Andrew Keller

I admit I don't have much bandwidth to contribute here, but it would be great if the end result of the WIP PR is that I could use `Unitful.Quantity` types...

That is likely a problem, yes; also, the type parameter `T` is restricted to `Real`. My use case is for validated math with Unitful quantities, so correspondingly I would also...

I am not sure I could pin it down to one particular design reason, but here are a few arguments in addition to the comment I linked above. One argument...

That's true, and certainly helpful in many cases. (Measurements.jl also works in this way, but not the other way around, if I remember right.) However, if a package defines a...

I don't think there is a perfect solution for dispatching on a `Unitful.Angle` type, because a) angles are dimensionless (see https://github.com/PainterQubits/Unitful.jl/issues/38#issuecomment-254578080) and b) one could always define new angle units,...

> The fact that `div(5u"m", 2) != div(500u"cm", 2)` is another reason why this should not be allowed. There's a lot of past precedent for this sort of argument, e.g....

Sorry for not getting back to you earlier. I like this idea a lot. I'm a little worried that we'd get collisions because someone else might define `Irrational{:μ}`, otherwise it...

Should this issue stay open given that #204 was merged, and if so what remains to be done for your needs?

Thanks for your report. I think it is an integer overflow issue... `c0` is defined as an integer to permit exact conversions. I agree this is not something most users...

For example: ``` julia> uconvert(u"m^2*H^-1*s^-1", (1.0Unitful.c0)^3*Unitful.ɛ0) 2.3856725796184712e14 m^2 H^-1 s^-1 ```