'

SC07 Day 3: Programming bits and atoms

A computer is a tool and a program is a thing, said Neil Gershenfeld, director of the Center for Bits and Atoms at MIT during the SC07 keynote on Tuesday. The audience was a little skeptical. "If programs are things," pointed out one attendee, "then bugs are consequential." Do we need to fear being outwitted, outsmarted, and outplayed by self-fabricated machines we helped create?

Cardboard creation

The SC07 keynote address was given Tuesday morning by Neil Gershenfeld, Director of the Center for Bits and Atoms at MIT. Neil challenged his audience to reconsider "obviously true" statements like "binary information is represented with two states". In light of current and future technological trends what if we relax these statements? He listed several of these tautologies and set out to reword them to be more correct in today's and tomorrow's world:

  • Computers come in cases -> computers come in rolls, buckets
  • Compilers optimize programs -> optimizations program compilers
  • Bits are zero or one -> bits are between zero and one
  • Internetworking -> interdevice interworking
  • Programs can describe things -> programs can be things

One of Gershenfeld's interests is cheap digital fabrication. The cardboard man in the picture was fabricated by his daughter and actually inspired the PhD thesis of one of his students. The killer app for digital fabrication, he says, is personal fabrication - things you can't buy at Walmart. And what if, instead of sending energy, computation, etc. around the world, we sent the means to create it? He shared his experiences with doing just that in isolated and developing countries.

According to Gershenfeld, people want to measure and modify the world, not just visualize it. As regular objects become computerized and interconnected at a smaller and smaller scale, we're approaching the nano-scale of biological systems. We're "in the moment", he says, on the cusp of a fabrication revolution.

A computer is a tool and a program is a thing, says Gershenfeld. But during the Q&A, an audience member pointed out, "If programs are things then bugs are consequential." True, responded Gershenfeld, but we need not fear digital competition. "It's nothing new," he said. "Biological systems have been competing successfully for millions of years." We may be forward engineering how biology works, he says, but it's already working for us.