1923: "Felsius"

This forum is for the individual discussion thread that goes with each new comic.

Moderators: Moderators General, Prelates, Magistrates

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25850
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1923: "Felsius"

Postby gmalivuk » Sat Dec 02, 2017 5:26 am UTC

chridd wrote:
gmalivuk wrote:
chridd wrote:
By the way, I just added Felsius here (under "less serious"), so now you can convert between Felsius and every other temperature system I could find when I first made that page.

Holy floating point errors, Batman!
Curious what specific floating point errors you're seeing? (I know there are some, but are there some that you're seeing that I'm not aware of? Or if I wasn't clear enough about the "radians Fahrenheit/Celsius" options? In any case, putting "switch to using rational numbers" on my mental to-do list.)
Or just round off after two or three decimal points.

The one I checked was 20 Fahrenheit, which gave something like 3.9999999999999984 Rømer.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

User avatar
da Doctah
Posts: 826
Joined: Fri Feb 03, 2012 6:27 am UTC

Re: 1923: "Felsius"

Postby da Doctah » Sat Dec 02, 2017 5:57 am UTC

I say just do away with numbers altogether and give all temperatures in descriptive terms:

"Tomorrow at sunrise it will be about room temperature, rising to hot enough to melt gallium in the outlying areas, body temperature in the city center. A cooling trend will bring overnight temperatures this weekend down to cold enough to need a sweater, and some parts of the valley will dip down low enough to make your nose start running."

User avatar
chridd
Has a vermicelli title
Posts: 789
Joined: Tue Aug 19, 2008 10:07 am UTC
Location: ...Earth, I guess?
Contact:

Re: 1923: "Felsius"

Postby chridd » Sat Dec 02, 2017 8:44 am UTC

gmalivuk wrote:Or just round off after two or three decimal points.
I can rationalize using rational numbers with the fact that I've already written a Rational class in JavaScript that I've used elsewhere. (Also, done. Hope I didn't introduce too many new bugs...)
~ chri d. d. /tʃɹɪ.di.di/ (Phonotactics, schmphonotactics) · she(?)(?(?)(?))(?(?(?))(?))(?) · Forum game scores
mittfh wrote:I wish this post was very quotable...
flicky1991 wrote:In both cases the quote is "I'm being quoted too much!"

sotanaht
Posts: 186
Joined: Sat Nov 27, 2010 2:14 am UTC

Re: 1923: "Felsius"

Postby sotanaht » Sat Dec 02, 2017 9:00 am UTC

chridd wrote:
gmalivuk wrote:Or just round off after two or three decimal points.
I can rationalize using rational numbers with the fact that I've already written a Rational class in JavaScript that I've used elsewhere. (Also, done. Hope I didn't introduce too many new bugs...)

Spoiler:
Image

User avatar
moody7277
Posts: 527
Joined: Thu Aug 14, 2008 7:06 pm UTC
Location: Extreme south Texas

Re: 1923: "Felsius"

Postby moody7277 » Sat Dec 02, 2017 2:00 pm UTC

Felsius...a temperature scale for the rels of us.
The story of my life in xkcdmafia:

Tigerlion wrote:Well, I imagine as the game progresses, various people will be getting moody.

ericgrau
Posts: 91
Joined: Sat Dec 13, 2008 7:14 pm UTC

Re: 1923: "Felsius"

Postby ericgrau » Sat Dec 02, 2017 5:00 pm UTC

I feel that Felsius gives me a more useful size degree-ish while describing-ish real world temperatures between 0 and 100. Yet 0 is, eh, close enough to freezing. 100 isn't anywhere near boiling but it's pretty dang hot and any hotter will burn my mouth.

However Rankelsienheit (the arithmetic-geometric mean of all 4 of course) is a much more fair universal system so I will reluctantly conform to it for the sake of uniformity. As must everyone.

tempest69 wrote:At least -40 is the same for all three.

I like that Fahrenheit has 180 degrees of liquid water. It makes Degrees seem like a reasonable term to use.

I hate "degrees Celsius " Centigrade is a much better word (meaning measured in hundredths), "degrees Celsius" make me think that someone hasn't really considered how this works, and are just parroting down a long line of parrots.

Ding ding ding that's how all standards work and why we should keep using whatever the heck we were using before. Because it's way easier than converting all our old stuff. The only legitimate reason to switch to Celsius is because the rest of the world uses it. Not any of that decimal water BS for slightly easier math that we don't even do by hand. Oh, and also why Centigrade can die in a fire. If you want language accuracy gtfo of English.

User avatar
pogrmman
Posts: 346
Joined: Wed Jun 29, 2016 10:53 pm UTC
Location: Probably outside

Re: 1923: "Felsius"

Postby pogrmman » Sat Dec 02, 2017 5:47 pm UTC

sotanaht wrote:Fahrenheit for weather, Celsius for water. It's what they were made for and what they are good at.

Think about the usable portion of the scale. Where I live 0°F and 100°F are about equally common. We very very rarely use numbers above 100 or below 0, but they do happen occasionally. In Celsius that would be a scale from -18 to 38, which is utter nonsense.


I agree about the weather. Sure, we break 100° about 25 times a year on average at home, but that’s reasonable. The average range of “normal” (going by the lowest and highest temperatures surpassed 10 times a year on average (this roughly equals 2 standard deviations in either direction)) is 31°to 102°. That’s more reasonable IMO than -1°to 39°(Though, surprisingly, it’s almost exactly a 40° range in Celsius!). There should be a much bigger difference between those two temperatures than is expressed in Celsius. At my school, those ranges would be -6°to 91° (-21°to 33°). I think the former is more reasonable again.

Celsius are too “big” for stuff like weather, IMO. 95°
and 100° certainly feel quite different — for me, the former is on the very upper edge of OK, but the latter crosses into hot (for sake of honesty, the heat index at home usually runs 3-5 degrees warmer than temperature and has a bigger effect on how it feels than the raw temperature). In Celsius, that’s not even a 3° difference.

User avatar
Flumble
Yes Man
Posts: 1959
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: 1923: "Felsius"

Postby Flumble » Sat Dec 02, 2017 6:20 pm UTC

Forget the compromise, superior temperature scale passing through:
SCSimmons wrote:Fiddled around with possibilities for a bit. I thought milli-electron volts per molecule (of an ideal perfect gas) looked pretty handy for everyday purposes. The unit turns out to be just under 8ºC and almost exactly 14ºF, so it's not very fine-grained; reporting to tenths of a degree [sic] would give you comparable precision to what you see in weather forecasts, but the comfort categories can be that wide & still work well. I'll call this unit the Maxwell ºM [sic]; he got a nice law named after him, but as far as I know, never a unit, the poor guy.

Code: Select all

ºM      ºC       ºF

30    -41     -41
31    -33     -27
32    -25     -13
33    -17       1
34    -10      15
35     -2      29
36      6      43
37     14      57
38     21      71
39     29      85
40     37      98
41     45     112


Water freezes at 35.2ºM [sic] and boils at 48.1ºM [sic] at standard pressure. And normal human body temperature is ... hold on, I know 98.6 ºF is an outdated rule of thumb, let me look up the real number & variance ...

Huh. Normal human body temperature is ... 40.00 +/- .02 ºM [sic]. :shock: This must mean something, right? :)

kaloo
Posts: 1
Joined: Sun Dec 03, 2017 7:00 am UTC

Re: 1923: "Felsius"

Postby kaloo » Sun Dec 03, 2017 7:09 am UTC

I really don't understand the value of Celsius. Why hasn't it been replaced with Kelvin outside of the scientific community. It's better for everything except the range of day to day numbers and if we're going to claim that matters then Fahrenheit's makes me feel better.

Petrograd
Posts: 2
Joined: Mon Jun 09, 2014 10:58 am UTC

Re: 1923: "Felsius"

Postby Petrograd » Sun Dec 03, 2017 8:24 am UTC

faubiguy wrote:
Petrograd wrote:The symbol should be Ð. D is between C and F, and the extra - helps show it's supposed to be half way to E.


Alternatively, it should be D♯, borrowing from musical notation.
True! That would work too.

I like the mixing of music and chemistry.

User avatar
PinkShinyRose
Posts: 829
Joined: Mon Nov 05, 2012 6:54 pm UTC
Location: the Netherlands

Re: 1923: "Felsius"

Postby PinkShinyRose » Sun Dec 03, 2017 3:03 pm UTC

pogrmman wrote:Celsius are too “big” for stuff like weather, IMO. 95°
and 100° certainly feel quite different — for me, the former is on the very upper edge of OK, but the latter crosses into hot (for sake of honesty, the heat index at home usually runs 3-5 degrees warmer than temperature and has a bigger effect on how it feels than the raw temperature). In Celsius, that’s not even a 3° difference.

So, are you saying Fahrenheit is somehow superior to Celsius because you don't have a use for every integer in a range for the weather? I don't get that argument...

To be honest, I don't get all this nonsense about 0°F to 100°F being the common ambient temperatures in their climate either; should everyone use a different scale depending on the climate they live in? Note I couldn't use standard in the previous sentence because there would not be one.

The reason why I (and possibly most outsiders) think the US is weird for using its traditional units is because most of the world (possibly >90% of countries) has switched to °C and a mixture of SI with metric units for things other than temperature (and pressure in medical settings, but that's just doctors being sticks in the mud); also to ISO216 (A-series) paper sizes.

rmsgrey
Posts: 3145
Joined: Wed Nov 16, 2011 6:35 pm UTC

Re: 1923: "Felsius"

Postby rmsgrey » Sun Dec 03, 2017 5:10 pm UTC

kaloo wrote:I really don't understand the value of Celsius. Why hasn't it been replaced with Kelvin outside of the scientific community. It's better for everything except the range of day to day numbers and if we're going to claim that matters then Fahrenheit's makes me feel better.


