The subject of this overview talk is the long-standing open problem of developing a 
suitable capacity theory for wireless ad hoc networks, particularly those with 
considerable dynamics, such as node mobility and bursty traffic.   We’ll briefly review 
why this problem is difficult, general, and important, and provide some (hopefully) new 
ideas on how to think about making progress.  I’ll summarize some notable recent 
research progress, particularly from the “non-equilibrium information theory” team 
participating in the DARPA IT-MANET program.