Papers from Mathematical Software II
CORPORATE ACM TOMS Staff · ACM Transactions on Mathematical Software · 1975
It became apparent in the late 1960's that a new aspect of mathematical computations was maturing.This aspect was christened Mathematical Software and revolved around the following concepts: algorithm design and construction, comparative evaluations of algorithms and programs, dissemination of mathematical software via libraries and systems, interface of users and mathematical computations, automation of mathematical computations.The first conference, Mathematical Software, was held at Purdue University in April, 1970, and its purpose was to focus attention on this topic and to provide a starting point for a more systematic development.Interest and activity in mathematical software grew rapidly during the early 1970's.First, the area is a natural one for investigation; it has many interesting questions.Second, it was increasingly realized that the traditional numerical analysis approach does not completely cover the problems that arise in mathematical computations and that material from symbolic mathematics and other areas of computer science must be included.Finally, awareness of the needs for high quality software and of the increasing costs thereof emerged.One consequence of all this was the need for a workshop on mathematical computations, and one was organized by the National Science Foundation and held in Granby, Colorado, in August, 1972.The general discussions there led to the formation of a small group to explore ways to give a better focus to mathematical software and, especially, to consider establishing a journal in the area.The group originally consisted of John Rice (chairman), Thomas Hull, Stuart Lynn, and Joseph Traub; Lloyd Fosdiek joined shortly after it was formed.They went through a long and evolutionary process of developing alternatives and exploring practical possibilities.The end result was the conference, Mathematical Software II, and the new journal ACM Transactions on Mathematical Software.The group, with the addition of Joel Moses, became the program committee for Mathematical Software II.In a rather open format, about 15 invited talks and 70 contributed papers were given.The first two issues of this journal constitute the proceedings of Mathematical Software II.All of the papers appearing here have gone through the normal refereeing process and all fall within the intended scope of this new journal.One result, however, is that these proceedings show only part of the activity of the