joule unit is always converted to newton meter

[Tey']

When using the joule unit, it is always converted to newton meter in the result. Is it expected?

Steps:

• Type 10 joule.
• Experienced result: 10 newton meter
• Expected (?) result: 10 joule

[Tey']

For @polwel I guess.

[Pol Welter]

This is on purpose. In fact torque shares the same dimension, but it must not be expressed in joules. For a similar reason hertz is reduced as well. There surely are some more examples which I can't remember now.

[Tey']

Thanks for the explanations.

If I understand that article correctly, Joule is used both for torque and energy, but in practice, people use Newton Meter for torque and Joule for energy (as Newton Meter sounds uncommon for energy). Would it be that bad to break the relationship between Joule and Newton Meter, and consider that, when a user uses Joule, it is for energy only as she would have used Newton Meter otherwise?

BTW, Hertz is not recognized by the evaluator so it can't be used, and I fail to see the similitudes with Joule.

[Pol Welter]

Joule is used both for torque and energy

AFAIK it's not. Which line in the wiki article made you think so? I chose newton meter over joule because expressing torque in joules is really confusing, but expressing an energy as the work of a force over a distance is fairly intuitive (assuming you know that this is the definition of the joule). Really it went like this: work (and consequently energy) is defined as the scalar product of a force and a distance. Hence the unit would be newton meter, but we decided to assign it its own unit, namely joule.

BTW, the unit system of the TI calculators (which I drew quite a bit of inspiration from) also handle joule this way.

Would it be that bad to break the relationship between Joule and Newton Meter

Yes :) Both are given by the product of a force and a distance. Since there is literally no way we can distinguish a cross product from a scalar product, torque and energy are really the same to SC (but not in real physics...).

Hertz is not recognized by the evaluator so it can't be used

Ah right, I didn't bother including that unit at the time :) Problem is frequency (hertz), radioactive decay (becquerel), angular velocity (rad/s) and probably a few more all share the same dimension. Too bad.

BTW, I was sick for a few days, hence my sparse activity. Just this morning I had some ~250 messages from you guys in my inbox. Anything especially important that I might have missed? From what I can tell, you did a great job :)

[Pol Welter]

Can I close this? It's good that this issue has been brought up, I am absolutely certain that many more people will complain over this, so we can link them here :)

[Tey']

Thanks for the explanations, it's more clear now. TBH, I've barely played with torques, but I've encountered joules multiple times in electrical engineering, so converting joules into newton meter for these usages does not look that intuitive to me.

Wouldn't it be possible to keep the unit given by the user in the expression and use it in the result as long as no conversion is needed? Something like 10 joule * 2 would result in 20 joule, but 10 joule + 10 newton meter would result in 20 newton meter.

[Pol Welter]

Should be doable, I'll think about it.

But with your idea you managed to point out a quite major bug to me... will report it in a minute.

[Tey']

Okay thanks.

[Tey']

Should be doable, I'll think about it.

Forget about it actually: the main goal of units is conversion anyway, so such a feature wouldn't make sense in the end.