Ever since I read K. Eric Drexler's Engines of Creation I'd been interested in nanotechnology and followed the relevant literature while trying to figure out, as is my wont, how to make a buck out of it. As one of the designers of Xanadu and an ardent supporter of the project, Eric became acquainted with Autodesk when we invested in Xanadu in 1988. When Eric saw a demo of HyperChem at a conference, he immediately realised that the fit between Autodesk and Hypercube, Inc., developers of HyperChem, was potentially very good. When Eric and Chris Peterson spoke to me about this opportunity, I finally saw not only an opportunity for Autodesk to establish itself in the scientific modeling market, but also a way to position ourselves to benefit from the advent of nanotechnology, if and when it emerged, for surely one could not design atomically-precise structures without a molecular CAD system, which is precisely what a molecular modeling package is.
In early 1989, I prepared this talk about the consequences of nanotechnology to help tilt the balance in favour of the HyperChem deal. The talk was delivered at the Autodesk technology forum on May 10, 1990, before an audience which included Eric Drexler, Chris Peterson, and Neil Ostlund, founder and president of HyperCube, the designer of HyperChem.
This talk has been reprinted by the Foresight Institute, in Micro-Times, and Mondo 2000.
These are indeed extraordinary times we're living through.
Few people are lucky enough to live at a time when their chosen field of interest becomes the center of a technological revolution that changes the face of the world.
We've all shared that good fortune; the success of Autodesk in so short a time is evidence of of how rapidly and how completely the evolution of computers has changed the way design and engineering are done and has reshaped the very terrain on which businesses compete.
Yet all of these remarkable events were predictable, at least in general outline, more than 25 years ago. I remember, like yesterday, the afternoon in the spring of 1968 when, after I first learned about the technologies used in the crude integrated circuits of the time, it hit me. There was no fundamental physical reason you couldn't put a whole computer on a single chip. It was just a matter of engineering, money, and time.
Now I certainly didn't know when it would happen, and I doubt anybody anticipated price and performance reaching their current levels, but the direction and the goal was clear. It made engineering and economic sense; there was no reason it wouldn't work; and each intermediate step along the way would clearly pay for itself in commercially successful products. That pretty much made it inevitable.
This is the kind of reasoning I'm going to be using in this presentation--projecting readily predictable trends forward and asking the question ``How far can we go, and what happens when we get there?''
Some of the conclusions that seem almost inevitable have profound consequences that are not just interesting, but important for companies like Autodesk who wish to grow and prosper in the coming years.
Editor: John Walker