Is mathematical software a legitimate research area?
John R. Rice · ACM SIGNUM Newsletter · 1981
Recently the chairman of a Computer Science department wrote to solicit my views "regarding the legitimacy of mathematical software development as a research area that can stand by itself". This question arose in that department's deliberations on promotions and tenure and they sensed "a clear split in the numerical analysis community on this question. The opposing view takes the position that mathematical software development is a sub-branch of the area of numerical analysis for which the software is being developed, and is not itself sufficient to establish research credentials for a numerical analyst." The following is my response to this solicitation.