The internet of the future will be an invisible network anyone can connect to with any device in any part of the world, one of its founders has said.

Dr Leonard Kleinrock, who as a graduate student at MIT helped developed the technology that made the internet possible, says he foresees a society not far from science fiction movies of the 1960s.

Kleinrock says that, just as in these movies, in the future people will have conversations with their computers, we will find embedded technology in our floors, in our desks, in our lamps, in our clothes, in our hotel rooms and so on. We will communicate with the room via sensors, actuators, processors, microphones and displays. In an intelligent room, you will be able to ask for a specific book, its content and where to find it. Information from the web will be presented via natural language speech, video, eyeglass displays and holograms.

Kleinrock believes future development will move closer to his original vision of the internet: it will be everywhere, always accessible, always on, anyone will be able to plug in from any location with any device at any time, and it will be invisible. Development will focus on high-speed communication between machines, devices, people and intelligent agents, regardless of what or where they are.

"Most of my early predictions are reality today, but what is left to be solved is that people are not able to plug in from any location with any device at any time, nor is it invisible," Kleinrock says of today's internet

The difficulty in moving around and connecting at different locations is a legacy of the early days. The internet?s TCP/IP protocol (the suite of communications protocols used to connect hosts on the internet) assumed the user, their device, network address and location were all fixed and tightly coupled ? an assumption that does not work in the modern mobile society.

Kleinrock says devices and technology will work together in the future without us noticing it. Required bandwidth and the way a person can be connected changes depending on the location. Today, a person has to adapt to local conditions and resources. In the future, the devices and the technology will adapt automatically to the user.

"The access technology for data services will automatically adapt depending on where the user happens to be. In a building, they may find enterprise wireless or wired networks available, but if they walk outside, they may automatically be connected to a 3G network and when they approach a wireless hotspot then they will likely be switched over to a WiFi access technology."

He says the mobile internet industry is well on its way towards mass personalization and true mobility, but interoperability is not working completely yet. Kleinrock says there are a couple of obstacles the mobile industry needs to work on.

"At the moment there is a war going on among operators. Larger ones buy out smaller carriers. Price is decisive to gain and keep customers, but what is much more important are services, content and footprint," Kleinrock says. "People are not inherently loyal towards operators. The operator which will provide the best and most useful service offering and content along with the broadest footprint will win the war for customers.""

Kleinrock says the manufacturing industry also needs to create small, battery-efficient devices with all the functions a consumer needs.

"Today we have too many devices: mp3 players, portable radios/CD players, mobile phones with cameras, PDAs, notebooks, laptop computers, smart credit card devices, intelligent wristwatches and so on. Many mobile phones are marketed as if they have all functionalities but often lack in performance in one or more areas. It results in people buying different devices for different functions. Most technical devices can?t communicate with each other and often they need different software. At the same time, we need to minimize the number of gadgets an individual carries around ? how many functions does a person really need, or want?"

As a graduate student at MIT, Kleinrock developed the basic fundamentals of data networking and uncovered the underlying principles of the networking technology that now drives the packet-switched internet. Packet-switching is a technology that breaks large messages into small fixed-size packets, each of which makes its own way through the network to be reassembled with the rest of that message at the destination; this provides one of the ways to gain efficiency in data networks.

"The research I did lay dormant for some years until the Advanced Research Projects Agency (ARPA), which had been funding computer research at universities around the United States until they recognized that these researchers would benefit by sharing computer resources if they were connected in a computer network, instead of granting every computer resource to each researcher."

With this in mind, Kleinrock's principles of a shared and open network could be born in a widespread implementation that would become the internet. The basic philosophy of the internet was that the community could share ideas and results from research among all community members. And it was not, as widely believed, a network created by the US Defense Department that could survive a nuclear attack.

"Scientists needed a large network that could handle data from a large number of users," Kleinrock says. "My early research described how we could use network capacity in the best way possible. The answer was to introduce distributed control and resource sharing. With distributed control, no portion of the network controls the rest of the network, but rather all network nodes participate in that control. This was a design that would allow the network to scale up while at the same time provide a resilient and robust network.  Resource sharing allows network resources to be shared among many user streams in a dynamic on-demand fashion, providing high efficiency in the face of bursts of traffic. One manifestation of resource sharing is packet switching, which happens to be the technology used by the internet.

On October 29, 1969, the Internet came to life when a message was sent from Kleinrock?s computer at UCLA (University of California).

"The purpose of that first message was to login to Stanford Research Institute (SRI) in Northern California from UCLA. We succeeded in transmitting the 'l' and the 'o' and then the SRI system crashed! Hence the first message on the Internet was 'Lo'."

The development of the internet since 1969 has been dramatic. One of the most important applications of the internet came in its early years when, in 1972, e-mail was introduced and immediately became the dominant application. It was then that Kleinrock first recognized that the power of the internet was human communication and not computer-to-computer communication.

"Indeed, it was e-mail that helped drive the use of the internet into the dot-com world in the late 1980s and early 1990s as the business segment saw the power of e-mail as it was used by the research scientists within their corporations," Kleinrock says. A second important event in advancing the penetration of the internet was the funding of the gigabit testbeds by the American government in the early 1990s.

"Al Gore, who was an eager proponent of the Internet, got the Bush (senior) administration to sign the High Performance and Communication Act of 1991," he says. " This Act allocated money for a high-speed network to be developed jointly by industry, academia and government. When, in 1993, the World Wide Web was introduced, it presented a very simple interface to the internet, and attracted many millions of internet users across the world, bringing us to the nearly one billion internet users today."

Karin Hanson