For subjective feel, both Celsius and Fahrenheit feel intuitive to people who've grown up with them. For practical use, Fahrenheit offers spurious precision, and tends to get rounded to decades or half-decades anyway for practical purposes; in Celsius, you can still see a 1 degree variation inside a room so it's still a bit more precise than is really needed, but people claim to be able to distinguish air temperature to 1C precision without needing a thermometer.

Personally, I prefer giving my preferred room-temperature as 18-20 rather than 65-70 (yes, I know those are slightly different ranges - no-one would give "64.4-68" as a temperature range for something like this except as a conversion of numbers they'd originally created as rounded values)

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 2646
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 1923: "Felsius"

Postby Soupspoon » Sun Dec 03, 2017 8:50 pm UTC

rmsgrey wrote:[Personally, I prefer giving my preferred room-temperature as 18-20 rather than 65-70 (yes, I know those are slightly different ranges - no-one would give "64.4-68" as a temperature range for something like this except as a conversion of numbers they'd originally created as rounded values)

Gas Mark 1/384ish? ;)

User avatar
patzer
Posts: 411
Joined: Fri Mar 30, 2012 5:48 pm UTC
Contact:

Re: 1923: "Felsius"

Postby patzer » Mon Dec 04, 2017 2:21 am UTC

kaloo wrote:I really don't understand the value of Celsius. Why hasn't it been replaced with Kelvin outside of the scientific community. It's better for everything except the range of day to day numbers and if we're going to claim that matters then Fahrenheit's makes me feel better.

"20 degrees" is a much easier value for everyday use than "293 degrees". Three digit temperatures for everyday use just feels clunky.

Also, more importantly: Celsius feels like the most natural system to use for most people (including me), due to what they've been brought up with.
If it looks like a duck, and quacks like a duck, we have at least to consider the possibility that we have a small aquatic bird of the family Anatidae on our hands. –Douglas Adams

drazen
Posts: 86
Joined: Fri Dec 05, 2008 12:35 pm UTC

Re: 1923: "Felsius"

Postby drazen » Mon Dec 04, 2017 10:21 am UTC

Heimhenge wrote:Why stop with averaging just Celsius and Fahrenheit? Throw Rankine and Kelvin into the mix too. Call it RKCF. I leave the formula as an exercise for the reader.


I was going to suggest Felvius, as the average of Celsius, Fahrenheit, and Kelvin.

Felvius should also be the official temperature scale of Festivus.

Celsius sucks for daily weather ranges. Fahrenheit has the very nice "in the 20s, in the 30s, in the 40s," etc., each of which generally has its own feeling, and is thus much easier to remember. For example, golf weather for me is 50s through 70s, or 80s if the humidity is low. My winter jacket comes out in the 30s, 40s if it's windy. If it's in the 20s in Celsius, that is multiple types of days and it doesn't really line up well with low/mid/high. Kelvin is for science and Fahrenheit is more useful for normal everyday use.

And besides, being from the US is often synonomus with NGAF about what the rest of the world thinks or does, so all the people who want to live like the rest of the world should move to Europe, what with its low population growth and "do it for Denmark" campaign. Then everybody gets their way! Unless their way is to control the whole world and demand everyone do things their way, but that's been tried before, but nobody was happy about it, and many folks had a rather terrible time of it for several years/decades because of the ideologies that tried it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25850
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1923: "Felsius"

Postby gmalivuk » Mon Dec 04, 2017 2:23 pm UTC

drazen wrote:If it's in the 20s in Celsius, that is multiple types of days and it doesn't really line up well with low/mid/high.
They only don't line up to your Fahrenheit-conditioned mind because you know for example that the "low 20s" in Celsius correspond to 68-73 in Fahrenheit, and you think of "high 60s" and "low 70s" as two different ranges.

Someone raised with Celsius might just as easily think "80s" in Fahrenheit doesn't line up well because it includes both the high 20s and the low 30s.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

drazen
Posts: 86
Joined: Fri Dec 05, 2008 12:35 pm UTC

Re: 1923: "Felsius"

Postby drazen » Mon Dec 04, 2017 3:05 pm UTC

and you think of "high 60s" and "low 70s" as two different ranges.


Not really. If you said high 60s or low 70s, I'd dress the same for both. Assuming a sunny day and no wind.

-20 C — 10 C = -4 to 14 F.
-10 C — 0 C = 14 - 32 F.
0 — 10 C = 32 - 50 F.
10 — 20 C = 50 - 68 F.
20 — 30 C = 68 - 86 F.
30 — 40 C = 86 - 104 F.

Too many decimals and crowded ranges in C, and less easy to talk about.

For me, 80+ F is right about where it gets too hot, but that pegs out to an awkward "26.67 C."
65-80 is a comfortable temperature with low humidity and light sun, but even rounding off it comes out to "18 to 26," which sounds off.
60-65 F is good for a light jacket, but that's "15.5 C. to 18.3 C"
50-59 F is time for a sweatshirt, which admittedly works out to a nice 10 C - 15C.
40-50 F is time for a winter jacket, but that's "4.4 C to 10C"
Below 40 F is where I start layering things under a winter coat.

