Premium
8.4.4 The Discovery Based Development Approach: A Process Aberration Or A Better Way To Develop Complex Applications?
Author(s) -
Shupp Jeffrey K.
Publication year - 2000
Publication title -
incose international symposium
Language(s) - English
Resource type - Journals
ISSN - 2334-5837
DOI - 10.1002/j.2334-5837.2000.tb00428.x
Subject(s) - procurement , reputation , process (computing) , quality (philosophy) , computer science , state (computer science) , domain (mathematical analysis) , risk analysis (engineering) , business , data science , operations management , engineering , political science , marketing , law , mathematical analysis , philosophy , mathematics , epistemology , algorithm , operating system
The world of complex software systems has been marred by a less than stellar track record. These systems have largely been the domain of civil or military government procurement, as opposed to commercial industry, since the former institutions can absorb not only the tremendous initial costs and long development time, but also the intrinsically higher risk of initial failure. However, the commercial world is beginning to move quickly toward much more sophisticated applications, and evidence suggests similar or worse rates of failures as complexity is increased. One such complex system has been under development for five years at M&DS. In early 1998, it was headed for a classic default of specified capabilities on all major fronts: runtime performance, functional stability and quality. This paper examines the circumstances that led to this state, and the success that has since been achieved. Most important is not the fact that a recovery had been accomplished, but rather how a new approach, called out in this paper as Discovery Based Development , was instituted to repair the damage and reputation in weeks instead of months or years. Armed with this technique, and applied earlier in the normal incremental or spiral developments that typify the commercial marketplace, a much higher success rate is quite possible for complex systems.