Wednesday, September 1, 2021

The Requisite Law of Variety



Now, this could be a bit stuffy, but it's actually quite readable: "The Requisite Law of Variety"

Reduced for a quick understanding, 'the Law' is an interesting concept to consider, to wit:
  • Essential Elements (E): these are outcomes, results, or other artifacts that you want to have some control over; or they are controls and mechanisms that you have to influence outcomes and results and disturbances
  • Disturbances (D): these are events or actions that impact E .... usually there are a lot of these!
  • Regulation (R) or controls: these are the means or actions to limit the impact of D's. Even in the Agile world, there are some controls or protocols to regulate the pace and rhythm
  • Passive capacity (K): in effect, buffers and elasticity that limit the fragility of the system and allow for some absorption of D without breaking E 
The Law is a bit mystical when expressed as math, essentially saying:
"the the variety -- or set -- of the E's that you need or value should be greater than
  • the variety -- or set -- of disturbances (D) reduced, absorbed or mitigated by regulation (R) and
  • other, or further, disturbances reduced by buffers or elasticity (aka: passive capacity, K)."
Practical project management
In other words, the practical advantage to a PM for having a big variety of Es is that Es overwhelm a smaller variety of Ds (or diversifies or dilutes the impact of Ds ) because big E implies a lot of different outcomes that are important, and a big tool set with which to work (*)

Risk management point of view:
So, whereas, some variety of Ds may have impact of some part of the Es, not all Es will be affected. Thus, the larger the variety of Es, the less impactful is a much smaller variety of Ds.

Value proposition point of view
If there is only one thing that is important to you as a project outcome, all your value is in that one thing. Consequently you have set up a vulnerability such that a single disturbance could wipe out your value proposition.  That is way too fragile!

And, 'having only one thing' violates the Law of Variety which intones that the variety of Es should be greater than the Ds that could impact E.

Anthony Hodgson puts it this way:

[The Law of Variety] ... leads to the somewhat counterintuitive observation that the regulator [or PM] must have a sufficiently large variety of actions in order to ensure a sufficiently small variety of outcomes in the essential variables E.

This principle has important implications for practical situations: since the variety of perturbations a system can potentially be confronted with is unlimited, we should always try maximize its internal variety (or diversity), so as to be optimally prepared for any foreseeable or unforeseeable contingency.

If you are familiar with the ideas of "anti-fragile" in system design, this last sentence is a good alternate phrasing for what makes systems anti-fragile, i.e. resilient


(*) In effect, the Law of Variety provides opportunity to define project success in several ways. If one of the Es doesn't work out, perhaps another one will. 

Some might say that with such a strategy, every project can be successful, or none a total failure: you just need to find a good E to hang your hat on. 

With such reasoning, it's not hard to conclude that 'project success' or 'failure' metrics are tricky. For more on this debate, look to the endless commentary on the Standish Reports of project success.






Buy them at any online book retailer!