For temperature, don’t forget the Réaumur scale which is exactly 4/5 of the Celcius. When I lived for a year in Switzerland, there was a thermometer outside one window of our apartment that gave temperature in both C and R and I realized that C + R + 32 = F.
In Norway that would be “give someone the pinkie and they’ll take the hand”.
Now if Norway had had the expression “missed it by a mile” it would have been a much worse miss since Norway had a much longer mile and we converted it in everyday usage to a “metric mile” of 10 km.
A person can get lazy working purely in metric. Bang out the math, look to see where the numbers fetched up, slap on the correct symbol and boom you’re done. If you forget a thousand here or there, what’s the harm? Now it’s a kilowhatzit.
When I started off studying chemical engineering at Canadian university, they’d give first-year students computational problems in a mixture of units, both metric and Imperial, with the explicit intent of slapping that bad habit out of our heads. If you winged it and missed a conversion, you were now off by 3.7 or something. Is that a fortnight/angstrom? A degree Rankine/pound? Who knows! You’re boned!
And thus we learned the importance of keeping track of all units at all times, because math errors can kill people.
(Also thanks for the reminder about the “barn”, because that’s one I’d forgotten.)
It demonstrates that you have to start arbitrarily from somewhere when you design a system. The second was, historically, derived from minutes, hours, and days: You divided the mean solar day into 24 hours, each hour into 60 minutes, and each minute into 60 seconds (with Babylonian number magic as the reason why 24 and 60 were chosen as base numbers). Later on, when the metric system (or, more correctly, the SI system) was designed, some unit had to be found for time as one of the fundamental quantities of the system), and the second was chosen for this purpose. For a long time, the second was, for these purposes, defined as the 86,400th fraction of the mean solar day; meanwhile, a new definition based on caesium atomic clocks has been codified, but that definition was calibrated intentionally so as to mirror the old definition of the second as a 86,400th mean solar day.
So it is true that the second is metric and days, hours, minutes, and seconds are not, but at the same time it is true that the second was derived from days, hours, and minutes.
All that Fahrenheit, Newton, Réaumur, Delisle stuff carries a distinct eighteenth-century vibe. Anyway, the point of a metric scale is not so much whether the numbers go up to 80 or 100 or 150, but that everybody uses the same scale. As for the size of a gallon or a pint or a ton in Canada versus in the USA, I would assume you get the smaller unit when buying, unless indicated otherwise 
The only place I ever heard of Réaumur was on the famous data visualization chart depicting Napoleon’s Russian campaign of 1812:
The temperatures on the chart are all given in Réaumur. Now I know how to convert them.
IIRC, there’s also an Einstein temperature scale. It’s logarithmic, so you can have any real number as a temperature.
Does that mean that negative (absolute) temperatures are represented as complex numbers?
Let me elaborate on that a bit. Originally, the day was divided into 12 equal hours, whose length varied during the year and the day into 4 watches. Incidentally, the ninth hour, around 3PM today, was dinner hour and we derive the word noon from some language’s word for nine. Later, the day and night were combined into 24 equal hours, whose length now didn’t vary during the year. Then a smaller division than hours were needed and they introduced minutes, meaning a small part (related to the word pronounced my-noot). Why 60? Well, I think it must go back to the sixty minutes in a degree of an angle, ultimately back to the Babylonians. When an even finer division became desirable, they introduced the “second minute” later shortened to “second”. Decimalizing all this would be a royal pain, much more complicated than decimalizing distance and mass.
France, of course, tried time decimalization a few times. It never took off.
France also tried to use the second as a basis for a new unit of distance, namely by defining that unit as the length of a pendulum that takes one second for half a swing. That pendulum has a length of almost exactly one metre, but that’s a coincidence; the meter was defined on the basis of measurements of the circumference of the Earth and is not related to the seconds pendulum.
Whatever way you measure it, a day is strictly related to the rotation of planet earth, and a year is the time it takes to orbit the sun. It’s not like temperature scales, Celsius, Fareinheight, Kelvin etc, where nothing (apart from absolute zero) is fixed and a degree can be anything you want.
You said those two things slightly differently. I doubt that was deliberate, but the distinctins matter.
in fact, both the day and the year are loosely related to the motions of the planet, for modern notions of “loose” and “tight”. In no useful sense is a day or a year the time it takes for the planet to make some motion.
There are many different technically distinct kinds of days and equally distinct kinds of years. They are meant to be invariant while the actual planetary motions are everchanging, both in predictable and unpredictable ways.
What do you mean here? I thought that the civil day, when the shops open, had to do with whether the sun was in the sky, which is exactly related to the motion of the planet. The civil authorities still think so, too, which is why there are (for now) leap seconds and we are not quite on atomic time.
In another thread I recall an amusing discussion of measuring or relating thermodynamic temperatures to gigabits per nanojoules… I suppose that for some weird systems this could be negative or infinite or zero…
OK, that’s an example of how complicated things can get. One perfectly reasonable definition of “a day” is indeed “the time from the first ray of one sunrise, to the first ray of the next sunrise”. But that “day” will vary in length over the course of the year, and also vary with latitude. So the next step would be to define “a day” as the average length of that time period. That works fairly well… but that’s about four minutes longer than the time it takes the Earth to make one complete revolution. And there are, in fact, clocks that are based on the time to make one complete revolution, instead of the average time from one sunrise to the next.
Years are even worse: I can think of at least nine different definitions of a year, some of which are variable.
Even in “normal” temperature systems, absolute temperature can be negative. Temperature is just the ratio of change in energy vs. change in entropy, and in some unusual systems that can be negative (though oddly, negative temperatures are “hotter” than any positive temperature).
Chronos mentioned a log-based temperature system, which makes sense if your temperature varies over a wide scale, say from nanokelvin to gigakelvin. But I found it amusing that a negative absolute temperature would be represented as a complex temperature in a log-based system.
The whole gigabit/nanojoule thing is amusing as well, and harkens back to the deep relationship between information and entropy.
There does seem to be something logarithmic about the temperature, or at least the inverse of the temperature, if you consider something like a Boltzmann distribution: if k is Boltzmann’s constant and T is the temperature of a system in equilibrium, then the probability of finding the system in a state with energy E should be proportional to e−E/kT
That’s a sidereal day and the extra minutes are due to the Earth’s travel in orbit.
A day (by which I do mean 24 hours) is the time it takes the Earth to rotate one time in respect to the sun.
Which number is different every single day if you look at it closely enough. That’s what I was suggesting upthread that @DPRK seemed not to get.
There is no single unequivocal unchanging absolute measurable quantity of time that can be referenced to the motion of a planet. For the precision needed in Mesopotamia, sure, a “day” is sunrise to sunrise.