I am slightly bothered by the notion that "blending in" has any chance of working in the modern era. Certainly, in the past, the entity carrying out the survei

I am slightly bothered by the notion that "blending in" has any chance of working in the modern era. Certainly, in the past, the entity carrying out the surveillance was likely to have a limited budget of time-and-attention, a limitation that could be gamed to evade analysis and pursuit as described. However, computerisation and data analytics change the game significantly. The cost of "attention" is reduced to such a significant extent that practically everybody's behaviour can be analysed, albeit superficially. It is a relatively trivial matter for the entity carrying out the surveillance to build a dossier on large swathes of the population: to identify and categorise their beliefs, their behaviours and their weaknesses. To score them according to how much of a threat they represent, or according to whatever other factors may be useful. The information-dominance landscape has shifted considerably over the past 10 years. I am greatly interested in seeing how our newfound industrial-scale-information-dominance is translated into real-world power: Presumably also by automated means and on an industrial scale. I imagine that this can be done subtly, by targeting specific influencers, inhibiting some selected communications and inserting or promoting others, thereby shaping discourse and the formation of political belief in a larger section of the population. Emerging geopolitical crises (Ukraine/Crimea, anyone?) provide an ideal petri-dish to see if the competing factions are deploying this sort of technology, and for assessing the effectiveness of these techniques.