Why Fahrenheit is better than Celsius. | Hacker News Reader

Why Fahrenheit is better than Celsius.

tumblr.com

28 pointscrasshopper55 comments

For weather, which I find more natural depends somewhat on the climate. In Denmark, the celsius system seems quite natural, even as an American. Freezing being at 0 C is a reasonable reference point; below it is cold, above it is okay, >10C above it is warm, >20C above it is grilling on the beach. I find it a bit less natural in Greece, though admittedly it may be different if I had grown up there. The main awkwardness is that I find myself mentioning specific temperatures more often; it seems that the colloquial equivalent for an expression like "low 70s" is "around 22".

I don't have a problem with people using whatever they want, though. In scientific and engineering work I'd stick to C, but I don't see great benefits from consistency in daily life; it's not hard to approximately convert in your head. The way I usually do it is to start at an approximate reference point suitable for the climate, like 0 C/30 F or 20 C/70 F, and then have the F temperature be twice as far from the reference point as the C temperature. None of those numbers are quite right, of course, but it's easy in my head to say that 5 C = 40 F or 80 F = 25 C, which are 1-2 degrees off but close enough for discussing the weather.

Only reason: Because Americans are raised used to it. Humans hate to change, learn new things or getting used to them goes against our comfort zone, humans rationalize almost everything according to what their emotions dictate them.

As a European, like anyone in the world that is not American or English, I have been raised using Celsious degrees, so I am on the other side of the coin.

Biased arguments are not useful. This is a debate instead of dialectic.

Having lived in Europe and the US, I see what he's saying. He's referring more to Miller's "seven, plus or minus two" guide for the human's ability to differentiate values along a single axis.

With Celsius, you're using an extremely narrow range in day-to-day life, the most common use case. Do I need to wear layers today? Hmm, there will be a high of 21, and a low of 15, so I guess I should grab a vest. If I ask someone the forecast, and they say,"around 20 degrees," I don't know if I'll need the vest, as that could cover 18 to 22 degrees.

Except in rare circumstances, you're not going to need to tell people it's going to be 40+ degrees. If you do, you don't really need the precision anyway, as anything above 30 degrees is probably going to be "hot."

Therefore, that 40-100 range (and beyond) is basically useful for cooking. Sure, it's important on a day-to-day basis, but you generally have specific targets for what you're cooking.[1] I'd argue Fahrenheit is even more useful in cooking, as the difference in one or two degrees Celsius can ruin your fish.

Fahrenheit, with a broader range, is easier to digest and more resilient to signal loss. "Forties" means "wear a coat," "eighties" means "call in sick and go play," "nineties" means "I wonder what's playing at the cinema so I can sit in air conditioned darkness."

Peace comes at -40 degrees, when C = F.

[1] Don't get me started on gas marks, from the people who brought you L.s.d.

I've started using Celsius in my everyday life. As an American, it's not the easiest thing to do, but I'm used to that from my use of the 24-hour clock and the metric system (even my car is in km/h). I prefer Celsius because it seems much less arbitrary. (A physicist-friend is trying to get me to switch to Kelvin, but absolute-zero has no implication on whether or not I should wear a coat when I leave my house.)

Fellow developers: please allow me to use the measurement system I choose to use. In my transition, I realized how much we take this stuff for granted and assume everyone does things the way we do it ourselves. It doesn't take much work to give people a toggle b/t Fahrenheit and Celsius, 12/24-hour clock, or English/Metric.

"The scientific community adopted Kelvin instead Rankine, and that pretty much sealed the deal for Celsius. Today’s familiar units such as joules and the calorie (the amount of energy needed to raise the temperature of a gram of water by 1˚C) assume we’re working with Celsius-sized degrees."

http://www.ericpinder.com/html/celsius.html

I'd say the primary difference for day to day weather use is cultural. If you grew up with F or C, that's what you're used to and is what's best for you.

I'm in Vancouver, BC. I'm really close to the US border and have spent a fair amount of time in the states traveling. Going to the states is always confusing as I don't have the experiences to know what their temperatures mean, nor do I have a map in my mind. I know what 0C, 5C, 10C, 15C, 20C, 25C, 30C and 35C feel like. It's no different that 30F vs 70F for the author.

I know freezing is 0C and boiling is 100C and the body temp is 37C. I have know idea what the corresponding values are in F, and I really don't care because it rarely affects me and if I need to know I can ask my iPhone.

The one place this isn't true, at least in Canada, is cooking. We still use the imperial system. The oven is always at 350F, and sometimes 425F. But those have no relation to my normal day to day temperature uses. They are just numbers I'm used to because I use them on a day to day basis.

Yes and:

- horses are better than cars;

- the telegraph is better than the phone;

- fixed line phones are better than cell phones;

- bulletin boards are better than the Internet;

- dialup is better than broadband;

- black and white TV is better than colour;

- standard def is better than hi def;

- windows 3.1 is better than Windows 7;

- and so on.

