We differ in definition of things then. In my world, if the repository is not generally accessible to the public, then the research is not replicable

Then I ask you to go to the little shaky green leafs on ancestry.com and call them up and you will find that over 90 plus percent have no sources/repositories -----which I take to mean that their data is not replicable and is therefore unreliable -- since it is not generally available to the "Public". I think one of the main reasons you don't see it there is because they want to keep it private