I have recently encountered an annoyance at work which fits quite well with my political philosophy, and I thought I would describe it, as it fits so well with my other posts "Examples From Another Field", "Object Oriented Programming, Apple Computers and Justice", "Some Libertarian Analogies", "Copyright as Politics" and "More Examples From Another Field".

You see, for a long time, computers were the realm of hobbyists, enthusiasts, and people trained in other disciplines, such as engineering and mathematics, and as such the rules were non-existent, or developed ad hoc. Even when standards organizations were created they were largely ignored, as those "in the field" recognized that a lot of academic standards were completely impractical for real world applications. In fact, the many "models", such as the CMM for software development, were usually only adopted by groups which could afford to be non-competitive, such as government contractors.

However, over time, as computer science developed into an academic discipline, and more and more software developers and network engineers came out of the academic world and were trained to think in terms of "best practices". And so, over time, these standards came to be accepted as the proper way to do things.

However, there is one problem. Many of these standards match some academic model of how things "make sense", but in practice many turn out to be impractical when applied to real world concerns. Confronted with faulty hardware, low end equipment, irregular traffic, imperfect programs, old software and a variety of other less than perfect circumstances, there are countless reasons to deviate from "best practices", but, too often, those raise don these standards are unable to accept this argument, making of the standards an end in themselves.

And that is where this fits with the arguments I have made elsewhere, such as in "The Right Way", "Deadly Cynicism" and "Utopianism and Disaster". There is an unfortunate tendency in humans, at least some humans, to make of "perfection" an ideal. Now, in itself that is not a problem, as striving for improvement is a fine goal. But, in many cases, this desire for perfection can be twisted, especially when it turns not into a quest for incremental improvement, but rather into a desire to impose wholesale the "right way" without consideration of practical circumstances.

It is the same problem I have with so much of liberalism (and also with some libertarians -- "Why I Am Not A Libertarian" -- and conservatives -- "Conservatism, Incremental Change and Federalism"), the desire to impose wholesale one's own concept of perfection. First, it shows a remarkable arrogance to imagine that one knows every possible nuance of the entire system to the point that there could be no improvement on your beliefs. ( "Appealing to Arrogance") Second, as individuals differ in their desires and needs, it is impossible to believe that your own concept of perfection would be suitable for every other individual. ("The Inherent Disappointment of Authoritarianism")  Not to mention the incredible conceit implicit in the rejection of previous systems, arguing that all past ages have nothing to contribute to your own understanding. ("Bad Economics Part 9", "Inversion of Traditional Values")

And that is why I have a problem with both these academically defined "best practices" and political utopianism. In the case of "best practices", it rests upon the belief that in the case of any given firm, a number of academic theorists know better than those with hands on experience what will suit their needs. Similarly, in utopian liberalism, it is the belief that the "best and brightest" know what everyone else should be doing, and should impose it on them for their own good.

Now, yes, there are times when standards can be good, but standards are best when they arise organically from the industry itself. When they are developed externally and imposed form outside, they tend to suit a particular academic theory more than any real needs of those working in the field. Similarly, political systems work best when the principles adopted are those that have arisen organically, through competing systems, over time and enshrined in tradition. Not that there can be no improvement on tradition, but there is something to be said for having stood the test of time, and resisting the challenges of all the best minds of the past.

But I have a feeling I am in a minority here. Just as during the technocratic era between the world wars, we are in an era where our faith in science has led us to absurdities, such as the belief that we can destroy the earth through simple every day activities, or that we cna somehow improve upon the free market through scientific management. It is a pipe dream ("The Limits of "Scientific" Management",  "Technocrats", "The Limits of Technocracy", "The Limits of Econometrics") But, until our mindset changes and we recognize the limits of "scientific management", until we recognize that individual freedom will always out-perform centralized control, I am afraid I will continue to hear about the benefits of standards developed by the "best and brightest", who must impose them on the rest of us for our own good.

POSTSCRIPT

Before anyone misunderstands my argument, I am not speaking here of simple standards, such as choice of measurements in engineering, or the FASB standards. (Though in some ways the FASB is a bit politicized and impractical.) Instead I am speaking of those sort of guidelines which say "you must never allow global variables in a program" or "GOTOs must never be used." These may be good general principles, but those who adhere to them as religious dogma often cause problems as they fail to recognize that there may be cases in which their principles need to be more flexible. There may be times when a GOTO is the easiest way to make a program work, or a global variable is the least confusing solution. And, rather than introduce needless confusion to comply with academic standards, it is sometimes better to violate them.