Elevator Pitch

What you learned

After a student takes this course, he or she might be asked what he or she learned in this course. During this semester we are going to build an answer to this question

In this course I learned:

(write out an answer here. Not too long. Something you could say in one minute.)

  • What complexity is, as it applies to information systems (see answers in development below)
  • To begin with, what is complexity and how it emerges?
  • Is complexity theory esoteric or does it gel with well known principles? How should you approach it to understand it better?

What is complexity?

Complexity is… (Come up with a one or two sentence definition that you're happy with. This should be suitable for a spoken answer to a smart but clueless colleague.)

  • Complexity is the "boundary between order and randomness" (Ruthen), measured not only by quantity but also by orderliness.
  • Complexity is an emergent phenomenon of increasing parts, their increasing interrelationships and interactions; giving rise to new properties.

And then provide an elaboration (in the event that the elevator gets stuck between floors and you have a little time).

  1. (Potential) Measurements of Complexity:
    • Computing (algorithm) - measure the shortest algorithm capable of reproducing the original data (Ruthen)
    • Computing (machines) - categorize computers into model classes (based on size and memory structure). measure the shortest program that reproduces the data within the lowest possible model class (Ruthen)
  2. How does Complexity emerge:
    • The parts interact on many scales that lead to complex behavior. We need to study the principles that govern how the new features and new properties appear as we move from one scale to another; as the complexity increases. It is important to realize that the processes that occur on each level are significant in itself and it is vital to study not only the whole but also the individual units because each has a relevance of its own.
  3. Is Complexity emergence Behavioral:
    • Yes. Understanding complexity may sometimes need focusing on the behavioral aspects of the agents to better reflect real-world situations. This was evident in how modeling competition among the agents in the digital ecosystem led to eventually recognize the cooperation requirements. It further led to recognize the relevance of concepts like tit-for-tat that influence the agent behaviors in the system to match it with real-world scenarios.
    • Thereafter, the digital ecosystems embraced the concepts of competition and cooperation; on how an agent formulates hypotheses about other players and adjusts them/drops them based on the rival actions; an adaptive play balancing the other agent actions and pursuing alternate paths subsequently to adapt to other agent moves. Here we can see the emergent and adaptive nature of digital ecosystems to better reflect the real-world.

Approach to Complexity

  • Studying complexity is not esoteric. It needs a simplified approach. How we get this? We need to take a decentralized view at complexity. AT&T was a complex organization intractable to manage. But the decision to split it up made it more manageable and to simplify the complexity of it. In our examples, the broader patterns of sand drop were studied by seeing what happened with one sand grain at a time. The focus was on local interactions and how the next sand grain triggered an avalanche. Hence the focus was on localized and decentralized view wherein we were focusing on one sand grain at a time. The simplification in studying complex system lies in the decentralized view.
  • The system never reaches back the same equilibrium again. It only toggles between randomness and order to subsequently toggle between two critical states. It moves from one critical state to another, thus creating a new baseline each time. The system thus has an adaptive self-organization. It is same as the economy after recession never recovering to the original growth rate. It only reaches a satisfactory growth rate relative to the current position.

Relevance to IT

What does this class have to do with research in IT?

(Come up with an answer here. It should be as you would speak it, not as you would write in a formal communication.)

  • Information Technology tools can be used to model complex systems, many of which occur in the business environment.
  • IT can be used for modeling complex systems wherein low-cost experimentation of alternate scenarios, adding/dropping of properties and decreased assumptions compared to mathematical models make the modeling more realistic. It thus increases the predictive power and in a faster way. Cheaper, Faster, Better. Sometimes, it is even about facilitation of experiments which are otherwise impossible to be credible.
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License