New Approaches to Programming
COLORADO UNIV AT BOULDER DEPT OF COMPUTER SCIENCE
Pagination or Media Count:
Programming is difficult and hard to learn. The evidence is all around us application programming backlogs that grow, not shrink scientists who must work through intermediaries to analyze data or automate experiments engineering students spending months or years to develop needed programming skills. Despite the proliferation of languages since Fortran, and many undeniably useful innovations, it is hard to point to any development within the programming mainstream that has had decisive impact in making programming not just a little easier, but much easier. If researchers are to make progress with this pressing problem it seems likely that they must divide and conquer. They cannot expect ideas that improve the lot of professional systems programmers to help the nonprogrammer scientist, since the professional programmer can afford a much bigger investment in understanding esoteric concepts and methods than the scientist who wishes to devote his or her effort to science, not programming. In this paper, the author focuses on the nonprogrammer, the person who wants to use the computer flexibly with minimum study of it. He will review ideas in programming technology that seem to have promise in making programming much easier to learn and do for the nonspecialist. Because of growing interest in user interfaces, including construction of visual representations of scientific data or of complex systems under analysis and because of his own research interests in these topics, he will pay special attention to ideas that can make the handling of interactive graphics and animation easier than it is now. This review will be subjective and incomplete. The authors aim in presenting it is to interest readers in this problem, and to learn from them about aspects of the problem, or pertinent research, that he has overlooked.
- Computer Programming and Software