A Simple Definition of Complexity
By Jim Johnson…
The other day, two of my master students at the Antwerp Management School asked me for a simple definition of complexity. My first answer was, “Complexity is like porn—you just know it when you see it. And I think it depends on the individual’s perspective. For me, brain surgery is complex, but chewing gum and walking at the same time is easy. For some brain surgeons I know, chewing gum and walking at the same time would be complex, while brain surgery for them is just another day at the office.” However, my students’ question got me thinking, so I asked two of the most knowledgeable people I know in the area of complexity.
Dr. Aloysio (Lou) Vianna, our South American Research Director, has a PhD in Complexity Theory. Lou says, "We all run after happiness and despise complexity. But both have something in common: they do not exist by themselves. Complexity is the main consequence of life! We admire life, but we do not understand it. Complexity is the astonishment caused to us by thousands of interconnected factors that continually interact with us. But our brains can only manage one "full hand" of factors at a time. We have to learn to simplify life.”
Dan Ward, author of The Simplicity Cycle: A Field Guide to Making Things Better Without Making Them Worse, says, “My definition of complexity is ‘consisting of interconnected parts,’ so something can have either a high degree of complexity or a low degree of complexity.” According to Dan, “high" and "low" degrees of complexity are context-dependent. “For example,” he says, “one hundred interconnected parts would be a very simple spacecraft—and a very complex pencil sharpener. Having a low degree of complexity basically means something is simple, so the best definition of simplicity I can offer is ‘consisting of few interconnected parts.’"
Every project in the CHAOS Database is rated for complexity. After a project profile is completed, it’s sent to an adjudicator and reviewed. The adjudicator will review the project for several traits or characteristics and assign weights or ranges. One of these traits is complexity. Complexity has a five-point range, from very complex to very easy. The adjudicator will look at several parts of the profile to assign the complexity range. These parts include:
· Number of users and types of users involved
· Number of executives/stakeholders involved
· Number of departments/divisions involved
· Opinion of the degree of difficulty by the person completing the profile
· Is this project breaking new ground, or has it been done before?
· How many people are working on the project?
· Are the members of the staff internal or outsourced?
· The skills of the project team
· The skills of the project sponsor
Before assigning a complexity rate, the adjudicator will consider the application type and review the open-ended worksheet questions. The adjudicator will also consider if the process and methodologies used in the development and implementation of the software are complex or simple. However, the two main considerations are the number of interconnected parts and the organization’s skills. Our adjudicators have reviewed hundreds, if not thousands, of projects. They have compared each one’s degree of complexity against that of other projects and environments. In other words, they know it when they see it!
See how complexity affects software project resolution. Get your copy of the new CHAOS Report: “Decision Latency Theory: It’s All About the Interval,” in our store.
Jim Johnson is a professor at the Antwerp Management School. He is also the founder and chairman of The Standish Group. Johnson has been professionally involved in the computer industry for over 50 years and has a long list of published books, papers, articles and speeches. He is also responsible for a combination of technical, marketing, and research achievements focused on mission-critical applications and technology. Johnson is best known for his research on project performance and the early recognition of technology trends. A pioneer in modern research techniques, he continues to advance the research industry through case-based analytical technology. Currently, Professor Johnson is experimenting with a new teaching technique known as “Nanoclassing.”