I spend so much of my time here criticizing the excesses of the political left, or even a few of the conventional right, that I sometimes forget that there are other philosophies which are equally wrong on one or more issues,a nd which ar epopular enough that I need to spend some time refuting them.

Recently I mentioned that the third parties, at least those of a libertarian bent, tend to suffer from a peculiar problem, seeing government as an enemy rather than a tool. Having written that, I realized that I never really elaborated on that topic properly. Oh, I wrote about libertarianism before, explaining why I believe federalism is a better approach, and even drawing an analogy between certain copyright schemes and the libertarian approach, but those essays were interested almost entirely in describing why the libertarian "top down" approach to freedom was wrong, not with their views on government.

On the other hand, I have written before about the view of government as a tool, but mainly to explain why it should be restricted to the functions for which it is appropriate, not to justify the existence of government. So, to correct this oversight, let me use two of my favorite analogies to explain why the libertarian vision of government is a dangerous one, while also getting in a few digs at some of my whipping boys, Wikipedia and Linux. (If this doesn't prove I am a geek, I don't know what will.)

But before I start, let me make some general statements. In the past, I have written that liberalism suffers from a specific problem, the view that individuals are incompetent, and the government all knowing. It is at the heart of the liberal philosophy, whether they admit it or not. Well, to be accurate, most liberals believe people are incompetent, a handful believe people are competent, but they are at the mercy of conspiracies from which the state needs to protect them. But my point is, without postulating either incompetence or evil, it is hard to justify the intrusive liberal philosophy of government.

Libertarians seem to take this philosophy and stand it on its head, granting individuals infinite competence and virtue, while portraying the government as either incompetent of evil. In their view, individuals are completely good until they enter government, at which point they become as evil as liberals believe those entering big business become. Now, I will grant it makes more sense to argue for governmental incompetence, than individual incompetence, but it is still wrong. Much as I might rail about government inefficiency, it is because we apply government to the wrong tasks, not because government is inherently inefficient. And, though I admit government tends to attract those who like to coerce others to do their will, that does not mean that government service either attracts evil or turns men evil.

In other words, I think the libertarian world view is as absurd as the liberal. But, I will let my analogies make my case for me. When that is done, we can discuss what a realistic approach to government and the individual would be.

My first example comes from the world of computer operating systems. For those unfamiliar with the topic, let me provide a quick summary. Unix developed at Bell Labs as an operating system. It was largely developed in conjunction with the C programming language. At some point, a free version was developed at the University of California at Berkley. I will spare you all the tales of the legal turmoil involved in freeing Unix from copyrights, but all the Unix versions derived from that free Berkley version are named something-BSD. Linux, on the other hand,w as developed from scratch, or mostly so. Linus Torvalds, or so the story goes, wanted a copy of Unix, but could not obtain the BSD code, so he developed his own variant, which formed the core of all the modern versions of Linux. Both continue to exist as free distributions, and there are innumerable variations on each.

Where my political analogy comes in is in how the code is maintained for each. Both allow users to see the complete source code, and both welcome user contributions, but there are serious differences in how that code is handled. (As well as in copyrights, but I dealt with that before.) Linux, basically adopts an anarchic model, allowing users to pretty much push out patches as they will, though some official distributions exercise some control over the release (more on that later). However, in general, Linux is similar to the "wild west" analogy the left loves to use. Anyone who wants to put out a patch can¹.

The various BSD distributions, on the other hand, tend to be controlled by a committee of editors. These editors schedule a release every so many months, on a regular schedule, accepting proposed alterations, voting on them, and testing them, then putting out a definitive release at a scheduled time. The various distributions differ as to the schedule, the mechanism of appointing the editorial staff, and so on, but in general BSD releases tend to have some form of central control².

So, what does any of this have to do with politics? Quite a lot actually. As with the GPL copyright, which is a cause often championed by the same people who develop for Linux, there is a suspicion of any centralized control or restrictions on development among some segments of the Linux community. They view the BSD style central committee system with suspicion, thinking that such editorial control means that important changes will not take place³.

