back to article Time Lords set for three-week battle over leap seconds

An upcoming International Telecommunication Union (ITU) conference is about to become an international battleground over whether or not to retain the leap second – the periodic adjustment of Coordinated Universal Time (UTC) so it stays in agreement with atomic clocks. The debate's expected to be so intense it will continue …

  1. Anonymous Coward
    Devil

    Why stop there?

    I think it's about time to convert to Metric time, no more division by 60, or 24, or 365.24.

    1. Your alien overlord - fear me

      Re: Why stop there?

      Napolean tried. Noone took any notice of that half pint (or .5 litre).

      1. Anonymous Coward
        Happy

        Re: Why stop there?

        "Noone took any notice of that half pint (or .5 litre)."

        Nepolean was actually pretty tall for the time. It was simply a myth created by the British press for propaganda purposes, and a misinterpretation of French imperial measurements that showed him a a short person.

      2. Velv
        Headmaster

        Re: Why stop there?

        "Napolean tried. Noone took any notice of that half pint (or .5 litre)."

        .284Litre. .5litre is (nearenuff) a pint

    2. BlartVersenwaldIII
      Unhappy

      Re: Why stop there?

      If you mean metric time where every minute is 100 seconds, I think south west trains have already switched to it.

    3. John Robson Silver badge
      Boffin

      Re: Why stop there?

      60 and 24 are actually nice numbers - they have many useful factors.

      That 364.25 is akward, but is the relationship between two fundamental properties of the planet (at the moment, one of those properties is gradually changing)

      Yes - I know (hope) you were being sarky, but too many people blindly agree with that kind of nonsense...

      1. JeffyPoooh
        Pint

        Re: Why stop there?

        JR wrote: "That 364.25 is akward..."

        365 (not 364) and .2425 (to be a bit more precise).

        The 0.2425 is from the Leap Year rule:

        Every four years (+0.25),

        but not every 100 years (-0.01),

        except every 400 years (+0.0025).

        And awkward is spelled 'awkward'.

        1. Brewster's Angle Grinder Silver badge

          Re: Why stop there?

          @JeffyPoooh

          1. In astronomy, you work in Julian years where there are exactly 365.25 Julian days of 86400 SI seconds in a year. So either JR is dumb or an astronomer.

          2. 365.245 is not exact either. My copy of the Explanatory Supplement to the Astronomical Almanac gives the tropical year as 365.2421897 Julian days in the year 1900.

        2. John Robson Silver badge

          Re: Why stop there?

          Wow - I made two typos ;)

          356.25 is close enough to convey the relationship I mean... If I typed 3.14 then you'd be able to fill in the rest.

          The missing day is because I don't work on Friday (pick one at random)

          Awkward was missing a "w" - That usually gets typed on that Friday....

    4. Richard 22

      Re: Why stop there?

      But base 10 is a dreadful system, chosen purely because we happen to have 10 digits. 12 is a much better base since it has a greater number of factors. It would be better to simply move the entire metric system over to base12

      As a software engineer I'd be quite happy with base 16 either.

      Hexadecimal time please!

      1. Blergh

        Re: Why stop there?

        I think it would be much easier to work with base 12 or 16 if they just created some new numbers for 11 to 15. Using letters A-F is a terrible way of doing it and too confusing because I already have a relationship to those letters in the alphabet. They need their own shapes and names.

      2. Doctor Syntax Silver badge

        Re: Why stop there?

        "But base 10 is a dreadful system"

        Quite true. The metric enthusiasts mock our old ponds and ounces. But given a pair of scales (just scales, no weights) would you prefer the task of dividing a pound of sugar into 16 ounces or a kilo of sugar into 10 lots of 100gm?

        Metric is so nineteenth century.

        1. Anonymous Coward
          Anonymous Coward

          Re: Why stop there?

          "Given a pair of scales..."

          Tried buying a pair of scales recently? You can usually find one eventually, mixed in with the electronic scales and the needle-readout ones. But they aren't exactly common. Or cheap.

          There are many reasonable arguments on both sides. That isn't one of them.

          1. Bucky 2
            Mushroom

            Re: Why stop there?

