Evolutionary technology development practices are all the rage. The various agile and extreme programming frameworks have captured the imagination of the commercial software world and are starting to become a major influence on the Defense industry. Based on the increasing usage of the word ‘agile’ in the Defense dialogue, I am expecting to see the type of aggressive oversubscription that has accompanied previous commercial crossover fads like Total Quality Management and Lean Six Sigma. But before DoD and the Defense industry make a wholesale commitment to agile methodologies (to the exclusion of legacy approaches), there are a couple of important points to keep in mind.
Without question, the agile idea lends itself to the type of threat scenarios currently confronting DoD. In the absence of a steady state operating environment, agile development, where capabilities are continuously optimized around the margins, makes a lot of sense. It gets back to Peter Sims’ “Little Bets” concept that I have discussed in previous posts – start with what you know and evolve in measured steps based on experiential feedback.
But the agile design philosophy doesn’t mitigate the need to think through a problem from start to finish. On the other hand, thinking through the problem doesn’t mean meticulously defining every operational excursion likely to be encountered during the life of a technology. Big Design Up Front (BDUF) is often mistakenly viewed as a counterpoint to agile design. To be sure, the fundamentalist interpretation of BDUF, where an excruciating level of planning overwhelms the adaptation imperative, is at odds with agile precepts; however, I would argue that this should not necessarily be the case.
A marriage of agile and BDUF is required to produce the types of solutions demanded by contemporary military problems. By way of analogy, the reason that jazz works is that it reflects a disciplined and unifying structure around which improvisation can take place. Done right, agile development incorporates a similar pre-considered discipline that can only emerge through the application of a BDUF approach. No amount of evolution can make up for a faulty initial design. And no amount of up front planning can anticipate the complexities of modern military conflicts. Agile and BDUF should not be viewed as competing methodologies within the context of DoD R&D; they are two sides of the same coin.