Friday, August 26, 2005

Discussion of Resilience and Consilience in Social Networks

Earlier this month Zenpundit posted an interesting article on socal networks, with a focus on resilience and consilience. Below is his entry, followed by my comments that he graciously posted on his blog.


A while back, Dr. Barnett and Critt Jarvis entered in to a "strategic alliance" between The New Rule-Sets Project and Enterra Solutions, which is the baby of Stephen F. DeAngelis to develop " Enterprise Resilience Management"(TM). It would seem to be at once a concept, a service and a systemic software tool for organizations to efficiently manage dynamic changes in regulations, security, information flow and market environment. From Enterra's website:"Resilient organizations turn security, compliance, information integration and business process management from non-strategic cost items into the strategic components of a sustainable competitive advantage. The positive benefits of Enterprise Resilience Management™ range from increased valuation, marketability and corporate responsibility to a lower cost of insurance and lower total cost of ownership. Additionally, ERM assists in lowering potential damage to an organization's reputation and critical assets. This helps to create internal controls and solutions that protect senior executives and organizations from legal liability."The target demographic are corporations, government agencies and militaries. I'm not qualified or familiar enough to discuss the software aspect but I find the focus on " Resilience" to be very important conceptually.

DeAngelis has written about his ideas on cultivating organizational resilience here and here. Like Tom, DeAngelis is a visionary writer so his pieces tilt toward shifting your perspective on old worldviews and like Dr. Barnett he understands that freely evolving complexity in systems has significant ripple effects - hence his making " resilience" the core of his philosophy.Why is this important ? " Resilience" in free scale networks refers to how resistant the network is removal of its nodes ( removing a node lowers the efficiency of the network by increasing the distance between nodes or disconnecting them entirely). Corporations, government agencies - all groups in fact - are networks. Because most formal organizations in American society still carry the structural and cultural legacy of the industrial revolution they tend to be hierarchical, vertically-organized, culturally-rigid and are less than resilient. Take out key actors - the " nodes" -and institutional paralysis ensues. Possibly collapse.

So the Enterra-NRSP partnership is really selling network efficiency and survivability. In PNM terms, engineering a robust defensive capability against System Perturbations that would allow an organization reeling from cascading effects to " bounce back" from an attack. As I said earlier, resilience a key concept and quality in terms of importance. But what about...offense ? Or expansion of the network or the network's radius of influence ? What about structuring an organizational network to gear its behavior, culture and strategic thinking in terms of
"Consilience " as well?

Consilience was a term rescued from obscurity by Edward O. Wilson, the famous sociobiologist in his book of the same name that means a " jumping together" or unity of knowledge. Consilient thinkers look for the common underlying Rule-sets in disparate phenomena ( all phenomena at their most ambitious) - like Horizontal thinkers they are seeing connections across domains but the interests of Consilient thinkers are directed at the root level - the fundamental laws, principles and axioms applicable to all domains. In Wilson's words:"The trend cannot be reversed by force-feeding students with some of this and some of that across the branches of learning; true reform will aim at the consilience of science with the social sciences and the humanities in scholarship and teaching "You can't get a whole lot more horizontal than that ! What would be the advantages of building " Consilience" in to a network's structure, system and culture ?

Survivability: Like resilience, a high degree of consilience in a network would be likely to improve the network's longitudinal prospects by adapting efficient non-zero sum Rule-sets.

Influence: By adapting principles, practices and concepts that other networks find analogous to their own, the message of the network has more memetic appeal by virtue of being more readily comprehensible.

Compatibility: As with communication and influence, common Rule-sets make potential cooperation, alliances and mergers with other networks more likely as well as more harmonious.
Adaptability: Members of networks with a consciously consilient culture are more apt to themselves become better horizontal and creative thinkers. Their
OODA cycle may be faster because they are all - collectively and individually - seeing farther and to wider horizon.

How consilience would be designed in terms of software applicatons is something far beyond my ken but it would seem to be a fruitful conceptual field to explore."

My comments:

"The idea of 'resiliency' is important in scale-free networks. While there are many nodes in any sort of complex network, whether social, business, electronic (i.e. Internet), biological (food webs, metabolic processes, etc.), or other, what makes a network scale-free is that some small number of the nodes have many more links than the vast majority of nodes (which only have a few links). These highly linked nodes are the hubs of the network, and in some sense are responsible for holding the network together.

From the standpoint of software, perhaps the biggest fear is the computer virus wiping out a company's computer network. Of course, the obvious choice is to hit the network servers and routers, which are the hubs. And these hubs are the most obvious parts of the network to protect. But what one cannot forget is that if nodes on the periphery are infected, it is very difficult to kill the virus completely.

Now add in Wilson's idea of 'consiliency.' How can a network make use of fundamental principles from a variety of fields to enhance the performance of the entire network? In everyday terms, to me this almost sounds like multitasking. One needs to have members of the network who have studied and are trained in multiple fields, or small numbers of individuals who know something about a lot of different fields...research shows this multitasking tends to *reduce* productivity if you take the individual route. I may be a bit off on this, but in network theory, there is a hierarchical structure to some real networks that was discovered in ~2002. There are naturally forming, self-emergent networks within networks. There is still a scale-free mathematical structure to the more complex networks, and they are now called modular networks. A large company does this by having different departments, which by themselves are networks of workers. But the hubs, department managers, perhaps, are the links between the departments (modules) to form an ever more complex structure. The Internet and biological cell are naturally occurring modular networks, and the more people look, the more this structure is found in real networks.

Modularity makes use of a variety of local information for the global success of the overall network. The fact that this occurs naturally through the evolution of many types of networks is intriguing. Perhaps this is what Wilson's intuition was telling him. If I were a manager, I suppose I would encourage interaction between my department and others, to cross-feed each other with our knowledge and find out how to push the boundaries of our business.

This is one thing I wish happened more in schools, as Wilson also suggests in education, because teaching techniques and methodologies can be used across disciplines and subject areas...this seems to be an efficient and effective way of promoting horizontal thinking, because teachers can break away from 'standard' ways of teaching our own subject and learn some new ways of teaching from someone else in a different department. We need to take advantage of the departmentalized, intellectually specialized modules in such networks in order to help find new insights and breakthroughs. "

No comments: