I am a proponent of a Martian timekeeping system where each day has 24 hours 39 minutes and 35.244 seconds and all the rest of time units have standard lengths. For example, at midnight your martian clock would show the time as 24:39:35 for a fraction of a second, then wound back to 00:00:00. The calendar then would track the Earth calendar as close as possible, with shortened months to accommodate for the longer days.
Basically, everything is roughly in sync with Earth and uses similar or identical units, except the time of day which is synced to the martian solar day, the only local time of relevance to the martians, for things like working light and circadian rhythm, solar panel irradiation etc. The "seasons" of Mars are of little relevance due to the very artificial conditions in which the martians will live, the weather outside is deadly all year long with slight variations in deadliness.
So the entire idea of keeping a martian calendar with funny month names doesn't really make sense, the only thing martians will care is the time of the next Earth transport and the dates on Earth so they can relate with their loved ones. As is the idea of having different lengths of seconds and hours, which will render almost all Earth equipment and measurements unusable, as opposed to a slight tweak in the software of their clocks to enable times past 23:59:59.
> The calendar then would track the Earth calendar as close as possible, with shortened months to accommodate for the longer days.
Why bother with months? Why not just count weeks? You could use something like the ISO 8601 week calendar – the year is divided into weeks. Weeks run Monday (1) to Sunday (7). Every year beings on a Monday – ISO 8601 has two years, the Gregorian year (starts on 1 January) and the ISO week year (always starts on Monday). The ISO week year always contains a whole number of weeks, either 52 or 53, so the first day of the ISO week year is often a few days before or after 1 January. 2023-W09-2 is an ISO week date, it is the 2nd day (Tuesday) of the 9th week of 2023.
So, here's my idea for a Martian calendar. The base unit is the Martian day (sol) of 24h39m35.244s. 7 Martian days is a Martian week. The calendar year is based on the Earth year, but it starts on the Monday of the Martian week closest to 1 Jan 00:00 UTC.
> You could use something like the ISO 8601 week calendar – the year is divided into weeks.
Note that ISO 8601 weeks can be non-intuitive: per the Gregorian calendar, December 30, 2019, was in 2019; but per ISO 8601, December 30 it was in Week 1 Year 2020. Tom Scott:
Despite this, I'm an advocate for martian self-determination when it comes to deciding their timekeeping system. Earthlings, despite their good intentions, should stay out of the matter categorically.
Everyone says this but reality never seems to work out that way when push comes to shove.
Will you be singing that tune when a religious minority you hate moves there, establishes a theocratic government, starts teaching women to read, hanging undesirables and troublemakers and regressing on pretty much every human right your ancestors died for, bonus points for starting a war with the locals.
That said, in the historical example I'm referencing the state was too weak to do much about them from afar and they did clean itself up after subsequent immigration diluted their culture over the following 100yr. God what a bunch of assholes they were in the meantime though...
I refuse to get involved in Martian politics. You can CIA them however you want though. I'm sure you can find just as many instances where foreign meddling and infantilization of people had disastrous consequences.
You'll be happy to know that you can appreciate their input on Earth matters as much as they'll appreciate yours on martian matters.
I agree; but unless we discover some inteligent indigenous species, Martian self determination can happen only if earthlings will go and live on Mars. And for that, they will need clocks, clocks made on Earth by other earthlings.
Cue sci-fi dystopia where the totalitarian Mars gov't gases (or somehow otherwise knocks out) everyone for 40 minutes while they search the place and haul off dissidents.
> I prefer KSR's "time gap". At midnight on mars, everyone just gets a 40 minute period where the clocks don't move.
For anything legal, scientific, etc., dealing with time or speed (tracking experiments, wage and hour laws, etc.) you need to be able to track time during the clock pause, which will lead to adoption (by social convention, if nothing else) of a standard method of recording times within the pause, which leads to clocks that display in that format times during the pause. “All clocks stop for 40 minutes” is not a stable equilibrium.
It's basically the same thing, but harder to achieve technically because many things assume clocks are moving all the time. For example timers and timeouts are frequently implemented as loops polling system time has reached a certain future value. Many things break or malfunction if a large number of things happen at the same time, for example logging, crons etc. Imagine investigating a cybersecurity event perpetrated in the gap and finding all file dates and logtimes as unreliable.
Probably, but even Windows would break in this scenario. It interprets the hardware clock as being in local time and derives UTC from that (you can change that with a registry change, but it isn't very reliable in my experience)
It's worth noting that even atomic clocks run on Mars faster than on Earth, due to the different gravity. So you might expect say half of the 50 microsecond daily gain quoted in this article for the Moon, 10 miliseconds per year or a full second per century, more than enough to make you your spaceship completely miss its intended orbit. And martians will definitely want to run their own atomic clocks, making sync very problematic.
So TAI is not the end all for what I expect to be a very complicated technical problem.
> It's worth noting that even atomic clocks run on Mars faster than on Earth, due to the different gravity.
The biggest effect is the Sun's gravity, followed by orbital velocity. The relative masses of the planets is negligible by comparison, though in online discussions that's almost all anybody ever seems to talk about.
I can't find confirmation through other sources, and can't credibly validate it myself, but the best estimate I found for time dilation between two clocks on Earth and Mars is roughly 5.6 parts per billion, or ~177 milliseconds peryear. See https://space.stackexchange.com/questions/33590/what-is-the-...
Wow, that's huge, and the Earth - Mars offset is likely perturbed by things like the motion of Jupiter etc. Keeping the time representations compatible and in sync will be a full time job for many people.
> Some special boffins on Mars would run atomic clocks, and worry about synchronisation with earth.
I think at some point, you have to redefine utc to be based on an estimated atomic clock at the center of the sun, because it's easier for mars and earth to estimate that, than for earth to go on its merry way and mars to estimate the time on earth. Relativism is fun?
How is stopping the clock for 40 minutes human friendly? I'm a night owl and there are many things in my life that happen between 00:00 and 00:40, can't imagine pretending they all somehow happen in the same moment.
I have had to make my computer believe that I was in Japan to play certain videogames.
There was also that fascinating time I had to roll back the calendar to use a crack for a Denuvo protected game.
IP doesn't have a requirement for clock synchronization. Certificate validation does need clock synchronization, but most of the time rough synchronization works; you only need precise synchronization if you present certiticates close to their not-before or not-after dates.
> each day has 24 hours 39 minutes and 35.244 seconds
> at midnight your martian clock would show the time as 24:39:35 for a fraction of a second
A similar idea could be to have exactly 24h 39min 35s each day (i.e. the last second before midnight is 24:39:34), then add a leap second every 4 days to compensate for the missing 0.244s, then skip it when the 0.006s errors cumulate up to 1s.
This just because in my opinion it is easier to live with leap seconds than with non-integer seconds.
Is the difference in quality of life so dramatic with a day wrapping on non-integer seconds when both proposals wrap at non-integer minutes and non-integer hours?
The real question is why stop at miliseconds - it seems inconsequential. Go all the way down and assume that as the hour/minute/second granularity all proved insufficient in the past, the same way milliseconds would become insufficient one day. Make the period parameterizable (i.e. not a constant), with only the current value at 24 hours 39 minutes 35 seconds 244 milliseconds.
Do you honestly consider modifying the core idea of the 24 hour day to be a "slight tweak"? I can't even imagine how many different things rely on that assumption.
(and I'm sorry, but what does "naming months and seasons in Mars" have to do with this?)
You have basically three options for timekeeping on mars: adjust the day, adjust the second, or give up on days syncing with the (apparent) movement of the sun.
The first option breaks mostly assumptions made in software and processes, the second mostly assumptions made in hardware, and the third breaks human's circadian rythm (a 23 or 25 hour rythm isn't a big deal, but it has to sync to light).
Of those three, the things that rely on 24 hour days seem to be the easiest to change. Though there is precedent both to messing with the length of a day (a leap second makes the day one second longer, making 23:59:60 a valid time) and the length of a second (google pretends like seconds are slightly longer as an alternative way to deal with leap seconds)
Sorry about that, I was just throwing somewhere at the bottom of the comment section an off-topic ideea I've been toying with. Didn't expect it will raise such interest and be voted towards the top.
Oh, I didn't realise that this proposal had been adopted!
I read[0] that "It will take about 50,000 years for a mean solar day to lengthen by one second (at a rate of 2 ms per century)" and "The [accumulated] difference between UTC and UT would reach 0.5 hours after the year 2600 and 6.5 hours around 4600" which I expect will be sufficient to take us through to the time where we will no longer need co-ordinated time or an IERS.
> and I'm sorry, but what does "naming months and seasons in Mars" have to do with this?
Time of day and calendaring and discussed together because they allow you to ask things like "what was the date and time on Mars prime meridian on Feb 2 2023 6:00 UTC" and get a recognizable date and time back. It's essentially a timezone, albeit a very strange one with a variable offset.
> modifying the core idea of the 24 hour day to be a "slight tweak"?
Ok, maybe an exaggerated hand-wave on my part. It's a slight software tweak on my desk clock. I was contrasting with the alternative of modifying a fundamental SI unit and more or less rebuilding from scratch all hardware sent on Mars. The fuckups alone resulting from confusing Earth seconds with Mars seconds.
