Visual Complexity
I recently found a great website at Visual Complexity. It is a list of networked data displays or visualizations. They are beautiful to behold.
Networks are everywhere and networks are extremely important to the human race. Networks define our supply chains, how airlines run, the power grid, the internet, disease propagation and the 6 degrees of separation between you and any other human being on earth. They permeate our lives and we are only now beginning to understand how to analyze them. The emerging science of networks can tell us how robust a network can be based on the 'type' of network and the 'kinds' of connections. It can inform us how a network will perform and characterize the kinds of behavior we can expect from it. When you consider the intricate web of connections that transport essential commodities, resources, diseases and ideas across the globe you begin to see the value of comprehending networks. Visual displays can play a vital role in that comprehension.
In 1997 Toyota lost a key manufacturing plant in a fire - known as the Aisin Crisis. The company was suddenly forced to respond to a supply chain crisis that could have killed a global multi-billion dollar company. The plant manufactured critical brake valves used in most of its cars. Running a just-in-time inventory system meant that global production would be completely halted. Because of a robust social/industrial network the greater Toyota group company employees were able to engineer, refit, and recover from what could have been a total loss. The Wall Street Journal noted "36 suppliers, aided by more than 150 other subcontractors, had nearly 50 separate lines producing small batches of the brake valve in less than a week. In one case, a sewing-machine maker that had never made car parts spent about 500 man-hours refitting a milling machine to make just 40 valves a day." Understanding networks means understanding how to recover from catastrophe or leverage extraordinary capabilities.
Scientists are discovering there are universal principles that apply to all networks. Electrical grids, the internet, social networks, food chains, and transportation networks can all be analyzed using the emerging science of networks. The science of networks has developed rapidly over the last half decade. A good primer is "Six Degrees: The Science of a Connected Age" by Duncan Watts. In this book, published in 2004, Watts presents a useful overview of how the thinking on networks has developed over the last half century. He and his colleagues have defined types of networks and developed the mathematics to explain the universal behavior and characteristics of networks. This book introduces the leading thinkers, their ideas, and the interdisciplinary way in which the science of networks has been advanced. It provides some powerful anecdotes of how our lives are impacted and influenced by networks. I was enthralled.
Like the science of networks, presenting and sharing information about networks is an emerging area of information design. The best way to display networks is still experimental and the Visual Complexity site gives us a lot to consider. It takes a long time to build generally accepted conventions for data display. Information design began with maps over 5000 years ago, and statistical data wasn’t presented in them until the 1100’s. The first graph was invented in the 17th century, and time-series and bar charts were not seen until the 18th century.[1] The paragon of network data display has not yet arrived and it will be some time before it does.
The Visual Complexity site offers a lot of beautiful designs – most of them created in the last few years. Interestingly, the utility of these displays is sometimes very limited. There are several reasons for this. The essence of a network is captured in the relationships with which it is constructed and network visualizations can be found that display hundreds, thousands and even tens of thousands of relationships. Network data is not necessarily reducible as is statistical data. It is hard to simplify without losing useful information. Also, networked data is frequently labeled. Because labeled data is conceptual it suffers from problems of accessibility, overlapping graphical and textual elements, and limited feature extraction.
Effectively communicating the important features of a network and the relevant concepts on display is a very difficult problem to solve. These are challenges faced by those of us in the business of strategic scenario analysis, project planning, contingency planning, and risk analysis. It will be fun to watch the technology of network visualizations develop over the next decade.
1. Edward R. Tufte – The Visual Display of Quantitative Information; Visual Explanations
Networks are everywhere and networks are extremely important to the human race. Networks define our supply chains, how airlines run, the power grid, the internet, disease propagation and the 6 degrees of separation between you and any other human being on earth. They permeate our lives and we are only now beginning to understand how to analyze them. The emerging science of networks can tell us how robust a network can be based on the 'type' of network and the 'kinds' of connections. It can inform us how a network will perform and characterize the kinds of behavior we can expect from it. When you consider the intricate web of connections that transport essential commodities, resources, diseases and ideas across the globe you begin to see the value of comprehending networks. Visual displays can play a vital role in that comprehension.
In 1997 Toyota lost a key manufacturing plant in a fire - known as the Aisin Crisis. The company was suddenly forced to respond to a supply chain crisis that could have killed a global multi-billion dollar company. The plant manufactured critical brake valves used in most of its cars. Running a just-in-time inventory system meant that global production would be completely halted. Because of a robust social/industrial network the greater Toyota group company employees were able to engineer, refit, and recover from what could have been a total loss. The Wall Street Journal noted "36 suppliers, aided by more than 150 other subcontractors, had nearly 50 separate lines producing small batches of the brake valve in less than a week. In one case, a sewing-machine maker that had never made car parts spent about 500 man-hours refitting a milling machine to make just 40 valves a day." Understanding networks means understanding how to recover from catastrophe or leverage extraordinary capabilities.
Scientists are discovering there are universal principles that apply to all networks. Electrical grids, the internet, social networks, food chains, and transportation networks can all be analyzed using the emerging science of networks. The science of networks has developed rapidly over the last half decade. A good primer is "Six Degrees: The Science of a Connected Age" by Duncan Watts. In this book, published in 2004, Watts presents a useful overview of how the thinking on networks has developed over the last half century. He and his colleagues have defined types of networks and developed the mathematics to explain the universal behavior and characteristics of networks. This book introduces the leading thinkers, their ideas, and the interdisciplinary way in which the science of networks has been advanced. It provides some powerful anecdotes of how our lives are impacted and influenced by networks. I was enthralled.
Like the science of networks, presenting and sharing information about networks is an emerging area of information design. The best way to display networks is still experimental and the Visual Complexity site gives us a lot to consider. It takes a long time to build generally accepted conventions for data display. Information design began with maps over 5000 years ago, and statistical data wasn’t presented in them until the 1100’s. The first graph was invented in the 17th century, and time-series and bar charts were not seen until the 18th century.[1] The paragon of network data display has not yet arrived and it will be some time before it does.
The Visual Complexity site offers a lot of beautiful designs – most of them created in the last few years. Interestingly, the utility of these displays is sometimes very limited. There are several reasons for this. The essence of a network is captured in the relationships with which it is constructed and network visualizations can be found that display hundreds, thousands and even tens of thousands of relationships. Network data is not necessarily reducible as is statistical data. It is hard to simplify without losing useful information. Also, networked data is frequently labeled. Because labeled data is conceptual it suffers from problems of accessibility, overlapping graphical and textual elements, and limited feature extraction.
Effectively communicating the important features of a network and the relevant concepts on display is a very difficult problem to solve. These are challenges faced by those of us in the business of strategic scenario analysis, project planning, contingency planning, and risk analysis. It will be fun to watch the technology of network visualizations develop over the next decade.
1. Edward R. Tufte – The Visual Display of Quantitative Information; Visual Explanations