You're not really addressing the arguments presented in the linked post, you're just being snarky. I can appreciate that, but the linked post makes some really good points.

Also, the converse of the point you seem to be trying to make with your litany above is also true: new things are not always necessarily better than old ones.

> You're not really addressing the arguments presented in the linked post, you're just being snarky.

Correct.

> ... the linked post makes some really good points.

No, it doesn't (IMHO), hence my (admittedly) snarky response.

The TL:DR version of the post is "I'm used to Fahrenheit and like saying and hearing 'in the fifties'". That's not an argument, it's an historical artifact (being what they were exposed and thus accustomed to) dressed up as preference.

> new things are not always necessarily better than old ones.

Who ever said they were? I cited specific examples. Please don't erect a straw man and put words in my mouth.

The theme of what I said can be described as:

- people resist change; and

- when someone describes something as "better" what they normally mean is that they're used to it.

The TL;DR version of the post is that celsius throws away 70% of the 2-digit numbers.

There are 2 types of fools in the world; one says "this is old, and therefore good"; the other says "this is new, and therefore better".

There are 2 types of people in the world: those who classify others into categories and types, and those who don't.

In my 45 years, I've yet to meet anyone that doesn't.

Everyone you've met is that judgmental?

No, everyone I've met puts peoples into categories and types. It's how human brains are wired (I've read, anyway). It's What We Do.

I think I'm married to someone who does not categorise people. Maybe she's the only one?

The only thing I know about Fahrenheit is that -40C == -40F.

Furthermore, Good case for Fahrenheit being more intuitive. The same points can be made for Celcius though, so the whole point is moot.

"A random temperature scale is better than another random temperature scale because I'm used to it."

Did you read the article?

This always bugged me too. Fahrenheit is more than double the precision of Celsius. The thermodynamic properties of water just aren't important enough in daily life to make sacrificing that precision seem worth it.

Similarly, measuring things in feet is more precise and easier to me than measuring in meters. Tenth of meters is awkward, and decimeters are just never used colloquially. 5 feet tall vs. 6 feet or 1.5 vs 1.8 meters? Yuck.

They're awkward for you, like feet and inches are awkward for me. When I hear people say they're "five eight" I have absolutely no idea what that means.

People don't say they're "one point seven meters" tall, they say "one seventy four" as in 174cm.

can agree with the first but the second is just silly. In the metric system you would say 1.76m as 'I am one seventy six' just like you would say 'i'm five eight' for 5' 8" in the imperial system.

Usually people would mention their height in centimeters, such as 183 cm -- which is more precise than feet and inches. But that's not the point. Metric is "better" because it's decimal and since you very often need to perform arithmetic on distances, having them in a format that is easy to do arithmetic on is a good thing.

Unless you're doing physics, you'll never do arithmetic on temperatures, and therefore the ability to more intuitively describing the weather is much more valid.

For the record, I'm European, and have no intuition of Fahrenheit temperatures. But the guy is actually making a valid point.

The Eric Prinder.com link makes a related point to yours: People often conflate "metric is better" with "Celsius is better".

Can you give an example of a situation where the precision given by an integer temperature in °C is insufficient but in °F is sufficient, and which occurs often enough to make adding a decimal inconvenient?

Anywhere in the 70's or 80's Fahrenheit.

Can you be more specific? For outdoors temperature we would just say 20 or 23. Nobody needs to add a decimal point, because the precision given by Celsius is sufficient.

For indoor temperatures, European electronic thermostats seem to use half-degrees C, so you can e.g. set the air conditioner to 22.5. Not really a huge problem, but it's interesting that the size of the European thermostat control increment is around 1 F.

You needed to use two sig figs. Fahrenheit just needs one.

When is the "better" precision needed? Because the article itself says low/high 70's. This would translate to 20-25˚C. Who feels the difference between 72 and 73 (22.2222-22.7777˚C)? If you need precision, you will need decimals, in ˚F or in ˚C.

And I don't understand what does it mean to "not use all the digits". They are used, not just for the narrow range he cares about which is the comfort zone in his particular region of the world.

In Belgium it is between 15˚C and 25˚c all the year with few extremes on either sides. Nobody need 3 digits to tell if it's warm or not.

In other words "in the 10's" and "in the 20's" are completely unhelpful statements. In Fahrenheit the range you named is 59℉ and 77℉. "Low to mid 60's" and "mid 70's" communicates with fewer significant digits. 50's and 80's tell you something significant with just one digit.

So Belgium is a perfect example of a place where Fahrenheit would be better.

Saying 23 (°C) is less significant figures than "low to mid 70's (°F)" is language lawyering at best. You're ignoring the significant digits expressed by "low to mid".