For software that runs on Mars, I think "Mars date/time" has to be a brand new set of APIs and data types. The existing date/time APIs/types are for Earth times only. The "local time zone" on Mars would just be UTC. That way there wouldn't be much possibility of bugs based on confusing Earth and Mars time scales – all pre-existing software would still work with Earth date/times, and anything that needed to deal with Mars date/times specifically would have to be modified to handle them.
I think you misunderstood my point - I was saying make both Earth UTC and some Martian time scale available, just use a new API and data types for the latter. Which one to use in any given use case will be up to the application developer.
Sure, the leap second is a thing. We still represent the day as 24 hours for humans. They didn't seem to be talking about an "under the hood" hand waving of a significant part of the day.
Nobody is going to be having real-time Zoom calls between Earth and Mars so I'm not sure how in sync it needs to be.
It does make sense to at least keep the second the same since that is a critical base SI unit that affects lots of other things.
One interesting problem will be that high precision atomic clocks will run at a slightly different rate on the surface of Mars due to the lower gravitational potential, but this is easily solvable as it is on GPS satellites.
> Nobody is going to be having real-time Zoom calls between Earth and Mars so I'm not sure how in sync it needs to be.
If a contract say "the payment must be received before March 1, 2123", then you have to worry about time syncing between the planets. How fast are fund transfers between planets?
Wouldn't that get incredibly confusing with the calendar always being slightly off, despite using the same names? At that point why not just stick to a integer day count for the date.
Having a non-integer number of hours in a day might also get confusing. It would be near-impossible to make an analog clock, for example, and stuff like dividing up a day into three shifts would be really cumbersome. For day-to-day life redefining the second might end up being a lot easier.
> Wouldn't that get incredibly confusing with the calendar always being slightly off, despite using the same names?
I think you can keep the calendar within +/- one day of a certain Earth location by carefully adjusting the length of the months. So it would be much like the dates vary on Earth between time zones, for example at the time I'm writing this the date in Hawaii is March 1st, while my local date is March 2nd. But less predictable, yes.
> It would be near-impossible to make an analog clock, for example,
Why would that be an issue? You still have the hours hand make a full rotation in one period, and you make markings at each hour. You just have separate markings at 24 o'clock and 0 o'clock.
Separate inner markings, like on most 24 hour earth clocks I find on Google images. When you've lost clean divisibility with 5, conflating the dials to make every hour house 2.5 minutes is already not a nice design.
Of course, you need mechanically handle the jumping ahead of the minutes and seconds arms.
What you get tends to look like a very precise chronometer, the type a Mars scientist would wear.
Still goes at 1 rotation per hour. You put the minute markings on another wheel to precess each day. The hour and minute hands won't form the same angle at the same time each day, but it will still tell the correct time.
The problem is, what do you do at 24:39:35, when both the minute and the second arm have more than a third of the way to go before zero?
Any mechanical design where units are not cleanly divisible would have to devise a way for the arms to "fast forward" over the gap and accept nonsensical results for a few seconds before midnight.
Like I said, the arm keeps moving the way it already was, it doesn't jump forward because time isn't jumping forward. You shift the position of zero over. So rather than zero being at the top of the clock, the minutes marks spin about a third of a rotation and now where the minute hand is pointing is zero. It's no different from having a "day of the week" dial on a watch which jumps forward one tick every day.
It seems to me that if you rotate the minutes dial in a quick motion before midnight, you have both the "nonsensical times" problem for a few seconds and also a massive usability problem throughout the day, since the minutes mark are not where people expect them, the minute numbering is upside down or nonexistent, and the visual cues of vertical = half hour intervals, horizontal = quarters are all gone, killing the one of the killer features of the analog clock.
I can't see how that solution is superior to fast forwarding just the much lighter and already mobile minutes arm, but perhaps I'm missing something.
Well it really comes down to what you are trying to represent - moving the minute hand works well if you are only concerned with subdivisions of the hour, but if you are trying to keep track of minutes, you want to know how many minutes have passed between one point and another, so for example if I am waiting for something to cook for 45 minutes, I want to wait for the minute hand to make 3/4 of a full rotation. If the minute hand jumps a significant amount, then that no longer holds true. By moving the marks instead of the hand, you keep track of the passing of minutes naturally.
I don't personally agree that it would be a usability problem - people already use wrist watches which aren't perpetually oriented up, they naturally find the zero point and understand the position of the hands relative to it. The only difference is that the zero point for the minutes will not generally be the same as the zero point for the hours. I certainly don't buy the "it's not the UI we're used to so it is a worse UI" argument. Mechanically, incrementing the marks is much simpler - I'm not even sure how you would get jumping the minute hand forward to work normally nonetheless how you would handle things like adjustments; but for moving the marks it's just another gear, which most clocks would have anyways for incrementing days. I don't think reading the time for the brief moment when the marks are in motion (and thus you know it's precisely martian-midnight) is really any issue; the only edge case I can imagine is a back to the future type situation where you are trying to determine a specific time by looking at a stopped clock (or more likely a still photo of a clock) and can't see that it is in motion, but the confusion would go away immediately when you see the hour hand pointed at the zero position.
To "jump forward" you can just stand: once you reach 24:39:00 the seconds arm stops over zero, waiting for another 35.2 seconds; at 21 seconds before midnight the minutes arm gets picked up by the seconds gear and advances one display minute per second of time, until midnight is reached (thus readable as a "seconds to midnight" movement in settings such as New Years). I have seen more complex motions implemented in wrist-watches so it seems doable.
I'm not yet entirely sold on the moving dial principle, I've placed my wristwatch upside down on my hand and trying to read it gave me a headache. The angle the hands make with the line of sight is indeed not very relevant, but their relative position to the fixed known elements of the dial feels essential.
That is definitely not true - I have an analog wrist watch (and am under 30). Also, there can be benefits to the use of analog clocks: in some circumstances they're quicker to read the time off of.
Have your circular dial divided in 24.658 sections then.
Nothing good can come from using a slightly longer second then what absolutely every piece of Earth hardware expects. Unless you are billion dollar rover mission where every bit or hardware is a custom design, it's imperative to use Earth seconds.
I love that feeling when the clocks go back in Autumn. This would give you that every day. My natural circadian rhythm is about 24.5 hours. I enjoy the dark, and this schedule would provide variety as our waking hours cycled in and out of phase with the sun.
And it would make scheduling regular Earth-Mars communications much simpler.
Planetary body should be a new offset category/type that defaults to Earth. Instead of tracking time as time:offset:date:calendar, we should do time:offset:date:calendar:planetTimeOffsetRatio. If calendarplanetTimeOffsetRatio is 1.02777778 (Mars:Earth days), number of seconds available in the day is increased by 2.777778%
Even for those few months, there is overwhelming reason to link the circadian rhythm of the mission with the solar martian day, see the rover missions; a drifting UTC mission time with no intuitive mapping of days to sols immediately raises massive logistic problems and human error risks.
I think it would be better to have two separate clocks, one tracking solar time with arbitrarily defined beginning, one UTC. Both would rollover after 23:59:59.
It's easier to explain that a second on Mars is longer because it's spinning slower than all the quirks of rollover at an seemingly random value.
The meaning of seasons on Mars would evolve with how we understand the impact of martian weather on daily life. Try to explain winter to someone living all his life on the equator.
For some reason I thought this was going to be about an alternative to daylight savings time and any suggestion of abolishing that always gets my interest.
It's not though, it's about giving the moon its own time zone. Which is about that least exciting moon based news I've ever heard although I'm sure there's interesting technical challenges (which I sincerely hope I never have to deal with).
> For some reason I thought this was going to be about an alternative to daylight savings time
I think I know the reason:
> and any suggestion of abolishing that always gets my interest.
When I read the title, I thought the same as you. It caught my interest because it sounded nonsensical so I was excited to see the reasoning and merits (or lack thereof) of the approach.
I'll send good vibes to that one guy from tzdata who'll for sure maintain a Lunar Time Zone database and literally everyone in the globe will rely upon it.
edit: Arthur David Olson and Paul Eggert, you're The Dudes, dudes.
> There are also technical issues to consider. Clocks run faster on the moon than on Earth, gaining about 56 microseconds each day, the space agency said. Further complicating matters, ticking occurs differently on the lunar surface than in lunar orbit.
At first this sounded like they were describing a mechanical issue with analog clocks, but they're not: time on the moon actually moves this much faster relative to Earth because Earth's gravity warps time more than the moon's does.
This is interesting… and I wonder how this plays out across different planetary bodies. It effectively means how we define a second will be different across the board.
Can anyone weigh in on whether the MKS standard of a second is based on something in free space, or here on earth?
If universal (based on free space) Would this mean we should all convert to/from this second, and treat our own second as relative?
It doesn't really work like that: a second is a second anywhere in the universe, but time at different places in the universe doesn't progress at the same rate.
ie. the issue is not with how we define the second. The issue is expecting that if two people in different frames of reference wait for one second, that they will continue to be "synchronised".
In fact it's worse than that, it's impossible to even define the concept of "synchronisation" for entities in different locations. You can only know that two things are synchronised if they are in the same location.
