I find it quite legit. Having a singe repo would hinder progress as more and more implementations are added.
Even JuliaStats followed a similar distribution (https://github.com/JuliaStats), and it looks neat and organized.
Even JuliaStats followed a similar distribution (https://github.com/JuliaStats), and it looks neat and organized.
Why?
There are clear drawbacks, but there are clear advantages as well.