In this <https://www.google.com/url?q=https%3A%2F%2Fgithub.com%2FJuliaStats%2FDistributions.jl%2Fpull%2F475&sa=D&sntz=1&usg=AFQjCNGAxBWP0wtmikthVFCNfbzZtAz-Eg> very long discussion (and maybe some PRs referenced therein).
I think the rationale was that it doesn't really make sense to have Normal, etc. take a supertype of Real. Maybe complex is a counterexample, but in general... On Tuesday, April 12, 2016 at 11:22:05 AM UTC-4, Jason Merrill wrote: > > On Tuesday, April 12, 2016 at 8:29:25 AM UTC-4, John Pearson wrote: >> >> AD numbers are subtypes of Real. This is not quite mathematically >> correct (there's a long discussion on GitHub), but in this case, >> practicality beats purity. >> > > Do you have a reference to the discussion about AD numbers being subtypes > of Real? I'm curious to read the rationale, since Julia already has a > Number abstract type that seems like a more obvious candidate for the > direct supertype of AD numbers. I'm sure there's a good reason--I'd just > like to read about what it is. >
