Conspiracy in Social Networks

In a Big Brother society, the thought police wants to follow a "divide and conquer" strategy by fragmenting the social network into isolated components. You belong to the resistance and want to foil their plans. There are rumours that the police wants to detain individuals that have many friends and individuals whose friends tend to know each other. The resistance puts you in charge to decide which individuals to protect: those whose friendship circle is highly interconnected or those with many friends. To decide you simulate two different attacks on your network, by removing $($i$)$ the nodes that have the highest clustering coefficient and $($ii$)$ the nodes that have the largest degree. Study the size of the giant component in function of the fraction of removed nodes for the two attacks on the following networks:

A network with N $=104$ nodes generated with the configuration model $($SECTION $4\mathrm{.}8)$ and power$-$law degree distribution with $=2\mathrm{.}5.$

A network with N $=104$ nodes generated with the hierarchical model described in Image $9\mathrm{.}16$ and ADVANCED TOPIC $9.$B$.$

Which is the most sensitive topological information, clustering coefficient or degree, which, if protected, limits the damage best? Would it be better if all individuals' information $($clustering coefficient, degree, etc.$)$ could be kept secret? Why?