The general perception is that internet cannot be controlled, regulated or shut down by any single individual or party simply because there are too many computers involved. There are billions of computers connected to the internet and it is virtually impossible for any entity to control each and every computer all at once.
Well, as past distributed denial of service attacks have shown, it is not necessary to wipe out each and every computer to control the internet. Each and every computer connected to the internet makes use of its connection with web servers to access pages on the internet.
If these web servers are targeted, the resulting billions of computers will automatically be disabled without any difficulty. Instances of distributed denial of service attacks in the past have shown that it is, at least in theory, possible to stop or disable the internet. Of course, it is practically impossible to do so because anti denial of service mechanisms are equally strong.
Another advantage of the internet is that any and every bit of information to safeguard systems can be distributed over a large number of computers and servers at practically zero cost. This is the reason why there is a conflict about whether the DDOS attacks should be treated with seriousness or not. As the latest attempt by those in support of WikiLeaks or claiming to be in support of WikiLeaks showed, a concentrated attack on the web servers of financial organizations can disable a significant part of the economy in absolutely no time.
Another significant aspect of this denial of service attacks is whether the inherent data present in computers and web servers is also at risk. As on date, no such attacks has successfully accessed the private and confidential data placed in web servers. Both MasterCard and Visa were at pains to point out that web servers were simply unable to handle the load and that data was secure despite the attack.