How to tell time
Nov. 22nd, 2018 10:34 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
kengrMoving off earth means a lot of things change. And if we not only get off earth, but manage to colonize other star systems, they change a lot.
On Earth we have our "simple" 24 hour clock, with 60 minute hours and 60 second minutes.
Those numbers are actually a lot better than most folks think. 24 is evenly divisible by 2, 3, 4, 6, 8 and 12. That makes it easy to split things up for shifts and the like. The sometimes suggested "decimal time" only lets you divide the day into 2 or 5 chunk without dealing with odd fractions.
And 60 is divisible by 2, 3, 4, 5, 6, 10, 12, 15, 20 and 30.
So a standard day will likely remain 24 hours/ 1440 minutes/ 86,400 seconds.
Now, on Mars or some hypothetical planet around another star the length of the "solar day" is going to be different. And it's important to not get the local day and the standard day confused.
I'm going to assume that we will follow the lead of the folks monitoring the various Mars landers and use sol for the local solar day.
We will want to divide the sol into useful periods. And do some *evenly*.
Why? consider colonists on Mars. The sol is an important unit, because many/most outdoor activities are going to depend on sunlight.
One commonly seen proposal is that we just keep hours, minutes, etc. That's kind of doable, but a royal pain.
Consider Mars. A mean Martian solar day, or "sol", is 24 hours, 39 minutes, and 35.244 seconds.
The guys on graveyard shift are not gonna appreciate having that extra 40 minutes in their shift.
And once you have more than one base on the planet (unless they are very close to each other, you have to deal with timezones. So that make the extra "40" minutes happen at different times of day depending on your longitude. Ick. That make things worse.
I've seen folks suggest that the extra time always be inserted between 2400 and midnight.
That means that the time difference between different bases will vary with time of day! Yeah, computers could handle that, but it'd still be a pain.
We could just use centisols, millisols, etc. But that runs into the same problems that decimal time did on Earth.
So, we need to divide the sols into a *convenient* number of chunks. We *don't* want to call the chunks hours, minutes and seconds as there are far too many critical things that depend on those terms having a fixed duration.
For example, how long a power unit or air tanks will last. Don't want to have to worry about whether it'll last for 8 standard hours or 8 "Martian hours".
So we need a different term for the subdivisions of the sol. Long ago (I've been thinking about this for something like 30 years) I came up with "peri (plural: peris) from "period".
David Brin came up with dura/duras in his Uplift books.
We will eventually need *several* names, because there may be multiple inhabited planets in a system. And it'd help avoid confusion if each had their own name for their hour equivalents.
How many duras you want to divide the sol into depends on the length of the sol. You'll still want numbers that divide up easily for shifts. But you'll (probably) want to keep duras not *too* far off from hours. This is a bit of a balancing game, and I'll go into suggestions on that another time.
When the sol is really short or really long we may want to combine several actual sols into the "legal" sol, or vice versa. Consider Luna. 28 days is a bit much to deal with. making the "legal" sol 1/28th of the actual sol would be a lot more convenient.
Though, come to think of it, we might want a different term for these fraction & composite sols.
Now let's go in the other direction. We need a term for the local year. Latin and other Romance languages are out because they are too close to annum and annual, which are terms we need to keep the current meaning of.
Greek proves to be an acceptable choice. Etos is the Greek word. Not apt to be confused with anything else (unless you are a Greek. :-)
The length of months varies enough on the standard calendar that we may be able to get away with re-using the term.
Weeks are a different matter. There are consumables and other things that are rated in weeks. Plus "week" gets *way* tied into various religions (I'll deal with *that* can of worms in another post).
Which brings up another matter. synchronization.
Let's use Mars as an example. Say it's 0000 hours on Sunday at both a base on Earth and one on Mars.
7 sols later it'll be 0437 on Sunday on Earth.
14 sols: 0914 Sunday
28 sols: 1828 Sunday
40 sols: 0223 Monday
That 's right. In 40 sols, Earth is more than a day ahead of Mars.
So there *will* be local clocks and calendars on any planet where you go outside with any frequency and where dayligh makes a difference.
