back to article Get READY: Scientists set to make TIME STAND STILL tonight

Time could be up for leap seconds later this year, if Tuesday night's addition of one second to the world's time proves to be a success. International weights and measures standards-keepers at the Bureau of Weights and Measures (BIPM) near Paris want an end to leap seconds and favour a single, consistent time standard. BIPM …

  1. Jason Bloomberg Silver badge

    Having a single time is a nonsense

    And we will eventually recognise that when the sun's coming up when we go to bed and are working through hours of darkness

    If some people want a time system which is contiguous, without leap seconds, then fine by me. I can see the benefit of that, but in the real, physical world that is doomed to failure. It is no more practical than abandoning time zones and having everyone work to GMT, UTC or UTZ, or whatever it's called this week

    1. John G Imrie

      Re: Having a single time is a nonsense

      We don't have a single time we have multiple times called time zones and these are added, removed and altered at the whim of the local politicians.

      If you find that UTC doesn't match your current daylight then change your local time zone.

      Don't muck around with UTC

    2. Sebby

      Re: Having a single time is a nonsense

      Nah. We're human beings--we'll adapt.

      Abolish "Mean Solar Time", and move to TAI ("Atomic" time). I can refer to any future time, correctly and accurately, always. Time goes forward at a fixed rate, all the time.

      Then drop time zones, and move to the 24-hour time format. The world is a big place, and there's plenty of understanding in it: Australians just happen to get up at 18:00, that's all. And no bloody "Daylight savings", neither.

      Then drop relative time: every time is expressed as an absolute, and people become good at mental arithmetic so we can express durations, and swiftly add them together, or recognise time periods such as 86400 seconds equating to a day.

      Now stop insisting on having significance of day and night: time is used at whatever interval suits, and day and night are merely observable events. Efficiency may very well be improved. The problem of natural illumination can be solved, somehow *, but in the event that it can't, then sunrise and sunset are merely accurately-recorded points on the absolute timeline, by which means relative offsets may be expressed for the purposes of coordination.

      Then switch to a decimal system of time units, of suitable complexity to allow very accurate time scales but of sufficient manageability for everyday use.

      Not interested? But it's the future! Ah well, thought I'd just suggest it ...

      * I'm blind, so I don't have this problem, happily.

      1. Andrew Newstead

        Re: Having a single time is a nonsense

        Unfortunately, those of us who are sighted are tied to a diurnal cycle of light and dark, some quite strongly. One of the things recognised recently is how blue coloured light (such as produced in monitors and LCD TV screens) can trigger the wake up response in humans, causing sleep problems for those using screens late at night.

      2. Electron Shepherd
        Boffin

        Re: Having a single time is a nonsense

        Time goes forward at a fixed rate, all the time.

        The General Theory of Relativity covers how the relative motion of two observers and the gravitational fields they are in affect how the passage of time is perceived by each of them.

        Start with https://en.wikipedia.org/wiki/Time_dilation

      3. Adam 1

        Re: Having a single time is a nonsense

        >Then drop time zones, and move to the 24-hour time format.

        Seems like a lot of effort and you aren't even going to get it decimalised.

      4. ravenviz Silver badge
        Coat

        Re: Having a single time is a nonsense

        Australians just happen to get up at 18:00, that's all

        They already have Christmas in June anyway.

        1. John Tserkezis

          Re: Having a single time is a nonsense

          "Australians just happen to get up at 18:00, that's all

          They already have Christmas in June anyway."

          Hey! No we don't.

          No wait, the shopping centres usually start advertising their Christmas crap at about June, so you're right after all. Sorry, as you were.

    3. Anonymous Coward
      Anonymous Coward

      Re: Having a single time is a nonsense

      There are already several (at least three) time systems without leap seconds:

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

      People who want a time system without leap seconds should just use one of those and leave UTC alone. Some of us like having a time system that remains in sync with day and night.

