Hackers have learned to leverage the enormous number of poorly protected home computers by turning them into a large distributed system (known as a botnet), making home computers an important frontier for security research. They present special problems: owners are unsophisticated, and usage profiles are varied making onesize-fits-all firewall policies ineffective. We propose a social firewall that collects security decisions and both user and usage characteristics, and provides users with personalized information to assist with allow/deny recommendations. To succeed, a social firewall must deal with at least three user behavior issues: why contribute private information? why make effort to provide quality information? and, how to prevent manipulation by adversaries? We sketch an incentive-centered design approach to each problem. We provide an economic model and some analytic results for a solution to the fundamental problem: why contribute? We show that an excludable public goods mechanism can achieve a better outcome than a system without social motivators.

Rick Wash and Jeff MacKie-Mason. “A Social Mechanism for Home Computer Security,” Presented at the Workshop on Information Systems and Economics (WISE), December 2008.

Download: PDF

Understanding how users choose tags can help researchers better understand how tagging systems can be used and how to design better tagging systems for the future. We developed a simulation of del.icio.us, a popular social bookmarking tool, that allowed us to simulate users choosing tags using one of four possible strategies for tag choice found in the literature. We then compared the resulting tag choices with empirical data retrieved from del.icio.us to determine which tag choice strategies would result in choices most similar to those seen in the real world. We were able to rule out three of the strategies as unlikely to be the primary means by which tags are chosen on del.icio.us.

Rick Wash and Emilee Rader, “Understanding del.icio.us Tag Choice Using Simulations,” Presented at iConference 2008, Paper Track. February 2008

Download: PDF

Humans are “smart components” in a system, but cannot be directly programmed to perform; rather, their autonomy must be respected as a design constraint and incentives provided to induce desired behavior. Sometimes these incentives are properly aligned, and the humans don’t represent a vulnerability. But often, a misalignment of incentives causes a weakness in the system that can be exploited by clever attackers. Incentive-centered design tools help us understand these problems, and provide design principles to alleviate them. We describe incentive-centered  design and some tools it provides. We provide a number of examples of security problems for which Incentive Centered Design might be helpful. We elaborate with a general screening model that offers strong design principles for a class of security problems.

Rick Wash and Jeff MacKie-Mason. “Incentive Centered Design and Information Security,” Presented at the First Workshop on Hot Topics in Security (HotSec). July 2006.

Download: PDF

Network worms, simple slang terminology for automated intrusion agents, represent a persistent threat to a growing Internet in an increasingly networked world. However, their evolution has been somewhat limited, and they still rely on the same basic paradigms, which contain fundamental flaws. We analyze the basic components of a worm and apply this analysis to three worms found in the wild on the Internet. We then proceed to analyze the limiting factors of existing worm paradigms and outline new ideas which we expect to become prevalent. These new worms will prove to be more difficult to identify and eradicate. It is our intention in sharing this knowledge to stimulate the development of strategies to detect and counteract the threat of smarter network worms.

Download: PDF, PS