Site icon IT World Canada

The Internet as an application

Sometimes significant things happen with little fanfare. This year, for the first time in history, the dollar value of U.S. carrier routers deployed in networks not part of the Internet exceeded the dollar value of those installed as part of the Internet. This doesn’t mean the Internet is going away (far from it), but it does mean that IP networking and the Internet are diverging in important ways.

The cause of the change is the thing pundits have been touting for five years or more: convergence. Carriers are deploying IP infrastructure to carry voice, frame relay, ATM and other services. The old public switched telephone network (PSTN) is being transformed, just like it was supposed to be. But it’s not being transformed into the Internet.

From the first, the Internet was an application that ran on the PSTN, using PSTN access resources, leased trunks and so on. If you wanted Internet service, you had to get access, and a phone company person showed up to install your phone, or give you DSL, or fix your lines after a storm. This service was part of what some call the bloated PSTN business model. All of it costs money. Now the same big carriers that brought us the PSTN are building a new public network, a network based on IP. The Internet is an application of that new network, too.

Every broadband consumer will have the application we call “the Internet,” and there will still be those who dial into it. Corporations will connect to their customers through it, send e-mails, all the things that are done now. So what’s the difference?

Payments and profits. The new and critical content applications such as IPTV will sort of work on the “Internet application,” but they’ll work better on specialized, partitioned, IP infrastructure deployed by the common carriers and paid for incrementally by users. VoIP will work on the Internet if you don’t mind problems with QoS and security. And the same goes for VPNs, application networks, grid computing, storage networking, all that nice futuristic stuff: They’ll work better for users who pay for partitioned IP handling.

Is that bad? It depends on whether you think having this new stuff is better than not having it, because without a mechanism for generating profits from deploying these new services, there won’t be any new services deployed.

This shift in IP balance of power has other effects, too. The common carriers don’t much like the IETF, and the feeling’s been pretty much mutual. The big carriers are looking to the International Telecommunications Union (ITU), the body the carriers have supported for years. The ITU will deal with the business issues of IP infrastructure, and it will do things that close networks to exploitation, provide for security and surveillance where legally mandated, and create formal interconnection standards among carriers with real settlement.

The IETF will still work on the Internet and also on end-to-end standards and issues, but the organization has lost the big carrier buyers…probably forever. IP vendors that tout the Internet model will lose ground to those that support the carrier notion of infrastructure IP. Maybe that’s why Cisco Systems Inc. has been losing ground to Juniper Networks Inc.

There will be people who say this is a bad thing, that the Internet should be open and free. But shouldn’t it also be secure and accountable? Watch over the next couple of years, because you’ll see IP applications that have all of the virtues of the Internet and few or none of the vices. You’ll pay to get them, of course.

Yes, the Internet is changing. But it can be changing for the better. The PSTN was old, inefficient, slow to respond to change. But it worked, in incredibly bad conditions, through economic changes, wars and disasters. It could have worked better, and it will work better — using the same IP the Internet gave us. What we need to do is ensure that the new public network that runs the Internet provides us the business model of the PSTN, and the flexibility, innovation and excitement of the Internet. 047401

Nolle is president of CIMI Corp., a technology assessment firm in Voorhees, N.J. He can be reached at (856) 753-0004 or tnolle@cimicorp.com.

Exit mobile version