Places where that's not true can happily continue on UTC.
On Earth we have our "simple" 24 hour clock, with 60 minute hours and 60 second minutes.
Those numbers are actually a lot better than most folks think. 24 is evenly divisible by 2, 3, 4, 6, 8 and 12. That makes it easy to split things up for shifts and the like. The sometimes suggested "decimal time" only lets you divide the day into 2 or 5 chunk without dealing with odd fractions.
And 60 is divisible by 2, 3, 4, 5, 6, 10, 12, 15, 20 and 30.
So a standard day will likely remain 24 hours/ 1440 minutes/ 86,400 seconds.
Now, on Mars or some hypothetical planet around another star the length of the "solar day" is going to be different. And it's important to not get the local day and the standard day confused.
I'm going to assume that we will follow the lead of the folks monitoring the various Mars landers and use sol for the local solar day.
We will want to divide the sol into useful periods. And do some *evenly*.
Why? consider colonists on Mars. The sol is an important unit, because many/most outdoor activities are going to depend on sunlight.
One commonly seen proposal is that we just keep hours, minutes, etc. That's kind of doable, but a royal pain.
Consider Mars. A mean Martian solar day, or "sol", is 24 hours, 39 minutes, and 35.244 seconds.
The guys on graveyard shift are not gonna appreciate having that extra 40 minutes in their shift.
And once you have more than one base on the planet (unless they are very close to each other, you have to deal with timezones. So that make the extra "40" minutes happen at different times of day depending on your longitude. Ick. That make things worse.
I've seen folks suggest that the extra time always be inserted between 2400 and midnight.
That means that the time difference between different bases will vary with time of day! Yeah, computers could handle that, but it'd still be a pain.
We could just use centisols, millisols, etc. But that runs into the same problems that decimal time did on Earth.
So, we need to divide the sols into a *convenient* number of chunks. We *don't* want to call the chunks hours, minutes and seconds as there are far too many critical things that depend on those terms having a fixed duration.
For example, how long a power unit or air tanks will last. Don't want to have to worry about whether it'll last for 8 standard hours or 8 "Martian hours".
So we need a different term for the subdivisions of the sol. Long ago (I've been thinking about this for something like 30 years) I came up with "peri (plural: peris) from "period".
David Brin came up with dura/duras in his Uplift books.
We will eventually need *several* names, because there may be multiple inhabited planets in a system. And it'd help avoid confusion if each had their own name for their hour equivalents.
How many duras you want to divide the sol into depends on the length of the sol. You'll still want numbers that divide up easily for shifts. But you'll (probably) want to keep duras not *too* far off from hours. This is a bit of a balancing game, and I'll go into suggestions on that another time.
When the sol is really short or really long we may want to combine several actual sols into the "legal" sol, or vice versa. Consider Luna. 28 days is a bit much to deal with. making the "legal" sol 1/28th of the actual sol would be a lot more convenient.
Though, come to think of it, we might want a different term for these fraction & composite sols.
Now let's go in the other direction. We need a term for the local year. Latin and other Romance languages are out because they are too close to annum and annual, which are terms we need to keep the current meaning of.
Greek proves to be an acceptable choice. Etos is the Greek word. Not apt to be confused with anything else (unless you are a Greek. :-)
The length of months varies enough on the standard calendar that we may be able to get away with re-using the term.
Weeks are a different matter. There are consumables and other things that are rated in weeks. Plus "week" gets *way* tied into various religions (I'll deal with *that* can of worms in another post).
Which brings up another matter. synchronization.
Let's use Mars as an example. Say it's 0000 hours on Sunday at both a base on Earth and one on Mars.
7 sols later it'll be 0437 on Sunday on Earth.
14 sols: 0914 Sunday
28 sols: 1828 Sunday
40 sols: 0223 Monday
That 's right. In 40 sols, Earth is more than a day ahead of Mars.
So there *will* be local clocks and calendars on any planet where you go outside with any frequency and where dayligh makes a difference.
