In 1965 few people outside Silicon Valley had heard of Gordon Moore. For that matter, no one at all had heard of Silicon Valley. The name did not exist and the orchards of Santa Clara county still brought forth apples, not Macintoshes. But Mr Moore could already discern the outlines. For 1965 was the year when he published the paper that gave birth to his famous law that the power of computers, as measured by the number of transistors that could be fitted on a silicon chip, would double every 18 months or so.
Four decades later, equally few people have heard of Rob Carlson. Dr Carlson is a researcher at the University of Washington, and some graphs of the growing efficiency of DNA synthesis that he drew a few years ago look suspiciously like the biological equivalent of Moore's law. By the end of the decade their practical upshot will, if they continue to hold true, be the power to synthesise a string of DNA the size of a human genome in a day.
At the moment, what passes for genetic engineering is mere pottering. It means moving genes one at a time from species to species so that bacteria can produce human proteins that are useful as drugs, and crops can produce bacterial proteins that are useful as insecticides. True engineering would involve more radical redesigns. But the Carlson curve (Dr Carlson disavows the name, but that may not stop it from sticking) is making that possible.



