Creating an Internet of Things
The Internet connects computers to one another. What the Auto-ID Center aims to do, in effect, is develop a network that connects computers to objects - boxes of laundry detergent, pairs of jeans, airplane engines. We are not creating just the hardware (RIFD tags and readers) or just the software to run the network. With the help of our sponsors, we are developing everything that is needed to create an "Internet of things," including affordable hardware, network software and protocols, and languages for describing objects in ways computers can understand. The system will be based on existing standards. It will be open to anyone, and it will be entirely free - just like the Internet.

Why is this necessary?
Well, it's not, any more than the creating the Internet was "necessary." But just as the Internet has made companies vastly more efficient by enabling them to share data, we believe that the ability to track individual items as they move from factories to store shelves to recycling facilities will make companies vastly more efficient. Companies, in fact, have long realized that capturing data about goods automatically and accurately would be a great boon to their business. Billions of dollars have been invested worldwide in bar code technology.

Many companies are investing in RFID systems today to get the advantages they offer. These investments are usually made in closed-loop systems - that is, when a company is tracking goods that never leave its own control. That way, they don't have to worry about their tags being read by other companies. But most companies don't have closed-loop systems. Since all existing RFID technology is proprietary (meaning there are no standards), if Company A tags a widget and sends it to Company B, Company B can't identify the product unless it has invested in the same technology from the same vendor as Company A.

Another problem is cost.
RFID readers typically cost $1,000 or more. Companies would need thousands of readers to cover all their factories, warehouses and stores. Readers typically operate at one radio frequency. If tags from three different manufacturers used three different frequencies, a store might have to have three readers in some locations, increasing the cost further.

RFID tags are also fairly expensive - 50 cents or more - which makes them impractical for identifying millions of items that cost only a few dollars.

Creating one, open global network for RFID means that companies can invest in systems and have confidence that the tags they put on their products can be read by retailers and other business partners. It means companies can share information about products and develop systems that automate many of today's labor-intensive operations. For example, when shipments arrive at a store from a factory, the computer systems of both companies can be updated automatically and an invoice can automatically be sent. It also means that manufacturers of RFID equipment can make equipment in vast quantities, since it will work with anyone's system, which will help bring down the price of both tags and readers. (There are other issues related to cost that we'll tackle later.)