BioEd Online

When people are given free rein, they create extraordinary things. That's what Neil Gershenfeld discovered when he gave them access to the right tools. From students to farmers to film stars, they have made milk analysers, computer interfaces for parrots and artificial jellyfish.

In his new book, Fab, Gershenfeld describes how he and his colleagues developed the desktop 'personal fabrication' laboratories (fablabs) needed to turn dreams into reality. Gershenfeld, a member of the Center for Bits and Atoms at the Massachusetts Institute of Technology (MIT), Cambridge, reckons fablabs will usher in a technological revolution that will rival or even surpass that in personal computing.

In 10 or 20 years time, he predicts, your computer and printer will be accompanied by a personal fabricator that makes whatever you want, assembled using the same digital logic that the computer uses.

And crucially, he thinks that this will not be simply a plaything for the rich. He has already seen many examples of how personal fabrication can address the needs of poor communities, allowing them to make things in which industry is not interested.

Dream maker

Personal fabrication, claims Gershenfeld, will reduce the wastefulness of manufacturing. It will build objects piece by piece, from parts that can be disassembled and recycled. It will make it feasible for manufacturing to supply a market of one: you.

How did this vision emerge? "I started a class at MIT called 'How to make almost anything'," he says. "It was meant to take tens of students, but we had hundreds begging to get in. They were saying things like: 'All my life I've been waiting to take a class like this'."

And what truly took him by surprise was that these weren't the usual scientifically savvy MIT students. "They were less technical people. They made the most inventive and quirky things, like an alarm clock you have to wrestle with to show you're awake. They were hijacking the MIT technical infrastructure for personal expression."

Tool time

But what exactly is this infrastructure? At this stage, Gershenfeld's tools are off-the-shelf instruments. There are cutting tools that use laser beams or high-pressure water jets, which can slice through almost anything. There are milling devices, rather like power drills, that can cut sharp corners and smooth surfaces.

The cutters are precise enough to make components that will snap together without glues or fasteners. There are also devices such as three-dimensional 'printers' that use laser beams to burn complex solid shapes into liquid polymers, or that build up objects from layers of fine powder.

But it's not just about making new shapes. Gershenfeld wants to let people build objects that do things: that take measurements, or send and receive signals, or move about. For that, he provides low-cost microelectronic circuits that control and compute.

Basic approach

"The world is covered with grass-roots inventors," Gershenfeld says. And there's no need, he argues, for development projects to be focused always on low-tech solutions. Now that his team has put the fablab's tools and computers into a package that fits on a desktop and costs around US$20,000, his vision is already reaching the developing world.

In Ghana, for example, fablabs are being used to devise cheap refrigeration units and solar-power collectors that boil water for power generation. Once installed, the fablabs can be cheap to run. "A lot of the raw material starts coming from stuff thrown out by the communities," says Gershenfeld.

Ultimately, he wants to develop fabrication systems that work 'digitally', putting objects together from discrete 'bricks' that could be microscopic in scale. This, says Gershenfeld, allows perfect structures to be made by imperfect fabricators. "Lego bricks let a child build something more precise than their motor-control skills would otherwise allow," he explains. "Lego error-corrects for them."

In this way, Gershenfeld sees personal fabrication following the same trajectory as computing. "The big fabrication machines at MIT are like the old mainframe computers," he says. "Our desktop fablabs are, in cost and complexity, like minicomputers. And we're headed towards the equivalent of the personal computer. We're halfway down that path."

Much of the impetus, he thinks, will come from developing countries, where practical needs are driving invention. "In London and Paris, we get the response: 'Why bother? We already have all we need.' But in the developing world, personal fabrication is really engaging people in science and technology."

Fab is published by Basic Books on 12 April 2005.