There is too much coverage in the press about the wonders of 3D printing and it’s a distraction from the real revolution, argues Neil Gershenfeld, the head of MIT’s Centre for Bits and Atoms. “The coverage of 3D printing is a bit like the coverage of microwave ovens in the 50s. Microwaves are useful for some things, but they didn’t replace the rest of your kitchen,” he said, speaking at the Royal Academy of Engineering’s Grand Challenges summit. “The kitchen is more than a microwave oven. The future is turning data into things, but it’s not additive or subtractive.”
He explained how the first computer was connected to a milling machine at MIT in 1952. “What has grown forward is a digital revolution in making things. It’s cutting, grinding, lasers, plasmas, jets of water, wires, knives, bending pins, weaving, moulding, extruding, fusing and bonding.”
For Gershenfeld, the real revolution of fabrication is much more fundamental: it’s bringing programmability to the physical world. He invited the audience to compare the performance of a child assembling Lego and a 3D printer. The child’s assembly of Lego will be more accurate than the child’s motor skills would allow — that’s because the pieces are designed to snap together in alignment. Meanwhile, the 3D printing process accumulates errors, perhaps due to imperfect adhesion in the bottom layers. Lego is also available in different materials, while 3D printers have limited ability to use dissimilar materials. Finally, a Lego construction can be easily disassembled.
For Gershenfeld, Lego represents the digitisation of material, while 3D printing is still an analogue process that draws upon digital files. The bricks enforce constraint and, as a result, accuracy. He explained how digital fabrication echoes what happened when analogue communication and computation became digital.
Claude Shannon at Bell Labs showed in 1938 that by converting a phone call into a code of zeros and ones a message could be sent reliably no matter how noisy the system, thanks to error correction. By converting the signal into code, it increased accuracy enormously.
Shannon’s research had been motivated from working with a giant mechanical analogue computer which used rotating wheels and discs and became more inaccurate the longer it ran. John von Neumann and colleagues showed that they could digitise data in computing as well.
Gershenfeld explained how just as we have digitised communication and computing, we must digitise fabrication by learning how to program the growth of materials so that the “code you put into them doesn’t just describe them but becomes the materials themselves”. “Digitisation of fabrication is where you don’t just digitise design, but the materials and the process. The computer program doesn’t just describe the thing but becomes the thing. That’s not a metaphor. It’s literal. It’s exactly the story with Von Neumann and Shannon.”
He then talked about the development of Fablabs around the world — centres with various digital fabrication tools allowing people to make anything they wanted. When he created the first one at the Centre for Bits and Atoms, he was swamped with students he didn’t expect to see who wanted to make things such as web browsers for parrots, an alarm clock you had to wrestle with to turn it off and a dress that defends your personal space. “Students weren’t making what you could buy in stores, but what you can’t buy in stores.”
If MIT is the mainframe computer, then the Centre for Bits and Atoms is the equivalent of the Programmed Data Processor (PDP) for fabrication.
Now more and more Fab Labs are emerging, akin to developing the internet for digital fabrication.
He said: “Online mass classes are just terminals plugged into the mainframe. The Fab Lab network is creating an academy where students have peers in workgroups and tools that are linked with online content and video. You can download the campus and design something in any facility and make it in any other. This blows apart the boundaries of lead institutions.”