I'm surprised nobody has mentioned using git-annex (and Git of course) to manage data (full disclosure: I develop an archival management app which keeps its dat

I'm surprised nobody has mentioned using git-annex (and Git of course) to manage data (full disclosure: I develop an archival management app which keeps its data in git/git-annex repositories). Of the seven key properties, git-annex gives you 1,2,3,4 and supports 5 and 7; 6 depends on how you store things. Git-annex supports identity in the sense that each clone of the repository has a UUID. You can choose to have a central hub if you like but you don't have to (surprise! you don't really need GitHub!). It comes with caveats of course: Binaries can be synced but only textual data is actually versioned, and once you put something in it will always be there unless you use something like git-filter-branch.