      If they do get rid of UTC it'll just get reinvented in a few years' time, by which time we'll probable have eight or nine competing systems.

      1. Real Ale is Best

        Re: Having a single time is a nonsense

        Obligitory XKCD

      2. Wzrd1 Silver badge

        Re: Having a single time is a nonsense

        I'll stick with UTC. One standard works, rather than nine competing standards to generate confusion.

        I get enough of that confusion at work, dealing with EDT, EST and UTC and remembering which device is using which.

    4. Anonymous Coward
      Anonymous Coward

      Re: Having a single time is a nonsense

      "Microsoft is going for the big bang approach. Azure servers will adjust to the new second based on the time zone they’re in."

      So Microsoft are the only ones doing it as per the standards / actual change in time! Everyone else seems to be bodging it.

  2. Little Mouse

    But will this answer the fundamental question of time...?

    What is "Clocks"?

    1. This post has been deleted by its author

      1. Martin Budden Silver badge
        Trollface

        Re: But will this answer the fundamental question of time...?

        It's a song by Coldplay.

        I hadn't realised that noise Coldplay makes is songs.

        1. Jedit Silver badge
          IT Angle

          "I hadn't realised that noise Coldplay makes is songs."

          Album's out and the songs are lame

          All of them sounding just the same

          Gwyneth Paltrow sucked my dick

          But I still whine till it makes you sick

          Singing "Nyeeerrrrrr .... waaahhhhh"

          (Continue for 500,000 more sales)

          The IT angle? You can't spell "SHIT" without "IT".

    2. Andrew Newstead

      Re: But will this answer the fundamental question of time...?

      Track by Steve Hackett on Spectral Mornings

  3. Simon Harris

    A completely preventable situation.

    If only there was an El Reg unit of time we could all use instead!

    1. frank ly

      Re: A completely preventable situation.

      I'd suggest that a fundemental unit of time could be the 'regond', defined as the time taken for a sheep travelling at its maximum speed in a vacuum to travel a length of one linguine. (Or maybe a sensible multiple of linguine to give a practically useful unit.)

      The advantage of this definition is that if any kind of 'leap second' type of correction is needed, you can do that by interfering with the sheep (as agreed, supervised and witnessed by an international commitee of time boffins).

      1. Alister
        Coat

        Re: A completely preventable situation.

        The advantage of this definition is that if any kind of 'leap second' type of correction is needed, you can do that by interfering with the sheep (as agreed, supervised and witnessed by an international committee of time boffins).

        I'm sorry, but interfering with a sheep is a criminal offence in most countries, and besides which it's sick, SICK, you Baaaa stard

        1. Will Godfrey Silver badge
          Happy

          Re: A completely preventable situation.

          You're only saying that because you can't find one that fancies you.

        2. David 132 Silver badge
          IT Angle

          Re: A completely preventable situation.

          Call it a RAM upgrade, then.

      2. GrumpenKraut
        Coat

        Re: A completely preventable situation.

        No such maximum exists. The supremum, though, does. It's c, see?

        But for sheeps consisting of a single phonton ... My coat, thanks.

  4. Slacker@work

    Time is an illusion...

    ...Lunchtime doubly so!

    Well somebody had to say it.

    1. emmanuel goldstein

      Re: Time is an illusion...

      time is not an illusion, but the flow of time is.

      1. lambda_beta
        Linux

        Re: Time is an illusion...

        If the 'flow' of time is an illusion then time is an illusion, since time is only measured by the 'flow' of events.

    2. Clive Harris
      Happy

      Re: Time flies like an arrow

      Time flies like an arrow, fruitflies like a banana.

      (Graucho Marx)

    3. Captain DaFt

      Re: Time is an illusion...

      Does time exist?

      I gravely doubt it.

      But gosh!

      What would we do without it? -: Piet Hein

  5. Bob Wheeler
    Pint

    A simple method

    Either it is Beer O'clock, or it's not and I can go back to sleep. What more do you need form any time system?

