Scientific Understanding of Policy Complexity - July 2016

Public Audience
Purpose: To highlight project progress. Information is generally at a higher level which is accessible to the interested public. All information contained in the report (regions 1-3) is a Government Deliverable/CDRL.

PI(s):  Ninghui Li, Robert Proctor, Emerson Murphy-Hill
Researchers: Jing Chen, Haining Chen, Matt Witte

HARD PROBLEM(S) ADDRESSED

• Policy-Governed Secure Collaboration -  Security policies can be very complex, in the sense that they are difficult for humans to understand and update.  We are interested in two kinds of complexity measures.  The first is a measure of the inherent complexity of a policy.  The second is a measure of the representational complexity, which is the complexity of a particular way to encode the policy.  It is desirable to have a scientific understanding of both kinds of complexity. 
• Human Behavior - Our policy complexity is based on how easy for humans to understand and write policies.  There is thus a human behavior aspect to it. 

PUBLICATIONS
Report papers written as a results of this research. If accepted by or submitted to a journal, which journal. If presented at a conference, which conference.

ACCOMPLISHMENT HIGHLIGHTS

• We have been studying SEAndroid policies, and found that unexpected policy composition is a root cause for many policy misconfigurations.  For example, when two policy rules grant two separate permissions to one program, this combination may enable the program to do things that were not intended or expected.  Since these two rules may be distributed in different places in the policy, such interactions are difficult for humans to find.