Network Parameters: Harmony

1.2.5 Network Parameters Sitting in between the system and object level, the network parameters are a powerful interface between the top-down and the bottom-up. The network level parameters help to unravel the complexities of the system level indicators, reveal complex system interactions, test and explore them, and find the object level drivers of these behaviors. This, in turn, gives concrete handles on finding the best intervention points to improve system dynamics. The network parameters also create a verbal construction set, allowing teams to specify and explore in detail where certain dynamic patters come from, or can be engaged. Network parameters are on a higher level of abstraction than object level items. Because of this abstraction, it’s easier to extract a universal of parameters for networks. This standard set is printed on the left page. While standardized, for each challenge, we still check if these fit or are relevant on a case-by-case basis, but they are a great start. You can also choose to use a simplified set. The standardized network parameters in SiD are split into the RAH system indicators. Their names are acronyms, taking the first letter of each of their paramters. While their names are awkward, this may help you remember which paramters are in each set. The Resilience set is called CRAFTDCCV, the Autonomy set SSCNE, and those dealing with Harmony are called PEAIE. They’ve been derived from existing network science literature, and then changed, adapted, and refined in practice over about a 10 year period. On the next pages in this section, we discuss all parameters in detail, but first, we shortly look at the nature of these sets of network parameters. The network-level is all about collection of connections. This means each parameter reflects on all of the relationships in the system related to that parameter. These relationships can then be split into the object level to be investigated in greater detail. For example, the resilience parameter of diversity reflects on the diversity of all the relationships in a system. You can split this ‘diversity’ network parameter further up with ELSI, to specifically look at the diversity of energy systems, or transport systems, financial relations, and so on. The playground of system dynamics On the network level we find more powerful systemic behaviors that remain hidden to us on the object level. On a network level we are no longer interested in a single flow, object, or element. It exclusively concerns itself with groups of connections throughout the system. The network is about ‘how’, rather than ‘what’. For example, it matters more to the network-level that people can communicate instantly over long distances, rather than what’s being said, or what device they use to communicate. The network parameters can be observed from both the top-down as well as the bottom-up. From the top-down, they’re evaluated on how they contribute to the RAH system indicators. From the bottom-up, they are constructed by looking at the various connections between areas of interest on the object level. Because they can be approached from both sides, they are extremely flexible. In addition, they derive much of their meaning and definition to the context of the challenge at hand. Combined with their flexible nature, the network level is often a playground for creativity. Quantifying network parameters N etwork parameters are used most commonly in a qualitative sense. However, if need be, they can be used to construct quantitative simulation models using agent based simulation or other techniques. For example, a top-down evaluation of resilience, by scoring each CRAFTDCCV network parameter in a qualitative way, can also be investigated in a quantitative way from the bottom-up. Each separate ELSI element can be quantified and related with numerical analysis, in relation to each CRAFTDCCV parameter.

