A few months ago -
I pitch a story idea to a national newspaper, and they say it's "interesting," but they don't actually commission an article.
I think it's interesting too, so I decide to write it anyway. If they don't want it, someone else will. (Eventually, someone else did.)
The story is this: the RepRap project, overseen by Adrian Bowyer from his office at the University of Bath, is nearing completion. The team have very nearly got the machine to make a copy of itself. And according to Bowyer, the whole replication scene is reaching a point similar to the point that computers reached in the early 1980s. Soon, he says, there will be RepRap homebrew kits, self-replicating themselves around the world.
These kits, like those early 1980s computer kits, will only be interesting to a small group of enthusiastic nerds.
Bowyer believes that the future of self-replication will follow a similar path. Just as homebrew computers became cheap consumer items found in every home, he thinks replicators will end up in every kitchen. The cost of making the replicators themselves will plummet, as the quality of replicated goods improves and the speed of replication increases. Neighbours will replicate parts for neighbours.
And, Bowyer is keen to stress, this will all be happening under the open source umbrella. Of course there will be commercial replicators from big-name brands; but alongside it there'll be a thriving alternative scene offering cheap replicated replicators, the design and concept protected by open source licences.
And there is the environmental angle. Bowyer sees no reason why these future replicators - as far advanced from their primitive beginnings as today's iPhones are from the Sinclairs and Acorns we grew up with - shouldn't be able to use household waste as raw material. Why not throw plastic waste into a hopper at the top, and leave it to stew for a few hours? Tomorrow, you can make it into a camera.
Oh yes, says Bowyer, consumer electronics are perfectly possible. The RepRap team have made basic circuit boards out of plastic and Field's Metal, which melts as a very low temperature. You can melt it in a cup of coffee.
He can't see why a camera can't be replicated, or any other common object. Household replicators will probably offer a menu of items, in a manner similar to the iTunes music store. You point at what you want, then wait for the machine to build it.
"There's no constraint on musical creativity now; anybody can write music and distribute it to millions of people on the internet. There's no need for anyone else to be the middleman. The same thinking applies: what is the point of anybody being being the designers of things and the people who use them?"
All this, of course, will have consequences.
"We want to do to manufacturing what the MP3 format has done to the music business," was what Bowyer told me in his lab.
The lab is a scientist's playground. It smells of melted plastic and chemicals. There are long benches with racks above. A Lego model sits on one of the racks. ("Lego is the most amazing stuff," Bowyer says with a grin.)
In the corner furthest from the door sits the RepRap. It looks ungainly, clunky, like something from the last decade. Indeed, the idea is older even than that. Bowyer says the concept was first proposed by David Jones in the Ariadne column, in New Scientist magazine on 3rd October 1974. "He meant it as a joke."
Bowyer goes on:
"This could have a huge effect in the poorest communities on earth, too. Not only could they make their own machine cheaply, but they could grow their raw materials. Polylactic acid can be grown from starch. So then your raw materials are renewable and biodegradable too. All you need is the energy to run the machine, which isn't much at all.
"Once you have machines that can make copies of themselves, the growth rate can be exponential.
"You have manufacturing without transport. You can also recycle much more stuff, and it's carbon neutral - in fact, you can be locking carbon up in the objects you make, which would make it carbon positive under some circumstances."
Bowyer's office - far from the lab, along many twisty and bewildering corridors - is typically academic. Not messy, but not tidy either. Bookshelves cover all available wall space, except for a few family mementos. His desktop PC is hardly up-to-date, but runs Ubuntu fast enough for him to get by. He shows me how they design 3D objects for RepRap. He shows me one of the prototype circuit boards, and it looks very rough, almost as if it's hand-made.
I ask him: "Would you say this is the most important thing you've done during your career?"
He shrugs, looks to one side and smiles. "I suppose this is the most interesting piece of engineering I have ever done."
I leave Bowyer and stride through the corridors of the engineering department. Outside, the ugly architecture of the University campus allows a chilly wind to howl down the narrow gaps between buildings.
Driving home, I'm confident that I can sell this story to someone. The RepRaps-as-1980s-computers sounds to me like the best angle. It's another computer revolution, ready to spring forth.
In the coming weeks, I email almost every contact I know. No-one's interested. I feel deflated, but hang on to the hope I can sell the story somewhere. Weeks pass, and the details fade from my mind.
Even more weeks later, I see the story appearing in various places, including some of those I'd pitched it to. At first this angers me, but I can't allow it to. This is how the business works. Sometimes you have the right idea at the right time, and sometimes you have the right idea at the wrong time.
(22nd July 2008)