Stories
Slash Boxes
Comments

SoylentNews is people

posted by hubie on Sunday November 13, @10:37PM   Printer-friendly
from the everything-is-connected dept.

Today the xkcd: Y2K and 2038 comic was published and this reminded me of the recent very good technical blog post Time is an illusion, Unix time doubly so... from Jan Schaumann where he explains how time is handled on different operating systems including some historical background.

A famous scientist and adventurer once said: 'time is not linear but something like "Wibbly Wobbly Timey Wimey"'. He has since been proven more correct than he ever imagined.

As you well know, on Unix systems we measure time as the number of seconds since "the epoch": 00:00:00 UTC on January 1st, 1970. This has made a lot of people very angry and been widely regarded as a bad move.

For starters, this definition is not based on something sensical such as, say, the objective frequency of vibration of a Cesium-133 atom, but on a convenient fraction of the time it takes a particular large rock to complete a full rotation around its own axis.

If you want to learning about any of this:

  • Initially the time was measured in 1/60ths of a second
  • At least one country has two different utility frequency
  • Why Linux will fail again on 23rd April 2262 even with 64-bit counters
  • How different operating systems behave around the beginning or end of the epoch
  • What will happen with positive or negative leap seconds

then click here and read this fine blog posting.

Happy reading and learning!


Original Submission

This discussion was created by hubie (1068) for logged-in users only, but now has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 0) by Anonymous Coward on Monday November 14, @12:06AM

    by Anonymous Coward on Monday November 14, @12:06AM (#1279577)
    Just been trying to get my head round atomic time, GPS time and UTC. Yep, it's a mess! And when Alaska was part of Imperial Russia, GMT+15, anyone? (Could be worse [nationaltrust.org.uk], though.)
  • (Score: 4, Funny) by Opportunist on Monday November 14, @12:30AM (1 child)

    by Opportunist (5545) on Monday November 14, @12:30AM (#1279578)

    2038 is about a year or two before my retirement (I'll be about 62 by then), so you bet I'm looking forward to it. This is gonna be a sweet little nestegg to carry me into retirement, baby!

    • (Score: 2, Touché) by Anonymous Coward on Monday November 14, @08:02AM

      by Anonymous Coward on Monday November 14, @08:02AM (#1279618)
      Username checks out.
  • (Score: 3, Interesting) by bzipitidoo on Monday November 14, @01:51AM (3 children)

    by bzipitidoo (4388) Subscriber Badge on Monday November 14, @01:51AM (#1279584) Journal

    Now that we know a lot more about the Earth's movement, maybe we should update our timekeeping? The way leap seconds are added is arbitrary. We could do better.

    The Gregorian calendar is based on the flawed premise that the Earth is eternally unchanging. But now we know the Earth's rotation is gradually slowing. Shouldn't be too hard to model that mathematically. Why not do so, and build that into our timekeeping systems? That should last until we become a type II civilization capable of greatly altering the Earth's rotation and orbit.

    Shouldn't have to do anything as drastic as they did in 1582, dropping a week and a half. In 1582, there was no Oct 5 through Oct 14.

    • (Score: 4, Informative) by Anonymous Coward on Monday November 14, @02:08AM

      by Anonymous Coward on Monday November 14, @02:08AM (#1279585)

      But now we know the Earth's rotation is gradually slowing. Shouldn't be too hard to model that mathematically. Why not do so, and build that into our timekeeping systems?

      The long term trend will be a slowing rotation, primarily due to tidal forces from the moon, but this is not necessarily true on short-term scales relevant to human activity. In fact, over the last several years the Earth's rotation has been speeding up year over year.

      Moreover, geological events like earthquakes can have significant effects on the earth's rotation and are (probably) not predictable with mathematical models like you describe.

    • (Score: 2) by Opportunist on Monday November 14, @10:35AM (1 child)

      by Opportunist (5545) on Monday November 14, @10:35AM (#1279625)

      We should create a conspiracy nuttery around these missing dates.

      • (Score: 2) by maxwell demon on Tuesday November 15, @05:51AM

        by maxwell demon (1608) Subscriber Badge on Tuesday November 15, @05:51AM (#1279817) Journal

        Yeah, clearly the illuminati want us to believe those dates don't exist because they did something very evil at those dates, and if that evil action is leaked, nobody is going to believe it because it happened at dates that "don't exist". :-)

        --
        The Tao of math: The numbers you can count are not the real numbers.
  • (Score: 3, Interesting) by Thexalon on Monday November 14, @03:41AM (7 children)

    by Thexalon (636) Subscriber Badge on Monday November 14, @03:41AM (#1279594)

    Just because those guys 5000 years ago didn't understand fractions, they picked numbers that were easily divisible, and that's how we landed on a 12-hour day and 12-hour night (later turned into a steady 24-hours), 60 minutes to an hour, and 60 seconds to a minute. And of course some minutes now actually have 61 seconds, how fun is that. ISO 8601 also has to accommodate a calendar system still influenced by Roman political considerations from approximately 40 BCE along with some religious considerations from the mid-400's CE.

