I mean, there's a big important point where you start getting solid things falling on your head and slick icy roads and ruined fruit-plants, so it ought to be at more meaningful spot, like 0.
I'm a scientist and an engineer, and in those applications C is often better. But in daily life, Fahrenheit makes a ton of sense.
20 C - a bit less than room temperature (68 F)
30 C - pretty hot (86 F)
40 C - very hot (104 F)
And going the other way:
10 C - a bit chilly (50 F)
0 C - cold (32 F)
-10 C - real cold (14 F)
-20 C - really really cold (-4 F)
When you hear a temperature in Celsius, don't try to convert it to Fahrenheit, just think in Celsius. 15 degrees is halfway between "chilly" 10 C and "room temp" 20 C. One degree of warming - say 26 C to 27 C - is pretty easy to conceptualize when you have those 10-degree reference points.
If it was always reported to the half point I think we would be able to relate to Celsius better. And if I was making a population wide change I’d ensure reporting was always to the half degree.
Those people are free to use whatever units they like. It doesn't have much bearing on the point that Fahrenheit, which is used in the United States, is a good match for the climate of most of the United States.
OTOH, as far as weather goes, C has the nice property that anything below zero is freezing temperature. In any locale where snow and ice is a thing, that's handy to know.
Yesterday it was +20 C, today it is +25 C, so it is noticeably warmer.
-10 C tomorrow, so you'll need your woolen cap and padded jacket. -20 C and add a layer of underwear underneath and maybe a scarf.
When you boil water for tea (100 C) and don't want the water boiling but still hot, aim for 95 C.
When the inner temperature of salmon is +46 C, it is perfect, if it is +56 C or more, it is thoroughly cooked and will not taste soft anymore.
Usually the relevant difference is 5 or 10 degrees.
I don't even want to think what those examples could be as Fahrenheit.
My point is it is eventually a matter of what you are used to, but Celsius is more logical, because it is directly tied to water freezing and boiling.
For the most logical measurements we'd use Kelvin, but it is not practical! "Oh, it is 273.15 K outside, might be black ice on the roads"
I must strongly disagree.
The freezing point of water is an incredibly important and pragmatic point of reference for hundreds of millions of people around the world, who live in places where water can freeze (or thaw) on its own outdoors. Not just for transportation, but also its affect on biology in agriculture.
While the boiling point is (thankfully) not important for weather-reports... Cooking! What would you do if I said to "simmer" something? You aren't supposed to go high enough to reach the obvious boiling point, so how are you supposed to know when you're close enough if you can't remember the magic number?
As far as boiling and cooking, I would argue that there is no magic number anyways, since boiling point is variable by elevation. At 7500 feet / ~2250 meters, water boils at 198°F.
That's an excellent point, let's examine the relationship between altitude and boiling point, in both American units and everywhere-else-in-the-world units [0]:
5000 feet -> 202.97 F
7500 feet -> 198.33
10000 feet -> 193.6 F
Compared to:
1000 meters -> 96.73 C
2000 meters -> 93.38 C
3000 meters -> 89.95 C
In both cases, the boiling points at altitude are magic numbers. The difference is that with Celsius, you can interpret degrees as "percentage of the way from freezing to boiling". Going from 100 C at 0 meters to 90 C at 3000 meters is immediately meaningful as a 10% decrease. With degrees Fahrenheit, that same 10% drop in boiling temperature that happens at around 3000 meters / 10,000 feet is 212 F to 194 F.
0: https://www.omnicalculator.com/chemistry/boiling-point-altit...
100 -> 90 is 10% decrease.
212 -> 194 is ~8.5% decrease (and I had to go out of my way to count that).