(no subject)

[sigmaleph]

The conversion between Fahrenheit and Celsius is a distinctly annoying unit conversion to do in one's head¹, because rather than being a simple linear map (i.e. multiplying the value on one scale by a number to obtain the value on another scale) it's an affine transform (it involves both multiplying by a number and adding a different number). Most other unit conversions are linear maps, for the straightforward reason that the units involved measure a quantity with an obvious zero point; everyone agrees what "zero length" or "zero mass" should mean, and therefore zero feet is the same as zero metres and zero pounds is the same as zero kilograms. Not so with temperature: 0°C is 32°F (the common point between the two systems is actually at -40), because when those scales where invented they set the zero point using different physical references.

Nowadays, we know there is such a thing as an objective point of zero temperature, and have based temperature scales like Kelvin on it, but would this have been practical back in the 18th century when every scientist and their cousin were inventing temperature scales? Guillaume Amontons had an estimate of absolute zero in 1702 at what would be about 33 K, which predates both the Celsius and Fahrenheit scales but is not as precise as you'd like. Wikipedia claims there's an improvement in 1779 due to Johann Heinrich Lambert that got as close as 3.15K, but since Wikipedia also claims Lambert died in 1777 I am not sure how much I should rely on that. Encouraging as those results might be, Dalton in 1808 lists a wide range of estimates all of which drastically overshoot absolute zero (i.e. place it much lower on the scale)². There was nothing approaching consensus on what zero temperature would be until much later.

Still, one can imagine a world where the discoveries happened in a different order, or the earlier temperature scales fell by the wayside as so many others have in our history, and scientists insisted that we should use rational temperature units which place the zero in its proper place. A world where maybe some people use Kelvin and some people use Rankine, or equivalents, but nobody thinks something like Fahrenheit would be remotely acceptable.

Would that be a better world? I don't know. But it sure would have slightly easier unit conversions.

¹ More annoying ones exist (e.g. decibels) but they are not as relevant to everyday life

²It might be an interesting exercise to read through detailed experiments and figure out if I, a person with the benefit of a modern education in thermodynamics I have not yet entirely forgotten, can figure out where they go wrong

Comments

[lunartulip]

I lean towards yes, better-world-wise. No more inconvenient for the people who are working within a given system, same as neither Fahrenheit-users nor Celcius-users seem to struggle with temperatures more than the other in the current world, and more convenient for people converting between systems. (Including people engaging in cross-cultural communication, people using the Ideal Gas law or other equations which require correctly-zero-placed temperature scales, and probably lots of other people I'm currently failing to think of as well.)

[nathanielbuildsatesseract]

You can get by with Kelvin (or Rankine, I suppose) for everyday life, so I'm inclined to say it would be better.

I don't think there's a way to re-index with absolute zero, and the 1-atm freezing and boiling points of water at nice round numbers, so there's not much point to updating past one of those.