This is a position paper describing an approach to the creation of the Future Internet. We argue that the new architecture must respond to two key challenges: (1) increase trust among Internet stakeholders and (2) provide cost efficient scaling of the network to new levels of capacity, number of users and applications. We argue that the solution is to redesign the Internet by gradually replacing IP with a carrier grade transport system. In practice such a packet transport system can be created based on Carrier Grade Ethernet. We call the resulting network Internet by Ethernet. We make some fundamental arguments and outline the research agenda that will open based on the premises that we describe.