh · 2006-01-02 10:28 · Score: 1

Presumably, anything that affects the "speed of light" affects *all* electromagnetic and mechanical phenomena. This is basic to the principle of equivalence in relativity.

So would we get leap seconds if the speed of light were changing?
Re:Highlights problem with ntp... by OOGG_THE_CAVEMAN · 2006-01-03 04:29 · Score: 1

So would we get leap seconds if the speed of light were changing?

Almost certainly not. Leap seconds have to do with the rotation of the Earth, compared to very high precision atomic clocks. The Earth's rotation is irregular, on the order of msec per day.

http://maia.usno.navy.mil/plot-eop.html

shows multiple plots comparing the observed irregularities compared with physical models for the tidal and atmospheric forces believed to contribute to these irregularities.

http://tycho.usno.navy.mil/systime.html

Has the basic length-of-day variation.

There is no reason to believe that the Earth rotation is particularly affected by a change in the free-structure constant, in a way that would not also affect atomic clocks. The idea that some basic change to physical constants is giving us leap seconds is pretty much nonsense.
Re:Highlights problem with ntp... by OOGG_THE_CAVEMAN · 2006-01-03 04:32 · Score: 1

Did you actually read the article that you linked to, you numbskull? Let me quote for you

To see why the confidence in the invariance of the speed of light is so high, we need to look at the history of its measurement,
Re:Highlights problem with ntp... by OOGG_THE_CAVEMAN · 2006-01-03 04:37 · Score: 1

Do you have any link to information showing that phenomenon for speed of light measurements?

I have only heard of such a thing for Millikan's charge-on-the-electron measurements.
Re:Highlights problem with ntp... by Floody · 2006-01-03 06:41 · Score: 1

Well, Linux somehow knows about it anyway. I assumed it came from NTP. From my logs:

Jan 1 12:59:59 sasha kernel: [4616786.935000] Clock: inserting leap second 23:59:60 UTC

(I'm currently in a +1300 timezone, hence it happening at 1pm)

Both myself and the OP were at least partially wrong w.r.t. NTP & leap-seconds. NTP actually does have a mechanism for transmitting a leap-second indicator, however it's not part of the ntp timecode. Each individual ntpd must pass the leap-second announcement along to its clients, however local host processing is platform-dependant (not all systems are capable of exact leap-second adjustments). There have been (at least two) different PLL implementations in the linux kernel; any relatively new kernel + ntpd is, obviously from your log, capable of atomically adjusting by one second so that the host's cpu clock never varies from the ntp code (which is sync'd to utc no matter what). Basically, the utc/ntp "time standing still for one second" event happened simultaneously with the receipt of the leap-second notification and ntpd told your kernel to make the correct adjustment. One the next timecode interval, the clock difference would then be no different than standard cpu drift that ntpd is already accounting for.

In this way, a scientific application, if endowed with the correct kernel interface, could detect leap-seconds and account for them for any intervals that may span the inserted/removed second. Thus, any remaining issues are solely the responsibility of application developers. ;)
Re:Highlights problem with ntp... by njh · 2006-01-03 10:38 · Score: 1

No sorry, I was paraphrasing. Perhaps I'm thinking of the charge of an electron instead.

There's a good write up on http://en.wikipedia.org/wiki/Variable_speed_of_lig ht
Re:Highlights problem with ntp... by njh · 2006-01-03 10:40 · Score: 1

There is no reason to believe that the Earth rotation is particularly affected by a change in the free-structure constant, in a way that would not also affect atomic clocks. The idea that some basic change to physical constants is giving us leap seconds is pretty much nonsense.

Oh good, I thought my world view was wrong for a bit. This makes much more sense, thanks.

Re:happy new year by kfg · 2006-01-01 04:03 · Score: 1

let's wish for 365 more days of everything.

If you had everything, where would you keep it? -Stephen Wright

Actually, that's an easy one:

Everywhere.

KFG

Re:happy new year by SpinJaunt · 2006-01-01 04:03 · Score: 4, Funny

More along the lines of 730 days if you include the dupes..

--
/. is good for you.

WOOHOO! by spineboy · 2006-01-01 04:03 · Score: 1

Yeah! - Mandatory overtime - I get time and a half! - Oh wait.......nevermind

--
..........FULL STOP.

Re:WOOHOO! by House+of+Usher · 2006-01-01 06:44 · Score: 1

Just think of the interest that you could make on this :-)

--
I am but mad north-north-west: when the wind is southerly I know a hawk from a handsaw.

time.gov by srblackbird · 2006-01-01 04:04 · Score: 5, Informative

I watched the time at Time.gov: 23:59:56 (UTC) =>23:59:57=>23:59:58=>23:59:59=>23:59:60!=>00:00:0 0
It was Amazing! This was the first time for me... *remebers where I was at that moment

--
"The test of the morality of a society is what it does for it's children." -Dietrich Bonhoeffer

Re:time.gov by Krimszon · 2006-01-01 04:08 · Score: 1

Do you have a replay?
Re:time.gov by 26242 · 2006-01-01 04:54 · Score: 0

That would have been pretty cool to see. Awesome pawsum!
Re:time.gov by Rytsarsky · 2006-01-01 05:29 · Score: 4, Informative

Screenshot: http://flickr.com/photos/jk3us/79978824/

--
God became man to enable men to become sons of God. -C.S. Lewis
Re:time.gov by Anonymous Coward · 2006-01-01 11:10 · Score: 0

*remebers where I was at that moment
You mean sitting in front of your computer watching the time tick over right?

Damn, and I was wasting my time doing silly things like getting drunk and yelling "Happy New Year!"

Re:Happy New Year by Shimmer · 2006-01-01 04:05 · Score: 3, Informative

"Happy New Year's" is short for "Happy New Year's Day".

--
The most rabid believers in American Exceptionalism are the exact same people whose policies are destroying it.

Re:Happy New Year by dawhippersnapper · 2006-01-01 04:05 · Score: 0, Redundant

Day is left out. Happy New Year's Day! ......

--
Freedom is fragile and must be protected. To sacrifice it, even as a temporary measure, is to betray it.

How did you spend the extra second? by House+of+Usher · 2006-01-01 04:06 · Score: 2, Funny

So during the correction of the clocks and the extra second being added, what did you do? Did you ponder world peace? The latest 0 day exploits for Windows? Where Microsoft is going with the .NET platform with version 2.0? Or were most of you transfixed on Times Square watching the ball drop getting close to someone you love?

--
I am but mad north-north-west: when the wind is southerly I know a hawk from a handsaw.

Re:How did you spend the extra second? by Anonymous Coward · 2006-01-01 04:15 · Score: 0

Adjust my clock. Duh.
Re:How did you spend the extra second? by jzeejunk · 2006-01-01 04:24 · Score: 1

Yeah ! I watched the ball drop and I was close to my *coughs* *coughs again* laptop, reading slashdot - my one true love. You make me sick!!

--
sarchasm
Re:How did you spend the extra second? by Anonymous Coward · 2006-01-01 04:52 · Score: 1, Funny

I personally re-enacted every Italian military victory.
Re:How did you spend the extra second? by Ruff_ilb · 2006-01-01 05:49 · Score: 1

I was submitting this story to /. =P

--
http://www.TheGamerNation.com/Forums
Re:How did you spend the extra second? by Random832 · 2006-01-01 15:14 · Score: 1

Given that the ball wasn't set to drop for another five hours [and one second], that hardly makes sense.

--
We've secretly replaced Slashdot with new Folgers Crystals - let's see if it notices.

old news by Viriatus · 2006-01-01 04:08 · Score: 5, Funny

Old news, from last year.