If the two entities are in the same frame of reference (ie. not moving or accelerating relative to each other) then you could argue that you just send a beam of light there and back, and assume the light takes the same time in each direction in order to synchronise... But we don't even know that light travels at the same speed in all directions! We've never been able to measure the "one-directional" speed of light, we can only measure the round-trip, so for all we know light could have a preferred direction in the universe and travel faster in that direction, as long as it is the same amount slower in the other direction. We would never be able to tell.
"The second [...] is defined by taking the fixed numerical value of the caesium frequency, ΔνCs, the unperturbed ground-state hyperfine transition frequency of the caesium 133 atom, to be 9192631770 when expressed in the unit Hz, which is equal to s−1.[1]"
That's the whole point, it does. Or, said differently, it behaves the same in whatever reference frame it's in, but observers in other reference frames can observe it being slower or faster.
Here's an example with very elementary particles, muons:
Cesium on the moon would continue to have the same frequency on the moon, but a cesium clock on the moon would run 56 microseconds fast per day relative to an identical clock on earth, because time itself moves differently.
So I'm not an expert, but this behavior actually would behave differently in different gravity wells. Due to relativity, while you would observe a cesium clock working normally on the moon, your Earth buddy would disagree with your clock because the buddy would be down in a larger gravity well.
Sortof. It would behave the same to the observer sitting next to the atomic clock no matter their local frame of reference. The point though is that we have atomic clocks on both Earth and (soon) the Moon, so we need some sort of standard for coordinating between them.
I really fail to see why they wouldn't just use UTC.
> There are also technical issues to consider. Clocks run faster on the moon than on Earth, gaining about 56 microseconds each day, the space agency said. Further complicating matters, ticking occurs differently on the lunar surface than in lunar orbit.
And all computer clocks need to be corrected for drift anyway. Standard mechanisms for dealing with unruly oscillators should have no trouble with that.
> Perhaps most importantly, lunar time will have to be practical for astronauts there, noted the space agency’s Bernhard Hufenbach.
> “This will be quite a challenge” with each day lasting as long as 29.5 Earth days,
And humans are adjusted for a roughly 24-26 hour day cycle, which UTC handily gives you. Whether it happens to correlate with the natural light/dark cycle, well, you've lost that fight here anyway, it's not like they'd stay awake for 29.5 days and then sleep 29.5 days. Same thing for the ISS: every "day" they get 16 sunrises and sunsets. Of course they'll use artificial lighting to regulate sleep.
Not a single justification for complicating things.
True, but UTC drift calculations on earth are different from on the moon. Fundamentally there is another time standard on the moon. How we account for that is up to us, but it makes sense to formalize this IMO. You could perhaps try to first derive a galactic or stelar time standard, then make both the moon's and earth's a derivative, but that seems even more complex in a way, though perhaps the most elegant in another.
Point is, the moon experiences time slightly differently than the earth, so having a clear reference will help synchronization, both to familiar geocentric standards as well as while on the moon.
I don't think we should define days around the petty need of humans. Even calendars are setup to map roughly to larger systems like phases of the moon / tides or orbits around the sun. The fact that an astronaught needs to sleep many times in a single "day" should not be a concern for the reference day/time, but could perhaps influence the time zones?
> True, but UTC drift calculations on earth are different from on the moon.
Hey reference clock, what time do you see? Oh, it's X? And I'm at Y? Okay, I'll adjust my drift correction factor to Z.
They might drift more than good modern-day clocks, and the latency to the reference clock might be higher, but the basic mechanism of correcting for the drift shouldn't be any different.
What's the difference between taking a good clock to the moon, where it runs 56 microseconds fast each day, versus having a less-tuned oscillator that runs 56 microseconds fast each day?
It's not drift. Time passes more quickly on the moon, to such an extent that there are 56 microseconds more each day on the lunar surface than on Earth. This is a an effect of general relativity.
For everyday human-scale stuff, sure this doesn't matter. For precise scientific and industrial use cases, it does.
I repeat myself: What's the difference between taking a good clock to the moon, where it runs 56 microseconds fast each day, versus having a less-tuned oscillator that runs 56 microseconds fast each day?
Clock slew is unacceptable for many use cases. For keeping system time on a UNIX server or for astronaut watches, sure. For many other things you need a precise and accurate count of how many micro/nano/pico seconds have passed. You can’t just have one second suddenly be N nanoseconds longer and expect sensitive experiments to not be affected.
If you're just talking about measuring e.g. the duration of an experiment, you don't need to sync to TAI for that.
For matching with TAI timestamps, no two atomic clocks agree with each other exactly anyway -- TAI is an average of some 500 atomic clocks. If you care to that level, you have to correct for your elevation on Earth already. So no matter what you do, if you care enough, you'll end up with some correction factor.
For most practical uses, you don't use an atomic clock, just a good steady oscillator, apply a correction factor to its output if you want to match with some other clock source (convert to TAI, etc), and use feedback like NTP to keep your clock on time. The biggest difference between moon and Earth in that is the latency to authoritative clock.
Computers don't really care about the current wall time. They just care about measuring time between events. If a piece of software about the current day of the week it's only because a person cares about it and wants to do something special on certain days of the week.
So, yes, time systems should be designed around the needs of people.
I believe time systems should be designed around the physics of the natural systems we inhabit. This may help us mere humans better harmonize with the rest of that which makes up our worlds.
This may sound a bit hippy dippy at first, but just consider how much light pollution effect other creatures, for a quick example.
Fully agree. I think some of the proposed lunar base locations have near constant sunlight for much of the year - rendering any mapping to period dark/light periods mostly irrelevant.
The impact of gravity is relevant of course. There is no reason not to use Earth as the reference however.
I was so convinced .beats would take over the world (or solar system in this case). I had their watch, I installed their task-bar clock plugin for Windows 98, I harped on about it to my friends... a surprising number of which actually agreed it was a good idea.
Then it just kinda fizzled out. I'm still upset it didn't take off.
It didn’t work in 1793 when it was mandated by the Convention despite the rest of the metric system taking roots.
It was unlikely that a rebrand by a greedy Swiss corporation would make it more popular.
Plus, current time keeping already use superior base systems. 12 and 60 are both highly composite numbers. Using a decimal system would be a step back. I wish we had symbols for 10 and 11 however.
Having your unit system in the same base as your number system infinitely outweighs any advantages derived from the having a higher number of factors or whatever. If you want to switch out number system, and then change the unit system alongside that, I'll hear you out. But skipping the number system and doing it at the unit level is always dead on arrival as an argument.
If it were the case, people would have switched massively to a decimal time keeping system.
The fact they did not is proof enough for me that using the same base than our number system isn't relevant. It makes sense intuitively because it's extremely rare that we have to go above 24 for hours and sixty for minutes while we divide these time periods fairly often.
I would gladly switch to base 12 just to have simpler multiplication tables (0, 1, 2, 3, 4, 6, 8 and 10 are really simple, 9 and A are fine while 5, 7 and B are annoying) but I am no fool and realise it's never going to happen.
Well, no, it’s not this argument because as you rightfully pointed everyone did switch to the metric system while absolutely no one switched to decimal time. I’m French by the way.
People do divide hours in twelfth, sixth and third all the time. Thinking in blocks of 5 and 10 minutes is very common.
Thinking in blocks of 5 and 10 units is very common in metric systems too. The fact that 10 minutes is 1/6th of an hour is not very relevant; that it's a good size unit for pragmatic use is.
In decimal, in general in decimal to write a fractions you need an infinite number of digits (with repetition). We have convenient notation for that repetition. A few special fractions can be written with finitely many digits.
Using base 12 or 60 you are in the same situation.
If you want elegance, you should cut down on the number of special cases.
Hence I would suggest to use a base like 11 or 59. That way more fractions will be covered by the general case.
Base 11 also works well for counting on your fingers. (Though base 6 works for that one as well.)
12 and 60 are both optimal cases for what you are describing.
That's what highly composite numbers are: the set of natural numbers having more divisors than all smaller ones. 12 is the smallest number having 6 divisors and 60 the smallest one having 12.
Both 11 and 59 would be inconvenient bases. They are primes.
> Prime numbers have no non-trivial divisors. So more fractions get handled by the general case, the case of infinitely repeated digits.
I don’t understand what you mean by special cases and general cases.
R\Q is dense in R so most numbers can’t be written as repeating digits at all whatever the base you are using and in every bases all numbers which have a finite representation also have a repeating one by subtracting 1 to the last digit and adding an infinity of the biggest available digit after (decimal representations are not unique).
Generally you want more divisors for your base. The more divisors you have the nicer the fractional part is likely to look like for common operations.
I am not talking about numbers in R. I am talking about representing numbers from Q in a decimal fraction. (Or in a fraction to basis 11 or 12 or 60 etc.)
(I'm not sure why anyone would want to bring up R? Why R and not eg p-adic numbers? Or algebraic numbers?)
General case: your fraction p/q will be expressed as some repeating decimal fraction, like 22/7 = 3.|142857 (where | marks the start of the repeat.)
Special case: your fraction p/q will be expressed by a decimal fraction that stops. Like 2/5 = 0.4 or so.
You can't get rid of the general case. No matter what base you pick, the representation of the vast majority of your fractions will have infinite repeats.
You can however minimize the occurrences of the special case by picking a base with fewer divisors.
