You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Looking at the list of available units, the only "neutral" label is "dimensionless", but it could be inopportune in cases when the quantity is not dimensionless (i.e. not obtained as the ratio of two homogeneous quantities) but rather not specified. For those case, "a.u." (arbitrary units) would be preferable.
It is easy to understand how you might take a dimensionless quantity and multiply by, say, volts. The result would also have the unit of volts. It's not clear what would happen if you multiply a.u. by volts.
Is there a use case where using "dimensionless" causes a concrete problem with math operations, or is more about just keeping track of the fact that the units are in fact arbitrary (for graphing the data and making axis labels, for instance)?
Activity
rgerkin commentedon Jan 21, 2020
It is easy to understand how you might take a dimensionless quantity and multiply by, say, volts. The result would also have the unit of volts. It's not clear what would happen if you multiply a.u. by volts.
Is there a use case where using "dimensionless" causes a concrete problem with math operations, or is more about just keeping track of the fact that the units are in fact arbitrary (for graphing the data and making axis labels, for instance)?
ddale commentedon Jan 21, 2020
My first impression is that any operation involving arbitrary units should yield arbitrary units. a.u. x volts = a.u.
gulpgiulia commentedon Jan 21, 2020
This.
rgerkin commentedon Jan 23, 2020
How would other operations be handled? If I add 3 a.u. and 5 mV, what should happen?
gulpgiulia commentedon Jan 27, 2020
I agree.