In this paper we explore the interaction between of Internet communications, activists, and the state in Egypt, Syria, Libya, Uganda and northern Sudan. This paper addresses the following problem: Under what conditions are authoritarian regimes able to disrupt Internet traffic in situations of a popular uprising, and what can be done to prevent it? We illustrate that there are three critical variables in this interaction: redundancy in communications, distribution of power across organizations and individuals, and state regulation. We argue for a more resilient, redundant network. We propose policies that can be implemented in more open states with greater influence on the development of the network. We illustrate that the same investments that will empower dissidents strengthen the Internet for commerce and government, and against unauthorized attacks.