Places where that's not true can happily continue on UTC.
no subject
Date: 2018-11-22 08:59 pm (UTC)So.. M-seconds would be fractionally longer than E-seconds. Although an M-year would be longer than an E-year because rotational period and orbital period are not the same.
It would be confusing probably... but if you think about it, people really don't think much on the time scale of years, just months and days.
no subject
Date: 2018-11-23 05:54 am (UTC)Altering the length of the second would *totally* screw with damn near every bit of science and engineering there is.
Hower, if you want to go that way, you can buy a clock that keeps Martian time:
https://www.bmumford.com/clocks/martian/
At $150, I'm not likely to buy one.
Though I also see some $15 binary clocks that keeps Mars time...
https://www.etsy.com/listing/235511151/martian-mars-time-clock-kit
no subject
Date: 2018-11-23 12:25 pm (UTC)But what about after the planet is fully self-sustainable, and is manufacturing their own equipment, building their own oxygen and fuel tanks?
Also (as an aside): Are you familiar with the YouTube channel Artifexian? He focuses on worldbuilding and conlangs for the purposes of writing sci-fi and fantasy, at just a deep enough level for the author to keep everything consistent, and for the reader to get that the story is set in an alternate world.
no subject
Date: 2018-11-23 02:28 pm (UTC)It'd be an insane amount of work, for no good purpose.
Folks would use sols and duras in everyday life, but use UTC for coordinating with Earth and other colonies, bases and ships.
You need a consistent set of time & date values so *everyone* can coordinate.
Not that much different from the way we are more interested in clock time (which tracks somewhat with mean solar time) than with *actual* solar time.
There'd be a local calendar as well.
no subject
Date: 2018-11-23 04:27 pm (UTC)So, as of this moment, an Earth second is the time it takes for the electron in the outermost shell of a Caesium 133 atom to oscillate 9,192,631,770 times. And this is a good standard, because it's consistent even off the planet -- anywhere there's Caesium 133 around, clocks can be synchronized.
But in terms of daily life: getting the kids off to school, catching the shuttle, arranging meetings, etc., why couldn't we just say "An Earth second is 9,192,631,770 Caesium 133 oscillations long, and Martian Second is 9,431,242,593.416707 oscillations long." (Which is, presumably, how those novelty Martian clocks actually work)?
It's not actually a different unit of measurement, since both the Martian and Terran units of time are based on the same fundamental particle of the universe (until we find a new, better, and even more precise constant to base our seconds on).
Although, as pointed out in this video from last year: Martian Evolution|Space Time, colonists on Mars may undergo rapid physical changes in responses to their new environment, and want to avoid contact with Earth as much as possible.
no subject
Date: 2018-11-23 09:24 pm (UTC)Frequencies (radio, tv and a lot of other things) are defined by the number of oscillations in a second. Any physical constant hat has time as a dimension would change.
Heck, the length of a *meter* would change because the speed of light is a *defined* constant, and the meter is *derived* from that and the second.
You are talking millions of man hours to refigure things and to no good purpose.
Using some sort of "native" time system with different names is much easier all around.
So Mars has a 24 dura sol. That'd make a dura 3698.97 seconds. That'd be 61 minutes 38.9685 seconds.
Duras are divided into subunits some way or another. And those units have a name. Or you could just go decimal from there.
decidura: ~6 min, 10 seconds
centidura: ~37 seconds
millidura: ~ 4 seconds
So: "My shift starts at 7.15"
"The show starts at 18.50"
"I'll be with you in a couple of decs".
"It'll be just a few more cents"
"It only takes a few mills longer"
And the clocks & watches that more or less keep Martian time, do so based on oscillators that are based on seconds. For example one watch that can be programmed to show Martian time uses 32.768 kHz oscillator. For regular time the processor inside that watch has a counter that counts the oscillations. every time it hits 32768, it increments the time display by one second and resets the counter.
For Mars time, they have a counter run up to a different number. And since that value isn't *quite* right, they use one value 5 times and a different value 2 times so the length of the Martian "seconds averages out very close to the right length.
The mechanical clocks and watches use different tricks.
no subject
Date: 2018-11-28 01:27 pm (UTC)I think you and I are coming at this from different angles, though: you, from an engineering perspective, and I, from a linguistic one.