  6. Anonymous Coward
    Anonymous Coward

    ...accurate for a period of 158 million years...

    It's accurate for 158 million years, and then suddenly loses its accuracy? I don't think so.

    Anyway, the obvious way to deal with this is at the point when you convert between UTC and display time, which we already do for timezones.

    Right now the offset between BST and UTC is 3600 seconds. That offset should change to 3599 seconds, and UTC should continue unchanged. The TZDATA files will get bigger, but not much, since all countries should implement the leap second at the same time.

    However it means you can no longer work out the GMT date by taking a UTC value and dividing it by 86400. So you need to persuade people to use a library call to do that instead. Of course, nobody will bother to do this, so various applications will do the wrong thing for a few seconds a day.

    There will be lots of minor annoyances. For example, your telco's billing system, which is supposed to give you free calls at 19:00:00 onwards, will only give you free calls from (say) 19:00:10 onwards, so you'll be overcharged once in a while.

    Once the offset builds up to a minute or two, and people are turning up to meetings at the wrong time, they'll be resolved.

    But if you're going to decouple UTC from the rotation of the earth, why not do it completely? Let's have UTC decimal time, with a UTC "day" defined as 100,000 seconds.

    1. the spectacularly refined chap

      Re: ...accurate for a period of 158 million years...

      Right now the offset between BST and UTC is 3600 seconds. That offset should change to 3599 seconds, and UTC should continue unchanged. The TZDATA files will get bigger, but not much, since all countries should implement the leap second at the same time.

      Careful, this is a simplification of a simplification. Many of these amateur suggestions are implicitly based on layers of simplification, knowing or unknowing, which introduce paradoxical effects at the margins. Doing the job properly is complex which is why you have a panel of experts spending so long debating how this should be handled - I wouldn't call myself qualified to add anything meaningful to their discussion but I've studied this enough to understand the complexities. In this case BST is not a 3600 second offset from UTC: is closer to 3601 seconds, and adding the leap second to UTC simply alters the offset to just under 3600 seconds. BST is a defined offset from GMT which the UK still uses in law at least. GMT is essentially an older version of UT1: it is defined astronomically and has no leap seconds.

      Simply adjusting the definition of time zones like that has dramatic side effects - in particular events that happened close to midnight in the past can suddenly move from having occurred on one date to another simply by the introduction of another leap second. It doesn't take too much imagination to envisage all sorts of issues this can raise in the field of contract law alone.

      The most elegant way of representing this I have seen on computer is simply to allow a second to be two seconds long during a leap second, i.e. allow space in whatever fine scale representation you use to accommodate e.g. 23:59:59 and 1500 milliseconds as the middle of the leap second as opposed to a 23:59:60.5 representation. Apps that don't care too much about precision time keeping simply get 60 second minutes and can ignore the whole issue, still being guaranteed that the date, hour, minute are always perfectly correct - the second too unless they try to get clever with internal representations. Apps that need ultimate precision can get the precise time any time they wish.

      Yes, it's still a bit of a fudge but it avoids many of the issues of the simplified alternatives. It also closely approximates the current treatment with periodic insertion of leap seconds.

      1. Brewster's Angle Grinder Silver badge

        Re: ...accurate for a period of 158 million years...

        From the top. The number of UT1 days is the number of transitions of IAU 1976 equinox across the Greenwich meridian (calculated by the proper motions of the stars in the FK5 catalogue using the IAU 1980 theory of nutation). The length of the UT1 day can vary from 86400 SI seconds by a few milliseconds so midnight doesn't happen on a whole second boundary.

        By contrast, UTC days are designed to occupy an integral number of seconds, generally 86400, with extra seconds added or subtracted to keep UTC within 0.9s of UT1.

        According to Wikipedia, the UK hasn't made up its mind whether GMT means UT1 or UTC so is there some ambiguity. But in all non-nautical contexts, I would expect GMT to mean UTC.

