The problem isn't about censorship, and *is* about control.
Basically, the US has absolute control over what TLD's can exist, and no-one else has any real say over it. Potentially, they could switch off the .COM TLD today simply by removing it from the root servers.
I can't go into the reasons why I think that the US should share control of the root servers, as that will turn into a political debate, but you should probably take note that the .XXX domain was stopped in its tracks by under 40 letters from a single ...um... group addressed to the US congress ... and that's not good. The internet is supposed to be free from political control.
If ownership can't be shared, then I can see the rest of the world implementing their own root servers in the same way that China already has, resulting in the internet becoming fragmented.
@Raven,
The DNS system is basically arranged in a tree-like structure,
For example,
Root server, knows all TLD servers below it.
Picking .UK, knows all domain servers below it.
Picking .CO.UK, knows all domain servers below it.
Picking .DEMON.CO.UK, knows all servers local to this domain.
When I request a name from within the .DEMON.CO.UK domain, I get the answer directly from my local DNS, being a part of that domain myself.
When I request MSDN.MICROSOFT.COM, the request will run all the way up the tree to a root server, and then move down the .COM branch, then to the .MICROSOFT.COM branch, which will send the requested IP to me ... but this only happens if the address isn't already cached along the way.
It's basically like asking a question, and either getting an answer, or being told 'I don't know, but I know someone who does'.
@IanT,
The US military sponsored University research into packet-based networking so that the network couldn't be destroyed by removing a single link (by, for example, blowing up a city with a nuke).
The packet switching concept was 'invented' early in the 60's. It wasn't until the late 60's that the military actually began paying for research, and researchers from both sides of the atlantic were involved. The first real precursor of the internet didn't appear until 1972, and the first application was ... email.
CERN were *not* the inventors of the internet. They were the inventors of the http protocol, i.e. the Web.
The term 'bug' is not limited to computers. I knew it was used during the war for radar problems, but a quick google shows me that it was also used during the 1800's for general electrical problems too.
