These thoughts are partially in response to movements like BLM but apply to the more general issues of how we perceive reality and the dangers of creating false (or mostly false narratives through mostly isolated incidents).
Black lives matter caught on quickly because of the very real issues with policing in urban areas. I can think of fewer mainstream policies that (IMO) are more unconstitutional than stop and frisk. If walking down the street makes you subject to search and seizure I don't see how that doesn't violate the 4th amendment. The war on drugs is a disaster, just like the "war on alcohol" was during prohibition. But those policies combined with the increased crime that is typically correlated with poverty create a large segment of the population that views the police as adversaries or enemies.
The issues with policing make it such that when a police officer kills a black man and the circumstances suspicious then confirmation bias kicks in to make the affected population believe the worst. We saw this happen recently in Charlotte, police killed a man, his family claimed he was unarmed, turns out he did have a gun and refused to put it down (
http://www.nytimes.com/2016/11/30/us/charlotte-officer-acted-lawfully-in-fatal-shooting-of-keith-scott.html?_r=0). But there were several days of protests (and some rioting) following the shooting. Ferguson (the starting point) is a similar story, Mike Brown almost certainly attacked the police officer before he was shot. But the BLM movement is ready to believe the worst about police officers because of confirmation bias and the adversarial relationship they have with police.
Social media makes these incidents "feel" closer to individuals. The victim is no longer some guy that the traditional media reports on which can feel distant, it's a friend of a friend on facebook. Two to three degrees of separation between an individual and the victim will make the incident feel more personal mostly because people will underestimate how big 2-3 degrees of separation in their online social networks is. This leads to a vastly overestimated risk analysis for this type of incident. We live in a country of 300+ million people, with over 40 million African Americans. Bad stuff is going to happen so someone pretty regularly. Police will have many interactions with young black men and 99.9+% of them don't end with the black man getting shot. But a narrative grew about the danger that police posed to young black men, that is ultimately dangerous to the police and black men. We have seen several police assassinations by people motivated by this narrative* that the police are running around rampant shooting black men at will. I'm sure that this will cause some men when interacting with police to act in overly defensive ways (that lead to aggression) that either cause them to be injured, killed, or incarcerated for longer than they otherwise would have been.
I have a feeling we are going to see more of these types of movements in the future. The human brain responds strongly to new perceived risks particularly when someone "we know" is harmed in a particular way. We could see a movement like this spring up against self driving cars even if they end up on the whole to reduce traffic fatalities.
The conclusion here is: there are real issues with policing (but almost never is it them shooting people for no reason), confirmation bias (b/c of those issues) leads a group to believe the worst immediately, social media and poor risk analysis lead us to vastly overestimate the risk which can lead to narratives that don't truly reflect reality and can be dangerous.
*The "left" needs to recognize the danger of these narratives and speak out against them the same way they would when conservative christian groups liken abortion to the holocaust and abortion providers to nazi death squads. Both narratives don't comport fully with reality and lead fringe members of the group to believe violent actions are justified and necessary.