In contrast, picking a base with more divisors, like 12 or 60 gives you more special cases to worry about.
(And yes, you can apply the trick you suggested, to sort of remove the special case by writing 0.4 as 0.3|9. Or for that matter as 0.4|0; but I'm not sure that last one buys you anything.)
Just to be clear: this is all a bit tongue in cheek and just talking about what system is mathematically more elegant, by having fewer special cases. In practice, having more divisors is convenient; though in practice just rounding after some number of digits is also fine.
ALL decimal, base 12 and base 60 expansions of rational numbers end in a repeating sequence. Just like base 11 and 59.
The only difference between 1/7 and 1/5 in decimal is that for the 1/7, the repeating sequence is 7 digits long (142857) and for 1/5 the repeating sequence is 1 digit long (0). In base 12, 1/5 has a 4 digit repeating sequence (2497).
With your ten digits, it's easy to symbolize eleven digits: zero, one, two, ..., nine and ten. Hence base eleven is very convenient.
Base six is probably most practical for this: You can count from zero to five on your left hand. When you go to six, you close your left hand (go to zero) and increment the count on your right hand by one. Two hands let you count from zero to thirty-five (inclusive).
I never heard of this before but now that I looked it up I think it's a good idea. People will probably still want to keep their local times but at least for coordinating international stuff we should have one standard time for the entire world. It's insane that we still have to look up how certain time zones correlate to our own local times for online meetings and such. This can all be improved. Remember back in the day there weren't even time zones, each city used their own local time. That's basically the situation we're in now from a global point of view.
For meeting coordination I think we will probably just converge around a defacto standard existing timezone like UTC.
We would all develop an intuition for what different times in UTC 'mean' (first thing in morning, after lunch, mid afternoon, etc) relative to the one we're in so we're not looking it up all the time. Maybe keep a UTC clock showing on relevant devices.
I don't think there's a need for a completely new system.
Currently you have to know all your timezones and do the conversions in your head if you travel a lot for example. Additionally the organizers of many events and conferences, even very international ones, do not use UTC. So no, UTC doesn't solve that problem in practical terms. If the entire world used UTC for their local time, then it would be a similar solution. I personally wouldn't mind that actually. It would take a bit to get used to but in the long run it makes much more sense in an ever more globalized world.
Same as the world adopting the metric system, and in science specifically SI-units. Surely you agree even though that was a bit of a hassle when countries changed over from their old unique units at the time, in the long run it was the right decision. Imagine what it would be like working in an international team always having to convert units of measurement. It's still an issue in engineering sometimes, because outside of science the US didn't make the switch and it's extremely annoying.
You have to know how the sun behaves at your local position relative to the universal time regardless of whether we call that knowledge "time zones" or something else. It sounds like what you want is for everyone to be more explicit about their time zone: instead of saying "the conference starts at 3 PM" they should say, "the conference starts at 3 PM UTC-6", right? From that you can easily derive that it's 9 PM UTC, and interpret that however you like for your current location.
Maybe you haven't experienced it yet but it's not as simple. I live in different countries and regularly have to deal with this. There's often confusion about time zones. Sure, it's "just a simple" conversion in theory but firstly the point is that ideally you should not have to do this at all and secondly no one calls it "UTC-6". You're lucky if they add a time zone at all, and when they do it's usually "EST", "Beijing time", "Eastern" (which can be Australian, American or whatever, your job to know from context), etc. Then you have to deal with DST, but only in some countries during some parts of the year. And don't forget the countries that add half hours, or Nepal which is extra special with an additional 3/4 because why not. It's utter madness.
Unless you're a savant who can remember all regions with corresponding time zones and all the exceptions, you'd be looking this up, which is what people normally do.
Beats per beat? I can think of more confusing language constructs.
Plus acceleration is metres per second per second and that isn't unitless.
One solution though would be to make use of one of the freed up words and use minute instead of beat. So we could say minutes per beat. Does that deal with your concern?
I was hiking in Edgewood Park on the SF Peninsula, and when I got to my car and took off my hat, this big fat tick came running out. He must have been the size of a coffee bean, or it sure looked that way in my moment of fear.
Is it true that they grow six times larger on the Moon?
And now we know the true explanation of the first paragraph of Seveneves:
The Moon blew up without warning and for no apparent reason. It was waxing, only one day short of full. The time was 05:03:12 UTC. Later it would be designated A+0.0.0, or simply Zero.
* I want to call my friend in Germany. What time is it for her? I guess I'm UTC-7 and she's UTC+1, so she's 8 hours ahead of me. I know that everybody around the world typically wakes up between 6 and 9 AM and goes to sleep between 8 PM and midnight, and she's an early bird, so she should be awake.
* They said the meeting was going to be at 8 o'clock, but they are in a different time zone, did they mean mine or theirs?
Without timezones:
* I want to call my friend in Germany. It's the same time here as there. It rises at 13:00 here, but she's a quarter of the way way around the world. I have to look up their typical waking hours. She's an early bird, so is she on the early side of that, likely. What time does the sun rise there? How much does that change over the seasons?
* They said the meeting was at 18:00. I know exactly when that is.
* If I live too far from the UTC, the day will change in the middle of my working day. My friend said they'd see me on Tuesday. Do they mean the day that had Tuesday at the beginning of it, or the end? What does "Tuesday morning" mean when morning is at the end of the Tuesday? Is "morning" the beginning of the solar day or the beginning of the clock day?
I agree with DST and 12-hour clocks, those are antiquated and mostly pointless. If you get rid of time zones, you don't simplify anything. You trade one set of complications for another. The complication is that you are interacting with humans that have different hours than you do, getting more different as you get farther away. Without time zones, you still have those complications, and you trade the set of time zone complications for another set of complications, by having your time completely divorced from your local daylight and working hours.
There are stupidities with modern time zones, but the concept exists for a good reason.
Yeah, I hate timezones with a passion but the more I think about them, the more I recognize they're the least bad option available. They correctly optimize for making most common cases trivial, while rare cases relatively easy.
I think it needs to be said here: cross-timezone scheduling and datetime math are extremely rare relative to people's daily experience. It may not be obvious to us here, because we're much more likely to be dealing with the unusual case - building distributed systems, having international users/customers, working for multinational companies or cooperating with companies from other countries, etc. Almost nobody ever does that, and it makes no sense to make everyone do extra mental arithmetic when thinking or talking about other places, just for the sake of a minority of white-collar workers who can't be arsed to add a second column to their Outlook calendar, or double-check when they have a Zoom meeting.
Getting rid of timezones would make most of everyday matters location-dependent. Timezones allow us to use the same numbering of times of day regardless of where any of us lives. 08:00 is morning. 22:00 is night. 12:00-14:00 is around when people have lunch. Etc. We can talk about our days in all kinds of ways (movies, plays, songs, novels, guides, etc.), retaining a precise numerical scale (vs. much less useful terms like "morning" or "afternoon"), and everyone everywhere can relate. Getting rid of the timezones makes this impossible.
Note the difference between talking about daily experience with people around the world, vs. doing things in sync with people around the world. Timezones make the former trivial, and the latter easy. Removing timezones makes the latter trivial, but the former hard. Hard enough that people would quickly recreate timezones for convenience.
I have no issues with timezones at all. Morning will be in the morning (as per the clock), wherever you are, and timezones are reasonably well defined. Without timezones you would still need a system defining what the local equivalent of "morning" or "evening" or "working hours" is, in whatever town the person you're calling is in - and I feel we'll be going back to how things worked before the railway chaos forced the introduction of timezones.
What I hate is DST. That's not only useless, it's problematic. Heck, my old mother never managed to handle the change this last fall, in her nursing home - meals didn't come at the optimal time, she got undernourished, it destroyed her sleep cycles, and her health went all downhill.
The worst problem with timezones IMO is that they are way too big [1], compared to what they should be if the world were equally divided into 24 hour "strips". Europe for example has CET which stretches all the way west to Spain, despite UTC+1 ending somewhere along the French-German border - so Spain and France are 1 hour "ahead" despite not needing to.
Add to that cultural and climate differencies and it makes it even harder to schedule things. Like there is not a huge difference between France and Spain in term of solar hour, but rhythm is different because people in the south are used to take their lunch break when at the hottest moment of the day and tend to stay outside and up longer because they are most pleasant in the summer.
Most people in europe keep scheduling meetings between 14 and 16 which is pretty much lunch time here in Spain so most of the time I am having lunch during my video calls or take a quick bite in between calls without a proper break.
This might be a “problem” for people who like arbitrary lines to define their lives. Keeping most of Europe on one time zone lets people from different neighbouring and culturally close countries work together without needing to think about time zones.
If keeping most of Europe on one timezone lets people work together without needing to think about time zones, so does unifying the entire earth on a single TAI or UTC "timezone".
No, that does not scale. Sunrise and sunset are at very similar times in CET countries, meaning they have similar business hours, and I can set up a meeting between Spain, Poland, and the UK (which is on UTC/BST, but that’s close enough) at 11:00 CET and everyone will be fine with it and well awake at that time. If we extended CET around the world, and wanted someone in New York to join our meeting, they would be furious because they would need to ~~wake up~~ (edit: be awake and at a computer) at 5:00 ex-EST, or 1.5 hours before sunrise.
