Software TestingUncategorized

Cynefin and Software Testing – Disorder

The odd shape in the middle of the Cynefin framework is the domain of disorder. Despite not looking a whole lot like the others, it is a domain in its own right and arguably the most interesting of the domains as it is (I find) less straightforward to understand than the others.

At its most basic, Disorder is the domain of not knowing which of the other four domains you’re in. In hindsight, I should probably have started with this domain first, given that it is probably one of the most commonly encountered. Anyone that has a rote response to any given situation might be described as being in a permanent state of disorder, or perhaps more charitably, they act as though they were in the domain that best suits them, even when a different domain would be more appropriate.

Disorder also forms part of the boundary between each of the other domains and in doing so, helps illustrate that moving between them is not necessarily straightforward. There is a shift in context as you move from one to the other where the appropriate actions change. If you are moving between one domain and another for the first time especially, you likely aren’t certain that this course of action will hold.

If you think about rolling software to production for the first time, you could look at this as moving from Complex to Complicated. You’ve likely taken an idea from inception where there are many variables to consider at once, through one or more prototypes and iterations to a point where you feel you know enough (i.e. your idea is well constrained enough) that it is sufficiently predictable to put out into the world. Interestingly, we sometimes refer to this as ‘releasing software to the wild’, and it’s a metaphor that works well in this instance, I think. Unpredictable forces will act upon this fledgling software and we have the hubris to think we know what will happen. Nonetheless, we huddle around our monitoring, such as it is, and await news of the success or failure of this neophyte collection of bits and bytes.

Disorder ought to be a temporary state. At the point where you realise you don’t know where you are, you look at the data and (where possible) the history and determine what action is most appropriate. It is possible to be lost or trapped in a state of disorder however. This will be recognisable by anyone who has worked with any company that favoured procedures over skills or who has seen ‘work to rule‘ in effect. What you end up with is a set of procedures that people pretend to follow to some degree, and a series of backchannels and alternate structures to make sure that work actually gets done.

The inauthentic adherence to procedure that I described in my post on the Complicated Domain, or indeed the initial reaction of the business owner in my first Cynefin post are examples of an unintended drift into disorder and in the latter case, wilful ignorance that causes them to stay.


More on Disorder:


2 thoughts on “Cynefin and Software Testing – Disorder

Leave a Reply

Your email address will not be published. Required fields are marked *