L3 Forwarding devices implement policy. The policies, in their most basic form, tell the forwarding agents where, when, and how to handle various classes of traffic. What happens when competitive entities need to interconnect their L3 devices in order to build a larger network? Does the current NAP model work well? Do peering agreements, as we understand them today, work and scale well?
What I am curious about here is the view from the big networks. A small number of networks carry a very large percentage of the traffic. I'd like to know what piece of their traffic actually crosses a public exchange, rather than being delivered to a customer or another big network over a private peering arrangement. The answer to this question will put some level of relevence on this discussion. If those who carry 80% of the domestic traffic exchange 80% of their off-network traffic privately then NAP architecture and growth stratagem don't really make a difference. Brian Horvitz