Do you have an example that is not from an oppressive regime that does not care about the happiness of its citizens? Do you have an example that does not have an unofficial time zone [0] observed by people in the easternmost part of the country?
I'd say that the biggest problem with time zones is that most people confuse them with time offsets (which unfortunately is what you are doing). A time zone is a geographical area. That area can run on different time offsets - "times". So the UK runs on UTC in winter, and UTC+1 in summer. It's wrong to think of the UK as being in the "UTC time zone".
This may seem pedantic, but the confusion causes real problems with software.
> I'd say that the biggest problem with time zones is that most people confuse them with time offsets (which unfortunately is what you are doing).
The problem is, say, I have a colleague in Spain. I ask him for a meeting today at 0800 CET, which for him means sunrise was at 0746, while I in Munich have had sunrise and got woken up by the cats at 0652.
Huh? For me it is, and has been like that for a very long time. I work with various customers throughout the day, and they are in various different timezones. When I'm not working I almost daily interact with family, again different timezone. I am always aware of various timezones.
They do indeed talk to people in other timezones as well.
We may have online meetings sometimes, with at least three timezones involved.
I can't see how one would only contact in touch with people in a single timezone, unless one only wants to have contact with people in continental Europe (excluding e.g. Portugal and Finland), or people in a single timezone in the US. Timezones are everywhere. The earth is a sphere, so.. can't be avoided.
If your work involves coordinating across timezones daily, you're part of the small minority cases I was talking about. Offline world is still mostly local, and almost all interactions almost everyone has, online and offline, are local too.
Firstly, that's not a small minority case, and secondly, in my small town there are people from around 100 nations, and they have family and friends from all over. And, as I said, even central Europe has 3 timezones just inside EU. It's not like most people only have contact with their 50 neighbours, and never go anywhere.
When I talk with friends, timezones come up reasonably often. It's not something people can walk around and not be aware of.
> * I want to call my friend in Germany. What time is it for her? I guess I'm UTC-7 and she's UTC+1, so she's 8 hours ahead of me. I know that everybody around the world typically wakes up between 6 and 9 AM and goes to sleep between 8 PM and midnight, and she's an early bird, so she should be awake.
> Without timezones:
> * I want to call my friend in Germany. It's the same time here as there. It rises at 13:00 here, but she's a quarter of the way way around the world. I have to look up their typical waking hours. She's an early bird, so is she on the early side of that, likely. What time does the sun rise there? How much does that change over the seasons?
This is actually the same thing. In both cases you have to know or look up that there's an 8 hour difference between the two of you.
The bits about the seasons and being an early bird apply equally to both cases, so it's not really relevant to the comparison.
It is not the same thing. I do not need to reason at all right now, I just look at the clock that shows times in relevant timezones. I know all the information instantly, without any calculations in head at all.
In the latter case, you do need to do math, because what you are getting is meaningless number and then you need to convert it to some kind of "how long after typical start of day" format to get information you actually want.
Ah, so you're saying you have a clock showing German time, and it did the "math" for you.
Imagine multiple world clocks, in the typical Hollywood command bunker look. Instead of each clock having hour hands pointing in different directions, they'd all have the same hour hand position, but different highlighted zones for waking hours (or "shift 1", "shift 2", "shift 3", for 24h duty).
"* I want to call my friend in Germany. It's the same time here as there. It rises at 13:00 here, but she's a quarter of the way way around the world. I have to look up their typical waking hours. She's an early bird, so is she on the early side of that, likely. What time does the sun rise there? How much does that change over the seasons?"
In both cases you need to do some maths. You know early birds wake up at 11 where you live. If you know she's 8h ahead, just add 8. So, 19 is when you should call.
For calling people nothing really change. Adapting when moving somewhere else is the tricky bit. If I move to another city or country I just adjust my clock and automatically know when things will be open, working hours, etc. Otherwise, whenever you move you have to keep mentally reminding yourself that the work day starts at 12 here, not 7 like I'm used to and so on.
> * If I live too far from the UTC, the day will change in the middle of my working day. My friend said they'd see me on Tuesday. Do they mean the day that had Tuesday at the beginning of it, or the end? What does "Tuesday morning" mean when morning is at the end of the Tuesday? Is "morning" the beginning of the solar day or the beginning of the clock day?
The problem there is the lack of specificity, and timezones don't help or hurt that. It's already not clear if "Tuesday morning" means my Tuesday morning or yours. Maybe they could say "Tuesday around 1400" and we would both know when that was without considering the offset.
> If you get rid of time zones, you don't simplify anything. You trade one set of complications for another.
You do make a lot of things simpler by not having to figure out offsets. Sometimes you still will, but it will be less often.
> It's already not clear if "Tuesday morning" means my Tuesday morning or yours
In 99% of cases of my communications it's perfectly clear because it's the same for both of us. The GP said that without a local timezone Tuesday Morning would be ambiguous even in a local context.
In the military, there is Zulu (+0 UTC/GMT) and Juliet time ("my" time). Every report with times should always indicate the timezone (whether Z or J or otherwise), e.g. in NYC 2pm would be 1400Y. Any recent/current military folks want to comment?
Is there a different single letter per time zone band? That’s almost elegant if so (as long as you ignore the many places with idiosyncratic time zones).
But seriously, in DST world, the numerical offsets (-6, ...) are confusing enough, adding yet another "simpler" notation would only increase confusion.
I did say almost elegant...! I just hadn't noticed before that the 26 Roman letters are just enough for 24 hour-wide timezones, which seems like a nice coincidence, so I wondered if anyone had tried it. There are way more than 24 different zones in practice, of course.
Unfortunately, there's too many timezones for just A-Z, but nautical time does make an attempt.
See https://en.wikipedia.org/wiki/List_of_UTC_offsets for the funny markings they decided to use when they ran out of letters.
As someone frequently half a world off from those I'm scheduling a call with, it helps to adapt a format that explicitly states [calendar date] [24hr time] my time / [calendar date] [24hr time] your time.
More often than not, one's morning or afternoon is late in the evening or in the middle of the night for the other person. So morning is almost unambiguously referring to the westernmost person and afternoon to the easternmost.
Instead if the working day begins at 23:00 (as would be the case for Pacific time) for me and 8:00 for you, but the meeting is already at 1:00/10:00, it is unclear if you mean "at 1:00 of Wednesday, which is on the working day that started Tuesday at 23:00" or "at 1:00 of Tuesday, on the working day that started Monday at 23:00".
Wow, there's clearly pros and cons for both! It's all about compromise and the way it currently works, works for the majority of people that don't even experience the downsides in their day to day life.
Like having to coordinate meetings around the world or figuring out what time it is for their friend on the other side of the world. Most people never deal with any of that. They just want it to be 12:00 and bright out when they eat lunch (or whenever they eat lunch).
I've literally heard nobody say they want it to be 12.00 and bright out when they eat lunch, only that they want it to be bright out.
Suppose we set the universal time to the current -12h meridian. Everyone works with that exact same time. If I receive a scheduled meeting at 15.00 I know exactly when to be there (wow, that's late for me!). If glance at the clock and see that it is 4.00, I know it's time for my afternoon tea, and that the sun is still out.
I think people would get used to this line of thinking very rapidly, and would scheduling would be arguably made easier, because the proposed time would not need to be parsed by anyone.
Your comment makes me understand why people would want to abolish time zones. However, my issue with that is that I think it would cause confusion when people in everyday life use the word “afternoon”. I won’t know if they’re talking about the time when the sun is getting lower in the sky, or if they are talking about when it is past 12:00 on whatever clock we are mandated to read. We would all need to ask people to clarify those things, especially if they don’t have the detail-oriented mindset for these kinds of things. Now, granted, you do already have to clarify these things when you are speaking with people around the world already.
> * I want to call my friend in Germany. It's the same time here as there. It rises at 13:00 here, but she's a quarter of the way way around the world. I have to look up their typical waking hours. She's an early bird, so is she on the early side of that, likely. What time does the sun rise there? How much does that change over the seasons?
You look up her profile in teleappnal, and it says she's online, so you give her a call. Easy. Or, you book a calendar appointment like anybody civilized, and she either accepts or rejects it. (Telephones are another abhorrent Victorian invention which we'll get rid of, and sooner than you think.)
> With timezones: * I want to call my friend in Germany. What time is it for her? I guess I'm UTC-7 and she's UTC+1, so she's 8 hours ahead of me
> Without timezones: * I want to call my friend in Germany. It's the same time here as there. It rises at 13:00 here, but she's a quarter of the way way around the world. I have to look up their typical waking hours. ... What time does the sun rise there? How much does that change over the seasons?
You aren't doing a fair comparison. In your "with timezones" example, you are considering it trivial to look up/remember what timezone Germany is in, but in your "without timezones" example, you are making it out to be more difficult to look up/remember what the "day-hours" of Germany would be.
In a world without timezones, you could just Google "what are they day hours in Germany" just like you can Google "what is the timezone in Germany" now. And in fact, it would be much easier to reason about - if you know that the time is currently 10:00, and the day starts at 18:00 in Germany, then you don't need to do any more math, you can just wait until it's after 18:00 to call her.
