In a story titled “Bush administration annexes internet,” The Register talks about how the U. S. Department of Commerce intends to “retain control over the Internet’s root servers indefinitely.” This is a reversal of previous U. S. and may in some way be a response to U. N. attempts to control the internet.
What is surprising to me is that such an approach works. I mean, I’m assuming here that we are talking about the DNS root servers, since I can’t think of any other root server. For all practical purposes, root servers are just those servers listed in a file of IP addresses that a DNS server goes to to look for an address if all else fails.
This list at root-servers.org lists the IP addresses and locations of root servers A – M. It seems as though a lot of the root servers are outside of the control of the U. S. Department of Commerce already, so I’m not sure how they intend to “preserve the security and stability of the Internet’s Domain Name and Addressing System” other than say they control it.
How hard is it to set up a root DNS server? Well, the control issue is probably over the actual root DNS domain – the ROOT-SERVER domain for the gTLDs. Any root DNS server needs to be published so that any DNS servers can get them. If the DoC thinks that it can prevent a new root server from being published, I guess they could try. I’m really not sure what they could do if another nation-state or a non-governmental organization or a network tribe were to set up a new root server, and publish the address of that root server in the hints file of BIND’s named or something.
Some day nation-states will wake up to find not only that they cannot control the behavior of those within cyberspace but also that they no longer have any capacity to control the behavior of those who establish cyberspace itself. Between now and then, I suspect that nation-states and global governance bodies will continue to try very hard to maintain their control over both behavior within cyberspace and that of those who establish it.