All of the problems with fake profiles are fixable if we just use the social graph. It drives me nuts that this a problem. I have a solution, which i'll share h

All of the problems with fake profiles are fixable if we just use the social graph. It drives me nuts that this a problem. I have a solution, which i'll share here. I'm sharing it because i hope SOMEONE can build this or share it with someone at a high level at FB or twitter. I have a mortgage in silicon valley and a young child, so i'm not in a position to take the time and risk to do this. But i really desperately want to see it in the world. All we have to do is use the social graph to verify each other, and follow 'verified' edges to determine trust in a third party. People can just tell fb or twitter 'yes i know this account', and that's all we really need. If I can't follow any 'yes i know this person edges' to a remote account, don't let me interact with that account. Shadowban them. It's THAT simple. This technique stops bots and it stops trolling by fake remote accounts. If someone claims "i know all these fake accounts", then we ban that person, for creating all the fake accounts. Fake accounts are easily identified after the fact; when no real person pays any price, they'll keep getting created. Yes it has the downside of temporarily slowing adoption. That's the main reason imagine twitter and FB haven't done this. They think us being harassed is less important than onboarding new people. https://s3.neyer.me/respect-matrix-slides.pdf https://s3.neyer.me/respect-matrix-slides.pdf