Kristen Nygaard

Coinventor of Simula and object-oriented programming

• Born: August 27, 1926
• Birthplace: Oslo, Norway
• Died: August 10, 2002
• Place of death: Oslo, Norway

Primary Company/Organization: Norwegian Computing Center

Introduction

In the 1960s, the Norwegian computer scientist Kristen Nygaard, in collaboration with Ole-John Dahl, developed a new programming language, Simula, and a new methodology for programming, object-oriented programming, that would result in computer programs that would more closely follow the way people work. This system, based on chunks of code (objects) that could be assembled and modified in almost limitless fashion, made it possible to create programs with greater power and to reduce the time required to devise code. In the first part of the twenty-first century, object-orented programming is used in countless business applications and is the basis for developing applications on the Internet. In addition to his technical contributions, Nygaard was an active and articulate participant in his nation's political process, seeking to keep Norway's social and economic foundations intact.

Early Life

Kristen Nygaard was born in Oslo, Norway, on August 27, 1926. His life, like that of all his fellow Norwegians, was profoundly affected by the 1940 German invasion at the beginning of World War II. The invasion and subsequent occupation by the Germans disrupted the country's educational system: Initially, classes were canceled; eventually, the Nazi occupiers imposed mass arrests of teachers. Nygaard nonetheless managed to complete school, graduating from high school in 1945, the year the war ended.

In that year, Nygaard enrolled at the University of Oslo, where he earned a bachelor's degree in mathematics. Nygaard, like all young men in postwar Norway, was obliged to perform national service. Unlike others, however, who might have served as soldiers or sailors, Nygaard worked at the Norwegian Defense Ministry's Defense Research Establishment. From 1948 until the early 1950s, Nygaard performed general mathematical support for defense projects. Continuing his education while working, he received a master's degree in mathematics from the University of Oslo in 1956. His specialty was Monte Carlo simulations (large-scale algorithmic computations based on large random samples to see not only all possible outcomes of a decision but also the probable outcomes).

Starting in 1952, at the Defense Research Establishment, Nygaard became involved in operations research (OR) the process whereby practitioners search for quantitative solutions to support the decision-making process. He would remain in the OR section, eventually heading it in 1957, until he departed in 1960. In that year he moved to the Norwegian Computing Center (Norsk Regnesentral), eventually becoming its head in 1962.

Life's Work

Computer programming in the early 1960s had neither the variety of languages (allowing a choice for developers) nor the power that languages existing in the twenty-first century have. Programming languages were based on the premise that a task to be performed would be broken into a string of procedures. In other words, a program was a set of instructions in linear form that would prescribe the entire process from beginning to end.

What struck Nygaard and his collaborator, Dahl, was that this style of programming did not necessarily capture how people work to accomplish a specific set of objectives (their workflows). Nygaard and Dahl began to think of creating programs that would respond to a way of modeling the set of instructions that better reflect how people work. Their response was to begin to create their own programming language. What they developed was released in 1962 as Simula I. Simula I was not an entirely new language; it was built on an already existing language, Algorithmic Language 1960 (ALGOL 60, designated as 60 because it was officially released in 1960). ALGOL 60 was the latest version of a language that contained a new feature, “nested functions.” Nested functions meant that a procedure could contain subprocedures within it as part of its structure. That ability would prove to be an important point of departure for what Nygaard and Dahl would eventually do. They would build on properties contained in that language as well as ideas about organizing code that had started to be developed by others in the late 1950s. Building a language on the capacities of another language was the method by which Niklaus Wirth would develop his own language, Pascal, also based on ALGOL 60.

Simula I and, more specifically, its successor, Simula 67, put into place a new model for how problems would be examined by analysts and how the analysis would be put into a working program by coders. The process would be known as object-oriented programming (OOP). OOP has been often referred to as code that is similar to Lego blocks. Just as Lego blocks can be assembled in different ways to make many different structures, OOP is based on chunks (blocks) of code that can be assembled together. These chunks or objects, can contain both the data and the processes that operate inside them. They can then be assembled in various ways (in other words, determining the relationships or instructions or communications among the blocks) to create the program needed to solve the real-world problem. What is most valuable about these objects is that after one has been created (called an “instance”), it can be stored, copied, and modified for use in other programs to solve other problems.

