The emerging "Internet of Things" - a UW experiment in progress

There's an exciting project going to start next year at the University of Washington's Computer Science Building. You can read about it in a press release they put out recently. Here's a wee bit of it:

If you need information, the Internet offers a wealth of resources. But if you're hunting down a person or a thing, a computer's not much help. That may soon change. Electronic tags promise to create what some call the "Internet of things," in which objects and people are connected through a virtual network.

To see what this future world would be like, a pilot project involving dozens of volunteers in the University of Washington's computer science building provides the next step in social networking, wirelessly monitoring people and things in a closed environment.

Beginning in March, volunteer students, engineers and staff will wear electronic tags on their clothing and belongings to sense their location every five seconds throughout much of the six-story building. The information will be saved to a database, published to Web pages and used in various custom tools. The project is one of the largest experiments looking at wireless tags in a social setting.

I first heard about the "Internet of things" concept from folks at the MIT Media Lab, in particular Nicholas Negroponte, who was focused on this before he got obsessed with the One Laptop Per Child concept. A hands-on practitioner there was Neil Gershenfeld, who wrote about "Internet Zero" back in 2004.

The Internet is appearing everywhere. Phones speak it, appliances process it, coffee shops and even coffee pots serve it. But we'll be doomed if your coffee pot demands the services of an IT department.

As remarkable as its growth has been, Internet implementations that were appropriate for mainframes are not for connecting everything everywhere.  Yet there are compelling reasons for a light bulb to have Internet access, ranging from construction economics to energy efficiency to architectural expression.

Accomplishing this with the cost and complexity expected for installing and maintaining a light bulb rather than a mainframe raises surprisingly fundamental questions about the nature of scalable system design. The success of the Internet rests on the invention of "internetworking" across unlike networks; the Internet zero (I0) project is extending this insight to enable "interdevice internetworking" of unlike devices.

I really like the name "Internet Zero", whose origin is in an impromptu jest during a conversation between one of the I0 people and a baffled Internet2 engineer, who couldn't figure out how you could afford to network all these kinds of mundane objects:

At an opening event one of the architects of the high-speed Internet 2 project kept coming back to ask how fast data could be sent through the building infrastructure. After being reminded that light bulbs don't need to be able to watch broadband movies, he was jokingly told that the emerging network of everyday devices was part of an Internet zero, not Internet 2. The name stuck. I0 is not a replacement for the current Internet (call that Internet 1); it is a set of principles for extending the Internet down to individual devices.

If a networking researcher had trouble understanding the value of slow networks, the building construction industry did not. Ever faster computer networks have paradoxically become less and less relevant to their needs. Unlike plumbing or power, both the network infrastructure and the workers required to plan, install, and operate it have become increasingly specialized. Countless "smart home" projects have sought (with limited success) to find killer applications for bringing this kind of connectivity into homes, neglecting the biggest one of them all: building the buildings themselves.

In the 1997 US Economic Census, the annual revenue from making computer and electronic products was $438B, compared to $858B for construction. Even if the hardware was free, the cost of installing intelligent infrastructure in all those buildings could add up to the size of today's entire IT industry. This presents a technological challenge that demands neither gigabit speeds nor gigabyte storage, but rather simpler configurations.

To me this is a classic example of the "less is more" concept in action. I worked on a project in the late 1980's for Andover Controls Corporation, where we chose Arcnet rather than Ethernet as the link layer protocol: Arcnet was significantly slower, but also significantly less expensive, and more than fast enough for the HVAC control systems ACC produced. I make no claim to responsibility for that decision, by the way; it was then and is now outside my areas of expertise. HVAC engineers are an eminently practical bunch, and if I learned anything from them, it was a healthy skepticism for the latest and greatest technologies.

The dicey economic situation prevailing in the "Bush decade" has limited the growth of the market for the Internet of Things, but technologies grow in a steady evolutionary fashion in spite of economic conditions. I expect the next upward trend in the global economy will see the widespread emergence of the Internet of Things in the commercial marketplace.