            Tried buying a pair of scales recently?

            If you're looking for scientific scales, then yes, that kind of precision is pricey. But for common uses, like cooking (we were talking about sugar), you can find them at reasonable prices at any kitchen supply store (fancy-ass name brand stores notwithstanding).

        2. NumptyScrub

          Re: Why stop there?

          Quite true. The metric enthusiasts mock our old ponds and ounces. But given a pair of scales (just scales, no weights) would you prefer the task of dividing a pound of sugar into 16 ounces or a kilo of sugar into 10 lots of 100gm?

          So remind me again, how many ounces to a pound, how many pounds to a stone, how many stone to a hundredweight, and how many hundredweight to a ton? Because none of those are the same factor as any of the others.

          Also there are 2 different hundredweights and 2 different tons.

          Metric, for all it's faults, is far less confusing when trying to learn the various measures; just learn the SI prefixes and you are good to go with mass, length, and volume right away ;)

          1. Martin Taylor 1

            Re: Why stop there?

            @NumptyScrub: "So remind me again, how many ounces to a pound, how many pounds to a stone, how many stone to a hundredweight, and how many hundredweight to a ton?"

            I'll tell you a funny thing. There are still lots of us about who learned how to deal with this system at school. It didn't kill us, and we still remember the numbers (16, 14, 8 and 20, since you ask).

            1. NumptyScrub

              Re: Why stop there?

              I'll tell you a funny thing. There are still lots of us about who learned how to deal with this system at school. It didn't kill us, and we still remember the numbers (16, 14, 8 and 20, since you ask).

              Oh I remember them myself, along with l/s/d currency, and inch -> foot -> yard -> chain -> furlong -> mile for length.

              Metric / SI units are far more straightforward and easier to learn, IMO :)

          2. John Robson Silver badge

            Re: Why stop there?

            @NumptyScrub - Metric, for all it's faults, is far less confusing when trying to learn the various measures; just learn the SI prefixes and you are good to go with mass, length, and volume right away ;)

            Well, except for units of mass...

      3. Mage Silver badge

        Re: Why stop there?

        Octal would be handier than Hex, for hands.

    5. ravenviz Silver badge

      Re: Why stop there?

      The issues of leap seconds does not go away with metric time.

    6. Anonymous Coward
      Anonymous Coward

      Re: Why stop there?

      I think it's about time to convert to Metric time, no more division by 60, or 24, or 365.24.

      That's actually 365.2425

  2. Herby

    And we have...

    Daylight "saving" time as well. Of course this changes at the whim of various governments, most recently as "energy saving", and before that at the behest of barbeque makers (and other outdoor activity people).

    If they do away with leap seconds, it is just denial that the earth is slowing down. Then "high noon" won't be.

    1. imanidiot Silver badge

      Re: And we have...

      Whoever it was that introduced daylight savings time deserved a swift kick in the crotch...

      I'm not a morning person at the best of time, that shift in the summer then makes me feel like crap for months afterwards. Then when my body finally starts to adjust we go back to "normal" time.

      1. Philip Storry

        Re: And we have...

        His name was William Willett.

        You're a bit late to kick him in the crotch, as he's been dead for just over a century.

        https://en.wikipedia.org/wiki/William_Willett

        However, there is a memorial to him in Petts Wood. I've been past it a few times, and was astonished as to how free of vandalism it is, all things considered.

        I just read the Wikipedia article, and discovered that the man is also the great-great-grandfather of the lead singer of Coldplay.

        It really is a ***ing miracle that the memorial hasn't been blown up by now, isn't it?

        1. imanidiot Silver badge

          Re: And we have...

          I know he's dead, that's why I said deserved instead of deserves.

          As for the memorial, the man is dead, no need to go vandalise a grave over something like this.

          1. Guus Leeuw
            Pint

            Re: And we have...

            you are indeed, Sir. Your respondent mentioned memorial, not grave.

            Friday an' all that. Beer time!

  3. Your alien overlord - fear me

    half an hour out by 2700? Tell someone who gives a f*ck !!!

  4. deive

    We've used seconds for over a thousand years, now we know they are not quite right. So what does half the world do? Bury their head in the sand and say "change is scary and expensive".

    We will care about this as a space-fairing species. Just a few seconds off on navigation to another planet could leave you stranded!

    1. Dave Bell

      A modest timekeeping proposal

      If spaceflight is the problem, let's switch to Kerbal Unified Nominal Time. (6 hours per day, 400 days per year)

      1. ravenviz Silver badge
        Devil

        Re: A modest timekeeping proposal

        Unfortunate acronym there!

  5. John Robson Silver badge

    So basically

    We are having a discussion to decide whether to keep a time standard that has only existed for a fairly short time (1765 - so less than the 600 years we are worried about into the future).

    Yes, I'd like Greenwich to continue to be the centre of time, but it is just an arbitrary point on this roughly spherical planet. It isn't even at 0 degrees according to the shperoid used by GPS (no idea what shperoid is used by GLONASS/GALILEO etc)

    Given that we can measure the variation in day length caused by the wind on the mountains - is "high Noon" really important? Cornwall is already 15 minutes "out", so we're talking about "High Noon" moving by ~600 miles in ~600 years (at our latitude)

    I'd do away with BST as well, we could all agree to get up an hour earlier or later during some months if we really want to... Office hours are 9-5 in the Winter and 8-4 in the Summer. There, how hard was that?

    1. JeffyPoooh
      Pint

      Re: So basically

      The Prime Meridian has been moving around a bit over the years. You can see the different slots for the different instruments over the years. And now the WGS-84 one is something like about 100 feet from the ceremonial brass track.

      edit: Wiki says, "The WGS 84 meridian of zero longitude is the IERS Reference Meridian, 5.31 arc seconds or 102.5 metres (336.3 ft) east of the Greenwich meridian ..."

  6. John Robson Silver badge

    Sorry - GMT wasn't recognised as *the* prime meridian until 1884

    So it's only 130 years old...

  7. Justicesays

    It would save alot of effort

    To just move the Mean every 1400 years, and everyone can adopt the timezone one along at the same time (or go a .5 after 700)

    Well, it would certainly save us a lot of effort, and those distant decedents will either be cavemen, cybernetic gods, or non-existent. In each case there wont be much problem for them!

    Unless all the cybernetic gods crash due to an undetected leap hour issue.

    1. ravenviz Silver badge
      Angel

      Re: It would save alot of effort

      Maybe just stop the clocks in 2700 for half an hour, I'm sure everyone could do with a nice cup of tea and a sit down.

  8. Christoph

    The simplest solution

    By the time that the clock drift gets important we will have much better space drives. Even if it's just a gigantic solar sail. Then we can adjust the planet's rotation to match the clocks.

    1. Brewster's Angle Grinder Silver badge

      Re: The simplest solution

      And what about the moon's?

  9. Yugguy

    Do not forsake me oh my darl...

    Ah, sorry, wrong High Noon.

  10. orion216

    In the late 17th century something unthinkable happened when they decided to link the Earth's rotation directly to stellar circumpolar motion and came up with a fiction known as 'solar vs sidereal' time where to attempt to reason out that the Earth turns once more often than there are 24 hour days across an orbital circuit -

    " It is a fact not generally known that,owing to the difference between solar and sidereal time,the Earth rotates upon its axis once more often than there are days in the year" NASA /Harvard

    Anyone who goes to the main Earth article in Wikipedia will see this horrible notion of 366 1/4 rotations per circuit thereby making a mockery of both astronomy and timekeeping -

    "During one orbit around the Sun, Earth rotates about its own axis 366.26 times" Wiki

    https://en.wikipedia.org/wiki/Earth

    Timelords indeed !, they are even worse than the Middle Eastern thugs wrecking havoc with an astronomical and timekeeping heritage that stretches back many thousands of years.

    Let me ask you all a simple question, one so fundamental to human nature that it almost impossible to avoid. What is responsible for the appearance of the Sun followed by the appearance of the stars within each 24 hours ?. if any of you come up with one more rotation than there are 24 hour days then welcome to the warped world of theorists and the awful 'solar vs sidereal' fiction which tries to split rotations to the Sun and rotation to the stars into separate motions !.

    1. Justicesays

      Well... it works like this.

      If the planet was tidally locked, such that there was continual day in some places and continual night in others, the planet , as a whole, would have to perform one complete revolution during a solar year for that to happen. I.e. if you look from above, and say Britain was the center of the day light side, once 6 months had passed it would be facing the other direction, as seen from the observer off the planet/sun complex itself. This "offplanet" position is generally located in reference to the "fixed stars".

      Thus your extra revolution.

      HtH

      1. orion216

        Re: Well... it works like this.

        The fraud is not the 'solar vs sidereal' time fiction from the late 17th century which tried to assert the Earth turns once in 23 hours 56 minutes 04 seconds and subsequently the attempt to create a mismatch between 24 hour days and the number of rotations within an orbital circuit, the fraud is recent as they now conjure up a new fiction where they try to introduce a non cyclical reference based on the notion of a slowing Earth and an idealized rotation once in 24 hours back in the year 1820 -

        "At the time of the dinosaurs, Earth completed one rotation in about 23 hours," says MacMillan, who is a member of the VLBI team at NASA Goddard. "In the year 1820, a rotation took exactly 24 hours, or 86,400 standard seconds. Since 1820, the mean solar day has increased by about 2.5 milliseconds."NASA

        http://www.nasa.gov/topics/solarsystem/features/extra-second.html

        The anchor for timekeeping is not only the single rotation that causes the appearance of the Sun followed by the appearance of the stars with each rotation but also how many times this event occurs within the confines of an annual circuit. It turns out it takes 4 annual circuits and an extra rotation known as a 'leap day' to determine where timekeeping and planetary dynamics converge to a close proximity. By simple division the 1461 rotations for 4 orbital circuits reduces to 365 1/4 rotation per orbital circuit as an inviolate proportion.

        If people wish to work through the reasoning where the 24 hour system and the Lat/Long system work together and originate in an even older narrative from timekeeping which includes the external references for the leap day correction corresponding to an additional rotation they will do themselves and this world a huge favor.

    2. Naselus

      "'solar vs sidereal' fiction"

      Don't be obtuse. If the Earth didn't rotate at all, then it would still rotate once in relation to the Sun during the course of it's orbit. As with most spacy things, you need ot think in terms of frames of reference.

      1. orion216

        Naselus

        The 'solar vs sidereal' fiction doesn't just assign a 23 hour 56 minute 04 second value to the rotation of the Earth, it extends it on to the idea that the Earth rotates 1465 times in 1461 days. If you want a frame of reference then begin with the cause behind the appearance of the Sun followed by the appearance of the stars with each rotation and the look for the external reference which locks the number of rotations into an orbital circuit.

        1. Justicesays

          Rotation has an absolute frame of reference

          Unlike linear movement, rotational motion is deemed to have an absolute frame of reference (fixed stars again), therefor any mention of "rotation" in stellar terms uses this fixed frame of reference (giving the 366.26 rotations per year of the earth using the method I detailed above).

          https://en.wikipedia.org/wiki/Absolute_rotation

          Any other frame of reference for this kind of observation is not a valid one, due to obvious implications for things like stellar/planetary radius, slingshots and a whole bunch of other stellar and planetological phenomena

          1. orion216

            Re: Rotation has an absolute frame of reference

            I assure you the appearance of the Sun at dawn followed later by the appearance of the stars is due to a single rotational cycle of the Earth, one rotation per 24 hour day and a thousand rotations in a thousand days - everything else is forensics including the mess known as the 'leap second'.

