The Case for Abstracting Security Policies / Jun 2003
My first ever academic paper, written with the expert guidance of 
"Untrusted code" is just as much a social problem as it is a technical problem. Looking for a complete solution is unrealistic: it is analogous to looking for a solution to crime in general. With this in mind, we do not claim that our proposed framework is a panacea. However, although a number of security problems remain (e.g. covert channel leakage), we claim that our system offers the potential to raise the security level of existing general purpose operating systems significantly.