7 ms·

I still remember the day I found out the "98.6 degrees" human body temperature was just an oversignificant conversion from "about 37 C, or maybe lower."

by sirsar 6y ago

JadeNB 6y ago
> I still remember the day I found out the "98.6 degrees" human body temperature was just an oversignificant conversion from "about 37 C, or maybe lower." Really? According to Wikipedia, the Fahrenheit scale (https://en.wikipedia.org/wiki/Fahrenheit#History https://en.wikipedia.org/wiki/Fahrenheit#History) is about 18 years older than the Celsius scale (https://en.wikipedia.org/wiki/Celsius#History https://en.wikipedia.org/wiki/Celsius#History), and I'm surprised that human body temperature wasn't included as a calibration. (But of course the fact that I'm surprised by it doesn't mean it isn't true!)
- _underfl0w_ 6y ago
  Could've sworn I remember hearing in grade school that one of then was actually calibrated with cow body temperatures. Don't remember which scale, though.
  - jcranmer 6y ago
    I've definitely heard that as well. Where it comes from, I think, is that cow body temperatures are about 100°F, and someone assumed that the scale must have been calibrated using 0° and 100° as specific endpoints. According to Wikipedia, Fahrenheit seems to have chosen 32°F as the reference point for the freezing of water [it's not clear why this number specifically], and found the body temperature of a human to be 64° more than 32°F [i.e., 96°F]--it being much easier to divide a unit distance in 64 = 2⁶ than to divide it in 100ths. That does bring up an interesting point, though: while we modern people may think of the metric, base-10 system as being much easier to work with since it's all about lopping off digits, trying to divide a unit length into tenths with high precision is actually far more difficult than halves or thirds. This is why pretty much every customary system involves a lot of units that are twice or three times the next smaller unit. The metric system isn't really feasible until you get the dividing engine [1], which doesn't show up until about the 1760s. [1]
    - JadeNB 6y ago
      Your [1] seems to be missing. > This is why pretty much every customary system involves a lot of units that are twice or three times the next smaller unit. I think easily of units that step up by a multiple of 6 (inches and the various time steps), but nothing immediately occurs to me where one unit is directly twice or thrice another. (Oh, except tablespoons, which are three teaspoons.) What examples am I missing?
- cyphar 6y ago
  Fahrenheit was initially intended to have 96°F be equal to body temperature (in fact it was one of the three fixed points in the temperature scale, designed to have 64 steps between the freezing point of water at 32°F and human body temperature). However after the introduction of Celcius, Fahrenheit was redefined slightly (with the freezing and boiling point of water being the fixed "nice" values for the scale -- to match the model used by Celcius) which resulted in human body temperature no longer having such a nice value. This also moved the 0°F value. So while technically Fahrenheit does predate Celcius and it did have a "nice" value for body temperature when invented, it was soon afterwards redefined such that arguably the value is just a conversion from Celcius. In short, you're both correct. :D
  - folli 6y ago
    Why did they come up with 96 instead of 100, a nice round number?
    - JadeNB 6y ago
      Sounds like the motivation was powers of 2, not 10.
    - jcranmer 6y ago
      Imagine you've just made yourself a thermometer by marking off the 0° and 100° points. Now delineate the 2° intervals, using only the tools you would have had available in the early 18th century. Now imagine making a thermometer by marking off the 32° and 96° points (64° apart). Now delineate the 2° intervals. Considering those two tasks, 64 seems a much better number to me (only have to divide unit lengths in half) than 100 (have to divide something by 5).
      - folli 6y ago
        Great explanation, thanks.

I still remember the day I found out the "98.6 degrees" human body temperature was just an oversignificant conversion from "about 37 C, or maybe lower."

by sirsar 6y ago

I still remember the day I found out the "98.6 degrees" human body temperature was just an oversignificant conversion from "about 37 C, or maybe lower."

JadeNB 6y ago

> I still remember the day I found out the "98.6 degrees" human body temperature was just an oversignificant conversion from "about 37 C, or maybe lower." Really? According to Wikipedia, the Fahrenheit scale (https://en.wikipedia.org/wiki/Fahrenheit#History https://en.wikipedia.org/wiki/Fahrenheit#History) is about 18 years older than the Celsius scale (https://en.wikipedia.org/wiki/Celsius#History https://en.wikipedia.org/wiki/Celsius#History), and I'm surprised that human body temperature wasn't included as a calibration. (But of course the fact that I'm surprised by it doesn't mean it isn't true!)

_underfl0w_ 6y ago
Could've sworn I remember hearing in grade school that one of then was actually calibrated with cow body temperatures. Don't remember which scale, though.
- jcranmer 6y ago
  I've definitely heard that as well. Where it comes from, I think, is that cow body temperatures are about 100°F, and someone assumed that the scale must have been calibrated using 0° and 100° as specific endpoints. According to Wikipedia, Fahrenheit seems to have chosen 32°F as the reference point for the freezing of water [it's not clear why this number specifically], and found the body temperature of a human to be 64° more than 32°F [i.e., 96°F]--it being much easier to divide a unit distance in 64 = 2⁶ than to divide it in 100ths. That does bring up an interesting point, though: while we modern people may think of the metric, base-10 system as being much easier to work with since it's all about lopping off digits, trying to divide a unit length into tenths with high precision is actually far more difficult than halves or thirds. This is why pretty much every customary system involves a lot of units that are twice or three times the next smaller unit. The metric system isn't really feasible until you get the dividing engine [1], which doesn't show up until about the 1760s. [1]
  - JadeNB 6y ago
    Your [1] seems to be missing. > This is why pretty much every customary system involves a lot of units that are twice or three times the next smaller unit. I think easily of units that step up by a multiple of 6 (inches and the various time steps), but nothing immediately occurs to me where one unit is directly twice or thrice another. (Oh, except tablespoons, which are three teaspoons.) What examples am I missing?
cyphar 6y ago
Fahrenheit was initially intended to have 96°F be equal to body temperature (in fact it was one of the three fixed points in the temperature scale, designed to have 64 steps between the freezing point of water at 32°F and human body temperature). However after the introduction of Celcius, Fahrenheit was redefined slightly (with the freezing and boiling point of water being the fixed "nice" values for the scale -- to match the model used by Celcius) which resulted in human body temperature no longer having such a nice value. This also moved the 0°F value. So while technically Fahrenheit does predate Celcius and it did have a "nice" value for body temperature when invented, it was soon afterwards redefined such that arguably the value is just a conversion from Celcius. In short, you're both correct. :D
- folli 6y ago
  Why did they come up with 96 instead of 100, a nice round number?
  - JadeNB 6y ago
    Sounds like the motivation was powers of 2, not 10.
  - jcranmer 6y ago
    Imagine you've just made yourself a thermometer by marking off the 0° and 100° points. Now delineate the 2° intervals, using only the tools you would have had available in the early 18th century. Now imagine making a thermometer by marking off the 32° and 96° points (64° apart). Now delineate the 2° intervals. Considering those two tasks, 64 seems a much better number to me (only have to divide unit lengths in half) than 100 (have to divide something by 5).
    - folli 6y ago
      Great explanation, thanks.