Rob Faludi

The networking visionary on introducing your toaster to your smoke alarm, bonding with plants and bringing the outside in.

Rob Faludi is a specialist in physical computing and networked objects. As a researcher in New York University's psychology department and Center for Neural Science, he has investigated the connections between visual perception, motor action and the physical environment.

You had your own company for 10 years in San Francisco. Why did you move back to New York and go into teaching? Couldn't get a decent slice of pizza. And I really missed winter. Mostly, though, I came back to New York to better understand people. My consulting company had been doing some cutting-edge work. In team meetings, we'd get embroiled in lengthy debates about how [a Web] interaction should proceed.

When we finally put our creations in front of real users, they'd frequently breeze through the parts that we thought would be hard and then screech to a halt, completely bewildered by some choice that everyone on the development team had assumed would be obvious. I wanted to know what was going on. The answers would span science and design, so I returned to school for two master's degrees, one in cognitive psychology and one from NYU's Interactive Telecommunications Program. When it was all done, ITP asked me to do some teaching. That's turned out to be a terrific way to keep on learning.

To continue reading, register here and become an Insider. You'll get free access to premium content from CIO, Computerworld, CSO, InfoWorld, and Network World. See more Insider content or sign in.

Rob Faludi is a specialist in physical computing and networked objects. As a researcher in New York University's psychology department and Center for Neural Science, he has investigated the connections between visual perception, motor action and the physical environment.

You had your own company for 10 years in San Francisco. Why did you move back to New York and go into teaching? Couldn't get a decent slice of pizza. And I really missed winter. Mostly, though, I came back to New York to better understand people. My consulting company had been doing some cutting-edge work. In team meetings, we'd get embroiled in lengthy debates about how [a Web] interaction should proceed.

When we finally put our creations in front of real users, they'd frequently breeze through the parts that we thought would be hard and then screech to a halt, completely bewildered by some choice that everyone on the development team had assumed would be obvious. I wanted to know what was going on. The answers would span science and design, so I returned to school for two master's degrees, one in cognitive psychology and one from NYU's Interactive Telecommunications Program. When it was all done, ITP asked me to do some teaching. That's turned out to be a terrific way to keep on learning.

Can you define "sociable objects" and describe how you became interested in creating them?Sociable objects are simply devices that share information with each other and with people. The well-socialized object knows when and how to share. It doesn't bother you with questions it could answer on its own. It's happy to socialize with the other devices nearby, requesting information and lending a hand when it can.

When I push [my toaster] lever down, if there's a crumb stuck in the coils, it isn't long before my fragile pre-coffee state is shattered by the piercing siren of my smoke detector. It doesn't know about the toast, but really it should. If it were sociable, as soon as it detected particulate matter in the air, it would query the toaster to see if it had been activated. That would tell it that in all likelihood, it wasn't detecting an unattended, middle-of-the-night fire but instead a benign morning meal. The price of low-power radio networking and the just-minted funding for smart home energy networks makes this sociable smoke-detector scenario entirely within our reach. I want one.

Do your studies in psychology and neural science help you facilitate different ways of human-object communication? Oh, it's all about the people for me! I deeply respect engineering and computer science because they are fundamentally human endeavors. What I aim to do is take the magic they produce to the next level and find ways that new technologies can engage us and enrich our interactions.

Can you briefly touch on the concept of mesh network? Broadly speaking, a mesh network is a collection of devices that are all connected to each other both directly and indirectly. Any one device can act as both a node and a router for other nodes. Together, the devices create a robust communications structure, one that adapts fluidly when a new device enters the network or another one is removed or fails. There are over a hundred ad hoc routing protocols for these networks, but the basic idea is a kind of egalitarian structure. It's terrific for making robust and highly flexible local networks -- kind of like a party where everyone is making introductions to everyone else and happily chatting away, changing their layout as people enter and leave the room. On the other hand, it's an inappropriate structure for very large networks because the overhead of routing everything to everyone rapidly becomes unmanageable. So usually network architects use mesh at a local level and then route to a more hierarchical topology for internetworking. That way, the local fetes can stay fun without devolving into a shouting match.

For more enterprise computing news, visit Computerworld. Story copyright Computerworld, Inc.