The further to the left or the right you move, the more your lens on life distorts.

Monday, November 04, 2013

Normal Accidents

In what has to be one of the most insightful and 'deep' analytical examinations of the Obamacare debacle—and by extension, big government programs in general—Richard Fernandez examines why overly complex and tightly coupled systems are inherently prone to failure. Referring to the work of Charles Perrow who wrote a book entitled Normal Accidents, Fernandez writes:
In Perrow’s analysis, the systems most vulnerable to “normal failure” have two salient characteristics. They are “tightly coupled” and “interactively complex”. These are precisely the kinds of centralized structures which bureaucracies love to construct.

A tightly coupled system is like a house of cards. If you move one card all the others must be adjusted to suit. “Strong coupling occurs when a dependent class contains a pointer directly to a concrete class which provides the required behavior. The dependency cannot be substituted, or its ‘signature’ changed, without requiring a change to the dependent class.”

The opposite of “tight coupling” is “loose coupling” which occurs “when the dependent class contains a pointer only to an interface, which can then be implemented by one or many concrete classes.” Loosely coupled components have ways of working things out between themselves, but they are not directly dependent on each other.

One of the changes that Obamacare has made to 1/6th of the American economy is to take a relatively loosely coupled system and tightly couple it.
It should come as no surprise that the president, who has neither the intellect nor the experience to understand systems, and his supporters, who believe in the fantasy that wanting something badly enough will overcome the technical complexities of actually building and implementing it, decided on exactly the wrong strategy for "fixing" healthcare in America.

To illustrate, here's a much simplified schematic of the ACA (Obamacare):

In software engineering jargon, the tightly coupled nature of this monstrosity leads to unintended side affects. A small error in one element of the system can propagate widely, resulting in significant impacts in other parts of the system. Any change made after the fact generates unintended consequences in other system components. Testing is difficult or impossible because of the interplay of components, and worse, the architecture of the system is not itself amenable to change or adaptation or extension. Large “tightly coupled” and “interactively complex” systems are an invitation for severe problems, and good software engineers spend much of their time and skill architecting big systems that avoid those characteristics. Politicians don't, and arrogant politicians who think they know what is best for the rest of us, don't even care. That's why the Obama administration spent 3.5 calendar years and well over 1,300 person-years of effort building something that does not work.

Eventually, of course, the builders of this monstrosity will cobble together enough fixes to make it work. It might happen on November 15th or November 30th or it might not. But here's the thing. In cobbling together the fixes, the geniuses who believe fantasy trumps reality have inadvertently introduced side effects that will come back to bite in the months that follow. That's not conjecture, it's an absolute certainly. And as a consequence, just when Obama's media hamsters begin celebrating "the fix" and the removal of "the glitches" in the Obamacare website, festering just below the surface are hundreds and possibly thousands of defects that will make a bad dream turn into a nightmare.

Sleep tight, Obama supporters, don't let the bedbugs bite. Bad things are coming, and you own every one of them.

Update ( 11/5/13):
As bad things stir in the darkness, just beyond the fringes of the Democrat's worst nightmare, those of us who labelled Obamacare as a disaster way back in 2009 can only shake our heads in dismay. Peggy Noonan summarizes the situation well:
ObamaCare is a practical, policy and political disaster, a parlay of poisonous P’s.

And it is unbelievable – simply unbelievable – that the administration is so proud, so childish, so ideological, so ignorant and so uncaring about the bill’s victims that they refuse to stop, delay, go back, redraw and ease the trauma.
But it really isn't unbelievable. Not if you've carefully observed this president and his inner circle of advisors over the past 5 years. It's both predictable and unavoidable.