The data clearly supported the hole in the middle, and only reconstructions with a hole did fit the data. The data support the hole and that one side is brighter, but not much more than that. Thus a blurred image does not make one believe that we know more than we do.