Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Is it shocking that the number of hours in a day or months in a year is not divisible by 10?

Doesn't it make sense for some things to divide by 3 and 4, like a year, so that a season is 3 months long, instead of 2.5 months long (as in the old 10-month year system, ending in December)?

Does it not make sense to use a human scale for certain things, like outdoor weather (as opposed to say, temperatures inside of machinery):

In Fahrenheit: 0: very cold to people; 100: very hot to people.

In Celsius: 0: not that cold; 100: I died 50 degrees ago.

The metric system seems to me to be forced onto things where it does not fit, when in fact it's fine to use more than one system when contextually appropriate.

After all, none of us think it weird to use bases 2, 10, or 16 for different things on our computers.



I've seen the temperature argument and it never has made sense to me. What does "very cold" and "very hot" mean? For me living in Spain, 40º is "very hot", but I've seen people living in northern countries complaining about 30º. Viceversa with "cold".

The only reason Fahrenheit looks like "a human scale" is because you're used to it. At least Celsius has clear reference points (everybody knows how cold is "freezing water" and how hot is "boiling water").

> Doesn't it make sense for some things to divide by 3 and 4, like a year, so that a season is 3 months long, instead of 2.5 months long (as in the old 10-month year system, ending in December)?

Programmers all around the world would first hate you for changing things, and then love you for making a consistent dating system.

But, seriously now, dates are a bit of a difficult system to change. You want to be able to refer to past dates without too much confusion, and also there are outside restrictions such as the length of a year compared to a day (not many integer divisors of 365 days).


I moved from a Fahrenheit country to a Celsius country, and let me tell you that knowing 0° was freezing and 100° was boiling did not help me decided whether I should put on a jacket when it's 17° outside.

(I now know how to interpret Celsius, but it involved memorising certain fixed points. The "30 is hot, 20 is nice, 10 is cold, 0 is ice" rhyme is a good one.)


And if I moved to a Fahrenheit country I wouldn't know what to wear when it's 60 degrees. That's the point, that both scales need some "getting used to". Fahrenheit is no more natural than Celsius in that regard, and also has the issue of not having reference points.


0F is the brine freezing point. The 100 is better - originally thought as human body temperature, now it's a mild fever.


As opposed to what? I put on a jacket at 19°C. That's 66.2 Fahrenheit. How does it help?


My point is that everyone saying "Celsius is so easy, look at how logical it is!" are missing the point that you still have to memorize a bunch of temperatures. Nobody in their right mind is going to say "Well, it's 20% of the way between water freezing and boiling, so I'll wear long sleeves but no jacket today".


Both celsius and fahrenheit are completely arbitrary, just use whichever one everyone around you is using and you'll be alright. You'll internalise the scale eventually.

It's the one part of the metric system that isn't innately superior to imperial/customary


Celsius does the same thing but with 5 cadence; what you explain is a habit - 70 Frankenstein being 21 science is just as helpful.

For me how it works - 0 is freezing point, 36 is body temperature (you learn that around 4-5y of age), and it's imprinted heavily. Half your body temp is about to begin feeling cold (in general). Picked body temp as it is 100F (originally 100F was meant to be the body temperature)


Fahrenheit (the man) was himself a physicist... he invented the mercury thermometer. That's pretty scientific.

So calling it "Frankenstein and Science" is a little silly, especially when we have an absolute scientific scale (Kelvin).

Actually, nobody complains that we use both Celsius and Kelvin, so what problem is there with a third scale for people, if the former two are for industry and science respectively?


>So calling it "Frankenstein and Science" is a little silly, especially when we have an absolute scientific scale (Kelvin).

That was unnecessary, an obvious joke about both, and how people feel about them, etc. C and K scale the same way, so C is just an offset K (or vise versa). Both C and F are arbitrary, of course - zero (as developer) being the phase change of the water does make more sense to me; again arbitrary.


> I now know how to interpret Celsius, but it involved memorising certain fixed points

But that's literally the argument given for Fahrenheit.


My point is that everyone saying "Celsius is so easy, look at how logical it is!" are missing the point that you still have to memorize a bunch of temperatures. Nobody in their right mind is going to say "Well, it's 20% of the way between water freezing and boiling, so I'll wear long sleeves but no jacket today".


