As a result of their unplanned, license-free nature, WiFi networks have grown quickly in recent years, giving users unprecedented improvements in wireless access to the Internet. But being “chaotic,” i.e. unplanned, they have grown to be victims of their own success: when eager users set up too many wireless access points in a densely-populated area, the resulting noise and interference hurt everyones throughput and connectivity. Cellular mobile telephone networks are planned carefully, but in order to expand coverage indoors, providers are turning to customer-deployed femtocells, thus incuring the drawbacks of chaotic WiFi networks. We propose a ground-up redesign of chaotic wireless networks, with new architectural contributions focusing on what information the physical layer should pass up to higher layers. We propose a new physical layer interface called SoftAoA that passes angle-of-arrival (AoA) information from the physical layer up to higher layers. Using this expanded physical layer interface, we will first investigate fountain coding and receiver-based rate adaptation methods to improve wireless capacity in the vagaries of the “grey zone” of marginal coverage. Second, we will investigate improvements to security and localization that can be made based on the profiling of incoming packets’ AoA at an access point. Finally, we will investigate how a chaotically-deployed network can mitigate the interference it experiences from networks not under the same administrative control, and manage the interference it causes to those networks. The result will be more scalable, secure, and reliable chaotic wireless networks that play an even more prominent role in our lives.