Dave Snowden

On fragments and lessons learning

RSS Feed

A transition through Bonn today to spend time with the German Development Agency before moving onto Zurich.   It involved an early morning departure from my least favourite UK airport Stanstead but it was a great day.  Talking with a group of people not afraid of a concept or two and with the general intent of changing things for the better.   One of the themes was knowledge management within the development sector, something to which I have returned on many occasions over the past decade.

The main points I made are worth summarising even though they will not be novel to anyone who has followed by work over the years.   So here goes:

  1. If you capture lessons learnt at a project or activity level then you are going to get it wrong.  In complexity terms it is the wrong level of granularity of learning which requires ideas and concepts to combine and recombine in different and novel ways.
  2. It is much better to capture lessons learning as you go while the experience is fresh.  That was key in my original design of SenseMaker® and has the benefit of allowing field notebooks to replace reports thus saving people time rather than adding a knowledge management time burden which is all too common.
  3. The ability to reassembly those fragments from multiple sources in real time based on the metadata allows for conceptual blending, the creation of novelty and the discovery of new ways of thinking or doing.  It also enables peer-to-peer information flow, unmediated by agency which is critical in the development sector.
  4. If you are doing formal knowledge capture within a project then you do it around key events or turning points within the project or activity, both what people say happened and what they think might have happened.   That can link to techniques such as The Future Backwards which is one of our side casting techniques.
  5. Summarisation into formal documents has value, but only when the use of the core information stabilises.   Premature codification has destroyed many a knowledge management programme as has an over focus on trying to make tacit knowledge explicit.   Tacit knowledge is better triggered by interaction between fragmented micro-narratives and our perception of current reality.

I could expand, but five is a good number to be going on with!