Re:Happy New Year by Paradise+Pete · 2006-01-01 04:08 · Score: 1

I think it comes from a mix of happy new year and new year's day. Kinda like people saying "realator" from mis-seperating the sounds in "real estate" to "reala-state."

Re:Happy New Year by User+956 · 2006-01-01 04:08 · Score: 1

It's, "Happy New Year." Why does nobody understand that it's only one new year? It's not "New Years" or "New Year's" but "New Year."

I think it's a contraction of "Happy New Year's Day" with incorrect punctuation. This being Slashdot, incorrect punctuation shouldn't surprise you.

Either that, or it's a conflation of "New Year's Eve" and "Happy New Year." Or, Taco's drunk, dictating articles, and he's slurring his speech. There's a veritable menagerie of possibilities; a smorgasbord of likely explanations, if you will.

--
The theory of relativity doesn't work right in Arkansas.

oh, shit. by CAIMLAS · 2006-01-01 04:08 · Score: 1, Funny

Oh, shit. I've got to reset my goddamn clock again.

--
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers

Re:Happy New Year by gilroy · 2006-01-01 04:09 · Score: 1

Blockquoth the poster:

Why does nobody understand that it's only one new year?

What, after 2006 the calendar just ends? Of course there'll be many more New Years, and the original poster just wanted to pro-actively wish that all of them are Happy.

:)

--
The Mongrel Dogs Who Teach

Re:Happy New Year by Paradise+Pete · 2006-01-01 04:10 · Score: 1

and of course when I say seperate, I mean separate ;-)

a question by yincrash · 2006-01-01 04:13 · Score: 1

since it changed for UTC time, does that mean on the east coast, it was 7:00:60PM? CRAZY

Re:a question by Anonymous Coward · 2006-01-01 04:42 · Score: 0

Actually 6:59:60 PM EST, but yes, that's essentially correct.
Re:a question by denelson83 · 2006-01-01 13:10 · Score: 1

No, it was 6:59:60 PM.

Leaplog by davydmadeley · 2006-01-01 04:15 · Score: 0, Offtopic

Jan 1 07:59:59 oracle kernel: Clock: inserting leap second 23:59:60 UTC

Re:Leaplog by Anonymous Coward · 2006-01-01 04:43 · Score: 0

Dec 31 00:59:29 j1.sjc..net xntpd[86356]: time reset 0.999834 s
Re:Leaplog by dotgain · 2006-01-01 11:11 · Score: 1

Jan 1 00:00:01 lp0 on fire!

mnb Re:Highlights problem with ntp... by Anonymous Coward · 2006-01-01 04:16 · Score: 0

That is why real men use GPS for time.

BBC TV yesterday by takev · 2006-01-01 04:20 · Score: 1

Did they actually cheer a second to early.
Or did the BBC compensate for the Leap second.
How would you show a leap second on an analogue clock that they used.

Re:BBC TV yesterday by Anonymous Coward · 2006-01-01 04:35 · Score: 2, Informative

On BBC 1, the clock was counting down and went 5-4-3-2-1-1-0. The people in the streets were just waiting for the bell of Big Ben to strike midnight -- there's no second hand.

It's not the 'ephemeris second' that's the problem by Jonathan+McDowell · 2006-01-01 04:25 · Score: 5, Informative

Original poster is slightly wrong - it's not the length of the 1900 ephemeris second,
it's the fact that the Earth, like all of us, is getting older and slowing down, so that
the 2005 "Earth rotation" second (i.e. 1/86400 of one spin of the Earth) is longer than
the 1900 equivalent and longer than the atomic time (SI) second. Instead of changing
the length of the second, it is currently deemed less painful to keep using the old
length and stick in an extra second every now and again.

Since this depends on the slop of the Earth's interior, it's not a fully regular and predictable thing - we might even have to remove a second one year.

Great by the+eric+conspiracy · 2006-01-01 04:31 · Score: 4, Funny

Gotta love those long weekends!

Re:Happy New Year by xenotrout · 2006-01-01 04:31 · Score: 2, Informative

REALTOR® is actually a registered trademark, which seems much like "realator" as you say. Although it does sound like it could be an "official" contraction, it's made up.
If you care to know, someone decided, it would seem, to make up a word in order to create "de facto regulation" of an industry--that is, anyone can call themselves real estate agents, but only those who get trademark license (possibly by passing a test on how well they understand real estate agency or REALTY) can legally call themselves REALTORs. And people are "supposed" to prefer a REALTOR over a real estate agent (at least that's the hope of the company owning the name REALTOR--I don't know whether it works or not).

Re:Happy New Year by Anonymous Coward · 2006-01-01 04:37 · Score: 0

"New Year's" is a contraction. It is short for "New Year's Grammatical Rant"

Working by isorox · 2006-01-01 04:42 · Score: 0, Redundant

Hey, I was at work at midnight, I demand 1 second's overtime!

Re:Working by FLEB · 2006-01-01 09:27 · Score: 1

Yeah, but you probably get paid by the hour.

--
Information wants to be free.
Entertainment wants to be paid.
You just want to be cheap.

Chuck Norris by Anonymous Coward · 2006-01-01 04:43 · Score: 0

The only reason we have a leap second isn't because of the malarkey that they presented. We have a leap second because Chuck Norris roundhouse kicked someone so hard in the face that it slowed the earth's rotation.

Re:Happy New Year by fimbulvetr · 2006-01-01 04:43 · Score: 1

If you look at the etymology of the word, you can get a good idea as to why it's spelled with an a instead of an e. The mere knowledge of where the word comes from will serve as an excellent mnemonic device for spelling it correctly when the times come. I myself suffered from the same condition up until about 2 years ago.

January 0 by thomasdelbert · 2006-01-01 04:45 · Score: 1

There was a January 0 in year 1900? Hmm found no mention of it here.

- Thomas;

--
___ This sig is in boldface to emphasize its importance!

Re:January 0 by at10u8 · 2006-01-01 05:05 · Score: 1

Have a look at an old ephemeris and see the large numbers of tabulations of quantities with respect to date. For the sake of ease of typesetting it was commonplace to have tables with dates such as January 0, 5, 10, 15, 20, 25, 30, and 35. That avoided the need to redundantly set more lead into the matrix for the name of the month o nevery single line of the tables. In the case of tables being interpreted by humans there was no expectation of raising some sort of input exception because in the full context the meaning was pretty obvious.

Expected Result by Anonymous Coward · 2006-01-01 04:46 · Score: 0

Hmm...on my Windows box, the following code...

wscript.echo (cdate("1/1/2006 11:58:59") - cdate("12/31/2005 11:58:59")) * (24 * 60 * 60)

...yields the expected result of 86400 (Funny what I've come to expect, having used MS products.)

This noon-ish to noon-ish period should propery wrap whether or not MS conversions take into account your timezone (mine is MST,-0700) or not (at least in parts of the world, like the continental USA).

How did your systems behave?

Re:Expected Result by bar-agent · 2006-01-01 06:26 · Score: 1

How did your systems behave?

Pretty much the same on my Mac:

> osascript -e 'date "Jan 1, 2006 11:58:59 am" - date "Dec 31, 2005 11:58:59 am"' 86400

--
i'd hit it so hard, if you pulled me out you'd be the king of britain [bash.org]
Re:Expected Result by dotgain · 2006-01-01 11:51 · Score: 1

A:\ is not accessable. [Retry][Cancel]

11 hours late, 1 second short by hostingreviews · 2006-01-01 04:47 · Score: 1

Thanks Taco. We could have used that extra second when it happened - 11 hours ago. 3... 2... 1... 1...

