After making a weather app that helps me switch between C/F temperatures, I came to the same conclusion as MKBHD: "Fahrenheit is better for weather..."[1] Fahrenheit roughly covers the scale of air temperatures experienced by humans where 0 is cold and 100 is hot with maximum precision for a given number of digits.
Unfortunately, MKBHD failed to eloquently articulate this. And I couldn't find the Twitter image that helped me realize why F was better for weather.
It's most apparent with air temperature, but both Metric/Imperial systems have their merits. Generalized like this:
- Metric: science-based, regular (powers of 10)
- Imperial: human-scaled, divisible (many divisors, including 3)
For example, in the discussion a surfer says "if you measure wave height in metres or cm it is not as accurate compared to the human body (a 6ft wave is under 2 m and we would not call a wave 182.88 cm high)"
>a 6ft wave is under 2 m and we would not call a wave 182.88 cm
This is subjective but we call it 1.8m and I don't really see a problem "seeing" it visually. 1.8 is just a bit above my eye level.
>Fahrenheit is also more precise
You could use a C2 scale where C2 = Celsius*2.
>it’s best to use a temperature gauge that’s suited to the air
I believe neither Celsius nor Fahrenheit alone can tell you how "hot" the weather is. So the advantage argued is not very valid:
Air temperature does not give a full picture of "hotness", we have to also include humidity ( that's why weather reports usually have a "feels-like temperature" next to air temp ). This is where we use the Wet Bulb temperature [1] to describe the sense of hotness in our surrounding. Wet Bulb Temperature can tell you the rate at which your body can cool itself.
The Wet Bulb temperature still needs a unit of measurement.
> Even heat-adapted people cannot carry out normal outdoor activities past a wet-bulb temperature of 32 °C (90 °F), equivalent to a heat index of 55 °C (130 °F)
So the F wet bulb temperature seems to make even better use of 2 digit readings than "plain" F. With 2 digits, C only uses 32% of the possible 2-digit readings.
> The wet-bulb temperature (WBT) is the temperature read by a thermometer covered in water-soaked cloth (wet-bulb thermometer) over which air is passed
It seems to be a direct measurement, not derived from anything. You might be able to derive WBT given the dew point and dry bulb temperature, but it would still need units like C or F.
Perhaps you were thinking of a different measurement?
I believe that's a shortcut to measure WBT directly. By it's nature, WBT are affected by other variables such as humidity and atm pressure, etc... , so it's a derived unit: Twb = F(T, H, P...).
As a derived function, you tweak its scale as much as you like, you can take T = C*2 and that solved your problem.
But surely deriving the WBT couldn't change the unit of measurement?
A measurement that has no unit is relative humidity. It is just a percentage. There is no way to measure relative humidity directly and get any other type of unit besides a percentage. Instruments that measure relative humidity directly measure in percentages.
edit: ah, you can use C2 = Celsius*2 whether you're talking about regular temperature or WBT. C2 is the 'derived' unit. Not WBT.
Yes you can use C2, but why not just use F which still has better precision than C2[1]? For low values of C2, you may confuse it with plain C...
"What actually matters in the end is a common standard. And since America is the outlier, they are the ones who must conform."
As someone who grew up under the Imperial/Fahrenheit system and had to live with both the Imperial System as well both Metric systems—the Gaussian/CGS and MKS (SI) Systems at school and operate/convert between them on a daily basis, I can attest that a lot of bunkum is talked by those Fahrenheit-ers who devise every conceivable device and contrivance to avoid biting the Metric bullet. It's rather pathetic really (it's a bit like climate science denial).
Every student at my school who did science would overwhelmingly attest that the Metric System is better and much easier in every way. I cover this in considerable detail in the other HN story on Fahrenheit/Celsius titled Celsius Needs a Redesign here: https://news.ycombinator.com/item?id=28526183
As I say there, the Metric System isn't perfect but it beats the antiquated US and Imperial systems hands down.
[BTW, you'll find a link there to the Wiki Imperial site, therein you'll find an image of a 'one US gallon gas can purchased near the US-Canada border showing equivalences in imperial gallons and litres'. This typifies how confusing and problematic the US's stubbornness is. Eventually, it'll have to convert to Metric or it'll be out-competed by other countries.]
