Dave Snowden

Boundaries as domain shadows

RSS Feed

I’ve been teaching a masterclass her in India to a mixed group and took the opportunity to push myself to get the boundaries as domains version of Cynefin to a point where it could be published for comment and start to develop. I’ve been playing with it for around three months now which a fairly typical gestation time for any new framework or idea. It’s a matter of making sure it passes the napkin test; is coherent to the theory and provides utility for sense-making. I managed it today by drawing the framework then simply adding or extending existing lines and it all came together in two stages, one before and other during lunch; Twitter followers will have seen the iterations.

I’d been thinking about creating a version of Cynefin in which the complex-complicated boundary becomes a domain. The reason for that is a lot of techniques, for example a sprint within Scrum, are boundary transitions rather than domain techniques. In general any linear iteration or experiment is such a transition rather than a domain technique. Complex is about shorter cycle parallel probes. Maintaining a project within the boundary domain is important to prevent premature convergence, so it needs to be more than a line. Over the last couple of days I’ve been thinking of it less as a domain more as a gradient or shadow – which is how it is drawn here.

So the convention is fairly simply, we have three new elements:

  • A dark shaded area generally clockwise of the boundary which is effect is a holding space, a shadow, a period where you have crossed the boundary but are not fully committed.
  • Green shaded is narrower, steeper, such transitions need a change of practice.
  • Red is a danger sign, a cross only if you really mean it.

Then within the shadow areas themselves:

  • Complicated to complex is a shift from linear to parallel experiments for example.
  • Shifting from best practice to good practice (Obvious to Complicated) opens up to governing constraints with variety not one way of doing things  ….
  • …while a transition the other way can be done more experimentally, withdrawal is still possible.
  • Entering Chaos is a definitive decision, working on the edge of chaos in the complex domain means the patterns are sort of there, far from clear but starting to form and formal safe-to-fail experiments should be more transitory in nature.
  • The Obvious to Chaotic boundary should be a barrier. climbing back up the cliff is hard.

Disorder is interesting, working on that

So open for comment ….

… and just to be clear, this is a Cynefin extension, not a new Cynefin




PS: My resolution to write a daily blog post stimulated by whatever came in from Gaping Void didn’t survive an accumulation of back email to 800+ and a pretty punishing travel schedule. But I like the idea and may return to it in June and use the weekdays for that leaving the weekend, where there are no mailers for other material.

  • Ron Donaldson

    I read some interesting material on liminal zones recently. I wonder if your shadow areas might be more like intertidal zones, sometimes like one adjoining area then other times like the other.

    • Dave Snowden

      In the sense of a ‘threshold’ matches some of the ways I was thinking but I’m being careful how I use the word liminal as it is too easy to get beaten up by anthropologists if you use the word lightly …. intertidal, or possibly the whole salt marsh concept ….

  • Sherry Johnson

    At the moment I’m imagining staircases of varying types between the domains. Obvious to complicated is a narrow stairway downward. Complicated to complex is a set of parallel stairs. Complex to chaotic is a set of switchback stairs that shift occasionally, a la Harry Potter. And between chaos and simple is a dodgy elevator whose pullies throw out sparks.

    • Dave Snowden

      Love the image, but I can’t imagine it passing the napkin test!

      • Sherry Johnson

        Can’t you just change the napkin test to the pocket-whiteboard test?
        I jest.

  • Akshay Anand

    You could apply the shadow area concept anti-clockwise of the boundaries. Is there a reason you chose not to do so? For example, I would suggest there is a danger of entering a holding pattern going from Complex to Complicated (e.g. due to FOMO), or from Complicated to Obvious (e.g. due to experts resisting oversimplification of their expertise)

    • davesnowden

      Going anti-clockwise represents a significant deliberate change, clockwise can be transitionary

  • Matt

    I find very interesting the shadow/gradient perspective, and I was wondering if co-creation processes in which several people are involved could be an example. More in detail: proposing different (but in parallell) UX/UI alternatives or prototypes before starting a Scrum sprint, or committing on different possible (even this time in parallel) experiments as a retrospective output. Can these two examples be seen as actions maintained into a boundary domain in order to avoid premature convergence? I’m asking this because moving from complex to complicated in these cases appears to me a gradual shift from something dark to something brighter, until we can finally handle it in the complicated domain.