--
Why would you trust a testimonial when choosing hosting?

The clock problems by Antony-Kyre · 2006-01-01 04:48 · Score: 3, Interesting

Did anyone notice the atomic clock problems that happened when the leap second occurred? Some atomic clocks were different than others. If I am not mistaken, and I don't believe I am, the leap second occurred at 23:59:60 UTC (yes, I typed that in correctly). I also flipped back and forth between like ABC and NBC, Pacific Time, and notice they were like 3 seconds or so different in their countdown clocks. What is up with that?

Re:The clock problems by DAldredge · 2006-01-01 05:11 · Score: 2, Informative

That may have been caused by different lenghts of time delay. not all networks use the same amount of time delay on their live feeds.
Re:The clock problems by Antony-Kyre · 2006-01-01 05:14 · Score: 1

And I thought atomic clock servers were suppose to be accurate. I guess I was wrong. I manually used a program to check my computer time against atomic clock servers. Some servers didn't recognize the leap second, and I really thought they were suppose to be accurate.
Re:The clock problems by DAldredge · 2006-01-01 05:22 · Score: 1

Their are different ways to measure time. Not all systems recognize the leap second.
Re:The clock problems by gfilion · 2006-01-01 05:35 · Score: 1

I also flipped back and forth between like ABC and NBC, Pacific Time, and notice they were like 3 seconds or so different in their countdown clocks. What is up with that?

I was watching CNN and they had a couple seconds difference between their on screen countdown and the Times Square Coke-sign countdown. It kind of scrapped the fun of counting down to 2006, CNN was allready at 0 when Times Square had 4 seconds left.

Yeah, I'm a dork, having my new year's fun scrapped by a 4 seconds discrepancy.
Re:The clock problems by Hillman · 2006-01-01 05:38 · Score: 2, Informative

Some networks use a time delay to filter out profanities. Also it might be because the feed you're watching goes thru a sat link before reaching your affiliate.
Re:The clock problems by PPGMD · 2006-01-01 07:03 · Score: 3, Informative

Sat delays along with buffer so they can dump profanity can build up the time difference.
Last year during the Superbowl it was noticeable at my house, I had 3 TVs tuned to the game, 2 via DirecTV and another using rabbit ears, the over the air broadcast was easily 2-3 seconds ahead of the DTV broadcast. This is one of the reasons that the Sport Betting Houses that allow betting up until play completion don't allow cellphones, because you could have someone watching via a faster source or at the game itself, feeding you what's going to happen.
Re:The clock problems by solitas · 2006-01-01 07:20 · Score: 1

Some networks use a time delay to filter out profanities.
Yah - and they missed one: http://thepoliticalteen.com/video/looselips.wmv

--
"It's time to take life by the cans." ~ Bender ("Bendin' in the Wind", ep. 3-13)
Re:The clock problems by isorox · 2006-01-01 07:45 · Score: 2, Informative

We (A national UK broadcaster) had a countdown clock (2 identical computers). It was linked from the corp's central ntp server (in turn via GPS), but generally it's a PC clock. Normal broadcast equipment runs off a time code generator that's slaved to GPS.

It wasn't far off the pips on radio 4 analog (ignore big ben etc) in the gallery.

After leaving the gallery, it travels down to one CTA, then another, then another, then gets split, and is sent via seperate routes to the main transmitter in Crystal Palace. Other regions outside London recieve their feeds via various routes.

DTT is sent seperatly, and the signal is encoded (with major delay) as a low bitrate (to the local transmitter), and a high bitrate (to nations/regions) which can be opted out of (I'm a little hazy on this bit).

Additonally DSat is encoded then uplinked to a satelite.

As a consumer, if you recieve digital there will be an additional delay in decoding.

The point is that the time might be fine in the studio, however it goes through a lot of equipment that adds a lot of delay, and different networks will have different delays, different equipment etc.

Now how do you line your studio up? Based on the studio output, the analog reception from a nearby transmitter, the analog reception from a 2nd or 3rd level slave, the digital reception from one of many different transmitters (each with different equipment encoding with a different delay), from satelite (with the 3/4 second bounce plus digital encoding), and how do you account for various decoders?

That's the problem for a normal new year broadcast. Add in a leap second (99% of equipment out there will not accept 23:59:60 as a valid time) and you've got even more problems.

In the UK the most accurate broadcast signal you'll get is the pips on Radio 4 LW.

If you want to know the real time, get a GPS system.
Re:The clock problems by ottffssent · 2006-01-01 09:16 · Score: 1

I also flipped back and forth between like ABC and NBC, Pacific Time, and notice they were like 3 seconds or so different in their countdown clocks. What is up with that?

Differing amounts of broadcast time delay?

--
High-speed Road Trip (18.000KPH)
Re:The clock problems by Anonymous Coward · 2006-01-01 12:02 · Score: 0

Think before you post. Any system that doesn't recognize the leap second is off from UTC by a very noticeable amount, something like 14-33 seconds, due to the accumulated leap seconds since the system was defined.
Re:The clock problems by Ambient+Sheep · 2006-01-01 12:12 · Score: 1

Any particular reason to choose Radio 4 LW over Radio 4 FM? The delay in adding RDS? Thanks in advance.
Re:The clock problems by dracocat · 2006-01-01 13:36 · Score: 1

Good luck trying to speak and place the bet within 3 seconds. Also, are you going to change your bet based off how far the ball is kicked off?

In any case, the reason you can't use cell phones is so that you can't take bets for other people over the phone and relay them.
Re:The clock problems by petermgreen · 2006-01-01 13:52 · Score: 1

1: 30 seconds off is not "very noticable" i doubt many pc clocks that aren't on a network sync system are set much more accurately than that.

2: any system that uses a time server will have had the leap seconds forced on it even if it didn't understand them like standard unix time doesn't (causing discontinuities in system time and similar nasties)

--
note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
Re:The clock problems by PPGMD · 2006-01-01 15:57 · Score: 1

3 seconds, not really, lets say you have someone on site, you have 1 second delay from location to central facility, 7 seconds for content filtering, 1 more second to get it to the stations or to the consumer if you are accessing the feed from the bird directly. If you are access via a station, you have another second for them to receive and broadcast it, or 2-3 seconds if it's going back up to DirecTV bird and to the facility. I figure that 10 seconds would be close to average, even if you access to feed off the satellite.
Re:The clock problems by Anonymous Coward · 2006-01-02 02:03 · Score: 0

I noticed this once when working in a sports equipment store where we had several TV's showing the same game. If you stood in a place where you could see two TV's at once, the one TV seemed to have several seconds of delay compared to the other, even though they were tuned into the same station.

I was quite perplexed by how this could happen. It occured to me that evening what was happening, the one TV was tuned into one broadcast tower, and the other to a different tower, probably hundreds of kilometres apart. The next day I adjusted the tuning on the one TV, and they were in sync again.

I have also noticed this on my home TV when I switch between the same station broadcast normally and also by satelite.
Re:The clock problems by Anonymous Coward · 2006-01-02 16:11 · Score: 0

Sorry about using the Anonymous Coward thing, as I just don't feel like creating a login at this time.

The reason you can see a difference in the time between the various networks is not just related to satellite delivery, but due to the digital MPEG-2 encoders. NBC's SD feed is MPEG-2 (digital) and the ABC SD feed is MPEG-2 (digital). The process of MPEG-2 encoding causes delays. The nets try and compensate, but don't always get it right. Hence the, I believe, difference that you saw. I'm currently in a unique situation where I can see the ABC east coast feed (which is still analog, as well as digital) and can compare all three (the third is the HDTV video). My local ABC affiliate is using the new digital sat feed.

