The high cost of software
Richard H. Thayer, John B. Slaughter, Barry Boehm, Judith A. Clapp, John H. Manley, James H. Burrows
Abstract
Richard H. Thayer, John B. Slaughter, Barry Boehm, Judith A. Clapp, John H. Manley, James H. Burrows
Abstract
In any large scale data processing development the cost of software development is three to five times the cost of hardware. It is estimated that in 1980's this development cost will jump to ten times that of the expenditures predicted for the hardware of that era. If the Government or industry is to maintain a ceiling on this cost, research must be conducted on ways to reduce costs in the various phases of software development such as design, specification, implementation, etc. For purposes of this session, "cost" will encompass not only the resources necessary to develop the software or transfer it, but also costs associated with delays in both delivery and "fixing," and resources wasted with software of poor quality. This session is an outgrowth of a tri-service/industry symposium entitled "The High Cost of Software" which was held at the Navy Post-Graduate Center, Monterey, California from September 17--19, 1973, in which service, industry and university experts attempted, for the first time, to get a better understanding of the costs of computer software and to make some recommendations as to what could be done about it. This session deals with the causes of software costs, the identification of both specific trouble areas and proposed R&D approaches that could be taken to eliminate the negative impact these areas have, and the particular ongoing research projects aimed at reducing software costs.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In any large scale data processing development the cost of software development is three to five times the cost of hardware. It is estimated that in 1980's this development cost will jump to ten times that of the expenditures predicted for the hardware of that era. If the Government or industry is to maintain a ceiling on this cost, research must be conducted on ways to reduce costs in the various phases of software development such as design, specification, implementation, etc. For purposes of this session, "cost" will encompass not only the resources necessary to develop the software or transfer it, but also costs associated with delays in both delivery and "fixing," and resources wasted with software of poor quality. This session is an outgrowth of a tri-service/industry symposium entitled "The High Cost of Software" which was held at the Navy Post-Graduate Center, Monterey, California from September 17--19, 1973, in which service, industry and university experts attempted, for the first time, to get a better understanding of the costs of computer software and to make some recommendations as to what could be done about it. This session deals with the causes of software costs, the identification of both specific trouble areas and proposed R&D approaches that could be taken to eliminate the negative impact these areas have, and the particular ongoing research projects aimed at reducing software costs.
Key concepts: Computer science, Software development, Software, Session (web analytics), Software engineering, Engineering management, Engineering, Operating system