    It's all a big mess, and will drive anyone nuts who really tries to wrap their brain around it.

    --
    The only thing that stops a bad guy with a compiler is a good guy with a compiler.
    • (Score: 4, Insightful) by mhajicek on Monday November 14, @06:23AM (5 children)

      by mhajicek (51) Subscriber Badge on Monday November 14, @06:23AM (#1279606)

      I think it's that they did understand fractions.

      But the problem we have is feature creep. When time was implemented, it met the design requirements very well. But then people keep asking more and more from it, and it can only be patched so much until it needs a fresh rewrite.

      --
      The spacelike surfaces of time foliations can have a cusp at the surface of discontinuity. - P. Hajicek
      • (Score: 3, Interesting) by gnuman on Monday November 14, @10:41AM (1 child)

        by gnuman (5013) on Monday November 14, @10:41AM (#1279626)

        The requirements are rather simple,

        1. Interval of time, called a second, is to be fixed. And now it's even defined to some fixed value.
        2. Interval of time, called a day or a month, is to have fixed number of the above interval.

        The problem is that day and month and year, they are actually changing. It's not because we didn't measure it properly -- they actually change due to way that the earth is changing. And so this mapping from #2 to #1 and the other way, that's a problem.

        As to a fresh rewrite?? I think US transitioning to metric system would be much simpler than telling people that 1 day is suddenly be referred to as something else and that it starts at a different time every day because earth rotation is not fixed but interval must be fixed.

        • (Score: 2) by mhajicek on Monday November 14, @06:35PM

          by mhajicek (51) Subscriber Badge on Monday November 14, @06:35PM (#1279711)

          Those were not the original requirements.

          Originally, one needed to know rough hours, and the seasons, to plan the work on the farm. Everything else is feature creep.

          --
          The spacelike surfaces of time foliations can have a cusp at the surface of discontinuity. - P. Hajicek
      • (Score: 2) by Thexalon on Monday November 14, @11:42AM (2 children)

        by Thexalon (636) Subscriber Badge on Monday November 14, @11:42AM (#1279633)

        So looking into it further: Base 60 does look like it was chosen for its divisibility into 1, 2, 3, 4, 5, and 6 (as well as a bunch of other numbers). But yes, the Sumerians did have fractions, but closer to 1800 BCE, while they had math a couple of millennia before that, so they probably went through a period where they didn't have fractions because that's not something you think about right away when you first start counting things.

        --
        The only thing that stops a bad guy with a compiler is a good guy with a compiler.
        • (Score: 2) by mhajicek on Monday November 14, @06:31PM (1 child)

          by mhajicek (51) Subscriber Badge on Monday November 14, @06:31PM (#1279710)

          It isn't? I collected some apples: let's count them so I can give you half.

          --
          The spacelike surfaces of time foliations can have a cusp at the surface of discontinuity. - P. Hajicek
          • (Score: 2) by Thexalon on Monday November 14, @09:12PM

            by Thexalon (636) Subscriber Badge on Monday November 14, @09:12PM (#1279749)

            Division, yes. Fractions, no.

            Let's say for your example problem there are 16 apples. We want to divide them evenly between two people, that's pretty easy: You put 1 in my pile, 1 in your pile, repeat until you're done. And you count both piles of apples to make sure they're the same.

            Now let's make it 17 apples. We'll do the same thing, but we'll notice that 1 of us has 1 more apple than the other. So we'll pull that out and make a new plan for that one. (i.e. discovering the concept of a "remainder").

            It's a bit of a leap to get from that to "OK, maybe we can cut the extra apple perfectly in half". And even more to generalize that if we are divying up among 3, 4, 5, etc people. And there's some interesting stuff to think about when you're trying to treat 4 quarter-apple pieces as the same as 1 apple, which they kinda are but kinda aren't.

            And it gets even harder to arrive at that solution when you're divying up goats rather than apples.

            --
            The only thing that stops a bad guy with a compiler is a good guy with a compiler.
    • (Score: 2) by bradley13 on Monday November 14, @01:15PM

      by bradley13 (3053) Subscriber Badge on Monday November 14, @01:15PM (#1279647) Homepage Journal

      "...didn't understand fractions

      That's not it at all. They picked 60, because of trade. If you ship 60 pots to the next city-state over, your dealer their can resell 1/2 of them, or 1/3 or 1/4 or 1/5 or 1/6. That's incredibly flexible.

      Nobody wants a fraction of a pot.

      --
      Everyone is somebody else's weirdo.
(1)