And conflating the two problems really doesn't help. Simply: Apps exfiltrating data can be prevented almost entirely using sane permissions, sufficient securit

And conflating the two problems really doesn't help. Simply: Apps exfiltrating data can be prevented almost entirely using sane permissions, sufficient security, oh and not willingly turning a blind eye, helps. That is about read permissions and entirely not what this news is about. Bots and troll-armies posting to polarise and increase tensions, are indeed a write permission problem, and superficially appear to be what this news is about. Except that as long as humans can post on FB, so can troll armies and bots. This only takes away capability for the small to medium scale ease of use. Any determined adversary wouldn't have been using these capabilities anyway because it's too easy to detect as not organic. Surely the automated blackhat and/or psy-op bot farms use automated browsers, headless browsers, key macro scripts, or whatever to simulate an actual human user interacting at human speed--just 1000s in parallel. I mean it's not rocket science and as soon as you're running a slightly larger operation it pays to just do it properly right away. This news changes nothing for the bot farms on missions that had to fly below the radar already anyway.