        FWIW, a linear regression of the bulletin-B data from October 2009 suggests the next leap second will probably be end of June 2019. (It could just creep in at the end of December 2018.) But the Earth has actually speeded up in the 21st Century (we're had 19 fewer leap seconds than predicted in the late 20th Century) so we could end up having to add a few in a flurry, if the Earth regresses to mean. On the other hand, if the speed up accelerated, then we might end up having to drop seconds in the next couple of decades.

        And if anybody thinks I'm wrong, please correct me because I'm working on our delta-T code right now!

        1. JeffyPoooh
          Pint

          Re: ...accurate for a period of 158 million years...

          The UK hasn't even made up its mind about prime meridians. Or maybe it has, several times. Or had its mind made up for it by others.

        2. mad physicist Fiona

          Re: ...accurate for a period of 158 million years...

          According to Wikipedia, the UK hasn't made up its mind whether GMT means UT1 or UTC so is there some ambiguity. But in all non-nautical contexts, I would expect GMT to mean UTC.

          The relevant legislation is clear - it means GMT. GMT is its own time standard which isn't really surprising given that it predates any of the Universal Time standards as the original world time system. The nearest UT equivalent is UT0 but even that isn't a direct substitute since the precise methods by which they are defined differ - the GMT definition is slightly sloppy for modern standards of precision.

          The ambiguity arises from the popular conflation of the GMT to mean UTC, UT1, or whatever else the user wants it to mean. "Basically equivalent for everyday purposes" gets morphed into "exactly the same" or even "equivalent by definition". GMT has no fixed relationship with any of the universal time standards. It's also the kind of incredibly subtle issue that I wouldn't trust Wikipedia at all for - everyone fervently maintains what they believe is correct and precise, even if as a poster here remarked they are actually using a simplification of a simplification.

          Confusing matters even more in the UK is that although legally the time is defined as GMT, most precision time sources, such as the speaking clock, MSF, GPS and the BBC time pips all actually transmit UTC.

          1. Brewster's Angle Grinder Silver badge

            Re: ...accurate for a period of 158 million years...

            The 1880 act doesn't define GMT in any sense. (Which is unsurprising as it pre-dates accurate time-keeping and our realisation that Earth's rotation rate is variable.*) So there just isn't a GMT "standard".

            If it came down to milliseconds, I concur that the pedantic interpretation would be UT0 measured on the true Greenwich meridian (not the WGS84 meridian). However if my learned friend insisted on that, I'd insist the measurement be made using a "second" that is 1/86400th the length of relevant day, since that was how the second was understood when the statue was written; and that the day be reckoned from midday, not midnight, since that was also the case in 1880; and furthermore, that the measurement be made using a clock appropriate for the period - i.e. one so inaccurate that milliseconds would be of no import!

            In reality I would suggest that any British landlubber referring to GMT, including our modern legislators, could be reasonably presumed to mean UTC+0. Particularly since, as you note, every effective realisation of "GMT" is UTC (while noting that GPS doesn't actually transmit UTC, but TAI-19 and the offset of this time signal from UTC).

            * Rather pleasingly, it appears to be de Sitter measured this - see the Explanatory Supplement to the Atronomical Almanac. 2.55

        3. Anonymous Coward
          Anonymous Coward

          Re: ...accurate for a period of 158 million years...

          "the UK hasn't made up its mind whether GMT means UT1 or UTC so is there some ambiguity"

          GMT is GMT. It's not up to the UK to define how subsequent derivates relate to GMT.

  7. davemcwish
    Thumb Up

    Time enough at last

    Kudos to the sub/picture editor

    1. JeffyPoooh
      Pint

      Re: Time enough at last

      No, I don't like that picture.

