Adding noise and/or padding is trivial, it would just be done on a per-file basis. Of course, this would also lose the protocol's identical file feature, but th

Adding noise and/or padding is trivial, it would just be done on a per-file basis. Of course, this would also lose the protocol's identical file feature, but that's not a regression vs v1.