A Math Genius Who Decided Not to Play It by the Numbers

Robers Lee Hotz, Los Angeles Times, September 30, 1996.

When the MacArthur Foundation gave physics prodigy Stephen Wolfram one of its coveted "genius" grants in 1981, it knew it had done something unusual. Wolfram, then a 21-year-old Caltech research associate working in a field called quantum chromodynamics, was the award's youngest recipient. Foundation officials could not have been more surprised by the end result: He abandoned academia and went into business. Wolfram, 37, now is the creator and prominent pitchman of perhaps the single most widely used mathematics program for technical and scientific computing. Mathematica—part computer applications program, part programming language, part symbol engine—has more than a million users, ranging from engineers, physicists and technical economists to musicians, textile designers and computer graphics artists. Wolfram Research Inc. in Chicago[sic], which Wolfram founded in 1986, today plans to introduce a major make-over of the product—Mathematica 3.0—which he hopes will broaden its reach to numerically challenged financial analysts and even high school number crunchers.

In an interview with the Times, Wolfram talked about research, business, intellectual property, the role of universities, money and the pleasures of a computing life.

Q: As one of the most promising young physicists of your generation, you were poised to pursue a distinguished research career. What prompted you to develop Mathematica instead?

A: Part of the motivation for building Mathematica was that I needed something like it for myself. Conveniently enough, the thing that I needed turned out to be useful for a million other people.

Quite early on, I was interested in doing experiments on computers. One of the things that held me up was I just didn't have the right tools to do what I wanted to do. I spent a lot of my days writing a lot of pieces of software to support these experiments that I wanted to do on computers. I realized this was silly. I was spending lots of time putting together tools that in some cases could be quite general tools, but I was putting them together for very specific computer experiments.

Maybe, I thought, there is a better way to do this.

As we continue to develop Mathematica, I am probably one of its most stringent testers because I am always trying to push the envelope by using it for my own scientific work. I am always the one reporting the most outrageous bugs.

Q: You started your design of Mathematica where most software developers end theirs—by writing a 1,395-page users' manual. Why?

A: If you can't explain it honestly in the manual, then you are probably making a mistake in the way it is designed and people will never be able to understand how it is ever going to work.

One of the things I found to be the most intellectually demanding in building big systems like Mathematica is this whole thing of starting from nothing and then having to build some kind of language and some kind of structure that a lot of people are going to live inside. Can you set that up in a way that is intuitive for people?

If it was built in the way people traditionally think of building applications programs, people would never be able to use it in practice. It would be this huge mass of disorganized, incoherent features. Without some kind of overarching set of principles, your pieces would never fit together.

Q: You continue to conduct basic research, investigating computer automata and artificial-life algorithms. How do you balance the demands of a software developer with a commitment to serious scientific work?

A: One has to be fairly efficient and organized. It has been my personal practice that I work fairly late at night, by which time the company has closed down. Everything is very quiet, and I am able to concentrate on a solitary activity. It is a strange contrast.

In dealing with a company, managing things, playing entrepreneur, it is a very frenetic existence, where you make 30 decisions a day and you are dealing with people all the time. That is real different from doing basic science, where really it is you and the computer and nothing else. It is a strangely solo activity.

Q: You went from Caltech to the Princeton Institute for Advanced Studies to the University of Illinois, and now you are on your own. Is it easier outside the university?

A: My view about doing basic science is that if you have no choice, then getting paid by a university is a fine thing to do. If you have a choice, there are a lot better ways to live.

In my life now, where I am a CEO of a company, the actual fraction of my time that I can get to devote to basic science thinking is probably much larger than the fraction of time that a typical senior professor at a university would get to devote to actual basic research. If you are a senior university professor, you are out raising money from the government, being on committees, teaching classes, and it is only in the extra bonus time that you get to do research.

I don't have to beg the government. I don't have to convince anyone at the National Science Foundation that what I am doing is not as nutty as they might assume or as the peer review system might say it was.

Q: Universities are trying to foster innovation and new technology, but there is a rich history of friction between academic entrepreneurs and their schools. From your experience, can the academic culture support high-tech entrepreneurs?

A: Well, they can avoid trying to squash entrepreneurial activities. Ten or 15 years ago, when I was at Caltech, for example, the idea of commercial stuff going on in an academic setting was viewed as completely horrifying.

Universities in some cases provide the way to get the R&D for a product done fairly cheaply, more cheaply than by getting venture capital.

There was a time when the professors and the university scientists were the top dogs, where people really expected innovation would come from that sector. Increasingly, as the years go by, the innovation is really tending to happen in entrepreneurial companies, rather than in universities.

Certainly the computer industry is a case in point.

Q: Over the years, you always have been unusually careful to spell out your rights to any intellectual property. How important are patents and copyrights?

A: One of the things that people often forget about intellectual property and its ownership is that the formal pieces of paper that describe the ownership of a piece of intellectual property are a small part of the battle in terms of having it turn into something valuable.

So many times I have seen people who say they have this very clever idea and who want to patent it and who think they are going to get rich from the patent. It does not happen.

In order to turn something from that intellectual idea into something real, it needs tremendous human effort. It needs tremendous emotional commitment...really caring about the product one is trying to make and really pushing it hard.

Q: What would you consider your most serious mistake running Wolfram Research?

A: Part of our market is selling to universities—maybe 25% of our revenues. When Mathematica first came out, academics were used to the idea that any software they cared about was free—at least to them.

I thought there was a serious market for Mathematica in the academic market. We had to dig in our heels and say this is going to cost you real money. People got very upset about that. To this day, people keep saying: "You guys charge too much and are so difficult for universities to work with."

It might have been smarter for us to start off with the lower price we ended up with, rather than digging our heels in so firmly.

It was hard to know. We were creating a new kind of market for software. We didn't know the rules, and they didn't know the rules either.

I was worried that if we started with lower prices we would have some downward spiral where it would end up at zero.

Q: What is more important to you—the technical elegance of the Mathematica program or the financial rewards it brings?

A: In the world of high-tech industry, the money becomes the main point for a lot of people. Take your company, puff it up a bit, take it public, cash out, retire. And then what? I have kept my company private and intend to continue doing that because what I am really interested in is the long-term intellectual achievement that our product represents.

The Mathematica system and the language it embodies will be around as long as I am. I think it will be a continuing fixture in the intellectual landscape. That is important to me.