So, I wouldn't make too much of the difference. Back when all network feeds were analog, the clocks were extremely near each other. Not anymore.

Mr Video
Re:The clock problems by Antony-Kyre · 2006-01-02 16:31 · Score: 1

That really takes the fun out of New Year's.
Re:The clock problems by isorox · 2006-01-05 08:56 · Score: 1

Any particular reason to choose Radio 4 LW over Radio 4 FM? The delay in adding RDS? Thanks in advance.

Nope, I doubt that the delay is less than the delay you'll get based on distance from the transmitter, however I don't really know much about that as the corporation I work for doesn't broadcast any more, we hire someone called "Red Bee Media", who send the signals to tranmitters owned by "Crown Castle". Only broadcast I've learnt about is digital (mainly DTT and DSat), and even then it's not a lot.

Re:It's not the 'ephemeris second' that's the prob by product+byproduct · 2006-01-01 04:55 · Score: 1

It is worth noting that those "leap seconds" amuse only some people. People who work on systems that can't afford a 1 second discontinuity (such as the GPS system) use a continuous counting of the SI second.

http://www.leapsecond.com/java/gpsclock.htm

Re:Happy New Year by Paradise+Pete · 2006-01-01 04:57 · Score: 1

REALTOR® is actually a registered trademark

Realtor has been a plain ol' word for many years. Do you have a reference?

Re:It's not the 'ephemeris second' that's the prob by at10u8 · 2006-01-01 04:58 · Score: 1

This is very true.

To within about one part in 1.E12 the ephemeris second is identical to the SI second defined by the cesium resonance. In 1977 the length of the second of TAI was changed so as to conform better with the preferred definition of the SI second. Before 1977 TAI and UTC ticked faster than they do now. Astronomers did not object to the change in rate of TAI because it was within the uncertainty of the original definition of the ephemeris second.

Celebrate with a musical ode to the leap second: by gonzo-wireless · 2006-01-01 04:59 · Score: 0

Create Digital Music want you to celebrate this second. Here's the URL, looks like a good project:

http://createdigitalmusic.com/index.php?option=com _content&task=view&id=1075&Itemid=44

Earth's rotation isn't constant by Anonymous Coward · 2006-01-01 05:00 · Score: 0

Interesting bit of trivia for you... Earth's rotation is slowing down. When you think about it, the moon exerting a gravitational force against the earth leaves friction between the water and the earth's bottom. So very slowly, the earth's rotation will become that of the moon (where one side always faces the moon) so that there will be no more slowing. When this happens, a day will be around 300 hours. Of course, when that happens, we'll probably also be consumed by the expanding sun.

A link for the curious: http://www.jb.man.ac.uk/distance/strobel/gravappl/ gravapplb.htm

And... by manavendra · 2006-01-01 05:01 · Score: 1

How many seconds were wasted talking and discussing that extra second? Domino effect, anyone?

--
http://efil.blogspot.com/

That's So, Like, 2005, eh by pipingguy · 2006-01-01 05:03 · Score: 1

Rob, we all saw that coming a few days ago. But /now/ it's important to *you*. If you'd just stop with the babymaking attempts, playing with your new Legos and ignoring dupes you could start working on the reply for the 1,000,000 Slashdot account.

Re:happy new year by FlopEJoe · 2006-01-01 05:09 · Score: 0, Redundant

happy new year, slashdot. let's wish for 365 more days of everything.

w00t!! First dupe of 2006!!!

Not everywhere on earth 2005 was a second longer. by Anonymous Coward · 2006-01-01 05:09 · Score: 0

If you live "east" from England, 2006 actually is already a second longer than it wouuld have been if there was no leap second. The leap second happened after new year's eve, and so for us 2006 will have one second more. Americans of course get to have a longer 2005, and since Americans don't know there is something outside their country, I guess the title is fitting for a slashdot article.

Being born on a leap year might sound bad... by RavenChild · 2006-01-01 05:16 · Score: 0

Just imagine being born on a leap second.

Leap Second Lovers Are Traitors Says Bill O'Reilly by Fishstick · 2006-01-01 05:17 · Score: 4, Funny

"This year's leap second is an assault on the American public," says commentator Bill O'Reilly. "The reason the leap second is even being proposed is because of America Haters, because of Iraqi hate mongers, and let's be honest, Shiites. Why would you add a second to the year unless you're an anti-American hate monger?
I remember liberals at a party saying, 'let's add a second to the year' and I was the only one who spoke up against it. Why would they want to add a second to the year? Because it gives them a second longer to hate Bush.

"Look, look, look, look. A leap second is a denial of everything American, of everything good, of everything moral. They're saying we need this second because the earth rotates on its axis and revolves around the earth, well this is the no spin zone. So we don't need a leap second. Though I would rather have a leap second than some of these hate-mongers who go around hating even their own ideas! They need to hate their own ideas so much that you have many liberals proposing the leap second, which is an idea that they hate, yet, they propose.

"I am so so so so upset with these people, who actually believe their ideas, yet, I have no hate in my heart. I am a simple guy, who only has my own true beliefs and a few products that are my cornerstone to fight against the leap second poobah. Let me say it aloud: Leap Second, leap second, leap second. Doesn't it sound ugly?

"Please, don't let these Darwinian leap-seconders, who believe that the planets revolve around the sun, who believe that rocks are sedimentary, igneous and stalactites, who are innocent dim-wit believers in a faith bordering on hating everything religious like trees and fruitcake, yet, who don't believe in John 7:12:45:67:89, have their say.

"But you know what I love? Dialogue. Rational dialogue which allows me to say that aliens from a Iraqi loving planet want to abolish Christmas by adding a leap second to the Darwinian anti-God year. Dialogue is what keeps the American system God-loving and anti non-God. It also keeps the anti-God loving non-Iraqi loving insurgent deniers able to voice their hideous so-called opinions over the American loving tolerant airways. And now let's take some calls."

Steve Martin

--

There is much cruelty in the universe, John.
Yeah, we seem to have the tour map.

Missed it? by david.emery · 2006-01-01 05:20 · Score: 1

I was expecting the leap second to show up in the big Dick Clark Times Square dropping ball (5..4...3...2...1...1...Happy New Years) but I must have blinked and missed it...

dave

Re:Missed it? by imthesponge · 2006-01-01 05:37 · Score: 2, Informative

It's inserted at UTC 23:59:59, which would be 6:59:59 PM in New York.
Re:Missed it? by david.emery · 2006-01-01 06:00 · Score: 1

Well, I guess that makes sense... Still I was hoping to see the ball stutter :-)

dave
Re:Missed it? by Anonymous Coward · 2006-01-01 06:55 · Score: 0

So that's why Dick Clark was talking so slow.

nerdiest. explanation. ever. by Anonymous Coward · 2006-01-01 05:20 · Score: 0

Super gay.

tsunami caused rotation change by zontroll · 2006-01-01 05:20 · Score: 2, Interesting

Are they also adjusting the clock because the Dec 2004 earthquake and tsunami caused a small change in the rotation speed of the earth? I believe it caused the rotation to last a few miliseconds longer for that day. There some info at the NOVA site on pbs.org:
http://www.pbs.org/wgbh/nova/tsunami/ask-050331.ht ml
(scroll down to the sixth question)

Re:tsunami caused rotation change by zontroll · 2006-01-01 05:22 · Score: 1

whoops, wrong URL. Here's the correct one:
http://www.pbs.org/wgbh/nova/tsunami/ask-050402.ht ml

GPS not synchronized? by Bloody+Templar · 2006-01-01 05:22 · Score: 1

From TFA: "The Global Positioning System (GPS) epoch is January 6, 1980 and is synchronized to UTC. GPS is NOT adjusted for leap seconds."

