Dr. Vinton Cerf - Internet, Infinity and Beyond - Page 4

Pages:  Previous    1  2  3  4  5  6  7  8  9  10  11  12    Next

So when Bob Kahn and I were doing the original design, we decided that we wanted the Internet packets to be insensitive to the underlying transmission and switching systems that carried them, so we didn't care whether the packets were carried over the radio lengths or optical fibre of coaxial cables or satellite channels and in order to commemorate the notion that Internet package should flow of any switching technology even the ones that had been invented yet I had a T-shirt may sit IP on everything.

And that's sort of what I've been doing for the last 35 years. The side effect of success however in getting this Internet protocol to work over virtually any transmission system is that I need a new T-shirt because people are assuming that it exists and they assume that they can build layers of protocol on top of it that so I think my new T-shirt will read IP under everything.

The whole idea is that people are building new infrastructure each time they create new layers of protocol which sit on top of the basic system. Another aspect of the Internet design is that the packets don't know what they're carrying, they don’t know whether they carrying video or audio web pages or part of any email message. And so the application is completely divorced from the underlying switching and transmission technology this is very different from the traditional world of telecommunications where an application is infinitely bound to a particular transmission medium whether it's television over the air or radio over the air or telephony over twisted-pair or television over cable the transmission system and the application are very closely connected.

The Internet cuts through all of that and says I really don't care how I’m being carried and I really don't care what I'm carrying, which leads to what we've been calling over many years the end to end principle which basically says the network doesn't know what the applications are but the edges do and because only the edges that know how to interpret the packets that are flowing through the network you're free as an inventor to try out new ideas, to put them up without having to get permission from an ISP to try your idea; so when Larry Page and Sergey Brin started Google they didn't have to get permission from some ISP or all the ISPs in the world or pay all the ISPs in the world in order to carry their traffic they simply put the system up and allowed anyone who was interested to try it out.

That freedom to experiment freedom to innovate is very, very important to the economic success of the Internet today and it's very important in my opinion to maintain that openness and that freedom.

Sometimes you hear arguments under the term “net neutrality” which expressed concern that the Controller of the underlying transmissions physical transmission system, could abuse that control by looking deep into the packets and deciding which package should flow easily, which one should be delayed and which ones might be discarded or otherwise interfered with and there have been threats along going along those lines United States from some of the local exchange carriers. Google and it's colleagues have been fighting hard to argue that we should have legislated protections to make sure that consumers that buy broadband access to the Internet have the ability to send packets or receive packets from anywhere in the Internet’s universe and the fact that physical carriage should not confer any advantage of higher layers of protocol and we hope that we can succeed. My impression is that you have an atmosphere here in Australia which supports that view and for those legislators who may be here in the audience I applaud that and I hope that you'll continue to maintain that posture.

Pages:  Previous    1  2  3  4  5  6  7  8  9  10  11  12    Next