A Sense of Place: Guest Blog by Nortel’s Visiting Fellow
You’ve heard me talk in previous posts about the value of the network shifting - becoming less about the ability to move bits and more about the ability to create a truly personalized, immersive communications experience. In this respect, “context” and “identity” are key areas of opportunity for innovation in the infrastructure space.
I’m pleased to share with you another guest blog from Nortel’s Visiting Fellow, Dr. Andrew Lippman, who offers his perspective on this topic. (Andy is one of the founding directors of the MIT Media Lab and is currently with Nortel as part of a year-long sabbatical.)
In the end, there are only two things that are critical in communications: identity and context. Both are complex, and neither is completely understood or well used. But they are the opportunities of the future.
Identity means who you are. While in reality there is only one of each of us, in practice we all present many different faces to the world. In some cases, this is a matter of persona - we act one way at the office and another way at the family barbeque, yet we remain the same person. Conversely, we maintain multiple identities as a means to, for example, insulate ourselves from different experiences on the Internet. John Yoakum, at Nortel, has given this a lot of thought and has insights into how one exposes, protects and manages these distinctions. He notes that identity is something you own and you assert to gain trust and foster interaction, and not something to be managed for you by others. The level of trust necessary for each situation varies significantly. The overriding objective is to establish just the right level of trust while preserving privacy and avoiding identity theft exposure.
Context is likewise a small word that covers a lot of territory. Some of it is based on pure fact - geographical coordinates are just numbers but where you are is a construct predicated on inference, assumptions and the intersection of more than one database. A particular latitude, longitude and altitude co-ordinate doesn’t reveal whether you are indoors or at the movies. Hal Abelson just ran a class at MIT on programming open phones that sparkled with the exuberance of youth. One student made his phone vibrate in the theater and ring loudly elsewhere (i.e., the phone changed the volume because it was aware of its location). Another student buzzed a reminder to get milk as he passed a food store.
Ultimately, we will make real-world mashups that tell you when to head for a market checkout line so you can catch the bus that is now three blocks away and will get you to the train station on time (something I talked about in my last guest entry).
But wait, there's more. Suppose we assume that even identity is a mashup, something that is computed on the fly as the product of a negotiation. Why flood the world with "dumb keys" and "smart locks?" What happens when they become equals? Here's a couple of examples …
Last year, I took my daughter out to dinner and we got "carded" while ordering wine. This may be a distinctly American problem, since we worry so much about alcohol, yet it makes no sense. "Carding" means asking for ID to prove you are of legal drinking age (i.e., 21). Fifty years ago, age was just about the only thing that was known about you, and it was reliably printed on your driver's license. We used it as a proxy for being responsible enough to drink, or to vote, or to see an adult movie. We can do better now. For voting, there are already communities that use a different age for local elections than they do for national ones. Clearly there is a difference between having wine in town with the family and rolling out of a jeep with six drunken friends after a college football game. In fact, perhaps my daughter's rights could be authorized by me, even though I might be 600 miles away. Distance is no longer a problem; if we can validate a toll pass at 30 kilometers per hour, I could certainly grant her real-time permissions as needed.
At the MIT Media Lab, we built a prototype ID that negotiated with a door to decide whether you should be allowed in - a smarter key. When you approached the door, it asked you via your cell phone who you were, and you could respond with either a name or a reason. You might be a student, or you might have a temporary authorization to enter as a surrogate for me, or your validation could be that you are attending a meeting, in which case, the door could tell you where the conference room is. The point is that we replaced a unitary response with a negotiation. Instead of telling everything to every door that asks, your phone asked the door what it really needed to know.
Where does it go from here, and how do we get there?
My guess is that we need to think "horizontally" and build open protocols versus continuing to think in vertical silos that are each a fully packaged application. This trend is well under way but it might benefit from a little formalization. Think of it as a general parts catalogue for the creation of services and negotiations. There are inroads being made in web mashups, social networks, and in sensing both the environment and the activities of people within it. These are a good start but are largely ad hoc.
A second trend is that the network to which we connect is opening up more and becoming less the focus of a service. As Chris Hobbs, of Nortel, has noted, increasingly we will subscribe to services directly rather than the network that delivers them. The Amazon Kindle and the SPOT emergency locator are the poster children for this enlightened view.
This is Hyperconnectivity in action - it heralds an era where we move past products and into architectures. Instead of customers, we will have partners who design the products with us and with our tools. To get there, we need research, reliability, and ways to scale on demand. This is as true for an insurance company as it is for a communications enabler, and Nortel recognizes that, both in the ways that it deals with its partners and customers, and in the structure of the company itself. That's what makes for innovation.
Older: 