Huh? Why? Wouldn't this eventually make GPS so far out of whack with UTC so as to be useless for time (admittedly not its primary purpose, but a useful one nonetheless)?

Re:GPS not synchronized? by tugrul · 2006-01-01 05:46 · Score: 2, Informative

The relationship between GPS and UTC is broadcast from the constellation in the navigation data stream that also transmits time stamps, satellite ephemerises, a variety of correction parameters and other stuff I don't understand/remember :)
Re:GPS not synchronized? by MDMurphy · 2006-01-01 06:13 · Score: 3, Informative

GPS time just counts intervals, and it started the count in weeks, days, seconds in January, 1980. The system is aware if UTC though, and one of the various messages sent from the satellites includes the UTC offset. So if you receive and decode that message you'd know the UTC time.

As of yesterday, the difference between UTC and GPS time is 14 seconds.
http://maia.usno.navy.mil/ser7/series14.txt
Re:GPS not synchronized? by rbrewer123 · 2006-01-01 08:27 · Score: 1

As other replies have stated, the GPS satellites also broadcast the difference between GPS and UTC time. It even has a flag in there that warns of an impending leap second up to a week in advance. Actually, the nice thing about running a system from GPS time is that you don't have to worry about leap seconds.

Re:Happy New Year by AddressException · 2006-01-01 05:23 · Score: 1

Realtor has been a plain ol' word for many years. Do you have a reference?

That's what these people want you to think in order to get people to use their services preferentially! It's most certainly not a plain ol' word!

http://tess2.uspto.gov/bin/showfield?f=doc&state=s 15t3v.4.115

Steve Martin says... by Cuthbert+Calculus · 2006-01-01 05:24 · Score: 0, Troll

...that Bill O'Reilly says that Leap Second Lovers are Traitors!

Re:It's not the 'ephemeris second' that's the prob by Jonathan+McDowell · 2006-01-01 05:25 · Score: 2, Insightful

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.

Why bother? by sgent · 2006-01-01 05:27 · Score: 1

Why not just do a "leap hour" every six millenia?

There is no real reason other than convienance to even add it in the first place. Yes, having a day based on the earth rotation makes sense. Even the minute/second /hour timeframe makes sense (froma an circumnavigational standpoint).

But to be honest, this just seems over the top. Who cares if "earth time" is off by 1 sec / yr from "atomic time". They are both relatively arbitrary units anyway as long as our sense of day/night doesn't get screwed up.

Re:Why bother? by Detritus · 2006-01-01 06:39 · Score: 1

Who cares if "earth time" is off by 1 sec / yr from "atomic time".
Anyone who is doing navigation based on the current time (UTC) and the positions of the Sun and other celestial bodies. You get the longitude from the difference between local time and UTC.

--
Mea navis aericumbens anguillis abundat
Re:Why bother? by NixieBunny · 2006-01-01 08:13 · Score: 2, Informative

The folks who are into time are now realizing that the leap second was a sorta bad idea, since it is inserted with little advance notice and is not deterministic.The leap hour idea is being considered and would indeed solve the problem.
The big argument about leap seconds is that the computer folks want to eliminate them, but the astronomers are upset because the useful bit of information that tells how many seconds TAI is off from UTC is transmitted in a field that can only hold a number up to +/- one second in some time transmission protocols. So after a missed leap second, asatronomical time wil lbe off from UTC by an unspecified amount, which could be bad for anyone who uses UTC time to do astronomy. (Since UTC is the time that's transmitted around the world, this is important.)

--
The determined Real Programmer can write Fortran programs in any language.

Words by Icyfire0573 · 2006-01-01 05:32 · Score: 1

Love the verbiage

Log of Atomic GPS Clock adjustment by BiggRanger · 2006-01-01 05:34 · Score: 5, Interesting

I did up a project on sourceforge.net a few years back to sync my computers with a GPS http://atomicgpsclock.sourceforge.net/. Below is a log of the activity, normally there is a +/- 0.016 or so second instability, but 18:59:59 EST (or 23:59:59 UTC) the Navy made a 1 second adjustment to the GPS system, and it's vibible in the log at the next scheduled sync (in bold)

2005.12.31 18:33:49 00032 GPS Status - Tracking: 3D
2005.12.31 18:43:27 00020 Offset: 000.016 Buffer: 13
2005.12.31 18:43:27 00032 GPS Status - Tracking: 3D
2005.12.31 18:43:49 00020 Offset: -000.031 Buffer: 13
2005.12.31 18:43:49 00032 GPS Status - Tracking: 3D
2005.12.31 18:45:15 00033 GPS Status - Tracking: No
2005.12.31 18:45:34 00032 GPS Status - Tracking: 1D
2005.12.31 18:46:48 00033 GPS Status - Tracking: No
2005.12.31 18:46:52 00032 GPS Status - Tracking: 3D
2005.12.31 19:01:43 00033 GPS Status - Tracking: No
2005.12.31 19:01:55 00032 GPS Status - Tracking: 1D
2005.12.31 19:03:45 00020 Offset: 001.016 Buffer: 13
2005.12.31 19:03:45 00032 GPS Status - Tracking: 2D
2005.12.31 19:13:45 00020 Offset: -000.016 Buffer: 13
2005.12.31 19:13:45 00032 GPS Status - Tracking: 3D
2005.12.31 19:23:43 00020 Offset: 000.000 Buffer: 13
2005.12.31 19:23:43 00032 GPS Status - Tracking: 3D
2005.12.31 19:33:43 00020 Offset: 000.000 Buffer: 13
2005.12.31 19:33:43 00032 GPS Status - Tracking: 3D
2005.12.31 19:43:30 00033 GPS Status - Tracking: No
2005.12.31 19:43:40 00032 GPS Status - Tracking: 1D
2005.12.31 19:53:41 00020 Offset: -000.031 Buffer: 13
2005.12.31 19:53:41 00032 GPS Status - Tracking: 3D
2005.12.31 20:03:39 00020 Offset: 000.000 Buffer: 13
2005.12.31 20:03:39 00032 GPS Status - Tracking: 3D

Re:Log of Atomic GPS Clock adjustment by Anonymous Coward · 2006-01-01 11:52 · Score: 0

Here's another data point: http://lists.freebsd.org/pipermail/freebsd-current /2006-January/059504.html
There are long discussions on FreeBSD's lists about the complexities of handling the leap seconds.
Re:Log of Atomic GPS Clock adjustment by Anonymous Coward · 2006-01-01 12:32 · Score: 0

Are you sure you're not looking at UTC instead of GPS time? GPS time has never had a leap second adjustment before. Doing one now would cause a huge number of problems. The sats do transmit the current number leap second, so some programs make the correction and display UTC. I expect that's what you are seeing.

times square crowd 1 sec off tv time by drDugan · 2006-01-01 05:54 · Score: 1

I watched the time square ball drop from the west coast last night on TV -- nbc.

of course, they printed "live" on the screen which was bs -- but did anyone else notice that the digital time on the screen and the seconds that the crowd were chanting down were 1 second off? obviously, I had way too little to drink this year to notice this, and no girl nearby to kiss.

I'm wondering if maybe this was due to some clocks on 2005 time and some with more accurate accounting?

