We concluded with the sentiment that surely, surely we had arrived at the apex -- if that is the right word -- of miniaturization.

Uh uh, sorry, nope, no way. There is a new science in the birthing, called nanotechnology, that promises to make all kinds of practical machines, including computers, out of -- are you seated? -- a few molecules.

Nanotechnology was first proposed by Nobel physicist Richard Feynman in the 1950s, but its current-day avatar is Stanford professor K. Eric Drexler, author of Engines of Creation (Doubleday, 1986) and Unbounding the Future (Morrow, 1991).

The advantage of very small computers is that, being small -- measured in nanometers, or billionths of a meter -- they would be very fast, and encounter minimal physical resistance. Smallness, Drexler says, is a power unto itself.

Usually, when we think of things at the molecular level, we think of manipulating chemicals -- a jug of this, a quart of that, and presto, things will sloppily combine into new things.

But Drexler talks of building machines one precise molecule at a time, setting one kind of molecule abutting another kind, until it comprises a kind of nanoprocessor -- a micro-micro-microchip.

Drexler's computers will not be "electronic" the way today's are. Today's computers are like hydroelectric dams, standing in the path of a cascade of electrons, collecting some power, but squandering most of it. The first nanocomputers will function more like old-fashioned adding machines, with countless rods, levers and wheels, all made of different kinds of protein -- and yet still, because of their scale, tremendously efficient, outperforming the fastest research computers of today.

Crazy? Drexler is considered by many microengineers to be the flake of all flakedom, and nanotechnology has been dismissed as nanoreligion. But Drexler himself plunges deeper into the tiny, confident his naysayers are today's version of flat-earthers.

Drexler's nanofunctions already exist in nature. The production line at Cray or IBM has a natural counterpart in enzyme activities. Instead of glue and solder, the molecular computer will utilize natural molecular bonding forces. Computer cables re-envisioned at the molecular level, may be comprised of collagen, or silk. A nano motor may propel a machine much as the flagellum of a euglena moves the one-celled creature forward -- or is it backward?

Nanotechnology today is more of a research direction than a reality. A hundred practical questions must be answered for nanocomputers to become a meaningful reality.

How do you keep the thing from getting stepped on, or smashed by a mote of dust?

How do you control a computer that small? You can't hardwire it. Radio signals? Light signals?

And of course, it would be nice to know up front what they would be used for. Drexler imagines a handheld nanocomputer that will be a million times more powerful than the most powerful supercomputer today.

An everyday item? A WordStar screamer? That doesn't make much sense. But remember that when the first radio-tube machines were rolled out in the late 1940s, the ultimate world computer market was pegged as maybe a dozen units.

Drexler breathlessly mutters about a world in which microscopic computers are let loose to solve some of the problems the world is buckling under -- problems of the environment, healthcare, food production, and Drexler's own idea of what a postindustrial world will work to produce. It gets pretty vague pretty quick.

But real things are happening. Scientists at IBM proved they can move individual atoms around in 1990. when they precisely aligned 35 xenon atoms to form an IBM logo -- a tinier level than Drexler ever talked about.

Scientists at Syracuse University have built the first molecular logic circuit, that, if you could see it, would look like a wishbone, with two legs and a neck. The gate is a single, complex molecule, the size of a hemoglobin molecule, four nanometers across.

Gradually, we are getting a grasp on molecular engineering. Today's scanning tunneling microscopes and atomic force microscopes will doubtless seem primitive one day, but today they are doing duty as molecular forklifts, sliding atoms this way and that, building nanostructures in the lab.

But the main job of building molecular computers will be done by other molecular machines, empowered and programmed to fix or replace whatever needs fixing or replacing.

All in all, nanotechnology is just about the darnedest thing you could think about. As soon as you think you grasp it, it skitters away again, unvisualizable.

But that's OK, Drexler says -- just because we can't hold it in our heads doesn't mean it isn't happening. The dawn of the age of nanocomputers is nearer than just about anybody thinks. Drexler himself puts it at a mere 20 years from now.

So if you want to make fun of microscopic computers, now is the time to do it. Put it off too long, and people will remember what you said. And then won't you look like a goof.

To ""Future Shoes"" home page