Select your localized edition:

Close ×

More Ways to Connect

Discover one of our 28 local entrepreneurial communities »

Be the first to know as we launch in new countries and markets around the globe.

Interested in bringing MIT Technology Review to your local market?

MIT Technology ReviewMIT Technology Review - logo


Unsupported browser: Your browser does not meet modern web standards. See how it scores »

{ action.text }

One of the fundamental design principles of today’s Internet is so basic and so important that few users have ever heard its name; they just assume its existence. It’s called “end-to-end,” and some disturbing new developments are putting it in jeopardy. The end-to-end principle asserts that information pushed into one end of the Internet should come out the other without modification: the Net should act like a big, fat, dumb, digital pipe.

End-to-end operates on many levels. When you try to download a news Web page, for example, the two ends might be CNN’s server and your browser. End-to-end dictates that the Internet shouldn’t modify CNN’s data packets as they move through the network. It thus guarantees that the page you receive is the same one CNN sent. Who could argue with that?

Many people, it turns out. End-to-end pushes a lot of power to the endpoints, but it also saddles them with some important duties. One such responsibility is security. If some hacker sends you an “attack packet,” it’s the job of the network to deliver that packet, no questions asked. Too bad if you haven’t installed the security patch. That sounds harsh, but it is preferable for users to have this kind of control than to cede it to network administrators.

For a good example of a network that’s not end-to-end, think of today’s cell-phone networks. When I call my friend Jesse’s cell phone, I call a phone number that’s out in San Francisco. But the network knows that Jesse is actually in Boston: the call gets routed out to California then back to Boston, and Jesse’s phone rings. All of this involves a tremendous amount of work on the part of the network-too much work for end-to-end. When I talk, the network takes my voice, compresses it, turns it into packets, and sends those packets down a low-bandwidth digital wireless network to Jesse’s phone. The quality of what he hears is determined by the network, not by our phones.

If the cell-phone network were end-to-end, my phone would use a registration server to find where Jesse’s phone is located. It would then open up a channel to his phone, negotiate with his phone to find a mutually acceptable voice compression scheme, and the two phones would start exchanging digital packets. Suddenly the network is dumb and the cell phones are smart.

So what’s the advantage of end-to-end? Innovation. With an end-to-end cell-phone system, Jesse and I could upgrade to a better voice compression system just by buying new phones: nothing else in the network would have to be modified. We could also add three-way or four-way or even five-way calling, just by sending out more packets. You can’t do either of these with today’s cell-phone networks.

Of course, if Jesse and I have end-to-end phones, we’re not limited to using cell-phone networks. We could just as easily use the Internet through wireless Net access at a university or a Starbucks. And that’s the real threat of end-to-end: by putting the intelligence in the endpoints, end-to-end turns the cell-phone network-or any other network-into a commodity.

0 comments about this story. Start the discussion »

Tagged: Web

Reprints and Permissions | Send feedback to the editor

From the Archives


Introducing MIT Technology Review Insider.

Already a Magazine subscriber?

You're automatically an Insider. It's easy to activate or upgrade your account.

Activate Your Account

Become an Insider

It's the new way to subscribe. Get even more of the tech news, research, and discoveries you crave.

Sign Up

Learn More

Find out why MIT Technology Review Insider is for you and explore your options.

Show Me