Distributions defined in Turing: is the plan to keep them here?

[DominiqueMakowski]

I recently spent some time looking for OrderedLogistic, assuming that it was in Distributions.jl, but actually it was defined directly in Turing

Turing.jl/src/stdlib/distributions.jl

Lines 147 to 168 in 4766fdd

	function OrderedLogistic(η, cutpoints::AbstractVector)
	return OrderedLogistic{typeof(η),typeof(cutpoints)}(η, cutpoints)
	end
	Base.minimum(d::OrderedLogistic) = 0
	Base.maximum(d::OrderedLogistic) = length(d.cutpoints) + 1
	# TODO: only implement `logpdf(d, k::Real)` if support for Distributions < 0.24 is dropped
	Distributions.pdf(d::OrderedLogistic, k::Real) = exp(logpdf(d, k))
	Distributions.logpdf(d::OrderedLogistic, k::Real) = _logpdf(d, k)
	Distributions.logpdf(d::OrderedLogistic, k::Integer) = _logpdf(d, k)
	function _logpdf(d::OrderedLogistic, k::Real)
	η, cutpoints = d.η, d.cutpoints
	K = length(cutpoints) + 1
	_insupport = insupport(d, k)
	_k = _insupport ? round(Int, k) : 1
	logp = unsafe_logpdf_ordered_logistic(η, cutpoints, K, _k)
	return _insupport ? logp : oftype(logp, -Inf)
	end

Out of curiosity, is there a reason not to move these distributions to Distributions.jl where their documentation might benefit from increased visibility?

[devmotion]

IIRC I moved a few of them (quite a while ago), and my impression has always been that they should be moved if they are of more general interest.

[yebai]

We are happy to move more distributions in this repo, and also DistributionsAD to Distributions.jl, particularly the autodiff rules currently in DistributionsAD.