Affordable Access

Letter from Paul B. Schneck, National Aeronautics and Space Administration (NASA) to Joshua Lederberg

Authors
Publication Date
Disciplines
  • Earth Science
  • Mathematics
  • Medicine

Abstract

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION GODDARD SPACE FLIGHT CENTER INSTITUTE FOR SPACE STUDIES NEW YORK, N. Y. MAIL ADDRESS INSTITUTE FOR SPACE STUDIES 2860 BROADWAY NEW YORK, N. Y. 10025 TELEPHONE UNIVERSITY 6-3600 29 January 1974 Professor Joshua Lederberg Department of Genetics School of Medicine Stanford University Medical Center Stanford, California 94305 Dear Professor Lederberg: Your letter of January 23 raises many interesting (and exciting) points and I will try to cover each. The Institute for Space Studies does intend to join the ARPA network, and expects to be operational in the Fall (paperwork and telephone equipment account for the delay). Our purpose is primarily to use the ILLIAC, though we would be delighted to participate actively in other areas as well. The original presentation of the paper did not include any timings because we felt that they would not be meaningful. In fact, timing was not done until the referee suggested it. The compiler is used locally to permit the mathematicians, physicists, and meteorologists to avoid details of coding and concentrate on problem solving. I have enclosed a copy of the original typescript which contains appendices and other details removed to shorten the paper for publication. The original conclusion (the referee found it too “big headed”) is perhaps a better indication of the true value of the compiler. We have also found the compiler to be of use in diagnosis and debugging of programs. An alert is raised when a dead variable is removed, and this is often an indication of clerical error which the production compiler (Fortran-H) does not detect. I have also enclosed a copy of another paper “A Survey of Compiler Optimization Techniques”, which points out (footnote 2) that users either did not exercise or did not understand *‘explicit control” when once available in Fortran. The use of the “external” statement as in the IBM Fortran-H (extended) compiler will be inc

There are no comments yet on this publication. Be the first to share your thoughts.