NAPSS—a numerical analysis problem solving system

  • Authors:
  • John R. Rice;Saul Rosen

  • Affiliations:
  • -;-

  • Venue:
  • ACM '66 Proceedings of the 1966 21st national conference
  • Year:
  • 1966

Quantified Score

Hi-index 0.05

Visualization

Abstract

It has long been a recognized goal of computer programming technology to permit problems to be stated in languages appropriate to the fields in which the problems exist, and to provide for the solution of such problems as stated without requiring the services of specially trained programmers and analysts. Systems that provide these capabilities are “problem solving systems.” Their languages are problem-oriented languages. Systems of this type have been proposed and developed in a number of areas of engineering and design. Examples are the COGO system developed at MIT 1 and the DACS system developed by General Motors 2. At Purdue University during the past year we have been working on a design for a Numerical Analysis Problem Solving System (NAPSS). This paper discusses some of the features of the language and system design, and an approach through the development of so-called polyalgorithms to the design of the numerical analysis packages that are the essential elements of the problem solving system.