      If you look closely, you'll see that the 4-o'clock position is marked with what appears to be "IV". But the correct (in the sense of traditional) indicator for 4-o'clock on Roman Numeral clock faces is, perhaps surprisingly to some, "IIII".

      So I don't like that picture.

      1. mad physicist Fiona

        Re: Time enough at last

        If you look closely, you'll see that the 4-o'clock position is marked with what appears to be "IV". But the correct (in the sense of traditional) indicator for 4-o'clock on Roman Numeral clock faces is, perhaps surprisingly to some, "IIII".

        I'd argue that it is one of the few clocks that you can argue is correct. Modern usage of Roman numerals would indeed make 4 IV. However, the Romans themselves didn't use the modern prefix-to-subtract notation so 4 would indeed be IIII as most clocks use. However, in that case 9 would also be VIIII, which I can't recall ever having seen on a clock. You can argue about whether it is right or wrong but it is at least consistent, as opposed to most Roman numeralled clocks which are unambiguously wrong regardless of which system you use.

  8. John Savard

    Fortunately, We Know the Title Was Only a Joke

    Otherwise I would have to severly take you to task for perpetuating the fallacy that leap seconds (or other similar things, like standard time zones) affect time itself, as opposed to the names by which we call certain intervals of time.

    We used to get along without leap seconds quite well, by adjusting the length of the second of civil time instead. That would be more convenient, apparently, for computers and the Internet, so we should either go back to the old system or at least have it available as an alternative kind of Internet Time.

    After all, we already have GPS time in addition to UTC.

    1. JeffyPoooh
      Pint

      Re: Fortunately, We Know the Title Was Only a Joke

      JS suggested "...adjusting the length of the second..."

      Hmmm. Simply change the '9,192,631,770 cycles of the caesium 133 etc.' to a different number. Perhaps 9,192,631,867 or thereabouts to give an extra second every three years.

      Of course, we'd have to update some PROMs in the time standards, ...not to mention adjust the value of all sorts of derived constants, recalibrate much of the test equipment and electronics on Earth, amend all the text books, etc.

  9. Doctor Syntax Silver badge

    "Both firms will implement a policy of cutting up the second and feeding it into the clocks on tens of thousands of their servers 12 hours either side of midnight."

    The don't need to do this with leap-year days so why do they do it for leap seconds. Just allow an occasional 61 second minute.

  10. lawndart

    says:

    Lu Tze, pop over to Paris and let them know Gardeners Question Time.

  11. My-Handle

    A single, consistent time standard... where?

    I can see why this approach would appear attractive. We have all computer equipment, GPS satellites etc all running off the same time system to minimise errors, and humans can adjust time as they see fit...

    However.

    Accurate though these atomic clocks are, time itself is -not- constant. A GPS satellite travelling in one direction is going to be experiencing time at a slightly different rate than a satellite travelling in a different direction, and at a different rate again than a point on the ground because they all occupy different timeframes.

    I'm sure such inaccuracies are less than the leap-second, but I'm also sure that a great deal of ingenuity is taken up to account for this.

    1. Joe Gurman

      Re: A single, consistent time standard... where?

      There is one. It's called TAI, the French acronym for International Atomic Time, which ticks on, oscillation of cesium atom by cesium atom, until the last syllable of standards-keeping boredom.

      Why we think human-usable time standards have to be that uniform, and not include leap seconds, is probably based more on shoddy programming for those GPS/GLONASS/Galileo systems than actual issues. At my shop, we've been dealing with leap second jumps in the UTC - TAI offset for 21 years, with no problems of which I'm aware. And there used to be more frequent leap seconds. Guess the earth's rotational slowing down is slowing down.

      As for GPS satellites traveling in different directions, don't worry; that's all in the time standards calculation.

  12. Bloodbeastterror

    Phoenix

    I believe that Phoenix doesn't follow the daylight saving clock changes in the US because the local golf course owners objected that the extra hour's sunlight would scorch the greens.

