Moritz Schauer

Yes, one has to be a bit careful, because one would either like to give `S` or a factorisation object for `Σ`, which of course informs about `S`, but is...

Do we use oplus in MeasureTheory? I had oplus as convolution in https://github.com/mschauer/GaussianDistributions.jl/blob/c358ae9cd074e5d7a29a68271f0032dceabc432e/src/GaussianDistributions.jl#L106

I think we should not do it, it feels like working against the grain of Julia, after all ``` ~$ exec '/Applications/Julia-1.6.app/Contents/Resources/julia/bin/julia' julia> I ERROR: UndefVarError: I not defined ```...

`a * dist1 + b * dist2` per se is a good notation to produce an unnormalised mixture of distributions. We should not mix operators for "distributions as distributions" and...

Can’t we have the entire thing as actual MixtureDistribution of Dirac at 0 and Poisson, or make that work?

So let’s make it ZeroInflated{Poisson}? For the same amount of code we get a couple of related zero inflated distributions, e.g ZeroInflated{NegativeBinomial}.

If we go with `+` and `-`, the docstring should mention ⊕ from #1445

The right approach is not to memoize but just to not constrain the type of the scale so if needed one can use a wrapper for the inverse scale (similar...

I am also pushing this view a bit. Imho most of the time this type system doesn't help but goes into the way of usability. Usually a single abstract supertype...

Reasonable uses are imho: indicating what `pdf` returns (a density or a probability vector or undefined) or whether `rand(D, 5)` creates a Matrix or a Vector.