So you agree that the argument for it "feeling more natural" is actually, "I've already memorised a number"?


No really. The "comfortability" points are basically separated by 10 degrees F, with only 32 degrees F being the special one.


Then again 17 is pretty close to temperature you are entirely unsure if you use jacket or not. On other hand it is pretty pleasant summer temperature. On other if it is windy it might not be so nice.


very hot - about 35.5C or so, when you cannot dissipate heat with sweating. 30C is just uncomfortable but totally fine.

Also an year is not even 365days.


> very hot - about 35.5C or so, when you cannot dissipate heat with sweating. 30C is just uncomfortable but totally fine.

That's not only subjective but also depends on humidity. 34º with high humidity can feel far hotter than 38º in dry weather.


In Celcius, 0 water freezes, 100 water boils. Very convenient for weather and cooking. For the rest, what is hot weather or cold weather is very subjective, so I don't really see any use to define a scale on the base of that.


Do people actually directly measure boiling water? I've never seen that. It's a self-apparent phenomenon. I've never heard of anyone measuring anything while cooking where it matters which temperature scale they're using.


It is really useful for calibration. Stick a thermometer in there and you can tell how accurate it is.


Unless you live at altitude 0, I wouldn’t do that if I were you.

Celsius and Fahrenheit don’t make much difference. That’s the only part of the imperial system which is not incredibly inferior to metric.


Calibration of what? The thermometer? When do you need to calibrate a thermometer at the temperature of boiling water and how does the temperature scale affect that in any way?


The fact that calibration points at sea level of celsius are very easy to produce. Just need some ice and then some boiling water. And I can't even get my hands on pure ammonium chloride. Just some mixed stuff.


Yes, calibration of the thermometer. For example with sous vide cooking, it can be useful to know how accurate your equipment is since recipes are often expressed in precise temperatures.


Thanks for the example, but I'm still not sure what that has to do with the scale since you only care about a single number in that case.

I need to look up sous vide cooking. I've never heard of it before. Seems pretty interesting.


You need at least two points to determine the slope of the thermometer, 0 and 100 as freezing and boiling point of water makes that reproducible.


well the boiling point depends on the elevation (pressure), so that tells you nothing at all.


The metric system is based mostly on water. 1 kg = 1 litre = 1000 cm3


For people serious about their tea, measuring 70, 80 or 90 Celsius is vital.


> Doesn't it make sense for some things to divide by 3 and 4, like a year, so that a season is 3 months long, instead of 2.5 months long (as in the old 10-month year system, ending in December)?

The metric system also uses twelve months per year, so what is your point?

> Does it not make sense to use a human scale for certain things, like outdoor weather (as opposed to say, temperatures inside of machinery):

Yes, Fahrenheit works 'better' than Celsuis for describing weather temperatures. But it does arguably worse at pretty much everything else, including other everyday occurrences like cooking or washing clothes.

As an aside, 12 would arguably be the better base for an everyday number system, but we are where we are (and contrary to what's commonly claimed, the imperial system isn't derived from base 12 either - 12 is used in a few places, but most proportions are historically grown and fairly random).


Celcius 0 in weather is pretty handy. If it is at that or below I know that I should be aware that it might be slippery so drive more carefully and check where I put my foot. Much below that those things aren't that big issue but better wear more. And above that somewhat I can survive outside okay.


I will admit celsius 0 is useful for farming (will there be a bud-killing frost?), but I have no issue with the use of celsius in industry; there are merits for each system, and we don't need to eliminate one to get the benefit of the other. We can use each in their context, like we do with many other things.


0C is mostly useful on the road, knowing when it becomes slippery - around 3-4C as ice can form above 0 as well


September to December contain the numbers 7 to 10 not because the roman year was 10 months long, but because they started counting in March instead of January. That's also why February is a bit shorter: They ran out of days in the last month of the year.


> In Fahrenheit: 0: very cold to people; 100: very hot to people. > > In Celsius: 0: not that cold; 100: I died 50 degrees ago.

Celsius 0 and below, warning there could be ice on the road.

Hot/cold is too subjective. My relation to heat/cold was not the same when I was living in north of France, Swiss alps and now in south of Spain.


Oh but that was part of the original plan!!

https://en.m.wikipedia.org/wiki/French_Republican_calendar

10 months per year, 10 days per week.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: