Lively and long discussions took place between Peter Naur and his guest Edgar G. Daylight at Naur's home in Gentofte, Denmark in spring 2011. The nuggets are bundled in Lonely Scholar Conversations first issue.
What an absolutely cool guy!
— Dennis Shasha, New York University
Fascinating... the interview is a very worthwhile contribution to documenting the history of the field, and will be of strong interest both to computer scientists and to professional historians.
— Robert Harper, Carnegie Mellon University
What mathematical rigor has (not) to offer to software engineers
Peter Naur wrote his first research paper at the age of 16. Soon an internationally acclaimed astronomer, Naur's expertise in numerical analysis gave him access to computers from 1950. He helped design and implement the influential ALGOL programming language. During the 1960s, Naur was in sync with the research agendas of McCarthy, Dijkstra, and others. By 1970, however, he had distanced himself from them. Instead of joining Dijkstra's structured programming movement, he made abundantly clear why he disapproved of it. Underlying Naur's criticism is his plea for pluralism: a computer professional should not dogmatically advocate a method and require others to use it in their own work. Instead, he should respect the multitude of personal styles in solving problems.
What philosophy has to do with software engineering
Though Peter Naur definitely does not want to be called a philosopher, he acknowledges having been influenced by Popper, Quine, Russell, and others. Naur's writings of the 1970s and 1980s show how he borrowed concepts from philosophy to further his understanding of software engineering. In later years, he mainly scrutinized the work in philosophy and mathematical logic & rules in particular. By penetrating deeply into the 1890 research of William James, Naur gradually developed his own theory of how mental life is like at the neural level of the nervous system. This development, in turn, helps explain why he always opposed the Turing Test and Artificial Intelligence, why he had strong misgivings about the Formal Methods movement and Dijkstra's research in particular.
Peter Naur was born in 1928 in Frederiksberg, Denmark. He graduated from Copenhagen University in astronomy in 1949 and was a research student at King's College, Cambridge in 1950-51 where he programmed the EDSAC in order to solve an astronomical problem. After a year's stay in the USA and a brief return to Cambridge, Naur spent the rest of the 1950s in Copenhagen. In 1957 he received his PhD in astronomy and in 1959 he joined the staff of Regnecentralen, specializing in high-level programming languages. He became heavily involved in the design and implementation of ALGOL60 and organized the ALGOL Bulletin. From 1969 until his retirement in 1999 he was professor at the Copenhagen University Institute of Datalogy. Since the mid-1990s until this day, Naur actively studies and writes about psychology and in particular about how human life is like at the neural level of the nervous system. He has received several awards, including the Computer Pioneer Award of the IEEE Computer Society in 1986 and the ACM Turing Award in 2005.
Conversations Volume 2011 Issue 1
Pluralism in Software Engineering: Turing Award Winner Peter Naur Explains
Paperback, 127 + iii pages
Published October 2011
Conversations ISSN 2034-5976