Do We Need Digital Turtles?

By Kevin Ashton

The RFID industry is losing its memory about the reasons tags should carry only IDs.

One of the current debates in RFID is about memory: Should RFID tags carry data or just identity numbers? The idea that tags should carry data has strong proponents in pharmaceutical and item-level tagging and e-passports.

This is nothing new. It was one of the controversies during the development of the Electronic Product Code. That discussion was especially interesting because it happened at the same time the Internet was taking off. RFID emerged before the Internet was widely available, and there was no easy way to share data. RFID tags were promoted as portable databases, carrying information alongside products. This was great news for chip makers, who charge more money for more memory.

But supply-chain RFID must be cheap. More memory costs more. Even if the extra money for each tag is nominal, it becomes prohibitive when multiplied by billions of tags. The Internet raised a tantalizing alternative: Put only an ID on the tag, and use it as an Internet address to point to information on the network. This would reduce tag costs and allow access to more information than could be held on a tag, and it would be easier to secure and be faster, more scalable and less likely to get outdated. All this, and it would deliver the same applications as tags with added memory.

But the appeal wasn't obvious to everyone. I went around in circles explaining why there was no need for tags to be like digital turtles, carrying everything around on their backs.

Opponents asked two questions. The first was, "What about places that don't have the Internet?" A site that couldn't access the Internet would also have no telephone, no cell phones and no way to access satellites—an underground bunker, sealed from communication. But if such a place existed, why would it need RFID?

The second was, "What if the Internet goes down?" This was a little more reasonable. Networks have outages, but if backup systems and hard drives were used sensibly, the impact of any outage could be minimized. Electricity has outages, too, but no one uses that as a reason not to use it.

In the end, ID tags won. Or so I thought. As new people enter the RFID world and old hands move on, memories are being lost—or rather they are in danger of being added to tags, as the debate starts again. What's missing is an understanding of how many billions of dollars more expensive it is; how much time it wastes reading and writing; and how hard it is to secure sensitive information stored in a low-cost RFID tag.

Meanwhile the reasons justifying the use of ID tags have become stronger. The Internet is now ubiquitous and even more reliable. It is hard to find a computer that is not networked and harder to imagine what it is for.

But for some, digital turtles remain attractive—these people are still unsure about the Internet. They may as well ask, as one Auto-ID Center sponsor once did, "Do you think the Internet will catch on?" Yes, I think it probably will. That's why I believe—more than ever—that ID tags are all we need.

Kevin Ashton was cofounder and executive director of the Auto-ID Center. He is the author of a soon-to-be-published book about RFID.