2012/Thriving in Chaos: An Introduction to Systems Thinking

From Open Source Bridge Wiki
Jump to: navigation, search

For centuries we have learned to solve problems with a linear approach. This originated with Isaac Newton in the sevententh century and assumes that everything in the world is connected through cause and effect. Systems thinking throws away that assumption and examines the universe as small pieces connected into a complex network. You will learn how a systems thinking approach can be used to solve problems.

Speaker: Alex Kroman

Return to this session's details

Contributed notes

Systems thinking in general

  • Deterministic models
    • Clear cause & effect
    • e.g. sewing machines
  • Complex models
    • Cause & effect are not clear
    • e.g. cities
    • There are way more examples of complex models than deterministic ones
    • but we frequently try to treat complex systems like deterministic ones.


Info on 7 different kinds of systems

1. General systems thinking

  • concerned with the relationships between things, not the things themselves

2. Cybernetics

  • How do systems regulate themselves?
  • e.g. Scrums (lots of different feedback loops to improve over time)

3. OODA loops

  • Observe, Orient, Decide, Act
  • How do systems get faster?
  • Useful to figure out where the hangups in a particular system are

4. Game theory

  • How do systems compete?
  • Prisoner’s dilemma problem
  • see also: Von Neumann

5. Cellular automata

  • Study of evolving systems
  • e.g. Game of Life
    • Depending on arrangement, can actually compute fairly complex equations (e.g. differential equations)

6. Chaos theory

  • Unpredictable systems
  • e.g. Butterfly effect
  • Optimize for flexibility
  • Aim for the edge of chaos
    • Add or subtract rules to get there
    • Don’t want to be too comfortable, or too chaotic

7. Fractal

  • Scale invariant systems
  • Mandelbrot
    • Cotton price sale charts
  • e.g. Romanesco broccoli, mountains, blood vessels
  • Take a small working system, repeat it at different scales


Rules (definitely missing some from slides):

  • Focus on relationships between things
  • Need both positive *and* negative feedback
    • Too much of one (or only having one) means an unsustainable system
    • e.g. predator & prey, or something less life-and-death
  • Need to be nice
    • But not too nice
  • Rules within the system must be simple