    1. Joe Gurman

      Re: Phoenix

      Arizona and Hawai'i don't go on summer (daylight savings) time, primarily for astronomical and economic reasons having little to do with golf courses. Arizona, for instance, is economically linked to neighboring California, which is one time zone to the west. Hawai'i is at such a low latitude that there isn't much summer-winter variation in day length.

      Indiana used to make DST a local option, so suburban counties near Chicago could match its time, while downstate farming counties could stay with something closer to mean solar time. In 2006, they switched to everyone following DST, but placed some counties in the northwest (Chicago suburbs) and southwest parts of the state in the Central time zone.

      Oh, and the huge Navajo reservation in northeastern Arizona (as well as parts of New Mexico and Utah) does observe DST.

      It's a big country,

      1. This post has been deleted by its author

      2. Bloodbeastterror

        Re: Phoenix

        @Joe Gurman

        It was a joke. I knew I should have added a smiley... :-)

        The first time I read it (in an otherwise fairly dry company bulletin) I actually did laugh out loud.

        But thank you for the info - I didn't know that.

  13. mr.K

    Why in the first place?

    Since this is such a rare occurrence and a second is so small, why do we do it? I have yet to hear any reasonable argument for. To get an offset of five minutes we need a millennium. So maybe people are a bit nostalgic (autistic) and need the time to be 12 when the sun is in the south. Guess what, it isn't. Daylight saving time offsets that by an hour, but alright lets look apart from that. The time zones are quite wide and usually expanding into the last one and the next one in some places, so you have up to half an hour offset there as well. But, let us assume you don't use daylight saving time and live exactly on a longitude divisible by 15 then sun still will be in the south at 12 only on four occurrences each year due astronomical time is not sundial time, but mean solar time and the offset called the equation of time is up to 16 minutes.

    Can anybody enlighten me as to why we use leap seconds? It seems to me that for it to become an offset greater than what we experience throughout the year anyway we will have to wait over three thousand years. Isn't that something we could leave for that generation?

    1. teknopaul

      Re: Why in the first place?

      Agreed, leap seconds make no sense.

  14. Tromos
    Joke

    Tackle the problem at source

    Wouldn't it be easier to speed the Earth up?

    1. TeeCee Gold badge
      Coat

      Re: Tackle the problem at source

      Great idea, I'm in.

      Which direction do you want us all to run in and when?

      1. Martin Budden Silver badge

        Re: Tackle the problem at source

        When we are out of breath and stop running the fix gets unfixed.

  15. alain williams Silver badge

    Seconds for computers, convert for humans

    Time in seconds should be how computers see it.

    Time broken up into minutes/hours/days is done for the convenience of humans - who cannot deal with large numbers - who knows that 2592000 seconds is a (30 day) month ? The time convertion routines will deal with leap seconds.

    Slight complications with some applications, eg bank interest is computed in days - be they 86400 or 86401 or 86402 seconds long. Programmers will get used to this and will cope, it will become part of the 'tradition'.

    Fiddle the issue now and we will just end up being hated in generations to come when they have a big problem to deal with - and their computers will not deal with them since that is not the way things will be done.

  16. Anonymous Coward
    Anonymous Coward

    100 years ago

    I just got back my Grandparents 'Westminister Chimes' clock from restoration (after having it for at least 10 years after my Grand Mother died at 97) - cost £280.00 to restore. Chimes every quarter of an hour. 100 years old, so it (hopefully) will see another 100 years now.

    Does it worry about an extra second? Nah. It just works (again).

  17. asdf

    why?

    Why the fuck didn't they get rid of these retarded leap seconds when the world was getting computerized back in the 1980s. In the next thousand years the amount of leap seconds combined will amount to less than 10 minutes. Is it really worth introducing literally trillions of edge cases that rarely get tested in systems we rely for our modern world to function? Luckily from what I am hearing this will probably be the last. Good fucking riddance. Yes code should properly handle it but guess what the industry is full of hacks and management that authorizes a bare minimum of testing due to cost. Naming and shaming is great until it is your mass transit system that shits the bed.

