Adding Methodological Testing to Naur's Anti-formalism

S. J. Meyer
DOI: https://doi.org/10.48550/arXiv.1208.3739
2012-08-18
Abstract:Peter Naur is the leading critic of formalist computing because of his extensive writings that disprove the now dominate characterization of human thought as cognitive information processing. Naur criticizes the ideological position that only discourse that adopts computer inspired forms are acceptable. Lakatosian philosophy of the methodology of scientific research programmes (MSRP) is added to Naur's studies to allow testing of computing theories. After discussing Naur's criticism of mechanical cognitive information processing, I show how to add MSRP competition to Naur's descriptive philosophy. Next, Naur's claim that computing can not become scientific until organizational issues involving ideological suppression of discussions of computing and human thinking are solved is corroborated by institutional suppression of my 1970s attempts to criticize structured programming (SP). Various problems in computing related philosophy are discussed. First, I argue that my MSRP based degenerating research programme disproof of SP is better than Naur's programming as a human activity, Demillo's social processes and Fetzer's unprovable causal nature. Three areas for post ideologically based computing study are discussed: computing as a path to rediscovering 19th century conceptions of infinity, axiom of choice testing facilitated by computing and relation to physical theory, and testing concrete complexity methods based on efficiency proof analysis.
Other Computer Science
What problem does this paper attempt to address?