by Beat Gerber firstname.lastname@example.org, March 1999
translated by Pedro Gonnet email@example.com
revised by Günter Dotzel [email deleted due to spam]
Grey is the color of the functional-oriented computer-science building at ETH-Zürich -- like the color of the cells that fill the heads of computer-scientists, but what a difference! Although the sober architecture of the building barely manages to stimulate the imagination, the brains at work in these bare rooms still produce genial ideas. Niklaus Wirth is one of the innovators. The professor at the Institute for Computer Systems set new scientific standards with his programming languages and computers. In the sixties, he based the then chaotic world of computer programming to a clear, comprehensive foundation.
With the programming language Pascal, which he finished in 1970, Wirth accomplished what made him famous world-wide. Pascal, named in reverence of the French mathematician of the seventeenth century, was the first programming language based on a clear and uniform structure, making it the language of choice for education. Pascal's great break-through came in the second half of the seventies, when the first micro-computers from Apple and Commodore hit the market. Countless schools around the globe installed Pascal on their machines...
Since then, Wirth has perfected his systematic approach and developed the programming languages Modula and Oberon as the successors of Pascal. But he was also interested in hardware design. Only when hard- and software are in tune, it is possible to create user-friendly solutions. With this in mind, he developed the computers called Lilith, Ceres, and Chamaeleon.
At the end of March 1999, Niklaus Wirth will vacate his office in the computer-science building, retiring at the age of 65 like all other Swiss civil servants. Yet Wirth was not the typical confederate civil servant. Uncompromising -- for some of his collegues sometimes somewhat puritan or even obstinate -- he followed his goal: to turn the computer into a productive working tool for everybody.
Systematic, methodic, and structured, these three academic-sounding terms always show up in his discourse when he talks about hard- and software, i.e., computer systems should be simple, understandable and clear. With these requirements, he sees himself far from the mainstream. He is convinced that "swimming against the tidal wave of main stream is one of the oldest tasks of a university."
Today's computer-science landscape looks like a uniform desert: a handfull of companies, like Microsoft, Intel, IBM, and Compaq control the global mass market. Yet in Wirth's view, today's popular software products are labyrinths, forcing users to move around with innumerable mouse clicks, finding the exit only out of dumb luck. The highly popular text-processing software, Word, is one example of a very complex system, with millions of lines of code, sprinkled with uncountable errors.
"Reliable and transparent programs are usually not in the interest of the designer", continues Wirth. The marketing departments of the software giants had noticed that potential customers are not lured by simplicity. Peoples are fascinated by cool features which nobody really needs, but can be shown-off proudly to friends. "The wishes of the users count more than their needs", he concludes.
Nevertheless, the trend towards faster personal computers and more complex programs seems unstoppable. Yet for Wirth, next to the technical limitations, there are also human limitations: "Our heads are no longer capable to process everything." Even the scientific community is showing signs of turning towards simplicity and clarity -- Wirth's conservative credo.
Wirth, being a sharp and critical analayst, is also critical of artificial intelligence. Although this is a fascinating field, it has been promising results for the past 40 years -- without success. "Constructing computers with human characteristics is nonsense", says Wirth. The possibilities of technology are generally overestimated, one of the reasons being the decline of physics and chemistry classes in public schools.
Sometimes the retiree feels like a caller without voice; even his two daughters weren't fascinated by a technical education and his son became a musician. "Maybe I was a frightening example", shrugs Wirth, who during his career consequently went his own way. With a diploma in electronical engineering from the ETH, he left Switzerland in 1959. At the renowned Universiy of California at Berkley, he graduated in software engineering and finally was an assistant professor at the Stanford University.
After eight years in the US, Wirth returned to Switzerland with his family, tempted by an offer from the ETH-Zürich to create a computer-science group. Despite the triumph with Pascal, Switzerland didn't seem to be a too fertile ground for computer-science. "Sometimes frustrating", says Wirth, who built the Lilith computer between 1977 and 1980. The powerful workstation was one of the first to have a mouse, a high-resolution monitor, and a graphical user interface. In comparison, at this time, the Apple II seemed almost traditional, equipped only with a keyboard as input device.
But the Lilith brought Wirth no comercial success. The industry was cautious and unwilling to take any risks and the project for a commercial production was quickly abandoned. "The opportunity to build a computer industry in our country was then lost", says Wirth.
Teaching computer-science in Switzerland was still in its infancy. While the US introduced computer-science courses already in 1965, the ETH only decided to follow suit in 1980, twelve years after Wirth's return to Switzerland. Many initiatives, which he and his colleague Carl August Zehnder submitted before, were ignored.
Yet such deceptions didn't stop Wirth from doing research and teaching. Landmarks of his work in the eighties were the Ceres computer and the Oberon operating system, which allowed an uncomplicated dialog between the user and the computer. Lately the computer-science pioneer has been working on the edge of hard- and software and has developed tools for field programmable semiconductors (FPGAs).
A career with seven honorary doctorates and many honors, driven out of curiosity and thirst for knowledge since his early years, comes to an end. Niklaus Wirth grew up in Winthertur, beside the high-school where his father taught geography. The only child found his inspiration in his fathers library. Filled with technical works, he found descriptions of turbines, steam-engines, locomotives and telegraphs -- the technology fascinated him. Yet the theory alone was not enough for the student, he wondered how everything would work in real life.
In a group of model airplane hobbyists he built airplanes of his own design together with friends - more than two dozens, the largest with a wing-span of 3.5 meters. In high-school Wirth was fascinated by chemistry. At home he built a laboratory in his basement to try out on his own what he had learnt at school.
His blue eyes twinkle as the professor laughs like a little boy while telling an anecdote about an unfortunate model-rocket experiment. He and a friend had not sufficiently compressed the fuel mix of potassium-perchlorine, sulfur, and carbon. The projectile missed it's planned trajectory and landed at the principal's feet, who had just come around a corner of the school. No disciplinary measures were taken.
"An unencumbered time", recalls the Universiy Professor, who during his academic career, nevertheless was able to realise most of his ideas. Fundamental research provided him with the much needed space to create new knowledge - without having to think about marketing any of the potential achievements. Tempi passati.
Time has changed in science. Professors are now managers and orientate themselves towards market tendencies, such as Anton Gunzinger with his super-computers. Wirth still has problems with commercialisation, having always pointedly preached for a clear separation between industry and education combined with research. Yet now the universities must be efficiently managed and profitable. That's the new government policy. "It's a devastating perspective, if university reasearch has to show short-term benefits," says Wirth sadly, favoring full academic freedom.
Yet the digital contrarian will remain independent -- in retirement. He wishes to find more muse, return to building model aircraft and reading. Above all he loves detective stories. Wirth can't get away from logical combination.
Most of the ModulaTor back-issues are available from http://www.modulaware.com/