The Internet is not a thing - it is a dynamic. The current Internet is a prototype based on the end-to-end constraint which meant any solutions and protocols couldn't depend on service providers within the network. The architecture of a LAN of LANs sharing a common backbone was sufficient for the prototype as the use of the 32 bit IP addresses that were assigned from a central source.

Today, as we see with the networks in our homes, we can do our own networking and are ready to removing the training wheels and interconnect the various networks without depending on a single "Internet". P2P protocols and applications such as Skype give a hint as to what post-Internet infrastructure would be. We'll still call the effect "The Internet" but without the illusion that another telecommunications service.

Applications like telephony using VoIP are simple if we can "just connect" and have abundant capacity. This new infrastructure is dynamic and intrinsically mobile since it is based on relationships and machines or particular paths such as broadband pipes.

We've become so used to the idea of telecom as a service in the style of the railroads that the biggest impediment to moving ahead is recognizing that we don't need depend on service providers to assure that we can communicate. It's just the opposite - the Internet has demonstrated that it's hard to prevent connectivity. Instead of moving ahead we vesting more of our effort in pipes controlled by providers and thus stifling the Internet dynamic.

Our challenge is to first understand connectivity apart from telecom. We can then focus on local connectivity and then interconnecting the local efforts outside the purview of traditional telecom. The key is understanding and a community built on common protocols.