I really like the approach - looking at the situation, and crafting one's approach based on one's knowledge of techniques that might apply.

Cynefin always gives me pause. Complexity IS uncertainty! Anyone who has studied statistical physics will explain that entropy is simply the log of the number of possible internal states of the system. Thus, the more complex, the more internal states. Uncertainty IS essentially the number of possible internal states, because if you know the actual internal states, then you eliminate the uncertainty and hence the entropy!

So it has always seemed to me that Cynefin is trying to distinguish two things that are not really distinguishable.

For example, suppose one is considering maintaining a bicycle and a car. Both have some uncertainty as to their current condition: if there were no uncertainty - if one knew exactly which parts needed repair or replacement, then one could simply do that - there is then no complexity in deciding what to do.

Now a car has more moving parts than a bicycle. A car of unknown condition is more complex. But if all of the car's parts are in perfect repair - if one has inspected each part - then there is no more complexity, because one has certainty. One can ignore the car's behavior because one has certainty about its operation. The complexity of dealing with it - repairing it - goes away.

Yes, the car is a more complex machine than a bicycle: but the repair task - the "project" - has complexity pertaining to scale of the unknown - not the inherent complexity of the machine.

Cynefin seems to be missing the relationship between knowledge (uncertainty) and the factors that lead to it.