If that is what you mean, then of course you are correct. There is an SI atomic second. I was saying there isn’t really a “metric day”, at least not one used for civil timekeeping. Universal Time, aka UT1 is known to be a non-uniform scale of time related to the rotation of the Earth (or sidereal time, or however you choose to regard it) and is totally affected by tides, the weather, seismic activity, you name it. It is used to set your watch (including for celestial navigation, when that is still performed), not for an unequivocal unchanging absolute measure of time, because it isn’t.
Sorry to be a little too cryptic / hyperbolic. I was trying to get @bob_2 to see the category error in considering “a day” or “a year” to be anything but fuzzy vague ideas. They aren’t a thing which can be said to actually exist.
But we can further refine the vague conceptual ideas of “day” and “year” into multiple competing definitions of increasing precision and rigor as @Chronos filled in a couple posts later and as you say just above.
For those who didn’t grow up learning only Celcius…
100C - Boiling
30C - Hot
20C - Warm
10C - Cold
0C - Freezing
As a Canadian in my 30’s, I’ve only been exposed to Celcius in media / tech / academia / etc, so F is entirely gobbledigook.
But even when exposed to F, there is never any need to convert between scales (F → C or C → F) as the context tends to be so precise as to never engage with my mental Celcius-clock/Celici-ometer/Celci-ograph(?).
For example, a recipe that tells me to "cook dish at ###F " might as well say “cook dish at 
359F”. Oven temps in F become just a dial that I turn the knob to, and doesn’t require mental engagement.
“Celsius.” That said, degrees F is all over the place anyway, you don’t notice it? At the very least, all Canadians are heavily exposed to American media. I grew up learning only metric but have a solid handle on the various points of the Fahrenheit scale by pure osmosis.
Without a rigourous spellchecking algorithm I simply would not be able to survive in the modern world.
As for exposure, I simply never been exposed to F enough to internalize what the numbers mean. I came from a household where C was used exclusively (in room temp, weather reporting, etc). As for American media, I mentally engaged with temperature numbers very little (just with the general sentiments). For example if I heard “wow, its ###F outside, put on a coat” I’d think “wow, its [insert cold temp] outside put on a coat”.
I honestly could not tell you what the numbers mean (perhaps its different for others of my age, but for me, I’m lost with F).
Deleted upon edit
This I think is taking pedantry too far. Indeed, there are several different meanings of the word “day”:
- The civil day, defined as the time when it’s bright outside.
- The sidereal day, defined as a full rotation of the Earth around its own axis.
- The apparent solar day, defined as the time it takes for the Sun to revert to the same apparent position in the sky (which is more than the sidereal day, because the Earth does not only rotate around itself but also revolves around the Sun and needs to rotate a little extra to bring the Sun back to the same apparent position on the sky, as seen from Earth).
- The mean solar day, which averages out seasonal differences in the length of the apparent solar days.
All of these are based on actual astronomical motions of the Earth, and all of them are useful. The sidereal day is useful for observations of objects outside the solar system. The civil day and the apparent solar day are useful in daily life because it is on which our everyday rhythm of darkness and bright daylight depends. The mean solar day is useful because our timekeeping depends on it. It is true that the length of the mean solar day changes; there is a slow but steady slowing-down. But the leap second mechanism is in place to ensure that the time as counted by atomic clocks is periodically brought back into sync with the mean solar day. So the mean solar day, based on the Earth’s motion, certainly is still the basis of our timekeeping, also in a scientific sense.
Pedantry is par for the course on this board and fully expected.
Sure, but it should at least be accurate. Claiming that no useful meaning of the word “day” is in any way related to planetary motion is not accurate.
Oh! I’m with you all day long.
In addition, Newton’s Law of Cooling describes the change in temperature of an object (or system) over time based on initial temperature, object temperature and time, with a similar dependence on exponents of e.
I hadn’t thought of the Hot/Warm/Cold temps that way. Neat!
For warmer regions, 40C is a bit of key temp, meaning “Blistering Hot”.
Kind of like how we say “It’s over a hundred” for our hottest days in the States.
So, we could do weather reports with the simplest of terms:
40C - Hotter than Hell (104F)
30C - Hot (86F)
20C - Warm (a.k.a. Comfortable) (68F)
10C - Cold (50F)
0C - Freezing (32F)
I’m not sure how much more precise I need in my weather.
And looking at the F values above, everything matches nicely.
For a chart about Celsius in Canada, @orcenio 's listing is notably missing the below zero section. 
Mr home thermostat is set at a comfy 22 degrees C
When Canada switched our morning show host came out with:
0 is freezing
10 is not
20 is pleasing
30 is hot
That’s all you need to know for everyday temperatures. I do readily convert between the scales though. For example, today’s high is predicted to be 22 which is getting up close to 80.
Perhaps you’d care to expand the scale downwards a bit? 
e.g.
-10: Southern USAians freeze solid. Canadians close the screen door.
-20: Northern USAians start whining. Canadians put on a windbreaker.
-30: All USAians have died. Canadians complain their Hortons coffee gets cold too quickly while standing outside the ice fishing hut enjoying the breeze.
-40: Even Canadians put on gloves and a hat.
-50: Sorry, @LSLGuy can’t even imagine this happening on a planet fit for human habitation.
That seems about right. 
Where I live, we have a 70 degree C range each year: we can usually expect a few days around -35 in the winter, and a few days around +35 in the summer.
It’s a bit unusual to have an 80 degree C annual range, but it does happen. A few years ago we had +40 in the summer and -40 in the winter. (Car died when it was -40, so I had to walk to work.)
I think -45 is the lowest I’ve ever been in. Not counting wind-chill, of course.
Indeed 40C is a key temp (infrequent up here, but often enough to note it).
100C - Boiling (212F)
40C - Hotter than Hell (104F)
30C - Hot (86F)
20C - Warm (a.k.a. Comfortable) (68F)
10C - Cold (50F)
0C - Freezing (32F)
The perfect scale C-F / cross border, universal temperture conversion scale for the Yanks (and clueless Canucks like myself).
I’m not sure I can interalize the F system thou. Those numbers just… FEEL random.
True, my scale/advice stopped at 0C because, for me, every single degree below freezing is important. But if I had to…
40C - Hotter than Hell (104F)
30C - Hot (86F)
20C - Warm (a.k.a. Comfortable) (68F)
10C - Cold (50F)
0C - Freezing (32F)
-10C - Skating temp (14F)
-20C - Stay inside (-4F)
-30C - Colder than Hell, even ice-skating is difficult at this temp. What are you doing? (-22F)
Well, if you had internalized the F system you’d use slightly different points that didn’t seem so random:
100 - Hotter than Hell
85 - Hot
70 - Comfortable
50 - Cold
32 - Freezing (no way around that one, or 212 for boiling)