(And, for the record, "low 20's" °C would be 68–73°F, which is a perfectly useful range, just with a slightly shifted boundary than what both of us are used to.)

|{low, medium, high}| < |{0,1,2,3,4,5,6,7,8,9}|

where |s| := the cardinality of s

Well, it'd be "low", "low-to-mid", "mid", "mid-to-high", and "high". So essentially dividing every 2°F, but not really, because the boundaries are fuzzy and quite possibly overlap.

Only slightly less cardinality, a phrase that takes longer to say, and much less precision.

Freezing temperature is important in daily life. It means roads are very slippery, extra caution when driving. Your house may need certain adjustments when it starts to freeze (careful with those pipes!). Etc. Freezing matters.

> Fahrenheit is more than double the precision of Celsius.

How do you figure? For any given interval, say, freezing point to boiling point of water; F is precisely 9/5 the precision of C, no? Which is close to double, but not more than double.

Or am I misunderstanding your point?

The article says we don't need to be scientific, that a scale going from the 1800's coldest temperature (ice houses) to the body's internal temperature is better than one going from one change of state of something we can easily access, manipulate and judge, to the other state change.

No. Celsius. For all I know it's another one of those silly anti-colonial things like driving on the wrong side of the road.

This argument is about significant figures, nothing else.

The argument that scientific relevancy is not relevant in the choice of measuring unit is somewhat...ironic?

I do however agree that this doesn't deserve discussion and falls into the category of "someone with too much time on their hands and nothing useful to do with it"

Growing up in Canada, in the generation I did, we tend to use both.... though metric is official.

For winter temperatures, we tend to work in C.

0 is freezing - it's convenient. We'll say it's +5 or +10 or -5 out... we know we're talking about C.

Occasionally I'll catch my parents, or even me, saying "It's like 90 degrees out" - we're talking F obviously - only used for pool heater and hot days... because 100F is damn hot.

As to better.... in that all measurement systems are provably equal, the metric one is the world accepted standard, so just go with it already, jeeze...

I'm trying to figure out how old you are if you use both. Older than me (33), as I haven't the slightest clue what the temperature in Fahrenheit is, but younger than my parents (mid-60s) who insist on talking to me in Fahrenheit (and I still never managed to pick it up!)

30 + celsius degrees * 2

I hope that's 30 + (celsius degress * 2), otherwise he's older than my parents!

I guess. Having next to no experience of using Fahrenheit, I'm happy to take other peoples' word when it comes to day-to-day usefulness.

It seems to me, though, that the relative merits pale into insignificance compared to the potential benefits of picking a standard, doesn't much matter which one, and being able to forget about converting between the two.

My real life is better without two different temperature scales. And I happen to have learned some chemistry.

It's all about what one is used to. Of Fahrenheit I only know that 100 is about body temperature. Above is too hot, below - could be okay. So yeah I prefer Celsius, which happens to be a little less of a boolean scale to me.

What's Boolean about either?

Boolean in the sense of: Is X Fahrenheit warmer or colder than the average body temperature?

Isn't Celsius equally "boolean" in that sense then, as in this comment: http://news.ycombinator.com/item?id=2313288 ? Above or below freezing.

Sure is. To some people. I was talking about my own perspective :)

This is stupid and not worthy of Hacker news. And a stupid graph with a "FFFFUUUU"-style clutter in it wont make me go "lol!" and change my mind either.

Go away. Keep this to reddit please?

I think there's a very relevant point here. Listen carefully to his arguments and set aside any attachment you may have to C or F. Personally, I lament our reluctance to switch to the metric system here in the US, but setting that aside, the author does a good job of relating why he likes Fahrenheit. He expresses how he's able to relate to it better than Celsius, and he makes some good points.

How is that applicable to HN? Imagine the author is one of your customers. Further, imagine that he probably represents the majority of your customers in the US. Now imagine that you're building a cool new social weather website. As an international standards fanatic, you decide that you're going to use Celsius on your new website, because "that's what the world speaks gosh-darnit, and US users are coming with us, like it or not!" raises fist

This is a neat little exercise in listening. Is it realistic? No, it's patently absurd. You won't find many people who would roll a US-targeted website using metric units of measure. The implications are too obvious. But what if we apply this rationale to the other decisions we make. What aspects of our craft are affected by our attachment to "correctness", even when our users cry out for something different? Ask yourself these questions:

What attachments do I have that I force on my users/customers?

What aspects of my craft confuse users the most?

Then let go of your attachments and try a different approach, measure the results, rinse and repeat.

I think what's interesting is "X has a useful property P that everybody talks about. But P is not what we want to optimize on."

Ok, I'm not even going to bother commenting...

Water freezes at 0 degrees C. At what temperature does water freeze in F?

Downvotes ftw?

Reply on news.ycombinator.com