And when you talk in F about the twenties, teens, single digits, and below zero, each of those describes a level of cold that a bunch of negative C numbers don't describe in any easy vernacular.

I guess I could go with 26+, 18-26, 15-18, 10-15, 4-10, and "sub-4," but that just seems awkward. I could cut it down to 25+, 15-25, 10-15, and 0-10, but then I would never remember exactly what I'll want for layers at what temperature. And it feels dumb to have most of the cutoffs be at the 5's rather than at the 0's.

Also this does not account for humidity or wind, so now we need a 17th competing standard which pegs everything to nice round base-10 ranges.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25850
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1923: "Felsius"

Postby gmalivuk » Mon Dec 04, 2017 3:28 pm UTC

drazen wrote:I guess I could go with 26+, 18-26, 15-18, 10-15, 4-10, and "sub-4," but that just seems awkward.
Of course it's awkward, but those would only make any sense as "natural" temperature ranges because you've already decided the Fahrenheit ranges on which they're based.

And my point is that it's equally awkward in the other direction, if someone who grew up with Celsius is trying to describe their "intuitive" temperature ranges in Fahrenheit.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

sonar1313
Posts: 89
Joined: Tue Mar 05, 2013 5:29 am UTC

Re: 1923: "Felsius"

Postby sonar1313 » Mon Dec 04, 2017 4:06 pm UTC

PinkShinyRose wrote:To be honest, I don't get all this nonsense about 0°F to 100°F being the common ambient temperatures in their climate either; should everyone use a different scale depending on the climate they live in? Note I couldn't use standard in the previous sentence because there would not be one.


It's simple. 0 is fucking cold. 100 is fucking hot. In between that are temperatures that probably 99% of humans are experiencing at any given time, barring a heat wave in India or something. Outside that is dangerous. Even though there are people who live in climates where those temperatures are rather common, those people still take measures to protect themselves from those temperatures.

gmalivuk wrote:
drazen wrote:I guess I could go with 26+, 18-26, 15-18, 10-15, 4-10, and "sub-4," but that just seems awkward.
Of course it's awkward, but those would only make any sense as "natural" temperature ranges because you've already decided the Fahrenheit ranges on which they're based.

And my point is that it's equally awkward in the other direction, if someone who grew up with Celsius is trying to describe their "intuitive" temperature ranges in Fahrenheit.


It might be. Still, a degree Fahrenheit is a smaller range than a degree Celsius, so Fahrenheit offers more precision and therefore converting F to C seems objectively more awkward than vice versa.

rmsgrey
Posts: 3145
Joined: Wed Nov 16, 2011 6:35 pm UTC

Re: 1923: "Felsius"

Postby rmsgrey » Mon Dec 04, 2017 4:25 pm UTC

sonar1313 wrote:It might be. Still, a degree Fahrenheit is a smaller range than a degree Celsius, so Fahrenheit offers more precision and therefore converting F to C seems objectively more awkward than vice versa.


"mid 60s" or "18" - which is more precise?

And it's not terribly useful to have measurements more precise than the variation in temperature across a medium sized room - I've seen reports that temperature can routinely vary by 1C across a room which makes giving a temperature more precisely than that not terribly helpful unless you also specify exactly where that applies...

sonar1313
Posts: 89
Joined: Tue Mar 05, 2013 5:29 am UTC

Re: 1923: "Felsius"

Postby sonar1313 » Mon Dec 04, 2017 5:34 pm UTC

rmsgrey wrote:
sonar1313 wrote:It might be. Still, a degree Fahrenheit is a smaller range than a degree Celsius, so Fahrenheit offers more precision and therefore converting F to C seems objectively more awkward than vice versa.


"mid 60s" or "18" - which is more precise?

And it's not terribly useful to have measurements more precise than the variation in temperature across a medium sized room - I've seen reports that temperature can routinely vary by 1C across a room which makes giving a temperature more precisely than that not terribly helpful unless you also specify exactly where that applies...


Mid-60s F or mid-20s C - which is more precise?

I don't know what everyone else assumes when they hear the temperature will be 21. I know that when I hear a temperature prediction given in Fahrenheit, about two degrees to either side is reasonable to assume. I guess in Celsius you could say, you don't have to assume quite such a range. But why not have greater precision when you can and let the differences be explained by the capriciousness of the weather? I think it makes sense that a ten degree is enough to slightly change the outlook of the day, not be literally the difference between entire seasons (as in, say, 15 degrees vs. 25.)

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25850
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1923: "Felsius"

Postby gmalivuk » Mon Dec 04, 2017 5:48 pm UTC

If there are differences regardless of the temperature scale (i.e. we don't use a scale where a "degree" is the same size as a typical capricious weather change), then the argument for precision pretty much goes away completely. Else why not use millikelvins?

10 degrees being either a change in the outlook for the day or a difference between entirely different seasons is also a matter of the climate you're used to. And for example the difference between 85F and 95F is (for me) quite a lot more than enough to "slightly change the outlook of the day". 85 means I'll want to spend some time in air conditioning after walking any significant distance outside, whereas 95 means I won't want to go outside at all unless it's absolutely necessary. 15 and 25 seems like a similarly significant change. It's the difference between being comfortable in regular winter clothes (and maybe even getting too warm if I walk a lot) and having my nosehairs freeze when I step outside.

But on top of that, there's also the completely arbitrary preference for what a 10 degree change means.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

speising
Posts: 2109
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: 1923: "Felsius"

Postby speising » Mon Dec 04, 2017 5:51 pm UTC

what does make sense is that i know i have to take care of slippery ice on the pavements when i hear we have a temp below zero. i also like that water boils at a nice round 100 (more or less. anyway, what's 212 for a crooked number?)

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25850
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1923: "Felsius"

Postby gmalivuk » Mon Dec 04, 2017 6:12 pm UTC

How much harder is it for a functioning adult to determine whether a number is below 32 than to determine whether a number is below 0, though?
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

drazen
Posts: 86
Joined: Fri Dec 05, 2008 12:35 pm UTC

Re: 1923: "Felsius"

Postby drazen » Mon Dec 04, 2017 6:36 pm UTC

gmalivuk wrote:
drazen wrote:I guess I could go with 26+, 18-26, 15-18, 10-15, 4-10, and "sub-4," but that just seems awkward.
Of course it's awkward, but those would only make any sense as "natural" temperature ranges because you've already decided the Fahrenheit ranges on which they're based.


I've only decided the Fahrenheit ranges on which they're based because that's basically where the comfort points are, assuming low humidity and little to no wind.

I've also said it doesn't account for wind and humidity. A 78 F day could theoretically be significantly more comfortable than a 72 F night, if the day is partly cloudy and dry, but the night has a dewpoint in the upper 60's/low 70's. A 40 degree day can be much more miserable than a 30 degree one if there's a bunch of 30mph wind gusts on the 40 degree day but the 30 degree one is calm.

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 2646
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 1923: "Felsius"

Postby Soupspoon » Mon Dec 04, 2017 6:56 pm UTC

Below zero: Ice if there's water to freeze, wrap up. (If double-digits, it's probably already in the midst of visible snowfall, so visually assess the wisdom of being outside.)
If I had to split, high single-figures is bitter, low single-figures quite manageable.

Single digits (positive): Nippy. Try to remember gloves. Have pockets without too much other stuff stuffed in so that hands can go in there intermittently (if no gloves) or gloves can go in there intermittently (when I'm happy about my hands, and/or need them unfettered).
High-single digits can even feel quite balmy, if we've just had a cold-snap.

10s: Golden zone. A bit cool at the extreme lower end, a bit too warm at the upper end, but "normal". Clothing depends mostly upon air movement and any precipitation, with little regard for temperature for most of the range, but will often be a shirt/T-shirt with light jacket/blazer (according to situation) until it gets uncomfortable with the top layer.

20s: Invariably too hot. There aren't layers I can really shed, once I'm down to the (T-)shirt. I have a light UV-proof walking shirt laid away especially for just such occasions (as primary and sole layer, that is; it's also occasionally my secondary layer around the 15°C time, when it's sunny and endangers my arms). Low 20s are sufferable, High 20s are insufferable.

30s: No. No no no no no. No no no no NO NO NO. NO! - Definitely no fun. If I can dive into a Server Room for as long as possible, I will. I have never enjoyed the high-30s, when they've happened.


For reference, I had a brief check of various thermometers I have laying around the house (analogue and digital - rarely actually checked, but inbuilt into devices like clocks and other mostly forgotten items I had once plotted data from) and I'm working at 13 (floor-level ground floor) to 15 (eye-level, first floor1) degrees in light clothing, and am about to put on a jacket, a different pair of trousers (mostly for the pockets) and shoes in order to head out to a shop through air that various forecasts pegged at 8°C. Haven't decided about the gloves, yet. (There'll be a hat. There's nearly always a hat, atop my head or stuffed in my waistband ready for deployment. Sun, rain, wind or snow. It keeps my hair dry, my eyes less blinded by the sun, keeps my hair unruffled by the wind (though not unaffected by the hat) and my choice of jacket deals with my torso and, especially, shoulders.)


I'm old enough to have used °F in my youth, but have since lost the vital mental imagery of things. I still think in stone for body weight, so both kg (most of world) and lb (US) body-weights need me to perform some arithmetic. Otherwise, I'm mostly metric with most things when it isn't doorstep-milk, pub bitter or road distances.

