S Y S T E M A N T I C S 5
V. SYSTEM DESIGN GUIDELINES: LONGEVITY
1. Systems develop goals of their own the instant they come into being.
The moment someone begins using a simple working system, they become part of that system... along with their goals for this new, larger system.
As the system grows further, becoming still more complex as more people use it, more goals are added. Some of these goals will be mutually exclusive; they can't all be fully satisfied. Eventually it becomes impossible to explain why the system as a whole does whatever it does. The system appears to take on goals of its own that are completely independent of whatever purpose the original system's designers intended.
It's important to realize that this apparent intentionality--the appearance of a system's having a mind, personality and will of its own--is just that: apparent. It's not real. Systems with many inter-system goals are so complicated as to seem to act almost randomly or chaotically... but the human mind is disposed to see patterns, even where none exist. (Have you ever talked to your car or your computer as though they had distinct personalities? Why?)
This is a natural perception of systems. At a low level, there is no illusion of system consciousness--the system's actions are few enough and simple enough to be comprehensible. Only when these actions accumulate and their macro-level effects are observed is the appearance of conscious, intentional behavior created.
Some cognitive scientists and philosophers wonder if the same might be true of the systems called human beings.
2. The longer a system exists, the more its primary goal becomes self-preservation.
Constituencies again. The more people who are served by a particular system, the greater the resistance to changing that system... and, bizarrely, this is so whether that system works or not.
As a system grows, serving more users, adding more internal "features" and requiring more support functions which will require support features of their own, the entire system itself begins to act as though it has a will to live.
Eventually--once the system has gotten very large and been around for years--all other system goals will have been crowded out to make room for the fundamental goal of self-preservation. No matter how the system or its goals are described, internally or externally, the one true goal of the system will be to preserve its own existence. All of that system's functions will be bent to serve that purpose, and all resources consumed by the system will be directed toward defending it from external threats.
Example: An expert scientific panel found "no association" between silicone breast implants and immune system complaints. But the U.S. Food and Drug Administration under David Kessler had already made its decision to prevent women from having access to implants. Since the science called into question the competence of the FDA to decide for citizens what it thinks is best for them, it became a threat to the existence of the system. As Kessler put it: "If members of our society were empowered to make their own decisions about the entire range of products for which the FDA has responsibility, then the whole rationale for the agency would cease to exist."
3. The longer a system exists, the greater its resistance to any fundamental change short of complete destruction.
Resistance to change seems to be a natural feature of all mature systems. At some point, the effort required to change becomes greater than the resources available to make that change. (When the system under consideration is a civilization, the historian Carroll Quigley's terms for this are an "instrument" which becomes an "institution." But these terms could be applied equally well to systems in general.)
When this occurs, the system's final insulation from its environment begins. Soon no reality will get through to the controlling elements at all. The closed positive feedback loop is complete. And the only way the original problem will be solved is by the creation of an entirely new, small, working system.
Example: Despite the lack of a military draft for years, the Selective Service Agency continues to exist, requiring that young men (but not women--where's the feminist outrage?) register for the draft, and punishing them for failing to do so. Even the Pentagon has said that this registration is no longer necessary.
But to add insult to injury, in 1997 Selective Service Director Gil Coronado began arguing that the Agency has "a natural opportunity to promote voluntary service to America in AmeriCorps." AmeriCorps, as you may or may not be aware, is the "service organization" created during the Clinton Administration. Consisting in theory of "paid volunteers" (an oxymoron obvious to anyone but a politician), in practice AmeriCorps workers were highly politicized, manning phone banks during campaigns in California and generally redistributing tax dollars so that a few people could feel good about themselves.
Converting the no-longer-necessary Selective Service Agency into a registration front for AmeriCorps is a prime example of how old systems refuse to change fundamentally in response to a changing environment.
Example: Despite the fact that only one agency (NASA) has any use for huge quantities of helium, it took the U.S. Congress decades to privatize the national helium depository... and the last time I checked, the federal tea-tasting board still existed.
4. Most temporary patches aren't.
As any experienced software developer can tell you, the quickest way to insure immortality for a small, working, useful program is to call it a "prototype." It will immediately be declared a public or corporate resource; its future growth will be taken over by a committee; and your name will never be seen on it again . . . unless it goes berserk and kills someone, in which case yours will be the only name associated with it.
I. General System Behavior and Morphology
II. Systems and Information
III. System Design Guidelines: Number
IV. System Design Guidelines: Size
V. System Design Guidelines: Longevity