Vanessa Spedding, Scientific Computing World (December 1996) 3.
Stephen Wolfram, pioneer in the field of complex systems research, creator of the Mathematica software environment and president and CEO of the USA’s Wolfram Research, spoke to Vanessa Spedding about the changing face of technical computing.

Q: What inspired your ideas in technical computing?

A: I was working in particle physics where a lot of the problems involved complicated calculations. I tried to use existing computer tools for doing these, but grew frustrated with them and decided to build better tools that would help make more efficient use of time.

I had developed the SMP computer algebra system by 1981, but then moved on to complex systems research. Here I was dealing with what I view to be one of the most fundamental questions in science: how did nature produce all this complexity and what tools could we build that could help deal with such problems?

Q: How did this lead to Mathematica?

A: I needed to do lots of calculations and complicated experiments. I was analysing how I spent my time—the majority of it was spent thinking what to do and cobbling together programs and tools that would enable me to do it. I realized it would be nice to have a unified, integrated system to do everything: graphics, algebraic calculations, documents and so on. The main conceptual achievement was that symbolic processing could be the underlying foundation. Nature’s complexity inspired the idea that computational systems could parallel intellectual activity.

Q: How has the program changed?

A: Creating Mathematica has been a continuous process. I’m proud of the fact that there were very few mistakes in the underlying structure of the first version: you can run Version 1 programs in Version 3. The second version deepened the program—it took its capabilities further. Version 3 adds more depth and also broadens it. You can now add programmable documents and do mathematical typesetting, for example.

Q: Where is it going next?

A: Mathematica is a unique product. It is more general than other technical software; most of the competition comes from Fortran. I predict that more and more programs will be written in Mathematica rather than in Fortran until the replacement of Fortran is complete. Mathematica will reach out to the point where more practitioners will make use of it in their technical work. The interface will not evolve much more over the next 10 years: there will be refinements, but the notebook interface, with its interactive, on-line documents, presents a natural way of working.