By the late 1960s, the basic premise of OOP was in place, although developments in the coming years would increase OOP's sophistication (too much so, according to Nygaard) and power. OOP was becoming the basis for new languages that were developed in the 1970s and 1980s and that became widely adopted in the 1990s. Apple's computer developers began to use it. The Xerox Palo Alto Research Center (PARC, where Nygaard worked while on sabbatical) developed the object-oriented language SmallTalk. Bell Laboratories would develop C++, another object-oriented language, using the already existing language C as it base, similar to the way Nygaard and Dahl had subsumed ALGOL 60 as the base for Simula. Java, which is widely used to develop Internet applications, is another object-oriented language.

Although he took a degree of satisfaction in his accomplishment, Nygaard was not entirely happy with how things had developed by the end of the 1990s. First, he thought OOP languages had become too complex. Second, he believed that the manner in which analysts and programmers were taught OOP was not satisfactory. In order to remedy that problem, he started to develop comprehensive object-oriented learning (COOL) to provide the basis for effective use of object-oriented languages. He had started on this project at the time of his death and it has been continued by others since his death.

Personal Life

Nygaard's life encompassed a great deal more than developing solutions to technical problems; he seems never to have thought of technology as a self-contained entity without social implications. The effects of science and technology on society informed much of his thought; he was keenly aware of economic, social, and economic developments in Norway. His activities included technical work to support and assist workers as well as actively participating in the Norwegian political system.

In the late 1960s, Nygaard performed work for the Norwegian Iron and Metal Miners so they could better evaluate computer technologies as they might affect how miners performed their jobs and use them most effectively. As part of this effort, Nygaard developed training materials for miners so they could better understand the impact of new technologies.

At the same time that he was providing technical assistance to the labor union, Nygaard became involved in national politics. He would remain visible, active, and successful until shortly before his death. In the 1960s, Nygaard was a member of the Norwegian Liberal Party, changing his affiliation in 1971, when he became an active member of the Norwegian Labor Party. A strong believer in Norway's welfare state, Nygaard throughout the 1990s strongly opposed Norway's entry into the Eurozone. His opposition was based primarily on what he saw as the fiscal restrictions that would be placed on Norway's management of its finances (particularly social welfare programs) if it were to join. The forced austerity measures on some Eurozone members in 2012 have proven him to be correct, at least from Norway's perspective. The proposal was defeated and Norway never joined.

Nygaard received several major awards. These included the Association for Computing Machinery's A. M. Turing Award and the Institute of Electrical and Electronics Engineers' John von Neumann Medal. He also won the Norbert Wiener Prize from the American Association of Computer Professionals for Social Responsibility and the Order of Saint Olav from the king of Norway.

Nygaard was married, with three children. He and his wife, Johanna Ur, were married in Oslo on January 27, 1951. Their first children, a set of twins, were born in May of that year. He died of a heart attack in Oslo on August 10, 2002, a few weeks after the death of his collaborator, Ole-John Dahl, whose obituary he had written.

Bibliography

Atkinson, J. B. “Object Orientation, Discrete Simulation and Simula.” Journal of the Operational Research Society 46.12 (1995): 1510–13. Print. A general history of Nygaard's efforts and their results in the areas of the Simula programming language, as well as the importance of object-oriented programming.

Berntsen, Drude, Knut Elgsaas, and Håvard Hegna. “The Many Dimensions of Kristen Nygaard, Creator of Object-Oriented Programming and the Scandinavian School of System Development.” History of Computing: Learning from the Past. Ed. Arthur Tatnall. IFIP Advances in Information and Communication Technology, Vol. 325. Boston: Springer, 2010. 38–49. Print. An account of Nygaard, his developmental work as well as his political activities.

Holmevik, Jan Rune. Educating the Machine: A Study in the History of Computing and the Construction of the Simula Programming Language. Dragvoll: University of Trondheim, Center for Technology and Society, 1994. Print. A historical account of Nygaard and Dahl's development of Simula and the further development of the object-oriented paradigm.

Waldrop, M. Mitchell. “Frustrated with Fortran? Bored by Basic? Try OOP!” Science, new series, 261.5123 (1993): 849–50. Print. In the early 1990s, when several early programming languages were still being used, OOP was beginning to be considered the next big thing in programming. This article describes OOP's perceived advantages.