Re:times square crowd 1 sec off tv time by iantri · 2006-01-01 06:43 · Score: 1

satellite transmission delays.. it's always this way. times square -> satellite -> studio/control -> satellite -> cable -> you (or in the case of direct-to-home satellite, it goes satellite -> dish/directv/expressvu/starchoice -> satellite -> you) Each of those satellite uplinks is gonna add another second of delay. So The studio is going to be getting the feed from times square a second after it happens, but it's a few more seconds later by the time you see it due to the transmission delays.
Re:times square crowd 1 sec off tv time by lowid+(24)+_________ · 2006-01-01 09:57 · Score: 0

From what I could tell, it was due to dick clark actually being dead and operated by puppeteers.

(They probably had a hard time seeing the official countdown, what with being behind the desk and all)

--
free music

Nope, this year. by MDMurphy · 2006-01-01 05:56 · Score: 1

Since the leap second was added at midnight UTC, it was already noon back on the international dateline. So the adjustment was 1/2 way through the first day of the year.

not exactly true.. by Anonymous Coward · 2006-01-01 06:14 · Score: 0

If you sync your computer to a UTC time source with NTP (the case for most of us, I'd imagine, since POSIX support for TAI and leap seconds is generally shitty.. djb's clockspeed program is designed for TAI, but I'm not sure what TAI source he syncs to!), your clock will only be wrong from the leap second until the next NTP sync. So for 99.9999% of computer users, they will wake up on Jan 1 with a correctly set clock.

If you are doing calculations with the times, you need to take leap seconds into account, of course. This isn't NTP's problem (this is a flaw that arises when storing times in UTC rather than TAI). NTP just transfers the time between computers. Of course if you use the Unix timestamps, which increase by 1 every second, you won't have a problem.

Here's an analogy: Imagine if you computer didn't have support for daylight saving time.. things would be weird every time the clocks jumped forward or back.. the computer's clock would chug along normally, completely missing the change, then NTP would eventually say "hey, you're off by an hour".. and the clock would be set to the correct time. If you did calculations that spanned the instant that the clocks changed, you'd have to adjust your results by an hour.. but again, if you used a monotonic timer like the Unix timestamp, you'd still be fine.

No problem in Times Square. by SheeEttin · 2006-01-01 06:22 · Score: 0

People seem to think that the leap second would have affected the countdown in Times Square. Actually, it wouldn't. Over here, the time is -0400. This means that if it was added at 11:59:60, it would appear at 6:59:60. So, unless they started the countdown five hours early, it would go exactly as planned. (And it did.)

last year the US proposed to cancel this update by Herve5 · 2006-01-01 06:25 · Score: 3, Interesting

I know a person in charge of geostationary satellite control, and she says this time adjustment will have imposed a large amount of satellite and ground station software updates.
She added that because of this among many other updates, there have been a formal proposal by the US, some months ago, to change the rules and abandon any updating before there is a full day (!) of delay, but the proposal was refused.

FYI, this 1-s correction is the first in 5 years, but there were 4 others in the previous 5 years.

Waiting for one day would basically mean renouncing for some thousand years, or more probably, waiting for the next civilization to come :-)

Hervé

--
Herve S.

Re:last year the US proposed to cancel this update by denelson83 · 2006-01-01 13:15 · Score: 1

The previous leap second was 23:59:60 UTC on Dec 31, 1998. So this leap second is the first in 7 years, not 5.

So what about 'seconds since the epoch' by meldir · 2006-01-01 06:46 · Score: 1

So, when can we expect an update of localtime()?

After all, the epoch will no longer end at 03:14:08 UTC on January 19, 2038; it will now end at 03:14:07 UTC on January 19, 2038.

But more seriously, I think time should be monotonic. Most Unix people let the hardware clock run UTC, and let the OS add time zone and daylight savings only when presenting local time to the user. All file stamps are saved in UTC.

Changing UTC is as stupid as changing your hardware clock to daylight savings and back.

Re:So what about 'seconds since the epoch' by Anonymous Coward · 2006-01-01 07:07 · Score: 0

So, when can we expect an update of localtime()?

If you use NTP to sync your clock to a UTC time source (most likely), you've already got the correct UTC time, and hence the correct local time. There might've been a period where your clock was off by 1 second, but it's safe to assume it's been corrected by now.

If you use, e.g., Linux, and you updated your glibc recently, the kernel inserted the extra second already, regardless of NTP, so for instance your log entries around midnight UTC were correct. (Assuming your clock was correct before midnight UTC of course).

After all, the epoch will no longer end at 03:14:08 UTC on January 19, 2038; it will now end at 03:14:07 UTC on January 19, 2038.

Yes, the epoch moved. However the underlying Unix time_t value hasn't changed, of course, just the intepretation.

But more seriously, I think time should be monotonic. Most Unix people let the hardware clock run UTC, and let the OS add time zone and daylight savings only when presenting local time to the user. All file stamps are saved in UTC.

TAI time (from the atomic clock) *is* monotonic..but the earth isn't consistent so we take TAI, add or subtract accumulated leap seconds, and we get UTC. Then we add or subtract a time zone offset and we get local time.

So if you want to be super-accurate, you should run your hardware clock in TAI, use the Olson time zones that add leap seconds (i.e., instead of "America/New_York", set your time zone to "right/America/New_York"), and sync via NTP with a TAI source (if you can find one!!).

And of course, store your times as TAI (or Unix time_t values, or djb's TAI64N values).

Changing UTC is as stupid as changing your hardware clock to daylight savings and back.

That's why you don't worry about it and let the computer do it.

Depends on your timezone... right? by catmistake · 2006-01-01 06:47 · Score: 1

If a leap second is added to every New Years during the 24 hour period of world wide New Years, wouldn't each consecutive timezone get added all the previously added seconds of the time zones that came before it? (such that by the time the New Year's reached Hawaii last night, their clocks are actually set behind like... 20 seconds... and this done between 11:59PM and 12:00AM?) One second isn't very dramatic... but at 20 seconds its kind of surreal.

--
The Admin and the Engineer

Re:Depends on your timezone... right? by Mozk · 2006-01-01 07:08 · Score: 1

No.

The time is shifted, not the amount of time it takes to rotate through a timezone. Your comment basically made no sense. :P

--
No existe.
Re:Depends on your timezone... right? by catmistake · 2006-01-01 09:05 · Score: 1

sure it does... It makes sence, I'm just not sure if this is what happened. I'll take it slow for you so you'll understand...
New Years arrives in New Zealand, and they add one second to the clock.
Then, one hour later, New Years arrives in the eastern parts of ?Russia... so if its one hour after New Zealand's new years, then its one hour and one second, actually, because of the leap second added in New Zealand. Then that time zone adds their leap second.
skipping forward to Greenwhich, England... the New Years arrives there 11 hours and 11 seconds later... then they add their leap second... (so New Years is 12 seconds later than normal)
skip ahead 5 hours and 5 seconds later to the New Years at EST on the East Coast of the US, and you have all the leap seconds added to the previous time zones, it amounts to, I think, 17 seconds. New Years occurs in the EST 18 seconds later than normal due to the leap second in the EST.
Are you seeing the pattern here? Each time zone is adding a leap second to the last year, and if the time zones are all one hour apart from each other, then the seconds will be cumulative. It makes sense... it is a valid question.

--
The Admin and the Engineer
Re:Depends on your timezone... right? by Anonymous Coward · 2006-01-01 09:46 · Score: 1, Informative

No. There is only one leap second and it was added simultaneously worldwide at 23:59:60 UTC. This means the leap second was at 18:59:60 EST or 15:59:60 PST.

