I don't know, I think if there is a bias or we only read things released by Russian spies then it seems really important to take that into account If every sin

I don't know, I think if there is a bias or we only read things released by Russian spies then it seems really important to take that into account If every single person's emails are at least controversial then it just means you are going to be a mob chasing anyone that was targeted, unrelated to if they are better or worse than everyone else (and it seems like any email is damning, pizzeria-thank-you emails are enough to cause people to think there is a vast child trafficking conspiracy and literally shoot up the place). If it was an unbiased sample it is fine to catch a random subset and set examples of them, but if you have a biased sample then you might be actually leading to concentration of power in whoever has the least scruples in hacking and leaking opposition emails if you aren't careful to take the sample bias into account.