Strategic Analysis of Component-Based Development

John Favaro

Intecs Sistemi S.p.A.

Via Gereschi 32-34, 56127 Pisa - Italy

Tel: (+39 50)545-111, fax: (+39-50)545-200

Email: favaro@pisa.intecs.it

URL: http://www.pisa.intecs.it/

Component-based development is becoming one of the dominant paradigms in software reuse research (along with generative reuse); and in the software development industry, it appears to be becoming the dominant paradigm. As always, the discipline of software reuse economics should contribute to the evaluation of the costs and benefits of software development according to this paradigm. But up until now, we have mainly concentrated on measuring benefits in practicing reuse during "business as usual". For example, this may involve the measurement of programmer effort saved, or reduced maintenance effort. Strategic benefits such as increased market agility or the ability to take advantage of new opportunities have not been addressed in a systematic fashion–yet these are thought to be among the most valuable benefits of the component-based approach.

Work to date in software reuse economics has not yet adequately addressed the integrated combination of strategic and economic analysis that will be necessary for dealing with the component-based development paradigm.

A number of "Reuse Maturity Models" have been proposed to characterize the increasingly important roles that reuse can play in an organization. As one example, the Reuse Capability Model developed at the Software Productivity Consortium, foresees four levels:

Opportunistic. Project-level reuse. Reusable assets are acquired according to the technical needs of the project development teams.

Integrated. Organization-level reuse. Repositories, standards, etc.

Leveraged. A product-line reuse strategy is defined. Metrics increasingly important.

Anticipating. Management creates new business opportunities by exploiting the organization’s reuse capabilities. Future, anticipated market needs drive the acquisition of reusable assets. New technologies are seen as strategic drivers. The reuse infrastructure is flexible enough to adapt rapidly to market evolution.

A large body of work has been carried out to develop metrics and economic models to measure the cost reduction and various quality and productivity factors that are relevant to the first two levels. Little work in strategic planning exists for these levels of the Reuse Capability Model, largely because it is not really necessary--reuse at these levels is usually aimed at the improvement of existing projects and processes. The Leveraged level introduces a shift in perspective from operational to strategic value. Still, the benefits of product-line reuse can be captured to a great extent with traditional measures already available from previous work in reuse economics.

At the Anticipating level, strategic value becomes dominant. This also reflects the current reality in software companies–it has been estimated that up to 70% of the value of software companies is strategic. Purveyors of component and application framework-based technology refer almost exclusively to the strategic advantages they confer on an organization.

I believe that an integrated approach to the economics of software reuse at the Anticipating level is necessary–and by extension to the analysis of component and framework based development. The integrated approach considers strategy and valuation together rather than separately. The basis for work in strategy is the discipline of Value Based Management (VBM). I am cooperating with Marakon Associates, the pioneer of this approach to business strategy. The basis for work in valuation is traditional Present Value approaches augmented by the discipline of real options. The integrated approach has been named Value Based Reuse Investment.

The work of Poulin to date represents the most comprehensive treatment of reuse metrics and economic models to date, and has been extended to consider the economics of product lines. Jacobson, Griss, and Jonsson write at considerable length about strategic aspects of component and framework based development. Lim writes about the shift in perspective from operational (cost reduction) benefits to strategic benefits of reuse. Withey has made a strategic analysis of product line investment, where the concept of options was introduced. Sullivan has explicitly considered real options as a broad basis for the economic analysis of such fundamental software design processes as prototyping.

[Favaro98] Favaro, J., K. Favaro, P. Favaro, "Value Based Software Reuse Investment," Annals of Software Engineering (5), 1998, pp. 5-52.

[Jacobson97] Jacobson, I., M. Griss, and P. Jonsson, Software Reuse: Architecture, Process and Organization for Business Success, Addison-Wesley Longman, Reading, MA, 1997.

[Lim98] Lim, W. "Strategy-driven reuse: Bringing reuse from the Engineering Department to the Executive Boardroom," Annals of Software Engineering (5), 1998.

[McTaggart94] McTaggart, J. M., P.W. Kontes, and M.C. Mankins, The Value Imperative, The Free Press, New York, NY, 1994.

[Poulin97a] Poulin, J., Measuring Software Reuse: Principles, Practices, and Economic Models, Addison-Wesley, Reading, MA.

[Poulin97b] Poulin, J., "The Economics of Software Product Lines," International Journal of Applied Software Technology 3, 1, , 1997, pp. 20-34.

[Stewart98] Stewart, Thomas A., "Marakon Runners," FORTUNE Magazine, 28 September 1998.

[Sullivan97] Sullivan, K.J., P. Chalasani, and S. Jha, "Software Design Decisions as Real Options," Technical Report 97-14, Department of Computer Science, University of Virginia, Charlottesville, VA, 1997.

[Withey96] Withey, J., "Investment Analysis of Software Assets for Product Lines," Technical Report CMU/SEI-96-TR-010, Software Engineering Institute, Pittsburgh, PA, 1996.

John Favaro (favaro@pisa.intecs.it) Intecs Sistemi, http://www.pisa.intecs.it/

I have been working in the area of software reuse economics since about 1990. In 1995 I began collaboration with Martin Griss of Hewlett-Packard and with the Italian telecom industry, in which we worked on domain analysis methods aimed at an improved framework and component-based development process. We used the Reuse-Based Software Engineering Business as a basis for this work. This work in domain analysis for CBD led me to reconsider software reuse economics from the point of view of strategic infrastructure investment.