1 I'll let part of the world puzzle about that, coming from their own illogical system. :P

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25850
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1923: "Felsius"

Postby gmalivuk » Tue Dec 05, 2017 1:55 am UTC

drazen wrote:
gmalivuk wrote:
drazen wrote:I guess I could go with 26+, 18-26, 15-18, 10-15, 4-10, and "sub-4," but that just seems awkward.
Of course it's awkward, but those would only make any sense as "natural" temperature ranges because you've already decided the Fahrenheit ranges on which they're based.


I've only decided the Fahrenheit ranges on which they're based because that's basically where the comfort points are, assuming low humidity and little to no wind.
I'd argue that's where you've gotten used to thinking of the comfort points because you're used to Fahrenheit. I'm 99% certain you can't actually tell the temperature by feel with enough precision that 5, 10, 15, 20, and 25 ºC wouldnt' work equally well for you as cutoff points if that's what you were more familiar with your whole life.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

rmsgrey
Posts: 3145
Joined: Wed Nov 16, 2011 6:35 pm UTC

Re: 1923: "Felsius"

Postby rmsgrey » Tue Dec 05, 2017 2:29 am UTC

gmalivuk wrote:How much harder is it for a functioning adult to determine whether a number is below 32 than to determine whether a number is below 0, though?

A non-zero amount.

How small a difference does there need to be for an optimisation to no longer be worthwhile?

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25850
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1923: "Felsius"

Postby gmalivuk » Tue Dec 05, 2017 3:12 am UTC

At a glance a negative sign might be missed more easily than you might not see whether a number has a 3 or 2 in it.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 2646
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 1923: "Felsius"

Postby Soupspoon » Tue Dec 05, 2017 3:59 am UTC

There was a dot-matrix printer, somewhere I once worked, where one pin in the (vertical, horizontally dragged) head didn't work. It was exactly the right(/wrong) one to make an 8 look like a 0/O1 (without obvious gap), amongst other oddities, and it meant that minus signs were effectively whitespace.

But things are not so low-res, these days. Also light green numbers (or background) for low positives are often differentiated from light blue digits (or background) for zero and below. I don't know how much dicficulty B/G colour-blind individuals have with that, maybe that's why there are often other clues (roundel 'number container' vs squared-edged one, or even hexagonal/snow-flakey; also reversed background/text darkness). But I'm rambling now. What was the point, again?


1 Sort of like...

Code: Select all

  ###        ###                 ###        ###          ###
 #   #      #   #               #  ##      #  ##        #   #
 #   #      #   #               #  ##      #  ##        #   #
  ###   ->         (c.f. with)         <-  # # #  (or)  #   #
 #   #      #   #               ##  #      ##  #        #   #
 #   #      #   #               ##  #      ##  #        #   #
  ###        ###                 ###        ###          ###

   8                                         0            O

...but with the apparent gap between verticals being thinner than the missing horizontal, than in this layout, due to pin-splodging and whatever the actual rasterised font pattern was..

typo
Posts: 30
Joined: Wed Jun 16, 2010 4:12 pm UTC
Location: East of the Don and west of the Rouge

Re: 1923: "Felsius"

Postby typo » Tue Dec 05, 2017 6:54 am UTC

gmalivuk wrote:At a glance a negative sign might be missed more easily than you might not see whether a number has a 3 or 2 in it.

Here in Canada (Toronto) the Official weather.gc.ca forecast puts "plus" in front of temperatures from 1 to 5, presumably to emphasize the difference from their "minus" counterparts. It seems surprisingly natural. Tomorrow's forecast is "Periods of rain ending late in the morning then cloudy. Wind southwest 40 km/h gusting to 60. High 9 with temperature falling to plus 4 in the afternoon." Negatives are always "minus" and not "negative" or "below", and wind speeds (like road speeds) are invariably "km/h" and not "kph". It's all in what you're used to, I suppose.

petercooperjr
Posts: 9
Joined: Mon Jun 19, 2017 4:12 pm UTC

Re: 1923: "Felsius"

Postby petercooperjr » Tue Dec 05, 2017 1:58 pm UTC

The main advantage of the zero point of Fahrenheit is that it's around "When water freezes even with salt" (as that's related to how the scale was developed). It's useful when driving to know that if it's around zero or below, even though the roads are salted it may still be rather icy. Knowing water freezes around 32 isn't actually as useful, since usually there's salt on the roads at that point.

But yes, arbitrary standards are all about what one is used to.

User avatar
pogrmman
Posts: 346
Joined: Wed Jun 29, 2016 10:53 pm UTC
Location: Probably outside

Re: 1923: "Felsius"

Postby pogrmman » Tue Dec 05, 2017 5:50 pm UTC

Admittedly, it is really dependent on what you’ve grown up with. My “comfort” scale is something like this. It’s pretty detailed at the high end because I have a lot of experience with those sorts of temperatures (it’s over 85°F something like 160-170 times a year at home on average), but not at the low end. Although I live somewhere cold for school, I have spent a lot less time out and about in cold temperatures. It’s assuming average humidities that I’ve experienced at these temperatures (so like 10% or less for cold ones and anything from 10% - 100% for the hot ones).