I mean, here in America, we're stubborn provincials, and still measure everything in inches and miles. Strictly speaking, though, we measure in units called "inches," those units are nonetheless defined by how many centimeters they are.
And the ancient British defined an "inch" as three mature, uncooked, barley grains lined end-to-end. The unit of measure has been redefined in terms of its reference points, and also changed its actual size from the time of the ancient Romans until today. But we have not changed the word for it (Much to the frustration of scientists and engineers, but we can still communicate with each other using ancient vocabulary, even among people who wouldn't know a barleycorn if they saw it in their soup.
So I think you're probably right that there will be new names for Martian units of time, at least for those on Mars who are in regular contact with their counterparts on Earth. I just wouldn't be surprised if, among those who are focused more on life Mars-side, the old names for time just get stretchier.
Also, practically speaking, I think it would make more sense to divide the units of time into twelfths and sixtieths, like it is on Earth, rather than dividing each dura in units of base ten; the French tried that, after their revolution in the 18th Century, and it just didn't work. Using the thumb of your active hand as a "pointer" you can easily count to twelve on your fingers, by counting the joints of your other four fingers. And then, you can keep track of how many units of twelve you've got by counting off on the fingers and thumb of your other hand. So using a combo of twelve and sixty is organic to the human species; no matter how quickly speciation happens between Earthlings and Martians, I don't the number of fingers or finger joints will change.
no subject
Date: 2018-11-28 10:14 pm (UTC)The hand motions aren't hard to lock in, and in Boy Scouts my patrol used it to keep track of paces on orienteering courses.
Yeah, 60 and 12/24 are useful divisions. I just used decimal divisions of duras it that example to show that it could be done workably, and to avoid coming up with two more unit names. :-)
As for old definitions of units, at one time the foot was 1/12th of the combined length of the first 12 men to come out of the local church on Sunday.
Lord's used this to manipulate the size. They wanted big feet for surveying the land for figuring out what was owed to the King, and small feet when it was time to figure out what was owed to them...
ps. part of the reason that "metric time" failed was that they divided the day into 10 hours. That's way too long to be useful. And the 10 day week meant more days of work in proportion to the days(s) of rest.
hmmm. Imagine an alternate universe where they decided to shift to base 12 after the revolution and did all the metric units that way.
no subject
Date: 2018-11-29 09:09 pm (UTC)Well, yeah, if you've got some Moral Aversion to Leisure, and think you should only have one day of rest per week -- nine days of work for one day of rest is really unsustainable. But if they had split the week in half, and had one day of rest every fifth day, that's more like what we have now. In my head, that's just a pinch more generous. But I don't have the brain, right now, to actually do the math.
Imagine an alternate universe where they decided to shift to base 12 after the revolution and did all the metric units that way
Indeed. Depending on what you choose as your frame of reference, base 10 is not inherently closer to Pure Reason than base 12.
Arguably, counting in American Sign Language verges on Base 5, but because it's in cultural contact with English, it slips toward base 10 after 25. (ASL is in the same family as langue des signes française, but I can't find a video for LSF counting tutorials).
Well ...
Date: 2018-11-23 07:41 am (UTC)This has started so many wars.
Re: Well ...
Date: 2018-11-23 12:44 pm (UTC)Re: Well ...
Date: 2018-11-23 01:35 pm (UTC)We also vary a lot in how far we can "adjust" our rhythms.
Re: Well ...
Date: 2018-11-23 07:49 pm (UTC)Re: Well ...
Date: 2018-11-23 09:30 pm (UTC)But (like folks on Luna) they'll probably run on UTC and just check the surface time/date when they need to schedule some surface activity.
Having a city that was *designed* around different schedules all being equal would be wonderful for us "natural night owls"
Re: Well ...
Date: 2018-11-23 09:34 pm (UTC)Re: Well ...
Date: 2018-11-23 09:56 pm (UTC)Be about as important as tide tables are to non-coastal folks.
We actually have about a 3 foot tide here in Portland. Even this far upriver from the coast!