Mr. Wig
Don’t Copy That Floppy (Official Video - Digitally Remastered) (by AntiSoftwarePirates)
(1992)
THURSDAY
Cleanliness
Around 1 pm the yearning to create reaches a maximum, then slowly wanes away as the humdrum workday proceeds. Leaving work early only serves as a reminder that beltway traffic can occur at any hour. And finally, by the time I arrive home, all hope of a satisfying creative endeavor is lost. So, I sit. So, I drink. So I futilely try to titrate my alcohol intake in the hopes of achieving the illusive balance between subtle inebriation and stimulated productivity.
But maybe, just maybe, tonight I clean up this shithole just enough to get something done tomorrow.
Reasons I Should Stop Reading about Network Theory
1) Today’s Journal Entry:
I had a disturbing dream last night, of which I remember next to nothing. When I awoke—and I believe this thought may have been completely unrelated—I realized that flash mobs are an excellent model of the bot-net of personas that HBGary recently proposed creating to influence social media networks. There was a great amount of concern in the HBGary paper about reducing “cross-contamination,” which is obviously pivotal to control. In the event that bots became cross-contaminated, they would be very easy to pick out of the group, and the entire network of personas could be easily identified as originating from one user. It occurred to me, that the most effective way to deploy a bot-net attack on a social network, would be to make sure the nodes (bots) of your network were evenly dispersed from one another, and that they exhibited weak ties (acted as bridges) to real users (target nodes). Any cross-contamination would destroy bridges (if two fake nodes “friended” a real node, that real user would quickly be able to detect an abnormality, in that the fake nodes would be incredibly similar—a human would easily be able to recognize a pattern of spam).
This priniciple is embodied by flash mobs. In order for them to be effective, the participants must be scattered within a group—not only this, but effectiveness and impact are also increased by scattering them temporally. If they all began performing a routine at the same time within a tight proximity, the outside group easily recognizes them as a troupe and the impact can be more casually disregarded. If they start with one node in a centralized (game theory ideal location) and then spread across the group and across time to participate, they maximize the reaction they will get out of the surrounding network. Obviously improv flash mobs have a completely different goal than the proposed network of personas HBGary wrote about, in that they are intentionally provocative and draw a great deal of attention. Certainly the HBGary bots would be much more subtle in their operation, geared towards introducing an ideology to a social group and engineering or pushing existing social norms with these bot-networks.
I thought it was fascinating that this complex and manipulative social network attack concept can be easily modeled (and in fact has been, almost unconsciously) by improv groups who are merely seeking to boost their exposure or doing it just for the lulz.