Below 20°F: Really fucking cold. I wanna stay curled up in bed. Need at least my down coat, if not more. It gets pretty far below 20 at school.

20°- 32°: Still really cold. Still bizarre to me that the highs at my school can be below freezing for days/weeks at a time. I need a coat, but not necessarily my down one. If it’s windy, it sucks. I’d much rather be inside. This is as cold as it gets at home.

32°- 40°: Cold. But bearable — especially if it’s not windy. I certainly need a coat, but only a light one. I’d prefer to be inside.

40°- 55°: Chilly. I could go with or without a coat, but it’s not that bad. Still not pleasant, but I could be outside.

55°- 72°: Cooler than I’d like, but pretty nice. No coat needed. Nice to be outside.

72°- 85°: Perfect. Nice and warm, but not hot. Nice in all humidities. Can be a bit brisk at the cooler end if it’s dry and windy. Great for spending time outside — but a bit chilly for water activities.

85°- 93°: Pretty warm/hot, but still nice. If it’s dry, it’s amazing. If it’s more humid, it’s a little worse, but not bad unless it’s excessively humid (like 80+%). Great for spending time outside and great for water activities.

93°- 97°: Kind of hot, but bearable. Still really nice if it’s dry, but starts getting sucky if it’s not. I’d still spend time outside — especially in the shade. Perfect for swimming and other water activities.

97°- 103°: Hot. I don’t like spending as much time outside, but it’s not the kind of heat where you get dizzy when doing hard work for only a few minutes. Still not terrible if it’s a dry heat, but starting to get unpleasant. Usually get sweaty pretty fast. Good for swimming. This is normal summer weather in my mind — sticky, hot, cicadas buzzing incessantly. Pretty terrible if it’s humid — like 60+%. On days like this, the evenings and mornings are usually awesome.

Above 103°: Really hot. Usually, when it’s this hot, the humidity isn’t above 40% or so, but it’s pretty oppressive still. Oftentimes, just a little physical exertion makes you dizzy and need to sit down frequently — even in the shade. I’m usually really sweaty if it’s this hot, and I’d prefer to be inside. Good for swimming and lounging in the pool (when a pool gets to 88°- 92°, it’s wonderful to lounge in). Hot when it’s dry. Not all that great no matter what.

RyYYZ
Posts: 1
Joined: Fri Jan 18, 2013 2:49 pm UTC

Re: 1923: "Felsius"

Postby RyYYZ » Tue Dec 05, 2017 6:46 pm UTC

Arguing for the superiority of one system or the other on the basis of what your used to and comfortable with is clearly wrong-headed.

Personally, having grown up during the time of the Imperial to Metric (SI) conversion in Canada, I'm most comfortable with warmish to hot temperatures expressed in Farenheit, and cool to cold temperatures expressed in Celsius. I guess that makes me a proponent of Felsius?

Leaving aside personal preferences, when it comes to weather, at least for those of us who live in temperate climates, the freezing point of water is very important. It's the difference between needing a squeegee or an ice scraper. The difference between wet roads and possibly icy roads. I've never come to terms with the freezing point being "32". To me it makes more sense to have this important phase change at the zero point. And then having the boiling point at 100 is just convenient, but not inherently so. Arguably a scale with freezing at 0 and boiling at, say, 200, might have been more useful for day to day use.

Certainly the derivation of the Celsius scale is a lot easier to explain.

Note that when it comes to physics and chemistry, neither of these scales, or even Kelvin or Rankine, is "natural". For example, raising the temperature of 1g of water by 1 degree (C or F, take your pick) requires an energy input of how many joules? How many BTUs? (for water, for one degree Celisus, there is a unit in which the answer is unitary - 1 calorie)

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25850
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1923: "Felsius"

Postby gmalivuk » Tue Dec 05, 2017 10:23 pm UTC

pogrmman wrote:
(so like 10% or less for cold ones and anything from 10% - 100% for the hot ones)
You sure about that? Cold air holds a lot less water, so a given amount of water accounts for a higher percentage of what the air could hold. 10% RH on a 90ºF day means the dewpoint is 26ºF, which means that same amount of water on a 26-degree day would give 100% humidity.

RyYYZ wrote:Leaving aside personal preferences, when it comes to weather, at least for those of us who live in temperate climates, the freezing point of water is very important. It's the difference between needing a squeegee or an ice scraper. The difference between wet roads and possibly icy roads.
Except no it isn't, because the air temperature isn't the road temperature, and the water on the roads isn't pure, and even the mostly pure water in the air can remain liquid in temperatures much colder than "freezing".

Note that when it comes to physics and chemistry, neither of these scales, or even Kelvin or Rankine, is "natural". For example, raising the temperature of 1g of water by 1 degree (C or F, take your pick) requires an energy input of how many joules? How many BTUs? (for water, for one degree Celisus, there is a unit in which the answer is unitary - 1 calorie)
But that's a lot less useful for anything that doesn't deal with (pure) water (at the specific temperature and pressure you used to define your calorie). At least Joules are consistent with the other units.

