RESEARCH IN MACHINE-INDEPENDENT SOFTWARE PROGRAMMING
Abstract
During the past decade the 'credibility gap' in software specification and construction has widened alarmingly, while that in hardware specification and construction has narrowed until the time and cost of specifying and constructing rather major hardware systems now can be confidently predicted. The lessons learned from the software 'failure' of many large-scale command and control systems, of major commercial and university time-shared and batch operating systems, and of attempts to provide 'complete' language facilities such as promised by PL/I have not yet resulted in the development of a basic program construction technology which permits the prediction and control of software costs and facilities. The cost of construction of programs of large size or complexity on a 'one-shot' basis is difficult to predict in any case. Of more importance, however, is the fact that even when a large program finally operates in an acceptable fashion the program is completely un-transferrable. The object of the current project is to develop solutions for these problems of software programming technology. The project activities consist of research into machine-independent techniques for programming computer software, of the design and implementation of certain specific language and system facilities, and of experimentation with the machine-independent specification and implementation of typical software components.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1968
- Accession Number
- AD0683362
Entities
People
- Carlos Christensen
- T. E. Cheatham Jr.