Hacker News new | past | comments | ask | show | jobs | submit login

> Ew, multiplicative temperature comparisons in unspecified units

In the previous paragraph he's said that water boils at 100 C




I'd expect a multiplicative temperature comparison to refer to Kelvin, not Celsius.


Or Rankine for a US reactor, surely? ;0)


Ah, TIL that Celsius is "non-multiplicative". From ChatGPT:

"When using Celsius as the unit for temperature, it is not meaningful to say that one temperature is a certain number of times higher than another. The Celsius scale is based on the freezing point (0°C) and boiling point (100°C) of water, which are not absolute values. Thus, the Celsius scale has negative values, and simply multiplying temperatures does not provide an accurate representation of relative differences.

If you want to compare temperatures in a more meaningful way, you should use the Kelvin scale. The Kelvin scale is an absolute temperature scale, with its zero point (0 K) representing absolute zero. In this scale, it is appropriate to say that one temperature is a certain number of times higher than another because the scale starts at an absolute zero point. To convert Celsius temperatures to Kelvin, add 273.15 to the Celsius value. Once the temperatures are in Kelvin, you can then make meaningful comparisons using multiplication or division."


Yeah, kelvin strictly speaking, but for reactors on earth (not in 3 kelvin outer space) with a baseline earth-ambient temperature, perhaps multiplicative C is valid, very approximately? I don't know, I guess I'm asking.


Temperature, especially in this context, relates to energy. The relative energy isn't 8x.

> boiling point is more than 8 times higher than water’s, so it can absorb all the extra heat

This is a bit like saying that a skyscraper is 8 times as tall as an apartment building, so the gravity is much weaker up there. Obviously not as extreme a mistake, but that's why people are jumping on it.


Specifying "times" in Celsius is fine... as long as you stay either positive or negative.

Implying that you can only specify multiples in Kelvin is being pedantic.


The Celsius scale is an "interval scale" while the Kelvin scale is a "ratio scale". One cannot take ratios of Celsius values as ratios are not even defined for that. https://en.wikipedia.org/wiki/Level_of_measurement


But not really.

100°C times two is not 200°C.

100°C times two is 473°C


Howso? If the conversion to Kelvin is additive (as opposed to something more complicated), and multiplying in Kelvin is fine, then how is multiplying effectively-unsigned in Celsius not fine?


Ok, another example.

Is 10°C 10 times as hot as 1°C?

No.

Multiplying kelvin is fine because kelvin is an absolute scale.


> No.

For a given baseline, perhaps yes?. Kelvin has its baseline inherent, for C you can pick what you like if you're clear about it. Perhaps. (Edit: and using only positive or negative temps as pointed out above)


Yes.

Parent's spaghetti examples weren't the most illustrative.

What I believe they're griping about is that "10°C does not have 10x as much heat as 1°C."

In actuality, "10°C has 10x the difference in heat that 1°C has relative to the freezing point of water."

So it's a comparison definition quirk, not a numerical one.


    c = k + 273
    x * c = x * k + x * 273
This leaves multiplied celsius off by a delta of (x - 1) * 273.


That's showing multiplied Celsius in terms of Kelvin, no?

  1 Celsius = 274.15 Kelvin
  10 Celsius = 283.15 Kelvin
Obviously 283.15 K != 10 * 274.15 K.

But that's not what's being said, per my lower comment -- it's 10x the difference relative to the freezing point of water, compared to 1°C.

The original upthread comments I were taking issue with were saying that multiplication has no meaning / is invalid in Celsius.

It does have a meaning, and it's mathematically consistent.

It may not mean what's being assumed, but nuance doesn't mandate "thou shalt never," IMHO.


When I say multiply by n, what I'm really doing is applying a delta of (n - 1) times my distance to the origin.

The origin with celsius isn't at zero, so you're prevented from doing what you were pretending was multiplication (at least in the most natural way of doing it).

Obviously you can negate your offset, multiply and reapply your delta.

It's a bit damned awkward though.


It's a bit the same as with time, multiplication of the time of the day don't make much sense.

I started the work at 1:20 pm, but my colleague started three times later, at 4pm.


> Specifying "times" in Celsius is fine... as long as you stay either positive or negative

127C is twice as hot as -73C though


<pedant>C is a measure of temperature, not heat.</pedant>




Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: