Self-organizing Rule Sets

Stephen DeAngelis

May 19, 2006

One of the things that attracted me to Tom Barnett’s work was his understanding of the importance of rule sets. My vision of Enterprise Resilience Management relies heavily on rule sets and their automation in order to reduce human errors, improve performance, and increase security. Most people have an inherent understanding of the importance of rule sets in the business world, and they assume that some governing body enacts and then enforces them — oversight groups like the World Trade Organization. When I talk about rule sets and standards as they relate to Development-in-a-Box, however, who makes those rule sets and how they are enforced is much less clear to people. This is especially the case since I do not envision a “governing body” that either oversees the process or enforces its implementation. Development-in-a-Box relies on the voluntary cooperation and collaboration of disparate groups; hence, the ambiguity about rules sets.

In an interesting post by John Robb on his blog Global Guerillas, he points out an interesting paper by a high school student who examines how even informal networks, without any oversight, can evolve rule sets. Robb writes:

Here’s an interesting paper, by Alexander Franks, a very talented young man who was a finalist in a national competition. He explores, through the use of genetic algorithms and small world connections, how dispersed networks generate a coordinated rule set in noisy environments. This is an interesting topic since it is not at all obvious how open source networks develop cohesive rules sets — this in contrast to hierarchical systems that can propagate rules through central direction. In sum, his work suggests that one or two widely held rules (greater than 50% adoption) provide the basis for the evolution of an entire set. All rules that have affinity to those founding rules evolve until they are widely adopted. All minority rules that do not have much affinity are flushed. This has interesting applicability to open source warfare. It suggests that the plausible promise (the idea that starts the open source warfare community) provides a center of gravity that attracts rules that advance it and repels those that don’t. Any additional work on this topic is welcome.

I envision exactly this kind of phenomenon emerging as Development-in-a-Box takes wing. In a future blog, I want to talk more about communities of practice and how they self-organize because I believe it occurs using a phenomenon similar to that described by young Mr. Franks. In the beginning of his paper, Franks quotes Mathur A. Moreira and his colleagues (Efficient System-wide Coordination in Noisy Environments, Proc. Natl. Acad. Sci. U.S.A. 101 (33), 12085–12090) who point out that there are many naturally occurring decentralized systems that operate by self-organizing rule sets, including:

Conventions and norms, social learning in animals and humans, as well as fads, rumors and revolts. Examples are also abundant in biology and medicine: the saltation model of human growth speaks of decentralized cellular coordination, as does the Dictyostelium’s transition to a multicellular developmental program.

I make no claims to understanding “the saltation model of human growth” or “Dictyostelium’s transition to a multicelluar developmental program,” but I do believe that systems self-organize and do so in an efficient way. This should come as no surprise to anyone who has read Malcolm Gladwell’s best-seller, The Tipping Point: How Little Things Make a Big Difference, or Howard Rheingold’s Smart Mobs: The Next Social Revolution. That is why I am optimistic that Development-in-a-Box will unfold in positive and effective ways without anybody holding a hammer over the heads of those who wish to participate.