The differentiation [between Amazon and OpenStack] is that OpenStack technology is driving an initiative, throughout the world, which can be adopted and molded in a non-proprietary way. If you, as a user, want to integrate some proprietary technology into OpenStack, you can do that very easily. It's basically an open system. The companies and brand names you trust are there to make sure the technology works in a reference architecture. The second part, is there's been a real shift in the industry about how end users are viewing their IT assets. They want to manage and deploy them using a hybrid model. In the hybrid model, they want assets and IT infrastructure in their own environment, but also want to be able to take advantage of the public cloud, and take advantage of traditional, managed hosting. Many end users come to us saying we have X number of data centers, and we don't want any more. They want to put their new workloads in our data center. But, they have to be able to talk to workloads in their legacy data centers. With technology like OpenStack, it's really built, and designed for a hybrid cloud computing environment.
–Rackspace, OpenStack, and How Texas Is Driving The Cloud Computing Market
Good interview with John Igoe who I used to work with at Dell from time-to-time. He had some good input for a "how to get things done at a BigCo" presentation I did at DevOpsDays a few years ago: hide out, and deploy chaff as needed. He's now VP of Private Cloud at Rackspace.