An Internet of Things? Yes! But Hold onto Your Hat
Friedemann Mattern is a professor at the Swiss Federal Institute of Technology and is Director of its Institute for Pervasive Computing. He is developing visions of a future in which all objects are networked.
In your new book, The Internet of Things, you describe a world in which nearly every object is intelli- gently networked with other objects. What makes you so sure that this vi- sion will become reality? Mattern: Very simply because it’s techni- cally doable! In the great technology trends — microelectronics, wireless com- munications, sensors, new materials — there’s no end in sight to their progress. The convergence of these technologies will almost automatically lead to “smart objects.” As these objects perceive their environment, process data and commu- nicate with other objects, the ‘informati- zation’ of our world will increase rapidly. This may look like a kind of invisible magic, but I wouldn’t call it intelligence, because objects lack the knowledge and interpretive faculty that people have — which isn’t likely to change in the next 20 years.
You make it sound as though all technical problems are solvable. What obstacles do you foresee? Mattern: With respect to pure technol- ogy, I am really optimistic — with one ex- ception: The energy problem. Progress on reducing power consumption and in- creasing battery life is not moving as fast as I’d like. For example, if we wanted to drop tiny autonomous sensors from air- planes to monitor the environment, we could hardly replace their batteries after- wards. Another open issue is what the in- frastructure will look like, for instance which wireless standards will work best. People will try to link sensors via ad-hoc networks like Bluetooth, for instance a lawn sprinkler with a sensor that meas- ures soil moisture. But to save water, a smart sprinkler would also need to obtain
the weather forecast from the Internet. For that, however, it would have to use a network, and that costs money.
And not everything that’s technically feasible would actually be useful… Mattern: That’s right! So far, no one has missed intelligent and networked objects. We can live without them. On the other hand, we’re getting used to calling up and starting our car with a keyless entry chip on a key ring. Our pursuit of safety, status, comfort and entertainment will cause many such applications to become ac- cepted practice.
When surfing the Internet, you make a conscious decision either to be net- worked or not. But in a world of in- telligent objects, this isn’t possible. Do you think users will accept that? Mattern: The “Internet of things” does in- deed blur the boundaries between ‘online’ and ‘offline.’ There are going to be areas where user acceptance will be essential. And this will be especially true for applica- tions that make it possible to locate and trace objects — and their owners. That’s great if you’ve lost your keys, but not so great if your spouse uses it to track you down while you’re having an affair, or if a government wants to snoop on its citizens.
In your opinion, will we be able to really rely on intelligent objects? Mattern: When things become smarter and more autonomous, they won’t al- ways behave the way we expect — that’s normal. But as networking becomes per- vasive, faults could have worldwide ef- fects. That’s why some people insist on being able to deactivate objects’ intelli- gence — just as you can switch off the ESP in some cars. But that’s easier said
than done, and in a pervasively net- worked world, where all things are inter- dependent, it may not always be possible or even advisable. That’s because net- working is what provides the real added value — just like the value of a human being is greater than the sum of its body cells. We must use fault-tolerant designs and appropriate caution in introducing intelligent systems to ensure that excep- tional situations remain manageable. Ground rule number one must be that our environment must always be able to function without help from intelligent objects.
The ultimate fate of the Internet of things will also depend on whether people can make money with it. Are there any business models yet? Mattern: In cooperation with the Univer- sity of St. Gallen, we’re working with sev- eral large firms to develop such models. In this connection, we’re interested not only in technical implementation, but also in security and privacy issues. “Pay per use” will certainly be a significant trend. In- stead of charging a flat rate, providers could charge usage-based fees, which we already know from our phone bills. For ex- ample: If a car can continuously report how many miles it travels at what speed, and where it’s parked overnight, the insur- ance company could compute charges for liability and collision coverage on an indi- vidual basis. Many customers would be able to save money that way. At present, however, a flat-rate mentality seems to prevail — not only on the Internet but also in the travel business, where package deals are all the rage. I’m very curious my- self about what kinds of business models w i l l d e v e l o p w i t h s m a r t o b j e c t s ! ■ I n t e r v i e w c o n d u c t e d b y B e r n d M ü l l e r .
Te l e m a t i c s
Traffic management centers, like this one in Berlin, receive information about congestion, major public events, construction sites and the public transportation system.
The Road to Telematic Travel
Telematics — the conver- gence of computers and telecommunications — is changing how we travel. Siemens is linking central- ized traffic management systems with terminals such as vehicle navigation systems and mobile phones, as well as an array of new services. The result is less stress and shorter travel times.
G ermany, summer 2006. Soccer fans will be out in force for the World Cup games — not only in the stadiums but also, unfortu- nately, on roads throughout the country. As a result, the Ruhr region, Europe’s largest conurbation, and greater Berlin will have to deal with an even heavier traffic flow than usual. The volume of traffic in these areas is already massive. Around six million people are on the road in the Ruhr region every day, five million of them coming in from sur- rounding areas.
To prevent massive traffic jams, especially during the World Cup, Siemens is developing and operating the Ruhrpilot traffic manage- ment solution in the Ruhr region — on behalf of ProjektRuhr GmbH — and the Traffic Man- agement Center (VMZ) in Berlin. “These traf- fic management centers are the meeting points of all traffic data, but that’s not all,” says Hans-Joachim Schade, head of business development for Intelligent Traffic Systems at Siemens Industrial Solutions and Services.
“Here, besides managing the flow of all pri- vate vehicles and public transportation sys- tems, we also offer passengers and drivers ex- tensive information to make their travel simpler, safer and more comfortable. Without state-of-the-art traffic management, a major sports event of these dimensions could no longer be held in a large urban area.”
Dr. Thomas Lackner, head of the Telemat- ics Business Competence Center at Siemens, is bullish on telematics. “The outlook for traf- fic telematics solutions is good,” he says. “We estimate that the global market volume for this technology in 2004 was around 27 billion euros, and that future growth will average six percent annually.” Growth opportunities for individual segments of this vast market will vary greatly between 2003 and 2007, from three percent for parking guidance systems and railroad automation to seven percent for passenger car navigation and toll manage- ment, and as much as 15 percent for truck toll and fleet management systems.
Pictures of the Future | Fall 2005