    1. JeffyPoooh
      Pint

      Re: why?

      "...literally trillions of edge cases that rarely get tested in systems we rely for our modern world to function?"

      Disagree. This is good practice for the spec writers, system designers and coder drones of the world.

      1. asdf

        Re: why?

        >Disagree. This is good practice for the spec writers, system designers and coder drones of the world.

        Yep the "experts" in Kazakhstan need all the practice they can get.

  18. Old Handle

    How about a compromise?

    Drop the leap seconds, but insert a leap minute whenever it drifts that far (should be every hundred years or so.) Considering almost all time zones are only accurate to the closest hour anyway, there's no real need to try and pin them precisely to solar time, approximate is good enough.

    The only downside I see (which might be significant) is this would make it a once-in-a-lifetime event for most people, meaning we're even less prepared for it when it happens.

  19. John Brown (no body) Silver badge

    The people who might be affected...

    ...are those high frequency traders who rely on microsecond accuracy for their profits. They could be screwed if they are on one time adjustment rate while others are on a different one. That would be a shame :-)

    1. Anonymous Coward
      Anonymous Coward

      Re: The people who might be affected...

      Yep hopefully Sony won't shit the bed over a leap anything like they have in the past. Of course their IT department serves mostly to make other corporate IT departments feel better about themselves so who knows?

  20. Stu 18

    Microsoft big bang = big fail

    In my experience with windows 2012 servers on a virtual environment, it is bloody hard to make the time stay accurate +- a lot more than one lousy leap second!

    Especially if you have active directory which competes with a myriad of built in 'ms knows best' mechanisms all in competition to set the clock.

    I presume they must use third party code to fix it on azure, because ms knowledge base can't even agree on what is the best thing to do.

    I'd wager that MS servers will be all over the place time wise for a lot longer than a second. However since they explicitly exclude 'real time' as a thing they can do, what do they care!

    1. TheVogon

      Re: Microsoft big bang = big fail

      "In my experience with windows 2012 servers on a virtual environment, it is bloody hard to make the time stay accurate +- a lot more than one lousy leap second!"

      It's the same as under normal Windows exact that Hyper-V also gives you an additional option to set the time continuously on the VM from the host

      "Especially if you have active directory which competes with a myriad of built in 'ms knows best' mechanisms all in competition to set the clock."

      No, it's the exact same single method across all of Windows / AD - which is based on NTP 3 with some enhancements from NTP 4. It's not meant to be highly accurate

      If you need really accurate time then NTP sucks. You might want to consider using PTP (Precision Time Protocol) - for which 3rd party implementation exist for Windows.

  21. Captain DaFt

    Sigh...

    Now I'm getting nostalgic for my youthful days on the net.

    All this bickering and japing over time, and no one's even mentioned the Time Cube.

    Ah well, no meme lives forever.

  22. Andrew Moore

    GPS...

    I thought all the GNSS systems adopted the GPS approach- they ignore the leap seconds so they are currently adrift by 24 seconds.

    1. Paul Crawford Silver badge

      Re: GPS...

      Of course they do - they don't have problems with this by design.

      Its only the ground based software that is implemented by folk who (a) don't know what they are doing, and (b) don't test things that cause problems.

      Fsck'em - why not have leap seconds +/- every week and occasionally do two in the same direction on consecutive events? That way stuff will be tested and fixed because the code monkeys can't argue "oh it only happens one per 2 years or so".

  23. Tweets

    The second in UTC is defined as the length of time taken for 9,192,631,770 cycles of the caesium 133 atom produced by the transition between two levels.

    So why can't we simply extend a second by a few cycles of caesium? It's not as if we would notice!

  24. ravenviz Silver badge
    Alert

    Won't they have to move Stonehenge round a bit as well?

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon