Features

The World Summit on Sustainable Development, held last year in Johannesburg, wouldn't have been an obvious place to look for computing controversy. Mostly, it was as predictable as these events always are: Robert Mugabe delivered a tirade against Britain; Tony Blair visited a slum area to offer photo opportunities; and George W Bush couldn't see that it had anything do with him so didn't go.

On the sidelines, pressure groups of every kind lobbied for whatever salvation they believe in. Some, like Greenpeace, made the news, while others went largely unnoticed by the media. Among them was the catchily titled Action Group on Erosion, Technology and Concentration (ETC, pronounced etcetera), which presented a manifesto to warn the assembled world leaders about the threat of nanotechnology.

Nanotechnology is the domain of sober companies like IBM, which is developing ultra-small transistors, molecular circuits and ridiculously dense storage devices. Surely not a global threat. Away from computing, nanotechnology is being used to create new sensors, programmable drug-delivery implants, microfilters for pollution control and even car windows that won't steam up. Again, all apparently praiseworthy activities. Yet ETC isn't a group of simple luddites, nor is it alone in voicing its fears of this very tiny world.

Bill Joy, co-founder of Sun Microsystems and one of the leading developers of Java, has become a vociferous campaigner against nanotechnology. Just recently, Astronomer Royal Sir Martin Rees also called

ADVERTISEMENT

for restrictions on nanotechnology in his new book Our Final Hour. Earlier this year, the University of Toronto's Joint Centre for Bioethics published a stern warning that nanotechnology is heading for the same public confrontations and protests that have met genetically modified crops. So, what's going on?

One of the biggest problems involves reproduction. Nanotechnology devices, to the extent that they exist at present, are mostly laboratory curiosities that require considerable effort and expensive equipment to fabricate and perform some task that is itself undetectable except through yet more effort and expensive equipment. This is of limited practical value. What nanotechnology researchers therefore dream of is some form of self-assembly or replication. If nanomachines could build themselves as well as do something externally useful, we could just sit back and enjoy the benefits.

What happens, though, if the little critters run amok and start making too many of their own kind, or create mutant forms that do undesirable things, or find that human flesh provides exactly the raw materials they need, or... (fill in your own nightmare scenario here)? How will we cope with that when we can't even see the agents of our destruction spreading insidiously across the globe? This has become known as the 'grey-goo problem' and is top of the nanoscare agenda.

How realistic, though, is this idea of self-replicating machines? The fact that all living things manage to do it proves it isn't impossible, but examples of inanimate self-replication are pretty thin on the ground. A couple of self-replicating machines were built back in the 1950s, but they were large and did nothing useful other than to establish that it could be done. Also, they needed a supply of manufactured component parts to work with; they certainly couldn't extract raw materials and energy from the environment, which is what would be needed for a grey-goo machine. So far, nobody has managed to create one of those, either on the normal scale of industrial machines or the nanoscale. The basic toolkit is being developed, though.