And they are right. By exercising central control, BSD does miss out on some changes⁴. On the other hand, it also helps ensure that any code implemented will not have unforeseen consequences either, which happens much more often with Linux releases. The chaotic Linux release methods mean that code often interacts in unforeseen ways, with one user's patch breaking a patch just released by another user, thanks to the lack of central control. Of course, to work around this, various private groups and commercial ventures have created Linux "distributions", sets of packages which are guaranteed to work together, in effect, bringing through the backdoor what BSD distributions do honestly.

All of which is a very round about way of getting to my point. The two camps work as perfect examples of the conservative and libertarian views of government. To the BSD developers and to conservatives, government is an essential tool, part of the process, perfectly suited to certain tasks. To the libertarians and the Linux developers, government is a necessary evil. They accept it for the kernel, but otherwise they strive as much as possible for complete anarchy, which they call freedom.

And the results show exactly why I argue that libertarian fear of government is untenable as a political philosophy. BSD systems work out of the box, thanks tot he government built into the development system, a user can buy a BSD version, plug it in and get it working. Linux doesn't even have an "out of the box" distribution. The distributions are set up by companies after the fact, effectively bootlegging government into the system. Without them, Linux would be a kernel and a mass of code, some of which works, some which does not, and much of which has harmful interactions. Only by sneaking in despised government can Linux even be crafted into a working system by any except the most ardent experts.

But, some will argue that I am presenting something of a caricature in my description of Linux, and I admit that I picked the most extreme of the FSF/GPL/Linux arguments, so let me look at our second example, Wikipedia.

I have written at great length about my complaints with Wikipedia⁵, so I will give only a brief summary here. Basically, Wikipedia is a system which sounds good in concept, but falls apart once one begins to think about the specifics.

Wikipedia is the ultimate expression of libertarian fear of authority⁶_,  each article is open for editing by anyone, and any changes can be removed by anyone as well. The idea being that, given a body of well meaning individuals, the article will be continually edited by people who wish to either correct errors or provide additional materials, coming ever closer to the truth.

Of course, there are so many problems with this idea, I don't know where to begin. Firs,t it forgets that there are vandals out there, who would vandalize a page just for fun, and would continue to do so no matter how often they are rebuked by contributors. That oversight alone makes Wikipedia valueless, as there is simply no way to know if, at the precise moment you read an article, whether it has been defaced by a vandal or not.

More relevant to our topic, there is also the problem that "true" may not be as clearly defined as the wiki-faithful think. They might argue for NPOV (their wiki-jargon for "Neutral Point of View"), but they ignore the fact that what constitutes point of view still depends on an agreement about an underlying philosophy, and an underlying agreement about what constitutes truth and evidence. Without any such agreement, what is one man's point of view is another's certainty.

Let us just take one example. Suppose you believe the Bible is literally true, and I don't. You could then write, "the world was formed in six days' and claim it is NPOV, as the Bible says it so it is true. I would argue on the contrary that it is false, or at least only a POV. The problem is, without a means to resolve this dispute, we have an irreconcilable conflict, and end up in one of those "revert wars" that plague Wikipedia.

To solve such obvious problems, Wikipedia imposed a loose editorial system on top of the structure, but they are so afraid of allowing anyone authority that they basically made all editors equal, inside or outside of their field, and thus created simply a second tier of revert wars, this one fight by those supposedly resolving the same problem at a lower level.

But that is the problem with Wikipedia. Just like the libertarians, wiki-believers fear authority. They go to great lengths to prevent allowing anyone final say over matters. With the end result of either endless conflicts or the absurdity of truth established by majority vote.

The answer, quite simply, is the one adopted by true encyclopedias. Rather than allow the writers free reign, they hire knowledgeable generalists to set overall policy, who then hire experts in fields to establish the standards for each section. By creating standards and rules these sort of conflicts are largely avoided, and when they are not, they can be easily settled as both rules and a final authority has been established beforehand.

And that, in sum, is my complaint with libertarianism. Many libertarians do not understand that government is a tool, and instead see it as a "necessary evil". So they tend to think in terms of fighting with government rather than using it for the proper purposes, and the result in libertarian thought is as defective as the analogous thought is in Linux development or Wikipedia.

Of course we need to make sure those who exercise government power do not abuse their authority, and we need to establish as system that keeps government authority as small as possible while still carrying out its needed tasks. But that does not mean we should become enemies of government. All it means is that we must recognize that in government, as in all fields, men can through malice or misunderstanding, abuse their positions, and we need to put in place common sense safeguards to prevent that.

And that, in a nutshell, is my view of the state. It is a tool, a necessary tool, to collectively assist in protecting ourselves from force, fraud and theft, as well as to facilitate the settlement of honest disputes. It can be abused and bring about great harm, but with the proper safeguards, it can be trusted to perform an essential function.
 
---------------------------------------------------------------------------------------------------

1. Linux actually does seem to exercise some central control. Despite all the rhetoric about freedom, they do exercise control over the kernel itself, with Linus Torvalds managing the releases of kernel patches.

2. Disclosure: I did do some development for FreeBSD, though none of my proposed patches were accepted. I spent quite a bit of time fighting for a specific patch to the hard drive controller code, but had no success in persuading the community it was necessary, as they thought existing workarounds were adequate. As far as I know, the error I tried to fix still exists in the new ATA driver that replaced the old wd drivers. It did as of FreeBSD version 5.2.

3. Granted, not all Linux developers, or even all FSF contributors, are of the "information wants to be free" dogma. Many are much more realistic, as the strong central control over the Linux kernel suggests. But there are quite a few who do hold such extreme views, and it is that sort of user who provides my example.

4. On the other hand, as each user has access to his own copy of the source code, he can modify his own version to his heart's content. What BSD does not provide that Linux does, is general access to the same channels of distribution. BSD limits what is an "official distribution". Users can provide patches, but they are not seen as official, while everything released for Linux has the same status, outside of kernel code at least.

5. My Wikipedia articles can be found here:

Stop Confusing Me With The Facts!
Mystery Quotes
Wikipedia?
The Failure of Wikipedia
Opinion Masquerading as Fact
Funny Numbers
What is Wrong with Wikipedia?
Endangered Species
Sterility of Formal Economics
Deceiving Themselves?
A Question About Language
Why People Don't Take Academics Seriously
Proof Positive

6. I am here forgetting the editor system which was layered on top to resolve some of the worst abuses. the original general concept did not include them, so I will forget about them for the moment.

----------------------------------------------------------------------------------------

POSTSCRIPT

It would be interesting to see an online version of the Wikipedia I proposed, one with a general editor who would set overall policy, who then appointed section editors, who would appoint those under them, and so on all the way down to the maintainers of each individual page. The system would require users submit pages for review. Other users could see proposed changes if they wish, but the official page would not be updated until approved by an editor. New pages would also require the approval of the appropriate editor before they could be created.

It would be quite a change from the Wikipedia we have now, but I think in the long run, provided the editors were of a high enough quality, it would produce far better results, and certainly would not have the vandalism and hoax problem the current Wikipedia does.

POSTSCRIPT II

For those who read my earlier entries on Wikipedia and bring up my complaint that editors are granted too much power, my complaint is not so much that they are granted excessive power, but that they are granted excessive power in areas outside of their specialty, and with no higher authority to serve as a check on them. So a person who showed promise editing a post on medieval romantic poetry can now make pronouncements about atomic physics without anyone correcting him. That is what I meant by their excessive power. And, strange as it may sound, that too is the outcome of the anti-authoritarian tendencies. As with most who distrust government, their solution is to give the upper echelons of government way too much power, but creating several equally omnipotent posts in the hope that they will serve as checks on one another. But that is the topic for another blog.