> If I live too far from the UTC, the day will change in the middle of my working day. My friend said they'd see me on Tuesday. Do they mean the day that had Tuesday at the beginning of it, or the end? What does "Tuesday morning" mean when morning is at the end of the Tuesday? Is "morning" the beginning of the solar day or the beginning of the clock day?
Again, you're assuming that language would not be different in a world without timezones. It's a completely unrealistic scenario, language always evolves to fit our real needs.
This is obviously a software problem that could be solved universally at the OS level.
We have all kinds of low-level protocols for coordinating all manner of things. Distilling a coherent system of agreeing on a timezone difference, between two communicating entities and presenting that to the user at any time would be quite basic.
We have location data for all, at all times. We have a local, accurate clock. There's little else we'd need. What remaining issues are all UX. Apple could roll such a thing out parallel on all their multitude devices and it'd be basically trivial.
Is it maybe just habit that we gravitate to a manual process, when an automated one would be rather easy in this situation?
On iOS, the existing clock app already provides a basic form, tapping the plus button allows the addition of multiple zones and provides useful additional context in the form of offsets and what day it is relative to the users local date and time.
That's halfway there, it would only need contact awareness to be able to provide a solution.
Apple already do have support for specifying the time zone in appointment times, and the Clock application can show the current time in different places. What woudl you want to add to this?
Attach the current, local time to all contacts, at the basic, fundamental level. If contacts are in a different zone, every app accesses an OS-defined UX, that displays a contacts local time in a ubiquitous, familiar fashion all users get used to.
The amount of people complaining here about how difficult it is to keep a handle on illustrates it's not a trivial mental calculation, even for HN. I would agree, especially when coordinating multiple, distant parties.
The takeaway is that we, humans, tend to reference things in non-universal frames of reference. Be it "morning" for time or "office" for location. Timezones can be seen as an emergent property of long-distance communication. Whatever the algorithm you choose the underlying problem with "inertial" frames of reference remains.
Every time I see someone propose scrapping timezones, they end up in one of two traps. Either unilaterally dictating that Australia, NZ, etc become fully nocturnal - or they end up re-inventing timezones with a different name.
I’m with you on the last two but a life without timezones sounds horrible.
So many issues, like having to relearn the clock when you are in another timezone. I’ll meet you for lunch at 06.00! Life is based around the sun.
Having some sort of agreed upon world timezone would be neat though, whether that was GMT or something less euro centric. So everyone would understand “we have a zoom meeting at 11.00 WT” instead of having to decide which timezone to refer to.
UTC is nice not because it is euro centric, but because there is a lot of sea on the opposite side of GMT and therefore it is pretty easy to place the date change line. There are multi-timezone countries on both sides of the oceans (USA, Russia, and the European Union basically counts as a third) so the date change line has to stay on the sea.
There is only one good alternative, which would be to place the date change line in the middle of the Atlantic (between UTC-3 and UTC-2) and the fundamental time zone in Japan or somewhere Australia (UTC+9/UTC+10).
Iceland and the Falklands would likely be respectively -9 and +12 from Japan.
Greenland is probably the best reason to keep Greenwich as the reference time zone. :) Most of the population (so to speak) is on -13 to -12. I have no idea if the people there have more ties to continental Europe and Iceland, or to Labrador.
Do you encounter people using UTC regularly? While it's the closest we have to a world timezone, in my experience people don't use it much outside of a technical context.
I usually see Americans treat Eastern Time as a standard, with references to Pacific sometimes. In the EU, where I am, CET is the usual timezone used for specifying a time internationally. And regardless of the location, it's very common for people to specify a time in their local timezone first and leave it to the other party to figure out what the difference is.
I haven't encountered situations like someone saying "let's have a discussion at 16:00 UTC" to set up an EU-US meeting.
My colleagues and I do this, but we're often involving 3, 4, 5+ timezones. Everyone knows their own offset from UTC, but I easily forgot whether Australia is on DST this month.
I agree to an extent, but I don't really see the problem either. If I book a meeting or something through any calendar application, the timezone will be fixed to the local timezone. Let's say I book it for 13:00 in my calendar, it will show up as a meeting at 14:00 in their calendar if they are +1 from me.
So whenever I specify a time with someone that could be in a different timezone, I would either make sure that I am using some application that solves it for me, or I would AT LEAST include the timezone.
Maybe I'm just not encountering the same problems that you do.
I'll meet you half way: let's place UTC next to the local time everywhere we can. It could be in parenthesis, in a tooltip, etc. It could get us all used to knowing our time in UTC and eventually help in cross-timezone communication.
If I need to call someone in their morning, I need to determine the same offset.
I have plenty experience with both ends using UTC regardless of where in the world they are. It's much easier. Either 0425 works for both or we need to adjust one way or the other. I still need to know that your schedule is four hours ahead of mine, but we don't need to do math to figure out what time 0425 is. The cognitive overhead is diminished.
That's a pro for having UTC, not a pro for having no time zones. I believe this is the main reason that going against time zones is flawed---people don't exactly know what they are claiming!
The point of the parent article was that it is impossible to make everyone work on the same time zone, as they will inevitably make up their own time zones. This is not even a hypothetical example, the article explicitly mentions western China [1] where the standard time zone was so off that everyone uses an unofficial time zone instead.
Without timezones, you need a lookup table for what daylight hours are for every location on earth. For convenience, rather than publishing a table for every latitude/longitude, let's group these locations along geographical borders nearby to longitude lines. We can call these grouping "zones", or in more broad contexts, "time zones."
Do you work with people in other timezones? It’s much easier to reason about what a reasonable time is for people when you know how it centers up with noon for them.
Sure, but I work across three timezones, and I can glance up to the top-right of my computer and see all three at a very quick glance, as well as knowing when people are likely to be having lunch, whether it's early or late for someone, and so on. I'm also now quite used to "Jerry gets up early his time" for certain coworkers.
I guess you could potentially have some sort of graphic to show the same thing for sunrise vs sunset ... but as you point out, the maths needs to be done either way, so why get rid of a system that works pretty well?
The math would've been a lot easier if GMT+0 had been somewhere in the Pacific ocean. Going from +1 to -7 is a lot more troublesome for most people than from +11 to +19.
Yeah it zigzags a bit but it hovers on the two sides of +/-12, placing it around 0 would mean half the world operating on a work week that is shifted by one.
So we'd need to adjust our terminology a bit. I'm sure the world wouldn't end because of that.
It's always amusing to me how opponents of the idea of abolishing time zones assume that everything will stay the same, except, well abolishing of time zones. Apparently in the moment we do so, humanity will be become incapable of adjusting even the slightest. It will be impossible to invent a couple of new terms to let people communicate easily about time. It will be impossible to replace time zone maps and time zone aware clocks with ones that show regular office hours in different regions, &c.
For the record, I'm not even really for abolishing time zones, but the fact that the word "tomorrow" would become slightly less useful is not a particularly significant reason.
We spend a huge amount of our time talking about and reasoning about time. Destroying the meaning of all of the language we've built up is pretty significant. More importantly, it basically ensures that all that terminology will be vastly different when you live in different parts of the world, where the day changes in the morning vs the middle of the solar day vs the evening vs the middle of the night. Weekday names will be entirely regional, and their implications will be as well.
usually (outside of global business meetings) when you are talking to a person about making plans you are both in the same time zone. So asking ‘your time or my time’ is unnecessary.
But if we got rid of time zones, you’d need new ways to distinguish between tomorrow meaning ‘after the clock ticks over 24:00’ bs ‘after the next night cycle’ because they would not be the same thing.
To be fair the same ambiguity occurs now if you are, say, parting ways with someone after midnight and you say ‘see you tomorrow’, but you actually mean later that same calendar day.
But I think losing the ability to make use, in everyday local conversations with people in the same place as you, of concepts like ‘tomorrow afternoon’, would be kind of a dealbreaker for time zone abolition.
I think even if governments around the world abolished time zones, tried to ban their use and forced a universal time, you would still end up with folk implementation of time zones that ends up being worse where people translate the universal time to and from the non-standard local time. You can see some of that behavior here: https://en.wikipedia.org/wiki/Xinjiang_Time
I was thinking getting rid of timezones would be crazy, but I work at a globe spanning all remote company and any time anyone discusses times it's always in UTC because everyone's in a different time zone and have no idea what each other's timezones are.
Just use Julian dates. This way you can also get rid of years, months, weeks, ... You only need to change the Julian period every now and then, ie. every 7980 years. See: https://en.wikipedia.org/wiki/Julian_day
The trade-off balance is not that big towards your desire to justify such a gigantic mentality shift for all humanity. In other words, it will never happen. (Ok, maybe the DST and the 24h clock only can go away at some point, but TZs will stay)
A reference time describes what standard of time we use. On Earth, for most human use cases, the reference time is UTC. UTC is measured by using International Atomic Time (TAI, from the French "temps atomique international") and adding or subtracting a prescribed number of leap seconds to account for the rotation of the Earth.
We don't use TAI in everyday use because it's constantly and continuously changing in small ways to account for the tiny accelerations or decelerations in Earth's rotation, and that's annoying for computers. So instead we use UTC.
A time zone is a different thing entirely. A time zone is a set of rules that describes how to convert from a reference time to a local time.
Most people think of a time zone as being a fixed offset, like "I use Eastern Time; that means UTC-5". This is wrong. A time zone is a _set_ of offsets, each of which applies to a _range_ of reference times. For example, we have a time zone called "America/New York". This describes, for any current or future datetime in reference time, what the correct _local_ time would be.
Here the scientists are doing two things: deciding what the UTC equivalent for the Moon should be (the reference time) and deciding what (if any) offsets should be applied to compute "local" time somewhere on the Moon (the time zone).
> We don't use TAI in everyday use because it's constantly and continuously changing in small ways to account for the tiny accelerations or decelerations in Earth's rotation, and that's annoying for computers. So instead we use UTC.
Wait, what? TAI is the one time standard that doesn't change. It's just a bunch of atomic clocks averaged together. As monotonic as human kind is able to manufacture, at this time.
UTC is TAI+varying offset in an effort to stay close to mean solar time.
The varying offset of UTC, called leap seconds, is what's annoying to computers. If we ran computers on TAI, they would be simpler!
And "data center time" tends to be something approximating UTC, with leap seconds smeared into continuous adjustments, just because UTC is annoying to computers.
> Wait, what? TAI is the one time standard that doesn't change. It's just a bunch of atomic clocks averaged together. As monotonic as human kind is able to manufacture, at this time.
That is not the complete story. The calculation used to average the clocks has changed over time, thus amending TAI. The "perfect monotonic" time you're talking about is TT (terrestrial time), not TAI.
Wikipedia about TT: "It is a theoretical ideal, and real clocks can only approximate it. [...] TT is indirectly the basis of UTC, via International Atomic Time (TAI)."
So, TAI is what we are able to measure. TT is a theoretical construct. Can't run computers on TT, can run them on TAI.
This is fine to do while we are still talking about space missions and quick landings, but the moment there is a base and people start to live there semi-permanently you won't be able to just pin it to UTC and call it a day.
Timezones are already bad enough in software. Having to additionally account for the moon would make people quit the profession.
> What we should really do is just get rid of all this crap and go with metric time that starts at an arbitrary moment and isn’t dependent on location.
What do you do about relativistic effects which cause time to move at different rates in different places, including on the Moon. From TFA:
> Clocks run faster on the moon than on Earth, gaining about 56 microseconds each day, the space agency said. Further complicating matters, ticking occurs differently on the lunar surface than in lunar orbit.
The really weird part to get your head around here is that in case of relativistic differences, it's actually time that's changing. So cesium still has the same frequency, the same number of events per second, but their second is not the same as ours.
When people talk about clocks running faster or slower, it muddies the topic a little because it makes it sound like a fault of the clock. The clock is accurately measuring the time it experiences.
My favourite trivia on this topic is that this is not hypothetical. If the clocks on GPS satellites kept earth time (eg, without accounting for relativistic drift), the entire GPS system would accumulate 10km of error per day. This is a genuine thing with measurable, real-world effects.
I think the point is that the clock on the moon runs faster from the point of view of an observer on Earth due to time dilation. I can't claim to understand it, but clocks "run slower" the deeper they are in a gravity well. To an observer colocated with the clock, it ticks accurately.
This isn't a new problem, though. GPS satellites have atomic clocks, and they drift 7uS or so per day due to time dilation.
Lunar gravity is about 1/6th of Earth gravity. Clocks slow down due to gravity (Einstein's relativity). Thus, clocks tick faster if there's less gravity.
Aside: we can measure this effect on Earth, between the foot and top of a mountain. Tom Scott has a video about exactly that.
Yes — relativistic effects are so fundamental that they affect every system/process (including atomic clocks ) and therefore any measurement scheme. That’s why we talk about “time slowing down”. The movie Interstellar has one popular depiction of this.
Because spacetime. Clocks run at different speeds depending on the gravitational field in which they operate. This is true for any two clocks, not just between earth and the moon. Satellites have to constantly correct their clocks for the same reason.
I think a more interesting argument is: Humans are 24-hour creatures (ok, maybe not precisely 24 hours, but close enough). That's how our internal clocks are wired. Its not nurture; its nature; and it would take generations living day-in day-out with a different cadence before we see biological adjustments.
Sure, our clocks are 24-hours-a-day because that syncs up well with the rotation of the earth. That's the scientists' argument, and it isn't why we use a 24 hour clock. The real reason is because it syncs up well with our biology. Its hard to imagine a world where these two things aren't synced up, but we're headed to some soon, and its likely that any clock which doesn't sync up to our biology (TimeAwake + TimeAsleep = 1 Day = 86400 seconds) will be rejected by anyone who isn't ordered by their commanding officer to use it.
Its like... just use UTC. Be done with it. The argument that "its an Earth Time, and we're Interplanetary now" feels like its coming from a sci-fi fan-fiction writer obsessed with The Expanse and solving problems before they exist. The most important priority for any timekeeping system is that its Useful. The second highest priority is that its Scientifically or Mathematically Beautiful, and this second priority is so far below the first one that we invented timezones and daylight savings time and everyone who has ever had to think critically about either of these things wants to jump off a fucking bridge but we tolerate it because they're Useful (and, we're getting rid of DST, because we've recognized that its Not So Useful, again, Usefulness drives everything).
Here's what's going to happen: they'll invent some kind of new Lunar Time with 29.35 earth-day long days or some other nonsense. Every clock on the wall at Moonbase Alpha will have both Lunar Time and UTC and probably US Central Time because Houston & NASA. The scientists who invent it will pat themselves on the back. The janitors who clean the place will have their shifts scheduled in UTC hours and the Gregorian calendar ("yeah man I'm off at 8pm and I don't work again till Monday, yeah I guess it is weird that its been night outside for two weeks yet we still say 'monDAY', I don't know man I just work here same as you let's grab some drinks"). In 350 years, the Martian Humans will declare independence and reject all forms of Timekeeping that Us Earthlings invented, including the one the ESA (hint: EUROPEAN space agency) tried to invent for them, because Politics. Can we just, I don't know, focus on GETTING to the moon first, before we start paying people to retrofit our microwaves with new clocks?
Idk we're planning these missions in the near/immediate future, so we need standardized times for those immediately future missions. Everything else comes later.
> Having to additionally account for the moon would make people quit the profession.
...and it will make other engineers rich when they dive into the complexity, wrap a clean API around it, and start a lucrative time-as-a-service company.
"Clocks run faster on the moon than on Earth, gaining about 56 microseconds each day, the space agency said".
I have no technical knowledge about clocks at all, but this sentence sounds odd to me, shouldn't there be a lot of different clock technologies that work in different ways?
Or are they saying this because there is only one "best" technology for NASA purposes and that one is off by 56 us on the moon?
According to special relativity clocks on the Moon will run slower because the moon is orbiting the Earth, so the clock has a higher relative speed. However, general relativity says the clock will run faster, because the Moon has a smaller gravitational field than Earth.
In this case the second effect will dominate, so even a perfect clock will run faster: it's not the clock which is fast, it is time itself. GPS satellites have to deal with the same issue.
Do you think there are any physical phenomena that can be used that might be sensitive to relativistic effects? Something like a quartz crystal that actually vibrates at different speeds depending on how dilated the current reference frame is? Maybe something external, like if our sun would be emitting some sort of signal at a regular interval that everyone could sync up against.
I wonder if, in the future, we will have a group of timekeeping satellites orbiting our sun. Maybe their orbit could be tweaked to have identical time dilation to Earth, and everyone can just sync off that.
Maybe we can derive some time keeping mechanism from the black hole at the center of our galaxy, or maybe we will just realize that our obsession with timekeeping is a silly terrestrial habit, and not really important at scale.
No it is pesky Einstein guy with his general relativity.
The bigger problem is that our timekeeping is earth based. The time definition should have been based on deep space, with corrections for time on earth.
So earth clock are slow with respect to clocks in deep space. Clocks on the moon are less slow are therefore faster than on earth
Not a physicist, but I was led to understand there is no universal frame of reference. You can't base time on deep space as if time was the same across all deep space, because it isn't. Right? A local time "correction" is easy when you compare against exactly one other point in space but if you want something like an universe-wide UTC that everyone can correct against, it can't be done. Although I don't really understand why.
Not a physicist either, but one practical problem is the speed of light. Time synchronization is likely to become very tricky if you have communicate with time sources that are hundreds of light years away.
> There are also technical issues to consider. Clocks run faster on the moon than on Earth, gaining about 56 microseconds each day, the space agency said. Further complicating matters, ticking occurs differently on the lunar surface than in lunar orbit.
As a programmer, I can feel a migraine coming on just thinking about the bugs this is going to cause.
I am just a fool who doesn't know anything, so take my thoughts with a grain of salt, but I think all the ideas shared here are too complicated and have too many breaking changes. I think a lunar timezone or martian timezone is a horrible idea. We need instead to have planetary offsets. How many insidious and horrible date/time bugs would we signing 22nd century programmers up for by trying to mishmash different planetary timezones together?
Keep it simple! If we added a celestial offset, it would work for the moon, mars, and any other planet we wanted to visit! Time could be stored as follows:
```
1) Time of day
2) Timezone
3) Celestial Offset (PlanetX:Earth seconds per day ratio)
4) Calendar (Gregorian, Japanese, Islamic, Lunar, Martian)
5) Solar System
```
I can understand that it's numerically simpler to start at 0h00 but... don't we miss something related to the sun ????
