In a community college building in Austin, Texas, the Army is doing something nothing short of amazing. Army Futures Command's Army Software Factory (ASF) is at once providing useful software products and taking the first steps toward addressing DoD's most pressing nongeopolitical issue, absorptive capacity.
Leveraging commercial software development practices such as minimal viable product, Army enlisted service members and officers work together in teams to produce products—but that just scratches the surface. Unlike the Air Force's Kessel Run, the work of ASF is all done by uniformed soldiers, assisted by contractors as mentors and instructors. When they rotate out, those soldiers take their knowledge, understanding, and wisdom with them, bringing indispensable absorptive capacity to the modern battlefield. These are not soldiers who will need contractors to tell them what can and cannot be done, they will lead from wisdom.
Absorptive capacity has withstood the test of time on the commercial side and remains a benchmark in leadership development.
Those two concepts, absorptive capacity and wisdom carry a lot of weight in innovation. Unlike six sigma, agile software development, model-based systems engineering, and a laundry list of DoD management approaches over the decades, absorptive capacity has withstood the test of time on the commercial side and remains a benchmark in leadership development. Since first published in 1990, the need for subject matter wisdom in leadership has grown to universal recognition as a key factor in innovative development at the speed of need. While this has become the standard in commercial industry, DoD has taken the opposite tack of outsourcing to contractors more and more of the top end of Russel Ackoff's cognitive hierarchy.
Figure 1: The Cognitive Hierarchy
Figure by Alyson Youngblood/RAND
Cognitive hierarchy pyramid showing in descending order: Wisdom, Understanding, Knowledge, Information, and Data with definitions of each on the right of the pyramid as below:
Wisdom conveys values and the exercise of judgement beyond the physical and temporal boundaries of the system.
Understanding is conveyed by goal-based explanations, answers to why questions.
Knowledge is what makes possible the transformation of information into instructions to control a system—sometimes called know-how. It’s the basis for designing a new system.
Information is processed data, answers to questions that begin with such words as who, what, when, where, and how many that derive directly from the data. Queries such as: write software code that does X or develop a proof that trisects an angle return information.
Data are symbols that represent the properties of objects and events.
Source: RAND analysis and Ackoff, R.L. (1999) Ackoff’s Best. New York: John Wiley & Sons, pp 170-172.
Please don't misunderstand, the Pentagon overflows with wisdom. It is the finest military machine the world has ever known. But when it comes to developing and integrating the bleeding-edge technology required for the modern, multi-domain battlefield, the requisite knowledge, understanding, and wisdom are seen as critical in terms of all three management dimensions: cost, schedule, and quality. Real geopolitical pressure combined with military cultural belief leads quickly to: Don't you know there's a war on? Addressing this urgent need leads to outsourcing because conventional DoD culture reflects the belief that it costs too much, takes too long, and the quality won't be as good if we develop and build in-house.
This is where ASF is getting it so right. The data show that it doesn't cost more, it doesn't take longer, and the quality is every bit as good when active-duty soldiers develop code. Soldiers make mistakes and even fail, a part of the learning process embraced in ASF and not many other places in DoD. While the concept of fail-forward is not a secret, it still hasn't broadly taken hold in the competitive military culture. These contractors don't do, they teach and help the soldiers to learn from those mistakes. The money spent on ASF contractors not only results in software products, but ultimately in reduced absorptive capacity as those soldiers rise through the ranks, bringing their wisdom with them.
ASF doesn't cost more, it doesn't take longer, and the quality is every bit as good when active-duty soldiers develop code.
Go to any defense site or publication and you will see references to machine learning, autonomous systems, and development at the speed of war. Currently, Pentagon decisionmakers require substantial outside support to make the critical decisions required for this environment. Is today's deployable AI the magic elixir it is marketed to be? What would a battlespace drone development loop look like? What are the actionable lessons learned from the 155mm howitzer deployment in Ukraine?
Policymakers need to ask if partial outsourcing of the top three-fifths of the cognitive pyramid is sustainable and, if not, does ASF hold the key? It isn't and it does.
