Premium
Programming Techniques to Expedite Communication of Scientific Models
Author(s) -
Waldman Sue E.,
Rickman Ron W.
Publication year - 1990
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1990.00021962008200020037x
Subject(s) - computer science , software portability , fortran , usable , programming language , code (set theory) , subroutine , documentation , standardization , modular design , variable (mathematics) , interfacing , operating system , world wide web , mathematics , set (abstract data type) , mathematical analysis , computer hardware
Computer programs describing scientific models aid in communicating models and their multiple components to other scientists. If several language‐independent programming practices are used, programs can be much more valuable communication tools than are many of the currently distributed programs. Text file input of parameter values and prompts permit simple modification of program algorithms and screen instructions without requiring recompiling. Standardization of variable name lengths to the lowest common denominator (which seems to be FORTRAN 77) of six characters increase the portability of code. Structured programming, subroutines, absolute abstinence from the use of GO TOs, and appropriate documentation provide organized, decipherable code and hence readable and usable programs. Language dependent practices of modular programming and passing of variables can simplify major model changes and program redesign. Programs written as communication tools do not necessarily execute in minimum time but can dramatically reduce time needed to understand a model and incorporate its code, without introducing error, into the code of larger models or simulations.