Systemic complexity and the need for simplification

I enjoyed this graphic that came from Donn Garrett via Lynne.

What is normal indeed! mention this now because it links to my last post, Has true innovation in Australian education declined?. There in a comment I wrote in part:

Current administrative systems are incredibly dense and complex. If you want to get a feel for this, try mapping them using Microsoft Project building in steps, decision points and estimated lags. It's actually surprising that anything gets done. It's not surprising that a lot of time is involved.

Then experimentation depends upon space to experiment in, resources to experiment with. The idea that systemic rigidities can impede new things is not knew. The term skunk works widely used in business, engineering, and technical fields to describe a group within an organization given a high degree of autonomy and unhampered by bureaucracy, tasked with working on advanced or secret projects. See http://en.wikipedia.org/wiki/Skunk_Works.

The problem in education is that scope for experimentation has narrowed, while its harder to find resources. There is no educational equivalent of a skunk works.

I suppose that I have always been a reformer. Here one of my weaknesses in a professional sense lies in the struggles I have had to ensure that my advice or my actions are set within the bounds of the current normal when I know that there is a better way.

There is very little point in saying to a client, a manager or a board or council that they could do things in a better way that falls outside the iron jaws set by normal processes and procedures, unless you can actually show them a path to get round those jaws. It's not helpful. It just adds to the frustration often experienced by people who know that things could be done better, but can't do anything about it. Rarely, and these have been the greatest satisfactions of my life, happenstance means that new things are possible. Then the dam breaks and real change occurs.

The forces of entropy are strong. Big organisations require rules - policies and procedures - to make them work, to allow the whole to hang together. This can conflict with a simple administrative rule, the need to push decisions down to the lowest practical level. The actual act of doing is best managed by those who have to do. The problem is that this creates mess, untidiness, lack of uniformity that can be difficult to accept or manage. How do you control something when the real act of doing is out of sight or indeed measurement? What happens when things go wrong?

One of the very simply messages that I have tried to get across in my writing is that complexity rises, action slows, with every additional step in decision processes. That was the point of my Microsoft project example. If you actually chart the various decision points and the interactions between them, you get a very complex web.

Take a simple decision to build a house in a particular area to meet a social need. That simple decision may need to go through more than sixteen major decision points and in so doing comply with a similar number of laws, rules and protocols. Once the contract is let, the physical act of building may not take time. But the time involved in getting to actually let that contract may exceed the building time by a factor of four or five or even ten fold.

My message is simple. If you want real improvement, forget in-principle arguments, although these are important; just chart what actually happens at a detailed process level. Then simplify. This will help you break free from the trap of the normal, to pose more fundamental questions.