The harmony set: PEAIE PEAIE are the five main network parameters that inform Harmony. These allow you to investigate to what degree the system has internal tension, or equilibrium within itself. These are the balance between peace and war, orderly and disorderly society, and pressure towards revolution and rioting. These are the hardest to quantify of the three sets, and the most influenced by social science. The PEAIE set consists of the following: Power Balance Expression Access Inclusion Equity Common Network parameters It’s interesting to note that several network parameters are much more commonly used than others. ‘Diversity’, for example, has been used for decades as a field of exploration in the social sciences. Because it is an existing and widespread field of research, the knowledge about how diversity affects the dynamics of systems is better known than others. For example, it’s widely proven that increasing diversity improves the strength of socio-cultural systems. Likewise, ‘Transparency’ is a common term in policy development areas, as well as in governance. The presence of more commonly known parameters helps us to see the power and wide ranging effects these parameters can have. But, we should take care to understand that this is not the case for all of these paramters yet. We’re less knowlegeable about the effects and dynamics of a paramter such as ‘redundancy’ in the evaluation of a social system. This, we should however take care not to miss out on network parameters that are less familiar and which may in fact bear powerful dynamics. Mapping Interdependency The power of each parameter becomes apparent when seen in the greater context of the system. For example, transparency of governance makes a country’s system work faster, better, and increase Harmony’s Equity parameter at the same time. Things that seem very complex can become rather straightforward when having the network overview at hand. You see also that the parameters tend to influence each other. Increasing one parameter’s score is bound to affect other parameters. To use network parameters effectively, the trick is to create an analysis system that provides an overview of all the important outcomes. For example, using an excel sheet to list all the parameters, their area of influence, and giving them a qualitative score can give a quick overview of a system’s dynamics. See page 276 for a table showing a qualitative network paramter analysis. This way you and your team can start to tweak the system and evaluate possible pathways for improvement. Clarity From Network Parameters For many, the network parameters require time to get used to. Once you have them internalized, we hope you’ll experience the clarity these beautiful tools yield. Within the network parameters lie hidden opportunities we can use to find radically better solutions for current societal problems. Some of these we’ve already discovered. For example, with the benefits that decentralized energy production gives to the autonomy and resilience of communities. a box of network glasses A handy mental exercise to practice and help evaluate the effects of network paramters on a system is to imagine a box of eyeglasses. Each pair of glasses represents a network parameter and allows you to ‘see’ the network in a different light. For example, imagine putting on the ‘diversity’ glasses, and look through them at a system such as a village or a company. Look at all the elements you see in the village, and what ‘diversity’ may mean to this. While you’re doing this, also imagine pushing the parameters to an extreme. For example, when investigating diversity, picture a village - all the houses in this village may be identical, or each may each be unique. How about people’s ages, professions, and so on? By zooming into diversity, elements it comprises of appear and in themselves help you understand the impact these elements have on diversity. Practicing this way is a quick and easy way to get a grip on the effects of network paramters on the dynamics of different systems. Learn to see the light Network parameters are most helpful for strategic decision making where complex systems are involved. You can easily un-stuck yourself when faced with complex situations, once you’ve become adept at recognizing and interpreting the network parameters. Learning how the parameters affect one another in a system tells a great deal about a system’s overall behavior, and allows you to instantly see possible effects of a certain measure, strategy, or policy decision. Because of its power and value, it’s worth to stress that it is worth practicing this for a while. To get good at it, try doing it on a daily basis. Practice looking at the network parameters on any complex system you come across. For example, while waiting at the train station, take the imaginary box of glasses with you, and apply them to whatever you see. What do the parameters mean when thinking about the train network? In the supermarket, think about the supply chains of the products, and the operations of retail stores. Think about all the bikes you see, and the transport network in the city. Use them when looking at a city park, natural systems, political networks, business relations, wherever you see them. Before you know it, you can start to ‘see’ the behavior of complex systems intuitively. Quantifying Network parameters The network parameters can be used to physically measure the performance of a network, if you attach formulas to each indicator. This is useful to get to grips with patterns in systems that feel counterintuitive, and computer programs can also help to better understand these patterns. Remember the lessons of complex systems when you do this though: you can model to learn about the system’s dynamics, but you cannot model to predict complex systems. To start quantifying, you can do this either by hand, or use more advanced agent-based modeling or systems modelers. Some of these are discussed in the tools section. To start working by hand, you can determine a simple formula to each network parameter, given a certain context. In the exercise on the next page we show some ways to do this for the resilience set. If you need some help with this, there’s a wealth of knowledge out there on how to do quantitative network analysis. Wikipedia is a good starting point for finding formulas, just type in the network parameters name, and you’ll often be able to find suggestions for formulas. For example, the Wikipedia article on Centrality lists five distinct ways of measuring this. On the next page are two simple examples, one from a bottom-up network approach, using a numerical approach, and one for a top-down, qualitative approach.