News your connection to The Boston Globe

(Illustration / Joel Holland)

How to make (almost) anything

MIT's 'Fab Labs' project aims to give ordinary people around the world the technology to design and make their own stuff. Is this the dawn of the age of 'personal fabrication'?

IF YOU ASK Neil Gershenfeld, there may come a day, perhaps not so far in the future, when we'll no longer need manufacturers to make our products for us. Gershenfeld, a physicist and computer scientist who runs the Center for Bits and Atoms at MIT, envisions a time when many of us will have a "fabrication center" in our homes. We'll be able to download a description of, say, a toaster -- perhaps one we designed ourselves -- to our computers, and then feed the designs and the raw materials into a personal fabricator. At the push of a button, almost like hitting "print," the machine will spit it out.

"In the end, fabrication [centers] will be just like PCs -- just technology that people have," says Gershenfeld, whose new book, "Fab: Personal Fabrication, Fab Labs, and the Factory in Your Computer," will be published in April by Basic Books.

This may sound like science fiction, but prototype versions of such fabrication centers, which Gershenfeld and his colleagues call "fabrication laboratories," or fab labs, are already up and running in Ghana, India, Norway, Costa Rica, and in Boston's South End. At least four other countries have shown interest in starting labs.

Earlier this month, a few dozen of their users from around the world met for the first time at MIT and at the South End Technology Center @ Tent City on Columbus Avenue, home to the Boston fab lab, to discuss the state of the technology behind the labs, how the current fab labs use their equipment, and what they'd soon like to build. Most said they want to work together with other users to solve problems they have in common, such as getting cheap Internet access into their communities. "The value is ultimately in collaboration," said SETC director Mel King.

So far, the fab labs -- which consist of about $25,000 worth of high-tech equipment and supplies, including a laser cutter, a vinyl cutter (normally for making signs but used here to cut copper for circuits), and a 3-D milling machine to make circuit boards, all connected to Linux-based computers loaded with open-source design and manufacturing software -- have mostly been used to teach kids technology, to let them play and explore, or to give adults a place to try out new designs for business ideas.

But the labs have the potential to become much more. The idea is that they can be empowering, especially in rural, developing communities, by giving people the ability to design and create the tools they want or need to solve local problems. In Ghana, users are trying to find an inexpensive way to build large solar energy collectors to turn the country's near-constant sunlight into power. In Pabal, India, a small community more than 100 miles outside of Mumbai, lab users developed diagnostic instruments to help fix tractor engines with timing troubles. And at the lab in Norway, users are working on GPS systems for boats and de-icing machines for windmills.

Ultimately, Gershenfeld wants to build a machine that can make any machine -- one that can "print" 3-D objects that include all the circuitry and mechanisms they need to move around, heat up, make noise, connect to the Internet, or do whatever it is they're designed to do. Such a machine -- think of the "replicators" on "Star Trek" -- doesn't yet exist, but Gershenfeld and others say there will be a version of it in a decade.

. . .

In 1998, Gershenfeld started teaching a class at MIT called "How to Make (Almost) Anything" as a way to introduce technical students to the expensive, industrial-size machines like laser and water cutters that he and his colleagues were using in their research. At first he gave formal lectures on each machine. But the students -- many of them artists, architects, or science students without a technical background -- "responded passionately to the tools," he says. Soon they stopped asking him for help. They worked alone and with each other to learn what they needed to build what they wanted -- things like a portable "scream machine" that saves your screams and plays them back (a kind of high-tech stress release); an alarm clock that won't shut off unless you prove you're awake by winning a game against it; a bicycle that recharges batteries when you ride it.

Though Gershenfeld liked his students' designs, he says, "they were making things for a market of one." He began to wonder how the fabrication tools could make a difference outside of Boston and decided that the mandatory outreach component of the National Science Foundation grant funding the CBA should be used to deploy field labs away from the "cluttered, intrusive technology" of cities. (That is not to say the Boston fab lab, which started a year ago, isn't important: Lab users, mostly inner city kids, learn valuable technology skills.)

Though personal fabrication research was still young, the technology had reached the point at which, for example, a relatively inexpensive ($4,500) table-top milling machine was precise enough to get down to the millionths of a meter (roughly the resolution of a CD player) necessary to build circuit boards with tiny components.

Technology, in other words, was becoming sufficiently cheap and sophisticated to make the labs useful. Gershenfeld describes the shift from large-scale, expensive machine tools to personal fabrication as analogous to the evolution that began 40 years ago from room-sized mainframes to personal computers. Instead of personalizing the ability to do digital computing, we're now able to digitize and personalize the ability to manufacture our own tools and machines.

As it currently exists, however, the technology imposes limits on what can be done. The fabrication machines used in the fab labs today can't produce anything larger than themselves. (The milling machine, for example, is the size of a printer.) The laser cutter cuts no longer than two feet. Nor can it cut very deeply: It would take a day for the laser to slice through an inch of plywood (they now use a saw when necessary).

The labs also face other, social challenges depending on where they are. In Boston, it is sometimes difficult to keep kids interested in learning after they're told, for example, they can't yet build life-sized robots. In South Africa, whose government is considering starting a fab lab, the challenge is apathy. "People don't want to be scientists or engineers anymore," said Riaan Coetzee, an information officer at South Africa's government-backed Council for Scientific and Industrial Research, at the user meeting in Boston. He believes a fab lab might get people excited about technology.

And there is the question of sustainability. The labs are relatively inexpensive to create, and so far MIT pays the start-up costs. But after a year or so, the labs are on their own. The ultimate goal is for the labs to be financially self-sustaining. To that end, Gershenfeld has met with people from the National Academies, the World Bank, and the World Economic Forum about funding, and though they like the idea of the lab, they all say it doesn't quite fit into their agendas.

"It's an animal the likes of which hardly anyone has seen," says Michael Jensen, director of Web communications at the National Academies. Gershenfeld, as always, has a novel solution: to create a different kind of funding organization, "somewhere between philanthropic aid, basic research, and business development," he says.

But the fab labs, Gershenfeld emphasizes, are research experiments and are still very much works in progress. So far, the reaction in the field labs has been encouraging. In Ghana, says Amy Sun, a grad student at the Center for Bits and Atoms who helped set up the fab lab there last summer, they ran an average of six classes a day for locals aged 4 to adult. She estimates that nearly 1,000 people came through the lab during her six-week stay.

The other encouraging sign is current lab users' desire to collaborate, even though they're in far-flung countries. Most of the labs, for example, want to build antennas for various communications purposes. Recently, the lab in Norway -- where farmers and engineers are collaborating to build a wireless radio network to track sheep and reindeer -- built an antenna and posted photos and instructions on the Web for the others to see.

Gershenfeld is impressed with how the lab users have taken the tools and made them their own. "I know how to do the technology," he says. "These labs are teaching me what it's good for."

Katharine Dunn is a writer living in Somerville. Email

globe graphic
A fab five
Pop-up A fab five
Since 2002, MIT's Center for Bits and Atoms has established ''fabrication laboratories,'' or fab labs, in Boston, Costa Rica, Ghana, India, and Norway. (Photos Courtesy of Center for Bits and Atoms)
Today (free)
Yesterday (free)
Past 30 days
Last 12 months
 Advanced search / Historic Archives