Software reuse: From library to factory

Abstract

Systematic Software reuse is a key business strategy that software managers can employ to dramatically improve their software development processes, to decrease time-to-market and costs, and to improve product quality. Effective reuse requires much more than just code and library technology. We have learned that careful consideration must be given to people, process, and technology. One approach to the systematic integration of these three elements is the concept of the software factory. At Hewlett-Packard Co., we have initiated a multifaceted corporate reuse program to help introduce the best practices of systematic reuse into the company, complemented by multidisciplinary research to investigate and develop better methods for domain-specific, reuse-based software engineering. This essay discusses our experiences. Key aspects include domainspecific kits, busi:-tess modeling, organization design, and technology infrastructure for a flexible software factory. T he phrase software crisis was first used in 1969 to describe the ever-increasing burden and frustration that software development and maintenance have placed on otherwise happy and productive organizations. Since then, managers have been looking for effective strategies to deal with software. Manufacturers of computer systems and instruments, such as Hewlett-Packard Co. and IBM, whose businesses relied mostly on hardware and mechanical engineers, today find that over 70 percent of their research and development engineers are working in the areas of software and firmware. Maintenance and rework account for about 60 to 80 percent of the total software costs. Systems take longer to produce than expected, and software is frequently on the critical path. by M. L. Griss Among the many solutions proposed to address this software crisis, the systematic application of software reuse to prototyping, development, and maintenance is one of the most effective ways to significantly improve the software process, shorten time-to-market, improve software quality and application consistency, and reduce development and maintenance costs. While many companies are developing proprietary software libraries, software reuse is not yet a major force in most corporate software development. We believe that this is largely because effective reuse depends more on socioeconomic than on technical factors at this time,4,5 while most users still concentrate on library or language technology.

Code Reuse in Open Source Software

“…Systematic reuse in software development firms requires years of investment (Frakes and Isoda, 1994) in order to create and maintain reusable code and other knowledge (Lim, 1994), populate repositories and libraries (Griss, 1993;Poulin, 1995), and provide tools for developers to identify and reuse code (Isakowitz and Kauffman, 1996). The organization's funding structure usually needs adaptation to coordinate reuse investments across the organization (Lynex and Layzell, 1998), because developers need to work in repositories and components that are not directly linked to a product.…”

Object technology and reuse: lessons from early adopters

“…8 Although reuse scholars have argued for years that achieving systematic reuse requires a host of organizational and process changes that go beyond technology, 9 it's not realistic to expect early adopters to know everything that might pertain to a technology they are adopting. Part of the burden of being an early adopter is the lack of understanding about the nature of the innovation being adopted, and the managerial practices required to achieve the desired results.…”

“…10 And for good reason: systematic reuse of this sort, unlike most advances in software development practice, may well be able to deliver the long-sought-after 10x increase in programmer productivity-at least under the right circumstances. But although OO provides some key technological enablers for systematic reuse, reuse consistent with this vision is rarely achieved: 9 I have not experienced a great amount of reuse, and I don't know anyone who has. It's still too early to tell how effective reuse can be...It requires a discipline and a support infrastructure.…”

The Evolving Object of Software Development

“…7 On the other hand, and echoing the broader literature, there is considerable The Evolving Object of Software Development Paul S. Adler concern over its bureaucratic nature. (In this ambiguity, CMM resembles the broader family of 'software factory' concepts of which it is a part; on the concept of software factory and the associated debates, see Cusumano, 1991;Swanson et al, 1991;Griss, 1993;Weber, 1997;Friedman and Cornford, 1989;Greenbaum, 1979;Kraft, 1997. ) In partic- The necessary process discipline is in place to repeat earlier successes on projects with similar applications.…”