    • Dave Snowden

      parallel experiments before a sprint would be complex (assuming they are a true ‘spread’), then the sprint is in the shadow zone – or at least as I envisage it

  • Andrew Wagner

    I see value in the discussion of transitions between the domains… I’m not convinced of the value of creating “transitional domains” themselves. They seem to add.. ahem.. complexity without value, but perhaps going down this road further will convince me. And certainly the transitions are what’s critical and worthy of diving deeper and developing further. A worthy experiment!

    • Dave Snowden

      More shadow or boundary/liminal zones than domains … And its for people who are already familiar with the framework

      • Andrew Wagner

        I see it reinforcing the idea of the cliff between Obvious and Chaos and reinforcing the curved lines verses 2×2 matrix. How does one draw a slope or ridge line–or cliff for that matter–in 2D without artistic skill? I suppose that’s the napkin test.

        • Stuart Masters

          You can use what we geologists use: a symbol e.g. an arrow, drawn tangential to the boundary and whose length represent the gradient.

  • Mitch Weisburgh

    I’m thinking that in going from one domain to another there is both a decision and a situation. You can decide to move from using tools of the complex domain to those of the complicated one. But also the situation itself could be changing due to the emergence of better tools and/or a change in environment (or something else). Should a model take that into account?

    • Dave Snowden

      Agree its both, what I am working on is the nature of a transition and the associated decision – is it a sharp change, deliberate act or can it be more transitory. Next iteration up soon

  • Dr Wendy Eford

    Right now I am thinking of my history as a white water kayaker and hydrotopography. I am a fan of metaphors and will work on sketching this. My all time favourite book on this is William Nealy’s Kayak: the animated manual of intermediate and advanced whitewater technique.

    Think different types of rapids on a river with lake as a source. I am not wanting to think of this as linear. but more of a landscape / waterscape. As any canyoner or rambler knows, up and down and around the river is possible if you don’t mind getting wet.

    Also think in terms of the dynamics of water flowing around a rock in a river – the cushion or pillow on the upstream side and the turbulence and still water behind the rock on the downstream side.

    Obvious to / from complicated

    Think lake to pebble race for the transition. Starts to break up the pattern between a still water section in small transitions across the boundary. Still able to walk back and forth as water is shallow /moving slowly and rocks stable enough.

    Complicated to / from complex

    Think grade 2-3 rapid. More of a scramble over larger rocks to get back and more difficult transition downstream. Accidents are possible. Greater commitment is needed but the amateur can make the shift so still safe to fail.

    Complex to chaos

    Think serious Grade 4-5. Scouting the boundary is essential as the risk of failure is much greater. hence more skill is needed. Studying the boundary gives a sense of the patterns which will deliver success in the transition into the turbulence. Commitment is the key to success but it is possible to drift into the current and end up in the turbulence without control. Retreat is possible but more technical.

    Obvious to chaos

    Waterfall. Looked great until you went over the edge. Scouting the boundary and knowing how you will hold your form in the transition and then land and recover is essential. Support for beginners is unlikely to be enough unless this is highly skilled and the circumstances set up well.

    The science of hydrodynamics would deliver more at a microlevel in terms of understanding how patterns within a smaller section of the boundary would operate.

    Buffer zone

    This is the place where the water hits the rock. It is a place of uncertainty but can be stable enough to observe from, though transition back into the current is not guaranteed. Once observations are made, going upstream for a run up is a better plan.

    Eddy fences, holes and many other features deserve more description than I will give here

    If interested, Nealy’s book will absolutely be food for thinking. https://uploads.disquscdn.com/images/e8293e39496543109396294ec25f0037bb07ef55210c9d8934db39ab1a9908d4.jpg

    • Dave Snowden

      I like the metaphor with the possible exception of complex to chaos which can be a controlled move. The link doesn’t work however