The only people who sould have the leap second at 23:59:60 local time and thus extend the final 10 seconds of their new year's countdown would be places where local time is GMT.
Re:Depends on your timezone... right? by AGMW · 2006-01-01 23:37 · Score: 1

New Years arrives in New Zealand
Kinda off topic, although this story is about the New Year's Eve extra second ... anyway ... cough ...
Why do people call it "New Years"? I think this is another Americanism that has crept in to the UK over the last few years. Surely, there's New Year's Eve, and New Year's Day (the days relating to New Year), but the time of year is "New Year" not "New Years".
Anyway, Happy New Year to all you Slashdottirs and Slashdotsons (as they'd maybe say in Iceland - the place, not the cheap UK frozen food outlet).

--
Eclectic beats from Leeds, UK
handmadehands.co.uk

MOD CHILD UP by Anonymous Coward · 2006-01-01 06:53 · Score: 0

at least read and/respond to it... what's up with all the extra seconds?

Re:It's not the 'ephemeris second' that's the prob by iabervon · 2006-01-01 06:57 · Score: 1

The discontinuity is only in the assignment of seconds of linear time (i.e., the number of seconds that have elapsed since some reference) to days and time of day. For purposes like GPS, where you don't care at all about making the time relevant to humans, the linear time is perfectly sufficient, and there aren't leap seconds or leap year days because those concepts are only relevant to the alignment of units larger than seconds. That site is a bit misleading, actually. GPS time doesn't use an hour:minute:second form. And the conversion of the GPS time values to human-readable clock values can include leap seconds, because the GPS signal includes a field for the number of accumulated leap seconds since GPS time started. So if you'd been watching the display on your GPS at midnight this new years, you probably would have seen it behave oddly as the leap second count field changed, most likely in the form of showing 12:59:59 for two seconds, because knowing to show 12:59:60 would require a bit of extra state tracking (you need to notice that the leap second count is different between two consecutive messages) and practically nobody is going to be paying attention and care.

Woohoo! An extra second of sleep! by Xaroth · 2006-01-01 06:58 · Score: 1

No wonder I feel so well rested!

--
That green slime had it coming.

Re:Happy New Year by Paradise+Pete · 2006-01-01 07:11 · Score: 1

If you look at the etymology of the word, you can get a good idea as to why it's spelled with an a instead of an e.

It's not a word I normally misspell. And in also speaking Spanish, there's no reason at all for me to make the mistake I did, as it's clearly an a not an e. I have no idea why I did it, and it bugged me when I saw it. :-)

Re:It's not the 'ephemeris second' that's the prob by CSfreakazoid · 2006-01-01 07:23 · Score: 1

Enter my expertise. I was a software engineer at NASA and happened to do some work on a clock manager software. The use of UTC actually causes less grief becasue everyone is using it. Managing several different times at any givin time could be disaterous. If something is calculated for one time, and executed in a different time, it will happen at the wrong time. Kinda like the inches/cm problems that were experienced in recent years. If you followed that you are of greater intelligence than I am. Guess that why I work for the government.

Re:Happy New Year by Paradise+Pete · 2006-01-01 07:23 · Score: 1

The link was expired. But doing a little searching I see that the all-caps version, REALTOR, is a registered service mark. I don't know what that does to the lower case version. Interesting.

So it probably bugs them even more when people say "realator." :-)

expanding earth? by peter303 · 2006-01-01 07:29 · Score: 1

Like a spinning skater extending and contracting arms to change speeds, the earth can change speed if its radius is changing. A large earthquake such 2004 Andaman can change the day a few microseconds if the fault throw has a vertical component. The Topex radar altimeter satellite has measured average sea level rising three centimenters in the past eleven years. This is thought to mostly due to thermal expansion rather than glacial melt or other causes.

WWV by spaceyhackerlady · 2006-01-01 07:50 · Score: 2, Interesting

>blockquote> I watched the time at Time.gov: 23:59:56 (UTC) =>23:59:57=>23:59:58=>23:59:59=>23:59:60!=>00:00:0 0 It was Amazing! This was the first time for me... *remebers where I was at that moment

I listened to it on WWV. They drop the 29th and 59th tick of each minute, and at 2359 UTC it sounded like (I counted the seconds myself):

tick (57)
tick (58)
silent (59, as usual)
silent (60, leap second)
BEEP (0000 UTC 1 January 2006)
titick
titick
titick
tick
tick

...and so on. The UT1 time correction went from -0.6 to +0.3 seconds. It's encoded in the double ticks.

Yes, I got a recording. Lame or what?

...laura

Re:WWV by Rytsarsky · 2006-01-01 08:00 · Score: 1

I wanted to get a recording, but decided to take that lame screen shot instead :) Could you post it somewhere so we can hear it?

--
God became man to enable men to become sons of God. -C.S. Lewis
Re:WWV by denelson83 · 2006-01-01 13:07 · Score: 1

I got a recording of it too. During the leap second, there was no tick, and the digital time code had a binary zero in it.
Re:WWV by Clueless+Moron · 2006-01-01 17:54 · Score: 1

Ok, nice transcription. But I call your lameness and raise it with this link: The 2005 Leap Second
So: does anybody have a video of it?

Satellite operations by d_p · 2006-01-01 08:23 · Score: 1

This was a real pain in the ass for satellite operators. We had a packed control room at 23:59:59 UTC.

Make sure the UTC offset in your GPS receiver or NTP server is changed from 13 to 14 seconds.

This article by Anonymous Coward · 2006-01-01 08:29 · Score: 0

Isn't covered by the copyright at the bottom of the page.

Does that mean I can dupe it on my own site?

Maybe this year will be like 1460 days...

Typo by evilviper · 2006-01-01 08:31 · Score: 1

Double "the" in article:

In order to ensure that the the atomic time ...

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant

Re:Typo by Anonymous Coward · 2006-01-01 10:13 · Score: 0

It's OK, that'll only take a second to fix...
Re:Typo by eobanb · 2006-01-01 13:28 · Score: 4, Funny

Double "the" in article:

In order to ensure that the the atomic time ...

No, that's a leap the.

--
Take off every sig. For great justice.

GPRMC lines by JeffGB · 2006-01-01 08:52 · Score: 1

This is the data stream from my GPS receiver, the leap second is the line in bold, (23:59:59 is repeated)
For other (accurate) clocks, it would have displayed the seconds as ...,58,59,60,00,01,02,...

$GPRMC,235955,A,XXXX.3905,N,XXXXX.6797,W,0.169,188 .3,311205,11.1,W*4B
$GPRMC,235956,A,XXXX.3905,N,XXXXX.6799,W,0.161,261 .1,311205,11.1,W*48
$GPRMC,235957,A,XXXX.3906,N,XXXXX.6800,W,0.209,298 .6,311205,11.1,W*49
$GPRMC,235958,A,XXXX.3906,N,XXXXX.6803,W,0.195,297 .5,311205,11.1,W*4F
$GPRMC,235959,A,XXXX.3906,N,XXXXX.6804,W,0.167,288 .0,311205,11.1,W*4F
$GPRMC,235959,A,XXXX.3907,N,XXXXX.6804,W,0.089,319 .4,311205,11.1,W*42
$GPRMC,000000,A,XXXX.3908,N,XXXXX.6803,W,0.248,16. 9,010106,11.1,W*77
$GPRMC,000001,A,XXXX.3908,N,XXXXX.6803,W,0.062,166 .4,010106,11.1,W*47
$GPRMC,000002,A,XXXX.3907,N,XXXXX.6804,W,0.172,177 .1,010106,11.1,W*49
$GPRMC,000003,A,XXXX.3907,N,XXXXX.6804,W,0.055,266 .9,010106,11.1,W*47
$GPRMC,000004,A,XXXX.3907,N,XXXXX.6804,W,0.148,281 .2,010106,11.1,W*4F
$GPRMC,000005,A,XXXX.3907,N,XXXXX.6805,W,0.103,240 .3,010106,11.1,W*4C

I removed my latitude and longitude since I don't want everyone to know where I live of course.