It seem to me that the "best" reference point would be "mid-day" (at 12:00), when the sun is perpendicular to the planet. It can be easily "guessed" and synced I think... like it has been for millenaries before the clocks
So I guess that instead of 0h00:00-24h39:35 I would prefer -Oh:20:17 to 24:19:18 (centered on 12:00). Or we can just decorrelate "absolute time" (in h/m/s) and "time of day" (in %) where mi-day is 50% (actual duration in H/M/S to be depending on the planet) and mid-year is 50% of the actual rotation around the sun...
They mean local lunar time, for coordination between actors on the moon. Lunar day length and even, due to gravity, second length differ from Earth (i.e., time moves at a different speed).
> They mean local lunar time, for coordination between actors on the moon. Lunar day length and even, due to gravity, second length differ from Earth (i.e., time moves at a different speed).
These two sentences aren't related to each other. Time moves at a different speed on the moon than it does on Earth. But that only affects interactions between the Earth and the moon. Two actors that are both on the moon are both on the moon; their Earth-based clocks will work without adjustment if their goal is to coordinate with each other.
English is an official EU language is one of the (few) working languages of the European Commission (French is also used a lot). And the United Kingdom remains a member of the European Space Agency. GMT overlaps with European time zones, with countries like France being big proponents things like UTC (and TAI, and the metric system in general).
He's referring to the current animosity between the UK and Europe. Both groups frequently go out of their way to make life harder for the other. The EU proposing the English GMT would be unlikely in this climate. Also FWIW I think only Portugal overlaps with GMT.
Also Ireland, which is important because of Northern Ireland. It would be politically inconvenient if the time in Dublin were ever to be different from the time in London.
What other language could the EU possibly use as a common language for communications between member states? English is the only language that you can use everywhere within the union and have a decent chance of being able to communicate with people (and this is true everywhere else in the world too).
Sure, although English is absolutely the dominant language in almost all of the Republic of Ireland, where in Malta the situation is similar to the Netherlands, where most people it's just a very proficient second language.
It’s rather humbling how many languages are spoken by individuals there - and my experience is that people there often have other languages in common too.
EUs English guidelines emphasized en_UK because that used to be the main audience to whom texts should be unambiguous. That requirement is no longer there, so en_EU (which defacto exists, although it has many variations) might now officially move closer to en_US, since that is what most cultural exposure is to for those to whom English is not a first language (i.e. almost everyone in the EU).
I therefore propose a half EU/half US flag pictogram to refer to English from now on.
A zone of what though? Time zones, are zones of our planet where we observe time in some coordinated fashion. If we're adding the moon, what part of our globe should it be considered a part of? This all seems pretty silly.
It’s not about creating an earth based zone for the moon. It’s about creating a system of recording time that works on the moon. Time doesn’t work the same on the moon as on earth, so why should they be linked?
Earth clocks are basically about the position of a point on earth, relative to the sun. The moon’s relationship with the sun is quite entirely different. So where is the equivalent of Greenwich on the moon? Does that even make sense? What is the equivalent of a day? Will there be weeks?
Earth clocks and moon clocks don’t run at the same time, due to relativity. And I guess the relationship between earth and moon time is far from linear depending on their relative positions and the sun’s.
So there is lots of interesting stuff to be worked out, and I don’t think it’s silly at all.
Why needlessly complicate things? Use GMT + 0, earth days, and earth weeks. Why introduce all this unnecessary confusion? There is no practical need to measure local time on the moon.
>Earth clocks and moon clocks don’t run at the same time, due to relativity.
I'm sure the 1/3000th of a millisecond per year won't be a problem
> There is no practical need to measure local time on the moon
Could it be useful to know when the sun would rise and set, or when Earth rises and sets, for the sake of having routines related to solar power or LOS communications? Having a Luna-native time-of-day might be handy for some things.
I think it makes some sense to come to an agreement, even if that agreement was to use UTC (TFA does mention that it's used for ISS).
> Could it be useful to know when the sun would rise and set, or when Earth rises and sets, for the sake of having routines related to solar power or LOS communications?
The sun rises and sets over exactly one (Earth) lunar month, 29.5 days. That's what the phases of the moon are, the progression of sunlight over the face of the moon.
Earth never rises or sets; the near side of the moon always faces toward Earth, and the far side of the moon never faces toward Earth. (This is known as "tidelocking".)
Two reasons. Firstly, human biology, it'll be better for the people living/working there to experience via things like lighting and having visual clues from things like clocks to regulate their life.
Secondly, for communication with those on Earth in a way that makes sense to both parties. We only use our system of time because it is useful (daylight saving might be an exception to that;) so why bother with a more "correct" version of that on the Moon? It's not like you'll be using the days there if it's light for one Earth month and dark for another.
Do they work with two time zones or time systems when you have a team on Earth and another Team on The Moon coordinating tasks? So the team on the Moon follows a Moon time zone and the team remaining on Earth follow a chosen Earth time zone and then they sync things up?
What is the use case for teams using two time zones/systems or does the earthbound team use the Moon time zone and is that better than the team on the Moon following an Earth time zone?
I'm not sure the rest of the world should look on complacently while Europe attempts to colonize the moon. This should be something for Lunarians to decide.
There's a whole lot of social conventions tied to local time, sunrise/sundown, and the circadian rythm in general.
Sure, it'd be possible but we'd have to concieve a lot of new cultural concepts to replace the ones invalidated by the shift to UTC, and find new ways to remind people that it's impolite to call people in the middle of the night.
In the long run, a list of time zones aren't that much more complicated to work with than a list of times when it's mid-day is in the major population centres on the planet.
Yeah, my point being that 1 minute passing is 60 seconds both on Earth and the moon. We have a Lunar calendar and the moon is tidally locked to Earth so it seems trivial to convert Earth time to Moon time. The terms "Earth time" and "Moon time" seem weird to me in themselves, in my view it's "Human time".
I understand having a separate time for Mars or other planets as they have their own cycles.
> The terms "Earth time" and "Moon time" seem weird to me in themselves, in my view it's "Human time".
- It's also weird the time on the moon would have to account for variations in earth rotatoin speed (as UTC does)
- Time on the moon actually goes slower due to relativistic effect - the effect is small (in tens of microseconds per day) but measurable.
Yes the article states that the time goes slower on the moon but 56 microseconds every day is negligible. (there is 1000000 microseconds in every second)
We can surely skip one second every 27 years on the moon if we are able to add 1 day every 4 years on earth.
Are relativity effects significant on the moon? I thought a rationale for this was the need to account for relativistic effects making clocks run at slightly different speeds on the moon than on the surface of the earth.
Is the "timezone" more of conversation protocol for earth time to moon time?
Current missions use a synchronised mission clock and don't last long enough for this to be a big issue right?
Yes, as the article states, clocks run faster on the Moon. About 56 microseconds per day.
Be that as it may, as far as I know, Newtonian gravity was sufficient for the moon landings. So I'm guessing this is more of a coordination issue between countries than a necessity for trajectories or such.
You age at the same speed. Stuff that is far away from you may age more or less rapidly, but that effect only exists at the point where some kind of communication travels between you and the faraway stuff.
Technically yes - but the difference amounts to 2 seconds per century. Your lunar clone would be younger than you, but the difference would be less time than you spent calculating the difference.
How about just picking a non-cyclical time? Just count the amount of seconds/milliseconds since 1 Jan 1970 where the duration of the second is synced to those on earth. Just have no official days/hours/years whatever, no leap seconds, no adjustments, no nothing. Those details can be filled in at will by whoever needs lunar time for daily use.
If you’re on the same planet and or moon thereof, it should “be” the same time wherever you go. Everybody should use UTC always. No more scheduling “wait did you mean EST or CT” bs, no more timestamp BS, no more BIOS —sysclocktohw. We’re evolved apes, we should use UTC.
Just name the zones after the predominant facing constellations / star at the epoch of this new standard. 12 or 13 zones might be a good amount? Too many?
This is perhaps the least political potential solution I could think of.
Did you read the article? The mass of the rock you happen to be standing on determines how fast time passes, so "UTC everywhere" would result in longer or shorter seconds depending on location.
why not just leaving it UTC? it's not like time zones make sense on Earth anymore so why start it? plus there are no Earth-like daylight patterns there that humans could possibly adapt their sleep to, lasting weeks at a time
Basically, everything is roughly in sync with Earth and uses similar or identical units, except the time of day which is synced to the martian solar day, the only local time of relevance to the martians, for things like working light and circadian rhythm, solar panel irradiation etc. The "seasons" of Mars are of little relevance due to the very artificial conditions in which the martians will live, the weather outside is deadly all year long with slight variations in deadliness.
So the entire idea of keeping a martian calendar with funny month names doesn't really make sense, the only thing martians will care is the time of the next Earth transport and the dates on Earth so they can relate with their loved ones. As is the idea of having different lengths of seconds and hours, which will render almost all Earth equipment and measurements unusable, as opposed to a slight tweak in the software of their clocks to enable times past 23:59:59.