629: How to Build an Android by David Dufty
629.892: Dufty, David F. How to Build an Android: The True Story of Philip K. Dick’s Robotic Resurrection. New York: Henry Holt, 2012. 259 pp. ISBN 978-0-8050-9551-7.
- 600: Technology (Applied Sciences)
- 620: Engineering and allied operations
- 629: Other branches of engineering
- 629.8: Automatic control engineering
- 629.89: Computer control
- 629.892: Robots
Of the many things to worry about losing, the android head of the late science-fiction author Philip K. Dick is perhaps the strangest (although, if you’ve read Ted L. Nancy’s Letters From A Nut, this wouldn’t even make it to the top ten). In 2005, a team of engineers was brought together at the University of Memphis to take the next step in human-robot interaction: to build a fully interactive and fully expressive head that could comprehend speech and synthesize its own responses.
They did. It worked. And then they lost it.
David F. Dufty’s How to Build an Android is the story of David Hanson, Art Graesser, and Andrew Olney and their quest to bridge the gap in human-machine communications. Up until now, robotic machines could only generate a finite number of responses to either typed or spoken input. Before the android was built, the team was working on a program called AutoTutor that could teach a user about a vast number of subjects (they started with physics). One of the best parts of the early work was a “proof of concept” device that integrated the tutor program into the sensory equipment and likeness of a Billy the Singing Bass toy.
When Graesser met Hanson at a conference and heard his ideas on robotics and philosophy, he was convinced that a machine could be built that would come uncannily close to feeling human. The Memphis team sought to integrate the works, interviews, and known publications about Philip K. Dick into an integrated database that could generate any number of answers from the stored “knowledge.” They then built a remarkable likeness of the author using Frubber, machined components, and tiny sensor motors.
Upon completion, the android head could hold conversations with passers-by (with a few amusing consequences) and was stable enough for a tour of the robotic conference circuit, including NextFest and ComicCon. But, when Hanson was transporting the head from Dallas to Las Vegas on his way to a presentation at Google Headquarters, he left it in the overhead compartment and never saw it again. Such a strange end to a wonderful tale.
Dufty’s tale is both whimsical and technical in a way that only an android creation tale can be. There were times when this probably would have made a better magazine article than a full-length book, but I enjoyed it nonetheless. The section on robotic philosophy, the Uncanny Valley, and the future of technology were slightly more interesting than the technical aspects of building the device, so if you’re looking for a lot of science jargon, you’ll have to go elsewhere. Oh–and if you happen to come across a strange, bearded, robotic head at a flea market or an auction, David Hanson would greatly appreciate its return.