Fractal stupidity

COMMENTARY--It is a midwinter Sunday evening on the M4 motorway and I’m heading west from London to Bristol. The headlights of the oncoming cars, and tail-lights of those in front, snake across the countryside far over the horizon. Six lanes of tightly packed cars are cruising nose to tail at 70–80mph.

I start counting and estimating. I reckon we are looking at around 60,000 to 80,000 cars an hour with an average of two occupants. It's a human pipeline transporting around 140,000 people every hour towards and away from London.

Of course the politically correct (PC) view is that I should have taken the train, and under normal circumstances I would have. But this evening's track repairs increase the normal one way journey of two different trains to a combination of four trains and two coaches. The journey time by train is also indeterminate and more expensive than using a car. So I don’t feel too guilty, even though I also have a driver so I can still sit and work the entire journey. In fact the difference in cost is marginal when taxis and other extras are entered into the equation. And I must be on time. I cannot afford to be late. So PC = 0, Common Sense = 1.

Our politicians and leaders believe that the train should be our primary choice and we must reduce the number of cars on our roads. Would it work? How many trains would we need, even if we could get them to be both reliable and on time (which we cannot).

An InterCity train might typically carry 700 people--and I am estimating here too. So on this Sunday we would require around 200 trains an hour between London and Bristol. But the best we might achieve would be one every 10 minutes--say 12 an hour--six to come and six to go. And at this level such a train service would only contribute an insignificant 6 percent relief. So is it remotely feasible that any train system could make a real difference? Not with today’s technology, for sure.

Many such instances of obviously incorrect doctrines and beliefs, plus the varied behaviors of people and managers in companies, schools, hospitals and homes have led me to conclude that human stupidity is fractal. That is, stupidity looks and behaves the same everywhere, in every organization and at every level of society. At some point a view is formulated, it gradually becomes accepted as holy writ and thereafter it becomes an unchallenged truism.

Let’s try a few more examples to illustrate my point. In modern management parlance it is a well-known maxim that 'what you cannot measure, you cannot manage'. As a result, we see control and measurement systems, performance monitoring and metrics have become all the rage. But look at how the numbers are manipulated and fiddled to satisfy social, political and financial needs.

So it is interesting to ponder. Is Enron any worse than governments that arrange their GDP figures, education and healthcare statistics to meet market and social expectations, to win voters and exert influence?

Adjusting exam pass rates to achieve a few percentage points improvement year on year soon sees compound interest taking a hand and employers moved to complain grade A students aren’t what they used to be.

Building health care systems that see staff and managers inherently needing to fiddle numbers to survive the political climate are equally irresponsible. To allow political forces to further manipulate waiting and operation times to show acceptable levels of service without actually addressing the real problems, soon sees more administrators than beds. Worse, the doctors, nurses and patients quickly realize the truth and start taking action to sidestep and sideline the unnecessary bureaucracy. The whole system becomes top-heavy, over-expensive and still fails to deliver as promised.

In the worst cases we see important decisions taken on the basis of manipulated or ignored information. And all too often people get hurt or killed as a result. Skimping on maintenance routines and failing to invest adequately in new and replacement plant are all common occurrences. Honest mistakes and accidents are difficult enough to live with; PC- and ignorance-induced tragedies are in another league.

It is always a temptation to take the actual data and manipulate it to the way you (or your boss) would really like it to be. My advice is always the same: Don’t do it, no matter what. Sooner or later someone is going to pay the price and it might be very high.

There is no immunity from the desire or opportunity to fiddle the figures and it may even be the ultimate temptation. But in science, technology and engineering there is not only a price, there is a fast response time to any such action. Fortunately we tend to use machines to do the job of data gathering and analysis and reserve human intelligence for the interpretation of the data. We are also pretty high on repeatability, peer review and multiple, parallel and often competing teams. If something goes wrong we tend to find it reasonably quickly. If someone makes an error or tries to adjust the figures they are generally identified quickly.

Somehow we need to get the same automation and discipline into our social fabric and systems so we can really manage with confidence. Until we do, we are all going to witness the cost of fractal human stupidity...

•  Mrs. Smith washing bottles and beer cans in hot water to be transported by car to the bottle and can bank--saving the planet by wasting 100-fold more energy than we can recover.

•  Transport policies that see 100-fold more pollution from traffic jams than would result for a wise road building program.

•  The continued exploitation of oil and destruction of reserves while R&D into alternative energy sources remains grossly under-funded.

•  Health and security programs that address non-issues and neglect the real risks to satisfy public opinion.

The list is endless--and fractal.

Originally typed on flight FR509 flying between Bristol and Dublin. Revised on FR292 en route to Stansted and finally dispatched to silicon.com from my home via a Wi-Fi link some time later.

biography
Peter Cochrane is a co-founder of ConceptLabs CA, where he acts as a mentor, advisor, consultant and business angel to a wide range of companies. For more about Peter, see: www.cochrane.org.uk. He writes a regular column for silicon.com that is archived here.