> -20 degrees Fahrenheit to 110 degrees Fahrenheit — a 130-degree range. On the Celsius scale, that range is from -28.8 degrees to 43.3 degrees — a 72.1-degree range
So, since in Celsius we (including the author) use a single decimal place we are comparing accuracy of 1 in 130 with 1 in 721! Maybe the author was a bit short on arguments.
14 comments
[ 4.2 ms ] story [ 32.0 ms ] threadUnfortunately, MKBHD failed to eloquently articulate this. And I couldn't find the Twitter image that helped me realize why F was better for weather.
It's most apparent with air temperature, but both Metric/Imperial systems have their merits. Generalized like this:
- Metric: science-based, regular (powers of 10)
- Imperial: human-scaled, divisible (many divisors, including 3)
For example, in the discussion a surfer says "if you measure wave height in metres or cm it is not as accurate compared to the human body (a 6ft wave is under 2 m and we would not call a wave 182.88 cm high)"
Here is another article that tries to explain this "hot take:" http://thevane.gawker.com/fahrenheit-is-a-better-temperature...
[1]: https://youtu.be/BMGrsOawKac?t=60
This is subjective but we call it 1.8m and I don't really see a problem "seeing" it visually. 1.8 is just a bit above my eye level.
>Fahrenheit is also more precise
You could use a C2 scale where C2 = Celsius*2.
>it’s best to use a temperature gauge that’s suited to the air
I believe neither Celsius nor Fahrenheit alone can tell you how "hot" the weather is. So the advantage argued is not very valid: Air temperature does not give a full picture of "hotness", we have to also include humidity ( that's why weather reports usually have a "feels-like temperature" next to air temp ). This is where we use the Wet Bulb temperature [1] to describe the sense of hotness in our surrounding. Wet Bulb Temperature can tell you the rate at which your body can cool itself.
[1] https://en.wikipedia.org/wiki/Wet-bulb_temperature
> Even heat-adapted people cannot carry out normal outdoor activities past a wet-bulb temperature of 32 °C (90 °F), equivalent to a heat index of 55 °C (130 °F)
So the F wet bulb temperature seems to make even better use of 2 digit readings than "plain" F. With 2 digits, C only uses 32% of the possible 2-digit readings.
I couldn't find anything for the low end...
It seems to be a direct measurement, not derived from anything. You might be able to derive WBT given the dew point and dry bulb temperature, but it would still need units like C or F.
Perhaps you were thinking of a different measurement?
A measurement that has no unit is relative humidity. It is just a percentage. There is no way to measure relative humidity directly and get any other type of unit besides a percentage. Instruments that measure relative humidity directly measure in percentages.
edit: ah, you can use C2 = Celsius*2 whether you're talking about regular temperature or WBT. C2 is the 'derived' unit. Not WBT.
Yes you can use C2, but why not just use F which still has better precision than C2[1]? For low values of C2, you may confuse it with plain C...
[1]: https://uxdesign.cc/celsius-needs-a-redesign-41709e048c4a
My parents grew up with Fahrenheit. It feels more "natural" and "better" to them.
I grew up with both. Both work fine for me.
The next generation after me grew up with Celsius, which feels more "natural" and "better" to them.
What actually matters in the end is a common standard. And since America is the outlier, they are the ones who must conform.
As someone who grew up under the Imperial/Fahrenheit system and had to live with both the Imperial System as well both Metric systems—the Gaussian/CGS and MKS (SI) Systems at school and operate/convert between them on a daily basis, I can attest that a lot of bunkum is talked by those Fahrenheit-ers who devise every conceivable device and contrivance to avoid biting the Metric bullet. It's rather pathetic really (it's a bit like climate science denial).
Every student at my school who did science would overwhelmingly attest that the Metric System is better and much easier in every way. I cover this in considerable detail in the other HN story on Fahrenheit/Celsius titled Celsius Needs a Redesign here: https://news.ycombinator.com/item?id=28526183
As I say there, the Metric System isn't perfect but it beats the antiquated US and Imperial systems hands down.
[BTW, you'll find a link there to the Wiki Imperial site, therein you'll find an image of a 'one US gallon gas can purchased near the US-Canada border showing equivalences in imperial gallons and litres'. This typifies how confusing and problematic the US's stubbornness is. Eventually, it'll have to convert to Metric or it'll be out-competed by other countries.]
So, since in Celsius we (including the author) use a single decimal place we are comparing accuracy of 1 in 130 with 1 in 721! Maybe the author was a bit short on arguments.