From Wikipedia,

A single point of failure (SPOF) is a part of a system which, if it fails, will stop the entire system from working. They are undesirable in any system whose goal is high availability, be it a network, software application or other industrial system.

The assessment of a potentially single location of failure identifies the critical components of a complex system that would provoke a total systems failure in case of malfunction. Highly reliable systems may not rely on any such individual component.

Though CIRCLE’s main ‘beef’ with large, centralized governments is the high risk (and historical precedence) that they turn against individual freedom and towards tyranny, another issue high on our list is the very idea that the centralization of action produces the most efficient results.  It’s  illogical at its core, at least to us engineers.  That idea of centralization leads to a design flaw called a single point of failure.  As defined above, SPOF’s are bad for any high reliability system.  They are to be avoided at all costs.

To design a large system that is to 1) run under all conditions, 2) be stable no matter the input, and 3) react in deterministic ways, with a single point of failure is ludicrous.  Just think about this hypothetical,

Imagine if we had a communication company that built their business around one cable that went between New York and San Francisco.  Let’s say they expanded their presence up and down the coasts but had all calls routed from one coast to the other over that one cable.  Would that be smart?  What would happen if a ditch digger in the Midwest hit and cut it by accident?  It wouldn’t be good.  The company’s repair crews, of course, would be on the East and West Coast as that is where the customers are, but the cut in the line would be in a small town in Indiana, or a rural stretch of Route 66 in North Texas.  Days would be required to get there and then more time needed to repair it.  No communication between the left and right sides of the country until it got repaired.  The media outlets would scream about how the company was unprepared for such a catastrophic event.

The media would be right but for the wrong reason.  It wasn’t the response that was the issue, it was the design.  They set themselves up for a relatively minor accident (incorrect placement of hole) to result in a complete outage.  Poor design trumps the poor response.

Yet, that is the path our elected officials in the federal government, and many in society, continue to follow.  Every issue is met with the need to centralize the solution.  It’s why the downturn in house prices resulted in a global financial crisis (don’t believe me?  Read The Big Short, The Devil’s Casino, and On The Brink, then get back to me).  Our federal government endorsed 3 ratings agencies, implicitly guaranteed over 50% of mortgages, and instituted policies within the banking system that resulted in a system with a single point of failure.  Nationwide decline in real estate prices?  Kaboom!

It’s the desired route by many when it comes to healthcare.  Instead of reducing the barriers of entry, allowing more to offer healthcare services (state certificates of need for new hospitals and clinics, high compliance costs, etc), the proposed solution is a single payer system.  One group, making decisions on what is allowable, determining treatment prices, providing best practices.  How sturdy do you think that system would be?  Not very, if you ask me.

Reaction to the gulf oil spill?  One central command that approves all offers of international assistance and course of action.  Yet, he doesn’t have the ability to communicate with everyone.  He doesn’t own the resources needed to collect the oil.  But hasn’t he (Thad Allen) been anointed by our president as “in charge”?  SPOF.  We have 4 governors waiting for various federal government agencies to approve things.  We have companies around the world capable of helping but sitting on the sidelines until EPA, Coast Guard, White House officials give the a-okay to act.  This actually isn’t one SPOF, it’s many.  And they are all connected to one basic assumption that central planning and centralized action can solve complex problems. [Don't get me wrong, coordination is needed and useful but that is absent in this instance.  It's the difference between an orchestra conductor and a puppet master.  We need a conductor, we have a puppet master.]

Want another example?  How about the water main that ruptured in Boston a few months ago?  We wrote about it here.  In the infinite wisdom of someone that appears to have very little to spare, it was decided that one mother of a pipe should handle all the water for 2 million people north of Boston.  SPOF.

And just last night, NPR reported on the ongoing issues with school lunches in the Boston Public Schools.  They want one provider for all schools within the system.  They spent much time writing the request for proposal only to get one response.  The problem is so complex that only one company felt willing to take on the challenge.  SPOF.  [ To make this even more entertaining, it turns out that the biggest issue is with 60% of the schools and that a nice solution exists for that segment, but the COO of BPS wants only one provider for all the schools.]

Either by luck or skill, our founding fathers appeared to be quite smart design engineers.  Understanding the risks of single points of failure, they came up with a great idea that respected the complexity of life, handing the responsibility to act in the hands of the individual, leaving the federal government in charge of defending borders and individual freedom.