Frames White

Main thing was in my last PR (#342) I really wanted to quickly see if it was Julia 1.0 failing or Julia 1.5. I wonder if the answer is we...

I am not keen on that particular solution. In general i want `@scalar_rule` to be less extra powerful.

> Note that AbstractFFTs seems to be a much smaller, and much slower-changing, package than ChainRulesCore. Especially if you include a test dep on ChainRulesTestUtilititesEtc, which is also changing. So...

I think most of the oens under 3 should be solved via implementing them for `ZeroTangent` without consideration for `project`ing. While there are many they are finite in number. and...

The case we have to worry about is functions with multiple inputs, some of which are zero, and some of which are not. Which as I say that, I realize...

`(Zero(), something)` should be `Tangent{Tuple{Int, Float64}(ZeroTangent(), something)`. Since `Tuple` is not a valid tangent type since it doesn't support `zero` or `+`. Though a lot of our methods do let...

Handled in the AD system before calling the pullback. like this line in Zygote https://github.com/FluxML/Zygote.jl/blob/1082ebd3aced63b99c4b6c2956a122ce6a37f97d/src/compiler/chainrules.jl#L94 and this is where we would change Nabla https://github.com/invenia/Nabla.jl/pull/189/files#r662148541

> While it's not too hard to contrive a Hermitian matrix with exactly repeating eigenvalues (e.g. a diagonal matrix), I've had no luck so far coming up with a non-Hermitian...

`tr` is an example that should be `FIllArrays`. Zygote does so.

I wonder if we need to open an unstream issue on julia itself about this.