> I don't think they'd have learned much from the visuals of a balloon

I suggest a balloon payload would be better aimed at monitoring (and establishing patterns of) RF traffic rather than visual data.