Spotlight session with speaker David Snowden, Cognitive Edge

Podcast of session will be made available through Dave's website: cognitive-edge.com.

Some key ideas from David's presentation:

• KM is ten years old. The question is... How do you create a knowledge sharing culture? Must understand that knowledge is only ever volunteered; it can't be scripted.  You can't measure whether someone has shared their knowledge. You can measure conformance if you want to force people to complete a form to share their knowledge.
• No one will share knowledge in anticipation of your need. Most knowledge systems are anticipation systems.
• What's happening now...People chatting with other people they trust in fragmented ways.
• The way people describe what they are doing is not actually how they are doing it.
• We can always know more than what we can tell and more than what we can write down. Hard to take tacit knowledge to explicit knowledge.
• Living oral histories: Working with knowledge in the field through stories, narrative base. Homo-narrans (rather an homo sapiens), people who tell stories, storytellers to make their point, make it stick (stickiness).
• Sense-making: How do we make sense of the world. so we can act in it. How do we re-route social systems to make sense of the world.  It's not enough to know that something appears to work, you need to know why, only then can you scale it, and understand the limitations of scale.
• Cynefin framework decision making framework for KM. Space for innovation. Don't confuse innovation with creativity. Creativity is a sub-consequence of innovation but not a cause.
• Hindsight does not lead to foresight - retrospective coherence & premature convergence.
• Mental models... get rid of that idea. Human beings are more complex. Human formulate concepts through pattern recognition not as a series of processes like machines. Much more dynamic patterns, more diverse experience that you can draw on. If someone says they are going to create a taxonomy... it'll be a failure. Remember that the word "taxonomy" is very close to the word "taxidermy." Everything is fragmented. Social computing as it is currently practiced is a chaotic system.
  *We need to move to pragmatic system that allow people to self-index their own material. The cost is very low. Two-thirds of our brain are associated with visualization. Knowledge is a great human capacity.

Types of knowledge systems:

• Ordered systems
  • System constrains agent behavior, cause and effect relationship discoverable and repeatable, prediction possible. Ordered system is on the edge of chaotic if go into catastrophic failure.
  • Hierarchical taxonomies, threaded conversations, in house facebook, etc.
• Chaotic systems
  • Agents are unconstrained, often deterministic at agent level, use of probability and statistics allows for predictability.
  • Free form tagging, assuming it will just work out, failing to manage, just increasing links (linked in)
• Complex adaptive systems (less known type of system)
  • System and agents co-evolve and system effects are therefore inherently unpredictable. Non-causal in nature. Highly sensitive to starting conditions and local interactions. System level effects are emergent and non-aggregative.
  • Applying constrains loosely. Fragments signified at the point of origin (or on access). Signified by name, key words, and free text. Signified by positioning on ambiguated forms. Signified by modulator scales.

Tell us your ideas and thoughts

If you attended David's session, tell us what you learned, how it might change the way you currently do something within your organization, or evolve how you view KM.

Sponsor: Thomson (spotlight session)