Hacker News Reader

Comment by ballenf | Hacker News Reader

Has a "metric" version of time ever been proposed? Seems odd to me that we're "ok" with divisions of 12 for time but not length.

I guess for science seconds are basically the only measure of time that usually matters?

View on HN

The French Revolution attempted a decimal calendar alongside their switch to metric. It's a fascinating artifact: https://en.wikipedia.org/wiki/French_Republican_calendar

Calendars are hard because you've got multiple inconveniently disparate sources of information you want to unite: universal constants on the one hand (such as the second being tied now to atomic vibration) and more "local" concerns such as the orbital position of the earth with reference to the sun, the moon, other neighboring objects in the solar system.

Vernor Vinge's novels, as one example, use metric prefixes and seconds exclusively (this is referred to as Metric time [1]), which is a fun thought experiment. The biggest complaints are that the units aren't necessarily great for human activities and definitely don't align well with minutes/hours/days/weeks/months.

Hectosecond is close to a minute, but slightly larger (1.666 minutes). Kilosecond is about 16.666 minutes long, which is the closest unit to an hour. It's almost a useful quarter-hour (but again it doesn't line up that well). But then you hit the order of magnitude wall and the next prefix up is Megasecond which is nearly, but not quite a fortnight (~11 days), and Gigasecond is just shy of 32 years. You long at that point for more prefixes between Kilo-, Mega-, and Giga- if you are using seconds as base unit at that point, especially living on Earth and trying to coordinate calendar weeks, months, years. (There are non-standard prefixes Myria- (10^5), also from the French revolution, and Hebdo- (10^7) which are almost useful enough here to beg them to be standardized.)

[1] https://en.wikipedia.org/wiki/Metric_time

Super interesting answer thanks!

There has been a proposal during the French Revolution with 10 hours a day, 100 minutes an hour and 100 seconds a minute. But it has only been in use for a few years beginning in 1792: https://en.wikipedia.org/wiki/Decimal_time

(10/24)(100/60)(100/60)=1.1574

Our biggest revolution yet: extending the length of the day by 16%!

Presumably, the second was shorter, rather than the day was longer.

EDIT: More specifically, if you take an average rotation of the Earth as 86164.1 seconds, and divide by 10, then 100, then 100, you get 0.861641, which would be the conversion from our seconds to the metric second in that scenario. Likewise, the metric minute would be 86.1641 of our seconds or about 1.43607 of our minutes, the metric hour would be 8616.41 of our seconds or 143.607 of our minutes or 2.39345 of our hours. The day would be slightly shorter than our day, reducing the need for leap seconds to keep our time aligned with the sun.

You could just define the second to be a 1/100,000 of a day and go from there. Aside from screwing with literally everyones perception of time, it could work.

But which day? Such a second would probably work for everyday purposes (just like to most of us a second is 1/86400 of a day) but it would be woefully inadequate as a universal fundamental unit of time. And defining the second in terms of something like "the length of the solar day at Greenwich meridian on Jan 1 2020" prevents anyone from ever reproducing the exact value based on their own measurements, which is one of the big reasons for defining fundamental units in term of physical constants.

This is especially true when you have things like large earthquakes[0] changing the rotational speed of the earth slightly.

[0]: https://www.nasa.gov/topics/earth/features/japanquake/earth2...

Yep, things leap seconds have been used for. The moon also constantly robs Earth angular momentum via tidal interactions.

But an even bigger issue is that the sidereal day (one 360° rotation of Earth as measured against distant stars) is not 24 hours but roughly four minutes less. And the length of the solar day also varies over the year due to the slight eccentricity of Earth’s orbit—days near perihelion are slightly longer than near aphelion. And then there are the higher-order effects caused by gravitational interaction with other planets...

One can arbitrarily define a second as anything and then say that one UTC day usually has 100000 seconds and some are longer or shorter. It is just a matter of scaling few unit definition constants by 10000/86400.

24 and 60 do make sense because they are highly composite numbers: https://en.wikipedia.org/wiki/Highly_composite_number

You could say the same thing about most Imperial units, but most people still prefer the metric system.

I would say most people prefer the one they learned first. For doing science metric is usually better. Easier unit conversions. I think inches/feet are better for construction as you have more options at what level of accuracy you want to work with and splitting things up unless you need to have a fifth of something.

And to think, we could unify Imperial and metric if we just shift the whole world to a duodecimal number system! :P

That's actually exactly why I like working in Imperial when woodworking. You're composing and dividing things by 2 all the time, so a base-2 fractional system is great.

12 and 60 are superior highly composite numbers¹ — as is 360, the number of degrees in a circle and the number of days in a year (with a little engineering work TBD).

Speaking of circles, SI hasn't fixed Hz vs rad, have they?

¹ https://en.wikipedia.org/wiki/Superior_highly_composite_numb...

> Speaking of circles, SI hasn't fixed Hz vs rad, have they?

I don't think so.

Anyone who hasn't heard of the Hz/radian issue should see the "Hz" definition in the units definition file shipped along with Frink - if you like finding interesting rants in unexpected places, it's delightful:

https://futureboy.us/frinkdata/units.txt

Appropriately, it now contains several outdated rants, showing vast impact these redefinition have: we'll all get an update to units.txt soon!

