> If combatants need to quickly spin up a digital mass movement, well-placed personas can rile up a sympathetic subreddit or Facebook Group populated by real pe

> If combatants need to quickly spin up a digital mass movement, well-placed personas can rile up a sympathetic subreddit or Facebook Group populated by real people, hijacking a community in the way that parasites mobilize zombie armies. Why do we and the author seem to keep dancing around the fact that Facebook must be refusing to garbage collect bogus nodes on their own social network? The company presumably still has access to its own social graph, to reams of data from site/app/like button/audio/webcam/key&mouse events, and whatever else it purchases from third parties. How in the world can a troll farm employee proxied through a hacked router in Branson evade detection? An employee who isn't cross-referenced in photographs, videos, phone calls, PMs, contacts, browsing patterns, games, purchase histories, smart tv ad beacons, and whatever else Facebook sniffs up? In light of that, the article seems like a distraction. While I don't know how to get social media companies to garbage-collect their graphs I know if they did it would diminish the fragility of the system. If that raises the cost of subsequent information attacks such that they must happen in a different domain that the social graph, that's not a Maginot Line. It's an improvement. Edit: same logic applies to the input coming in from ads. Facebook presumably has just as rich data on the organizations buying them as it does on its userbase. Prune the fronts and information attacks move again to a different domain.