Architecture as Jenga

#entarch In my post Where were the architects at RBS? (June 2012), I quoted a banking sofware expert comparing banking systems to the game of Jenga. I now want to expand upon the Jenga analogy.

Casual observers looking at a large and complex human activity system can easily convince themselves that there is a lot of inefficiency and waste. So it looks as if we can save huge amounts of money simply by taking out all the unnecessary pieces. But this is a bit like a game of Jenga. If you try to remove pieces without understanding the overall structure, you are likely to cause the whole thing to collapse in a heap. This seems like a pretty good reason for doing some kind of business or systems architecture.

However, complex human activity systems generally don’t just stop working. The people in the systems usually find a way to keep things running, although with a lot of hidden costs, including personal stress.

This has a lot of bad implications. Firstly, it is bad for the people and the working environment; among other things, excessive stress is damaging for people and their working relationships. It is bad for the architecture, because it results in a lot of additional structure (props and supports and workarounds) to keep the systems up and running. These props and supports and workarounds often end up costing far more than the original system, so it can be pretty counter-productive in terms of the cost-saving objectives. And finally, it is bad for the decision-makers, because the full consequences of their decisions are often unclear, and they may go on to repeat the same pattern.

I’m not saying that we shouldn’t eliminate inefficiency and waste – but in order to do it properly and safely, it helps if we have a proper understanding of the overall structure and behaviour of the system – in other words, an architectural view.