According to the feedback we've got, the usual goal is to see combined stats from multiple sources. I also have a suspicion that having a DB with multiple connections in a cloud storage wouldn't end well.
It depends if those cloud sync folders do atomic swaps of the files once they do a sync update or if they update in place as your reading the file. The key is making the 'other computers' dbs readonly from the perspective of the client view rendering app.
