Tech talk: Convergence of the bits and the atoms

The next 10 years will see IT's data world and the physical world moving closer together.

The next 10 years will see IT's data world and the physical world moving closer together.

If the PC changed the IT landscape in the 1980s, and Internet did the same thing in the 1990s, what can we look forward to for the next 10 years? My vote would be for the convergence of bits and atoms - bringing IT and the physical world closer.

Many IT systems contain data about tangible things, such as vehicles, packages, people and places. But nine times out of 10, the information originated from a keyboard and will end up on paper or on screen.

Of course, we have had computers in embedded systems and process control for a long time, but they have tended to be seen as distinct from mainstream IT.

Over the next decade, I believe this is going to change, enabled by developments in sensor technologies and intelligent wireless communications that will make physical interfaces affordable for large-scale deployment.

The key to sensor developments is the ability to use techniques developed for semiconductor manufacture to make tiny devices that will respond to almost any chosen physical phenomenon, be it mechanical, electrical, chemical or biological.

Because of similarities in their manufacture, they can be combined with computing power and IP communications capability in the same device, and will potentially be subject to the cost reduction curves of Moore's law. Similarly, the use of cheap, integrated computing power has enabled devices to share a radio communications channel, and through developments such as GPS, to be aware of their location.

In terms of applications, many of the basic ingredients are already there: mobile phones, satellite navigation systems in cars; remote-read identity chips in dogs and horses; road congestion monitoring networks; and soon, perhaps, widespread availability of Bluetooth and 3G mobile.

New applications will come from using these technologies together in new and exciting ways, and from getting the cost down to a level where they can be deployed on a massive scale.

Radio frequency identity tags have been used for many years, but they are rapidly becoming smaller, cheaper and potentially more sophisticated. In their most basic form they contain simple radio transponders which can be interrogated as they pass a tracking point and will respond with their identity - a form of electronic barcode. More sophisticated tags have some processing power, and can store information, such as where and when they were previously interrogated.

In principle, they can incorporate sensors and record information about their environment - for instance, whether they have been exposed to shocks or extremes of temperature. Equally, they could be linked to GPS modules to track their location, and cellular radio to communicate over greater distances.

Developments at this end of the market are providing increasing functionality in pocket-size modules in the £100 price range. They can be incorporated in containers and vehicles, or even be built into clothing.

At the bottom end of the scale, developments are pushing down the price of the most basic devices to the point where they are becoming a serious contender to replace retail barcodes.

Researchers at MIT now see "penny tags" as a realistic proposition within the next few years, making them affordable for many more products. One day, perhaps, the entire contents of your supermarket trolley will be scanned and paid for without even stopping at the checkout, but the technology still has a long way to go before that will be practicable or economic.

Alan Shepherd is director of technology exploitation at Consignia

Read more on Data centre hardware