Yeah, I don't think it's really an issue. Physicists only convert things to hours/days/years when disseminating results to a broader audience, when doing the actual physics everything is microseconds or megaseconds or whatever.

In scientific literature we do typically use one particular preferred time unit (second being the SI units) and then exponentiating properly (eg optically driven metal to insulator transition in V2O3 occurs in 1.3 picoseconds (10^-12 s), desorption of atmospheric gasses from CoPc thin film in vacuum takes ~10^5 s (~1 day)).

Heinlein's "Starship Troopers" has the space marines using "kiloseconds" and "megaseconds" conversationally, and it was always tricky for me to convert those to hours and minutes in my head accurately.

Odd choices... since 1000 seconds is ~17 minutes and 1M seconds is 4.6 hours.

> 1M seconds is 4.6 hours.

Really? :)

I think the parent meant 4+6i.

  $ units --verbose
  Currency exchange rates from FloatRates (USD base) on 2018-10-25 
  3070 units, 109 prefixes, 109 nonlinear units

  You have: 1 megasecond
  You want: 1 day
          1 megasecond = 11.574074 * 1 day
          1 megasecond = (1 / 0.0864) * 1 day

Gee, I never knew about units, thank you. Very cool. It can convert meters/s to furlongs/fortnight, cm^3 to gallons.. Write fractions like '1|2'. And you can add your own units to the units file.

Also in the book "A Deepness in the Sky" by Vernor Vinge. In the beginning of the book there was a convenient chart for converting to our customary earthly time units.

And in Dukaj's "Perfect Imperfection" [1] occasionally multiples of Planck time are used (with metric prefixes), as the post-humans and "out-of-space computers" can operate on very small time scales.

[1] https://en.wikipedia.org/wiki/Perfect_Imperfection

I'm sure measuring "days" becomes silly when you're not spending a significant amount of time on a rotating spheroid orbiting a single star.

Rounding up to 4Ks is a reasonable substitute for an hour. Also gives you the convenient 15min ~ 1Ks or "a quarter of a metric hour."

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

> Instead of hours and minutes, the mean solar day is divided into 1000 parts called ".beats". Each .beat is equal to one decimal minute in the French Revolutionary decimal time system and lasts 1 minute and 26.4 seconds (86.4 seconds) in standard time. Times are notated as a 3-digit number out of 1000 after midnight. So, @248 would indicate a time 248 .beats after midnight representing 248/1000 of a day, just over 5 hours and 57 minutes.

Yes, for science, seconds are sufficient (and metric). For everyone else, convenient computing of fractions is more important when dealing with time, so nobody seems to be interested in changing the time units: https://en.m.wikipedia.org/wiki/Sexagesimal https://en.m.wikipedia.org/wiki/Duodecimal

Years (or more specifically millions of years) gets used sometimes when talking about astrophysics, evolution or continental drift. (I am not a researcher and definitely not in any of those field though, so this may just be in material aimed at non-scientists). But I guess there's enough orders of magnitude of different that they rarely cross over so no one cares about the odd ratio, similar to J vs eV.

Seconds are the metric unit of time if I understand it correctly. It doesn’t make any sense to say “kiloseconds” or something like that because days are the time that the sun is up (ish), months are roughly based on moon cycles, and years are approximately the time it takes us to go around the sun.

It’s also historical and pretty hard to change the basic units of time.

FWIW, while scientists don't usually use kiloseconds, it's pretty common to see 1.437e4 seconds, rather than 3h 43min 20sec, used as units in time series etc. Similarly one uses milliseconds, femtoseconds, and soforth.

Metric had time from the start. However it was work 4 days, 1 day weekend. However the loss of personal time was very unpopular so it never caught on.

I'm not sure why a metric hour didn't catch on.

I think I'm missing the joke. Mind explaining it?

No joke, except I was referring to decmial time not metric time. https://en.wikipedia.org/wiki/Metric_time https://en.wikipedia.org/wiki/Decimal_time

a 7 day week gives two days weekends every 5 days, so you quickly lose personal time on the 4 day week.

Yep.

In Vernor Vinge's fiction, spacefaring humans even measure time in kiloseconds and megaseconds. 1 ks is slightly more than a quarter of an hour; 1 Ms is about eleven and a half days.

If we invent a system without timezones or daylight savings time I will adopt it in a heartbeat. I don't even care if it's not 10-based.

just set the timezone of your devices to UTC, and you're done! Actually getting other people to schedule meetings with you in UTC is going to be a bit trickier though.

Interestingly, a week can roughly be approximated as 10k minutes (with an error of less than 1%).

And a year is PI * 10^7 seconds (within 0.5%).

That's because the Earth's orbit is within 0.5% of a perfect circle.

(Just kidding.)

It makes 'more sense' to change the metric system to be base 12, because base 12 is very nice.

And really, you can use kilo seconds or terra seconds just as easily as micro seconds. Its just harder to relate to as human time-scales.

Reply on news.ycombinator.com