And BTUs are also unitary, if you use pounds instead of grams: 1 BTU is the energy required to raise one pound(-mass) of water 1ºF.
---
At the end of the day, the only compelling way to support metric instead of USCS is a simple argument from popularity. Pretty much everyone else in the world uses it, so communication and commerce would be more convenient if we used it as well.

Everything else basically comes down to personal preference or to the fact that the metric system is based on powers of 10 of the base units, which can be done just as easily with "customary" units once you agree on which base units to go with.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

rmsgrey
Posts: 3145
Joined: Wed Nov 16, 2011 6:35 pm UTC

Re: 1923: "Felsius"

Postby rmsgrey » Wed Dec 06, 2017 12:32 am UTC

The general rule for units is that for specific situations where you're only dealing with one unit, imperial measures are generally better for everyday use. Once you move away from everyday human uses, or need to convert between different quantities, SI units are generally better. And it's generally better for people to only need to know one set of units rather than needing to have an intuitive understanding of two or more sets of units in order to do any sort of science/engineering - or, worse, try doing science/engineering with no intuitive understanding of the units used for science/engineering.

And, yeah, SI ain't perfect - it's pretty unlikely that anyone would stumble into the perfect system of units before putting together the theory needed to understand what a perfect system of units would need to be, which would require a system of units in order to make the measurements needed to establish the theory. Overall, SI does a decent job and the hassle of switching to an more-or-less entirely new system would be even worse than that of getting everyone to use the same system in the first place...

User avatar
somitomi
Posts: 474
Joined: Fri Nov 06, 2015 11:21 pm UTC
Location: can be found in Hungary
Contact:

Re: 1923: "Felsius"

Postby somitomi » Wed Dec 06, 2017 11:42 am UTC

RyYYZ wrote:
Personally, having grown up during the time of the Imperial to Metric (SI) conversion in Canada, I'm most comfortable with warmish to hot temperatures expressed in Farenheit, and cool to cold temperatures expressed in Celsius. I guess that makes me a proponent of Felsius?

I think that makes you a proponent of the little known Celsi-nheit system, where temperatures below the freezing point of water are measured in Celsius, temepratures above it are measured in Fahrenheit and the freezing point iself is defined to be 16 °C-F.
This system has the quirk that values 0-15° and 17-32° are undefined, which is of paramount importance in making confusing calculations.
"It's like he's his own alt account. Saves all the logging in and out"
they/them/theirs = he/him/his ❖ If you want to use something else out of dadaist spite, I won't mind.
✆ Hello? This is Forum Games Discord, what is your emergency?

Peaceful Whale wrote:Stop quoting me!

sonar1313
Posts: 89
Joined: Tue Mar 05, 2013 5:29 am UTC

Re: 1923: "Felsius"

Postby sonar1313 » Wed Dec 06, 2017 6:43 pm UTC

RyYYZ wrote:Arguing for the superiority of one system or the other on the basis of what your used to and comfortable with is clearly wrong-headed.

Personally, having grown up during the time of the Imperial to Metric (SI) conversion in Canada, I'm most comfortable with warmish to hot temperatures expressed in Farenheit, and cool to cold temperatures expressed in Celsius. I guess that makes me a proponent of Felsius?

Leaving aside personal preferences, when it comes to weather, at least for those of us who live in temperate climates, the freezing point of water is very important. It's the difference between needing a squeegee or an ice scraper. The difference between wet roads and possibly icy roads. I've never come to terms with the freezing point being "32". To me it makes more sense to have this important phase change at the zero point. And then having the boiling point at 100 is just convenient, but not inherently so. Arguably a scale with freezing at 0 and boiling at, say, 200, might have been more useful for day to day use.

The problem is that in real world weather applications, there really is no universal water freezing point.

By this I mean that just because the weatherman says it's 30 (or -1) outside, it doesn't mean you'll encounter ice; and just because he says it's 34 (or 1) outside, it doesn't mean you won't. Sunny and 28 will melt snow faster than cloudy and 35. Bridge freezes before roadway and all that. What falls from the sky depends not only on the temperature at ground level (which they report) but the temperature in layers above. And if the temperature is reported at 34 in one place, it might just be 31 nearby. In real life uses, when they report the temperature as a certain number, there really isn't a bright, black-and-white line between icy and not icy - it's more of a transition zone and you have to know that if they say 34, it could still be icy, and if they say 30, it might not be. Which in turn means, to me, that a system that creates one isn't ideal.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25850
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1923: "Felsius"

Postby gmalivuk » Thu Dec 07, 2017 5:20 am UTC

Yeah, a lot of cities have official records and reports that come from their airports, which frequently have different weather conditions than other places in the city. (For example BOS is on perfectly flat reclaimed land sticking out into the harbor, which tends to make it cooler in the summer, warmer in the winter, and windier all year than most of the rest of the city.)
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: Bing [Bot] and 93 guests