            The fundamental unit of timekeeping is not the 24 hour day, hours,minutes or seconds nor is it the rotation of the Earth as the present academic dummies like to believe, it is the proportion of rotations that fit inside an orbital circumference. It is an extremely old observation which uses a specific astronomical event even though the observation was noted in the number of days needed to keep the festivals drifting through the seasons -

            ".. on account of the procession of the rising of Sirius by one day in the course of 4 years,.. therefore it shall be, that the year of 360 days and the 5 days added to their end, so one day shall be from this day after every 4 years added to the 5 epagomenae before the new year" Canopus Decree 238 BC

            The reason that Sirius skips an appearance by one day/rotation after four cycles of 365 rotations is that rotation and orbital motion are two separate motions. The first appearance of Sirius defines the Earth's orbital position in space and the number of times it takes the Earth to turn to reach that position. We use full rotations and days as a gauge so the fractional difference of a 1/4 rotation and the corresponding orbital distance traveled each circuit doesn't show up until the 4th cycle of 365 days/rotations where Sirius skips an appearance by one day/ rotation.

            https://www.youtube.com/watch?v=eeQwYrfmvoQ&list=PL-AwEPTQXEwutkW2QQKNtRJGBFyA41wbt

            Whether any of you care to know it or not, we format the parent cyclical observation that there are 1461 rotations in 4 annual circuits into the manageable calendar format of 3 cycles of 365 rotations and 1 cycle of 366 rotations.

            If you get into trouble with the ins and outs of these things, and I haven't even got into the 24 hour system and the Lat/Long system, then anchor your beliefs in the fact that the appearance of the Sun followed by the appearance of the stars is a single sweep of the Earth's rotation.

            1. Justicesays

              Re: Rotation has an absolute frame of reference

              "If you get into trouble with the ins and outs of these things, and I haven't even got into the 24 hour system and the Lat/Long system, then anchor your beliefs in the fact that the appearance of the Sun followed by the appearance of the stars is a single sweep of the Earth's rotation."

              Or, you know, use the actual physics as I pointed out, I don't really have trouble with my "in's and out's" and I don't have some kind of deeply held belief that the world revolves around me...

              Are you a follower of spaceology or something?

              1. orion216

                Re: Rotation has an absolute frame of reference

                The people promoting the 'leap second' have a deep belief that there is one more rotation than there are day/night cycles across an orbital circuit and that is intellectual nadir for any society.

                The Earth's Equatorial circumference is 24901 miles and split into 15 degrees of geographical separation amounts to 1037.5 miles and one hour's time difference - this is known as both the 24 hour system and the Lat/Long system as both are organized around the Earth's daily rotation. If circumstances were normal, the fact that the rotation rate of the Earth is 15 degrees per hour and at the Equator that corresponds to 1037.5 miles but this is no normal era, nay, it is close to being dystopian.

                The 24 hour system is derived as an average of each natural noon cycle given that each noon cycle varies in the length of time it takes the Sun to cross the observer's meridian to the next noon. It is because the word 'average' and 'constant' share a common meaning that the average 24 hour day substitutes for 'constant' rotation at a rate of 15 degrees per hour or 4 minutes for each degree of rotation.

                The Middle Eastern thugs blowing up ancient temples look like gentlemen compared to what is going on currently with the heritage of human timekeeping and its links to astronomy and terrestrial sciences with this 'leap second' business merely a continuation of vandalism.

  11. Ed 13
    WTF?

    Until we become a spacefaring race...

    ...and a significant fraction of the population don't live on the surface of the earth, then most of the race will synchronise their day with the rising and setting of the sun. So keep the leap seconds.

    It's the systems we invented that don't quite keep step, so they should be adjusted.

    1. Paul_Murphy

      Re: Until we become a spacefaring race...

      and a significant fraction of the population don't live on the surface of the earth, then most of the race will synchronise their day with the rising and setting of the sun

      That makes no sense - the rising and setting of the sun? where - in space?

      Once (if) we are a space-faring race we will need to have a time system that is robust enough to handle communication lag over light seconds and minutes, that people can agree on and will mean the same thing where ever everyone happens to be.

      What's needed is a far away slow pulsar that can be used as a metronome, or to discover some thing similar to a half-life of reality.

      Answers on postcards would almost certainly not be good enough :-)

      1. ravenviz Silver badge
        Boffin

        Re: Until we become a spacefaring race...

        There are also added complications with spacetime, how, or if, to take time dilation into account.

  12. John Savard

    Obvious Answer

    What did we do before we had leap seconds? Well, we used seconds that were ever so slightly longer than the SI second, or the second of Ephemeris Time. We can go back to that system, with time signals informing people of the length of the second of civil time for the current year, and with a separate time signal for atomic time.

    Of course, the old system might be amended somewhat to make it easier to adjust modern digital time sources to follow it.

  13. billse10

    "midday at Greenwich will be half an hour off"

    no, it won't. It'll still be midday at Greenwich.

    hello from guess where ...

  14. Quotes

    Why don’t we adjust the length of 1 second by a tadge?

    1. Anonymous Coward
      Anonymous Coward

      Because we've defined it in terms of oscillations in the caesium atom and that's tied to the speed of light in vacuum.

      1. cortland

        " tied to the speed of light in vacuum."

        With hemp twine!

    2. Anonymous Coward
      Anonymous Coward

      Clearly your tadge does not last as long as mine. ;-)

    3. Ed_UK

      "Why don’t we adjust the length of 1 second by a tadge?"

      As Brian said, it's been linked to a specific time-period of an emission by caesium. It's fair to ask why.

      I imagine the Time People were looking to link the second to something which fitted several criteria:

      1: It has to be really, really close to the earlier value of the second.

      2: It has to be reproducible, so that other peeps can have their own accurate seconds

      3: It has to be really stable; not drifting or jittering over time.

      The caesium emission fits these criteria, but the rotation of the earth cares little for our precise seconds and its period even has small, random fluctuations. So, every now and then, we have the leap second to keep our clocks in agreement with the earth's rotation.

      Executive summary: We now have a very accurate second but a rather inaccurate planet.

  15. CAPS LOCK

    It seems we need two words...

    .. because there are two types of second , one the varying solar second, and two the equivalent to a fixed number of oscillations of a natural phenomenon. I propose the name 'Wibble' for the fixed version. Alternative proposals invited...

    1. Martin Budden Silver badge
      Pint

      Re: It seems we need two words...

      With 100 Wibbles in a Wobble, obviously. Wibbley-wobbley timey-wimey!

      Actually your suggestion is the best: ordinary people can measure what time they can knock off for a pint on a Friday afternoon by using the same system we have now, with leap years and leap seconds as required to keep beer-o'clock at the same point in the day*. Meanwhile, astronomers and other boffins can measure accurate passing of time using a new set time measurement which never gets adjusted. Two systems for two purposes.

      *although it's always beer-o'clock somewhere!

  16. GX5000

    2070...The year South Korea also predicts its population will be extinct...and our clocks will be off by thirty minutes if we all ignored this (as if). Can we just calm down and stop hunting checkbooks for grants over a small software adjustment that can be automated, even if it means being off for an insignificant amount of time for a few days ? Oh the Stock Exchange scams, oh right, it's all about Money and scams isn't it.

  17. Doctor Syntax Silver badge

    This conference...

    ...what time does it start?

  18. Anonymous Coward
    Alert

    Does anyone really know what time it is; does anyone really care?

    The leap second is silly. If we must adjust clocks, we should add a leap minute every century.

    At the current rate of 1 second every (approximately) 18 months, the 2100 leap minute would put us slightly ahead and subsequent leap minutes would have us gradually falling behind. This could be adjusted with the addition of several leap minutes at the next millennium.

    This will give programmers a predictable algorithm to use for time and 85 years to change their programs. Okay, lets be real, they'll start working on it in 2099 and Y3K will still be a crisis of epic proportions.

    I'll await my MacArthur genius grant.

  19. swm
    Joke

    Let's Be Consistent

    Of course the (very accurate time-keeping) GPS satellites don't use leap seconds and seem to do all right. They even have to adjust for general relativity effects. So let's just use GPS time.

  20. cortland

    How precise do we have to BE?

    I don't much care about leap seconds when I reset a cuckoo clock from Daylight time.

    "measure it with a micrometer, mark it with chalk, and cut it with an axe"

    Anyone who leaves his accounting and management programs vulnerable to one second errors deserves what happens. If you INSIST on mm accuracy from a bloke with a spade you may find he's accurate enough to lay it, as we say here, up alongside your head.

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