Re:Leap Seconds are Unamerican! by cybermage · 2006-01-01 10:12 · Score: 1

Apparently, none of the moderators bother to follow links. The linked page is a parody written by Steve Martin.

--
Some people have a way with words, and some people, um, thingy.

Newspapers couldn't resist the hype by Stan+Vassilev · 2006-01-01 10:18 · Score: 1

We all know what it was like for Y2K, and although nuclear bombs didn't fly out randomly from military bases across the world, now newspapers couldn't resist not to hype up the friggin' leap second again:

http://www.suntimes.com/output/news/cst-nws-sec01. html

Quote: "If you don't get all the clocks synchronized when the leap second occurs -- you could have potentially interesting effects," Chester said. "The Internet could stop working, cell phones could go out."

YEA. SURE!

Re:Newspapers couldn't resist the hype by Detritus · 2006-01-01 11:04 · Score: 1

Even without leap seconds, it isn't unusual to see serious glitches in software when the time rolls over at the end of the year. I used to see it every year with software that predicted the orbits of satellites.

--
Mea navis aericumbens anguillis abundat
Re:Newspapers couldn't resist the hype by Stan+Vassilev · 2006-01-01 11:16 · Score: 1

Looks like a piece of bad software to me ;)

When you take complete date with seconds, minutes, days etc. etc. instead of just a timestamp, you have to implement much better logic that implements rolling of seconds into minutes, minutes into hours etc. then months into years, also account for leap years and so on.

If the programmer got lazy and implemented logic up to months, you'll experience odd behaviour each new year's eve.

But nothing is odd on new years eve, if you take the timestamp (timestamp is usually a flat number of milliseconds that have passed since specific point of time, say 1980) there's no even such thing as new year in it, everything flows smoothly.

And as for the leap second, well all devices that work in networks have to account for latency, drop outs and so on which means they the Internet doesn't have One Global Super Accurate Clock which if desyched can break.

So I just call BS on that quote in my previous post (plus it's Jan2 and noone reported something big weird happen).

Re:Leap Second Lovers Are Traitors Says Bill O'Rei by Brandybuck · 2006-01-01 10:32 · Score: 1

Who the hell modded this "informative"? Funny, maybe, but informative it is not.

--
Don't blame me, I didn't vote for either of them!

1900 January 0 ? by Anonymous Coward · 2006-01-01 11:04 · Score: 0

1900 January 0 ?

Isn't this the same date as 1899 December 31 ?

Re:1900 January 0 ? by Anonymous Coward · 2006-01-01 11:54 · Score: 0

No, it corresponds to a typo in an otherwise incoherent attempt to sound technically informed.

Bank It by Packet+Fish · 2006-01-01 11:14 · Score: 1

Who's the big wig who suddenly decided we all had to observe the leap second at midnight December 31st? You can all adjust your your clocks to account for the leap second if you want, but I'm banking mine. I'll just observe it some time later when its convenient, like when I need one extra second of something. Plus, until then, I exist one second in the future relative to all of you. And yes, it is a cool future, but no, no jet packs yet. ^_^

Obligatory Three Dead Trolls in a Baggy joke... by chris_eineke · 2006-01-01 12:28 · Score: 1

So you are one of those Twelve-o-clock flashers!

--
"All you have to do is be fragile and grateful. So stay the underdog." Chuck Palahniuk, Choke

For a second? by PhYrE2k2 · 2006-01-01 12:43 · Score: 0

You adjusted the clocks in your house for the second? Wow- I'd be surprised if they didn't drift by that in a week anyway, as I'm sure the clock on your oven/microwave/vcr isn't as accurate as you might imagine... not to mention the lack of accuracy of you setting it from a wrist-watch/phone/etc.

-M

--

when you see the word 'Linux', drink!

Re:For a second? by Anonymous Coward · 2006-01-01 16:14 · Score: 0

Most clocks I've seen don't even let you set the seconds, just hours and minutes. Come to think of it, the only clock in my house that you can set the seconds on is my computer.
Re:For a second? by devilspgd · 2006-01-01 17:08 · Score: 1

And you call yourself a nerd?

--
Give a man a fish, he'll eat for a day, but teach a man to phish...
Re:For a second? by Bush+Pig · 2006-01-01 17:42 · Score: 1

I can set the seconds (kind of) by interpolation on my 18th C case clock. Of course, there's about 2 minutes slop in the minute hand ...

--
What a long, strange trip it's been.

Re:It's not the 'ephemeris second' that's the prob by Guppy06 · 2006-01-01 15:16 · Score: 1

"People who work on systems that can't afford a 1 second discontinuity (such as the GPS system) use a continuous counting of the SI second."

The leap second is an SI second, as are all the other seconds in UTC. The only difference between UTC and TAI is the method those seconds are counted in. It's SI minutes that UTC doesn't always go by.

And just because it is "more important" for you to use TAI doesn't mean there isn't somebody else for whom it is more important to use UTC. But if you insist on having a hissy fit over a change that can be accounted for automatically if you were willing to prepare for it (leap second information is broadcast in machine-readable code on longwave radio frequencies by NIST), then stop trying to count minutes, hours, days and years entirely and start talking about hectoseconds, kiloseconds and megaseconds instead. Pick a convenient epoch and start counting, and then you don't even have to worry about the changes that have even been done in TAI over the years.

Moderators on drugs? by Anonymous Coward · 2006-01-01 15:57 · Score: 0

That garbage was given points?

> before there is a full day (!) of delay

Considering we're talking about one hour of adjustments every 6,000 years, that's 144,000 years before it's off a full day. Do the moderators really believe that was the solution proposed?

Re:It's not the 'ephemeris second' that's the prob by Fess_Longhair · 2006-01-01 16:14 · Score: 1

Since this depends on the slop of the Earth's interior, it's not a fully regular and predictable thing - we might even have to remove a second one year.

The dominant high-frequency change to the length of day is primarily atmospheric, with inter-annual variations from the oceans (El Nino). These changes are on the order of milliseconds.

Re:countdown - only in GMT zone by Jetson · 2006-01-01 20:37 · Score: 1

10,9,8,7,6,5,4,3,2,1,0,-1

The leap-second was at 23:59:60 UTC, so for most of us the countdown at midnight for the local new year went ahead on schedule.

In fact, I was sitting at my computer at 16:00 local (GMT-8), and watched as the UTC time went through 23:59:58, 23:59:59, 23:59:60, 00:00:00. It was strange seeing "60", considering how many programs I've written that had to avoid rounding 59.5 up and causing an error...

So I spent my leap second doing what any real geek would do -- watching that particular second come and go. I guess I would have to admit that "I enjoyed every second of it". :-)

Same with POSIX timestamps by nicuramar · 2006-01-01 23:27 · Score: 1

The same goes for standard UNIX time stamps (following the POSIX standard), which of course has the advantage that each day has the same length and the disadvantages you described above for NTP. Even consulting a table can't make previous timestamps 100% correct since the POSIX and NTP way is to repeat the same number two seconds in a row when a positive leap second occurs. This can make past timestamps ambiguous. Should a negative one occur, they will skip a second instead.

Re:Happy New Year by AddressException · 2006-01-03 03:07 · Score: 1

I hate the USPTO site -- there's no way to get permalinks to the information! >:|

Refund request... by InvisibleSoul · 2006-01-03 12:02 · Score: 1

I would like my 2 minutes and 0.9 seconds back please.

Slashdot Mirror

Leap Second At The End of 2005

269 comments