4 ms·

... and this friends is how you inadvertently start a revolution. you snuff those nodes, though nodes notice and go to the public about it via other channels. t

... and this friends is how you inadvertently start a revolution. you snuff those nodes, though nodes notice and go to the public about it via other channels. the resentment builds. revolution ensues. There is fake news and then there is fake news. Absolutely fake news is easy to identify and snuff out. What's harder to snuff out is a religion or an ideology, fake news that people believe because they are convinced that it's in their best interest to believe it to be true -- i.e. Communism will free us all. When those interests converge then as a government you have problems that are difficult to solve and trying to engineer the network can easily backfire and lead to revolution. The point is never to get to that point, not try to engineer the situation at the end. https://www.youtube.com/watch?v=wWhtcU4-xAM https://www.youtube.com/watch?v=wWhtcU4-xAM

otakucode 8y ago
I completely forgot one big part of the social-graph-manipulation model of establishing a forever-government. It's not targetted. This isn't aimed at selectively suppressing certain groups, or stopping something once it's spreading. It's an architectural attack on the structure of the network. It is unlikely that anyone would notice this because of a few factors. First, the important 'bridge' nodes are almost always poorly connected generally. (I picture a bridge node person like a member of a biker gang who keeps in contact with his elderly aunt, providing a way for ideas to spread from a biker gang to her knitting circle.) If they complain that they think someone is interfering with their ability to communicate with this other group which, by definition, their in-group is unconcerned with... it is unlikely the group would care much. Also, the 'interference' would optimally be as innocuous-seeming and quiet as possible. The fact that you only have to manipulate a small number of connections to have large effects on the networks overall average distance between clusters supports this. And yes, it should go without saying that this is a monstrously unethical and terrible thing to do. It would probably have a lot of unforeseen consequences. I've been wondering if it would change how things like fashion trends or fads occur and if monitoring that might be a way to detect it. It would seem intuitively that it might lead to echo chambers and amplification of ideas within clusters since whatever moderating influence the slight connection to a different cluster provided would be removed.... but I haven't seen any research about that and I'm not sure how to simulate it without it just being a self-fulfilling prophecy.

otakucode 8y ago

I completely forgot one big part of the social-graph-manipulation model of establishing a forever-government. It's not targetted. This isn't aimed at selectively suppressing certain groups, or stopping something once it's spreading. It's an architectural attack on the structure of the network. It is unlikely that anyone would notice this because of a few factors. First, the important 'bridge' nodes are almost always poorly connected generally. (I picture a bridge node person like a member of a biker gang who keeps in contact with his elderly aunt, providing a way for ideas to spread from a biker gang to her knitting circle.) If they complain that they think someone is interfering with their ability to communicate with this other group which, by definition, their in-group is unconcerned with... it is unlikely the group would care much. Also, the 'interference' would optimally be as innocuous-seeming and quiet as possible. The fact that you only have to manipulate a small number of connections to have large effects on the networks overall average distance between clusters supports this. And yes, it should go without saying that this is a monstrously unethical and terrible thing to do. It would probably have a lot of unforeseen consequences. I've been wondering if it would change how things like fashion trends or fads occur and if monitoring that might be a way to detect it. It would seem intuitively that it might lead to echo chambers and amplification of ideas within clusters since whatever moderating influence the slight connection to a different cluster provided would be removed.... but I haven't seen any research about that and I'm not sure how to simulate it without it just being a self-fulfilling prophecy.