Simplify, Simplify : Need Now Is for Technology to Be Less Complex
In 1962, Thomas S. Kuhn, a Princeton University historian of science, threw the scientific community into upheaval with the publication of his landmark book “The Structure of Scientific Revolutions.” Kuhn argued that the history of science was not, as previously believed, one of uninterrupted incremental progress building on all other advances, but rather was punctuated by conceptual “revolutions” that were essentially the blue-sky insights of specific individuals.
What happened between these epoch-making “paradigm shifts,” said Kuhn, could be called “normal science,” research that simply elaborated on the historic new paradigm until it was discarded for a new world view.
The same might be said now of technological innovation. Increasingly, it seems, what really counts as innovation--what builds entire new industries--is someone’s simple “knock on the side of the head,” as opposed to the marshaling of legions of researchers and engineers.
A good example is the World Wide Web. The Web, now the obsession of Wall Street investors and corporate executives in a dozen different industries, was dreamed up by a British systems developer, Tim Berners-Lee, while he worked at CERN, the European Union’s high-energy physics center in Geneva. Berners-Lee designed a so-called hypertext system to keep track of the constantly changing personnel at CERN, he says.
*
What has made the Web explode over the last couple of years is that HTML--or hypertext markup language, the codes used to describe and display text, images and links to a computer screen--is almost unbelievably simple. Not only are the basics of HTML coding easy to pick up in a day or less, but the system works on any machine with a browser program, and HTML documents can be prepared in any word processor. Grade-school kids are whizzes with HTML, and so are English majors.
The developers of Java, what could be another breakthrough for the Internet, apparently understood the importance of simplicity. Java, a new programming language that is especially effective for developing Internet software, isn’t nearly as easy to learn as HTML. But the Sun Microsystems programmers who developed it say they wanted “to build a system that could be programmed easily without a lot of esoteric training.” They deliberately left out some of the more confusing state-of-the-art routines of programming software.
“Advanced training in computer science today doesn’t mean that you’ll be prepared to develop the next big breakthrough,” says Terry Winograd, professor of computer science at Stanford University.
In fact, says Winograd, computer scientists are undergoing a curious change of status. Although we celebrate those who helped develop the Internet as pioneers and visionaries, many of the people behind significant computer innovations today are from other disciplines.
“What’s important to the information revolution now seems to be content, not programming,” Winograd says. “Programmers are still important, but they’re becoming less like superstars and more like plumbers--necessary but invisible workers.
*
As Kuhn suggested, it takes an individual stroke of simple genius, the kind that is rarely predictable, to put a hinge in history and set us off on a new paradigm. Galileo simply reversed Aristotle’s theory of inertia; Newton’s calculations were straightforward but breathtaking; Einstein’s formulas were perhaps the simplest of all.
In the computer field, significant breakthroughs can probably be counted on one hand. Mathematician John Von Neuman had a simple idea: the central processing unit with stored memory. Jack Kilby of Texas Instruments put three electronic components on one piece of silicon. Apple co-founders Steve Jobs and Steve Wozniak put a computer into a small wooden box. Other innovations are notable in the history of computing, but it seems to take a relatively simple proposal to change the world.
Simplicity is not merely an aesthetic or a good predictor of innovation. It helps chop away the unneeded complexity that has built up in periods of “normal” investigation. Einstein’s famous simple formula, E=MC2 , helped resolve many complex mathematical contortions physicists had produced in their attempt to explain data that was inexplicable using the old paradigm of Newtonian physics. The graphical interface PC made computing simpler. HTML and the Web helped relieve the complexity of using the Internet, which had been only for gearheads before the clickable interface of hypermedia came along.
*
What we need now are simpler computer and network architectures and simpler software. In their accumulated complexity lurks unpredictable effects, as chaos theorists are constantly reminding us. As we get more dependent on computers and networks, we’ll need breakthroughs that will make them simpler, more reliable and more comprehensible. In fact, doing complicated things simply is perhaps our biggest challenge, in nearly every domain of life. As Einstein said, “Things should be made as simple as possible, but not any simpler.”
It is the Zen-like quality of that insight that is at the heart of innovation, not the mountains of money, the immense institutions and the armies of people that chase the elusive “state of the art.”
Gary Chapman can be reached by e-mail at gary.chapman@mail.utexas.edu.
