The Fab-ulous Tech in Our Future
11/11/2011 8:00 am EST
Few people on the planet have a better sense of what the next leg of technological innovation will bring than Neil Gershenfeld, writes Josh Wolfe of the Forbes/WolfeEmerging Tech Report.
Professor Neil Gershenfeld is the Director of MIT’s Center for Bits and Atoms. His unique laboratory is breaking down boundaries between the digital and physical worlds, from creating molecular quantum computers to virtuosic musical instruments.
Dr. Gershenfeld is also the originator of the growing global network of field fab labs that provide widespread access to prototype tools for personal fabrication, and directs the Fab Academy, the associated program for distributed research and education in the principles and practices of digital fabrication.
He is the author of numerous technical publications, patents, and books including Fab, When Things Start To Think, The Nature of Mathematical Modeling, and The Physics of Information Technology, and has been featured in media such as The New York Times and The Economist. He is a Fellow of the American Physical Society and has been named one of Scientific American’s 50 leaders in science and technology.
How do you describe the work you do at MIT?
I direct the Center for Bits and Atoms. We work at the boundary of physical science and computer science, where many of the most compelling technical questions cross over between hardware and software.
Within that realm, one of the broad things we’re working on is digital fabrication: ultimately, how to
make something like the Star Trek replicator.
In the opposite direction, we’re also revisiting how computing works, to better align it with physics—for performance scalability and usability.
At MIT, we also run a fabrication facility where we can make anything on any length scale, and we launched the fab lab network, which has grown up to about 100 smaller fabrication labs located around the world.
Where do you draw the boundary line between the physical and digital worlds?
For me, that line actually makes no sense. Fundamentally, in our model of how the universe works, we can’t separate bits and atoms. I’ll give an example using the world of fabrication.
I believe we will see a transition take place in four stages: the first stage is computers (bits) controlling machines (atoms)—that began all the way back in 1950, when MIT connected the first computer to a milling machine to make aircraft parts.
The second stage is machines that can make other machines or parts of themselves. We’re in the middle of that phase right now.
Stage three is the next big thing, coming soon, when we’ll be transitioning from analog to digital materials. And in the fourth and final stage, we’ll lose machines entirely and move from externally programmable to internally programmable materials. The heart of that whole transition begins with moving from analog to digital materials.
How do you explain the concept of digital materials?
In digital fabrication, a program wouldn’t just describe the thing being made—it would actually become the thing.
By embedding codes into materials themselves, we can correct errors during construction. That may sound trivial, but it has the same importance as the difference between an analog and digital telephone or computer.
For example, imagine playing with Legos. Legos operate on the same principles we seek in digital fabrication: Lego bricks have a coordinate system, so you don’t need a ruler when you assemble them. Legos correct errors when snapped together; they are more accurate than you are.
They are made out of functional materials, with a range of types of bricks, and the work is reversible. When you’re done, you don’t put it in the trash—you disassemble it.
Would it be fair to compare the concept of digital materials to how nature assembles structures?
Without a doubt, it’s the same in spirit, but the domain is different. The same principles apply to molecular biology, for example, in the construction of proteins by ribosomes.
What’s different is that we’re going to do it on length scales and with materials that aren’t available in organic chemistry. We’re developing applications in electronics and in aerostructures, and many areas that don’t have analogs in nature.
Have you worked with these types of digital materials in your own lab?
We’re working to develop both materials and applications. One example is in high-performance structures like composite airplanes, which typically require ribs, skins, spars, and stiffeners.
But when all of these parts come together, what’s essentially formed is a single volume material—not a coarse bunch of separate parts. By using digital fabrication concepts we can make a similar structure from lighter, stronger, higher-performance materials.
Think about your computer monitor screen. The image is decomposed into pixels that can do anything, and that is what makes it powerful. This is a very close analogy with making materials where you can control each dot, and thereby control the overall material property.
There’s a lot of excitement right now around 3D printing. Is this an important technology?
3D printing is more revolutionary for people who don’t use it than for people who use it regularly. 3D printing is just one of several fabricating tools that are used both in our MIT facility and in smaller fab labs in the field.
Within that range of tools, 3D printing is used perhaps 20% of the time. It’s a slow process, and the material properties aren’t great. And while 3D printing is additive versus subtractive, the materials are still analog: instead of cutting material, you squirt material.
The much bigger transition is from analog to digital in materials. That’s the real revolution coming.
Onto the fab labs—when someone walks into one of these facilities, what do they see?
A carefully-tuned and slowly-evolving subset of the best-used equipment that you’d find in MIT’s own fabrication facility. They’d see a laser cutter to make 2D parts for press-fit assembly of 3D structures, a numerically-controlled milling machine to make parts for furniture, or boats, or whole houses.
Then there’s a precision, micron-resolution milling machine for molds and circuit boards, a 3D printer for things to be 3D printed. There’s also a sign cutter for cutting flexible materials for signs, but also printing masks or flexible circuits.
They’d also notice materials for molding and casting, and finally a whole host of design tools and videos to help a member use all of the equipment at their disposal.
What kind of projects have people made in fab labs?
A fab lab project from Afghanistan was recently in the news—they used a fab lab to create a city-wide Internet, with high-gain antennas and linking radios.
Another fab lab in Barcelona won a People’s Choice Award in the European solar decathlon with a very ambitious project—they developed a complete solar house, including the furniture and power-control systems, and made it all in a fab lab.
The fab labs all share capabilities and they keep evolving. People work together as a network to share projects and property processes, and support new business platforms and educational platforms for the network.
Right now, we’re racing to spin off a foundation to keep up with the growth of these labs—it’s been a kind of viral spread.
You’ve been at the forefront of this movement for a while, as described in your book, Fab. What was the inspiration for that book?
In Fab, I tell the story of how I wanted to go to a vocational school to weld and fix cars, but I was told, “No, you’re smart; you have to sit in a room.”
It seemed punitive—nobody could explain to me why I wasn’t allowed to weld. I worked at Bell Labs and when I tried to work in a workshop, they said “No, you’re smart, you have to tell somebody else what else to do.”
What I see happening today, through 3D machining and microcontroller programming, is the recovery of the type of self-expression that was available in the Renaissance, like painting or writing a sonnet.
What do you think people can look forward to in the area of DIY fabricating over the next decade?
Exponentials always have this rhythm where nothing seems to happen, and then everything happens. The first-generation DIY machines were not good machines, but I’d predict that over the next year, high-performance DIY machines that make their own parts will come onto the market.
At the rate we’re going, high-performance assemblers will displace 3D printing within five years. That’s a reasonable time scale for the distribution of DIY tools into places like the town library, giving each community access to these capabilities.
Twenty years from now, in this deeper research roadmap, we may go from externally programmable matter to internally programmable matter, and that’s really the Star Trek replicator phase.
And that’s when everything changes?
Well, no…the lesson we learned from the Internet is that everything changes not at the end, but back in the days when the technology is being invented.
Most of the real changes happen well before the technology reaches its final form. I didn’t understand that at first—I kept waiting for the attention to calm down so we could finish the research.
Even though the fab labs are imperfect, and far from a final form, you don’t have to wait for the research to finish to figure out how to build the business platform and how to build the education platforms, and how to live, work, and play when anybody can make anything.