Friday, October 2, 2009

Axes, Leaks, and Loops.

It's been a bit since I promised those posts. I've recently started teaching, and got the job and the cirriculum at the last minute. Coming up with my lesson plans, getting them posted to the district website, marking papers, all of these have been a little time consuming. But the operative part of all that is I've got a job now. One that might even let me write on occasion. Going forward I'm going to try for something modest - one a month, essay wise. I'll probably post a bit more often just griping, but I can't be sure how often I need to gripe, so...

At any rate, I recently had a chance to reread three essays which influenced my thinking. Individually, they're influential. Read together, they interact in interesting ways.

The first one, and one that I only recently found for the first time, is Polsky's essay on Leaky Abstractions. It's worth a read, but if you're short on time, I'll sum up the pertinent points. Any model we make of reality is going to be imperfect. Mental models used to ease interaction with reality, abstractions, are no exception to this general rule. When abstractions are different to the subjects they're modeling, they 'leak' at the points of difference. A common maxim about this idea is that you mustn't push an analogy too far.

The next essay that I've reread recently is about an abstraction which is more leak than not. Pournelle's Political Axes describes the failures of the false Right / Left political dichotomy fairly well. It also points out a better way of examining political positions, by plotting each person or group based on a pair of unrelated but demonstratable values. Effectively, to refer back to leaky abstractions, he points out the leaks in the Left / Right dichotomy, then suggests a far less leaky alternative.

The final essay... Actually, I'm lying. It's not an essay, it's a series of nigh incomprehensible PowerPoint presentations, which have some supporting documentation. At any rate, the final thing that I reread recently was a synopsis of Boyd's OODA Loop. Now, the Wiki isn't the synopsis I read, but it's a fair synopsis itself, and I can't find the one I read. The OODA loop describes all human cognition. We take in informatin (Observe), we assign meaning to the facts we have gathered (Orient), we ascertain the correct action (Decide) and then we Act. One way in which human beings can 'short circuit' is to 'get inside their own OODA loop'.

That last is described in the Wiki, but I'll summarize. When you can cycle through that whole process fast enough that your opponent is still Orienting, he has to re-Observe to account for your new Action, then re-Orient. Meanwhile, you're cycling again. This often creates a state of confusion and demoralization in an opponent. That is known as 'getting inside an opponent's OODA loop'.

Now, the longest part of the loop is the Orientation phase. What the training of certain military organizations (including the USMC) tries to do is to shorten that Orientation phase to near nil for combat operations. Soldiers and Marines are trained in the appropriate response to any given situation, until recognizing the situation (Orientation) happens on a split-second, subconscious basis, and the Action occurs primarily from 'muscle memory'. Against any opponent who is not trained in the same fashion, this will do exactly what you'd expect. The opponent is caught in a continual state of Orientation, at least until the Soldier or Marine has a chance to Do Unto Them.

The related concept of 'getting inside your own OODA loop' is a Bad Thing. In short, it involves extending your Orientation phase until the likelihood of completing Orientation before the environment changes of its own accord. Once Orientation is that long, reaction is the only option, and even those reactions will typically be slow in coming. One major cause of slow Orientation times has been identified; if a person's worldview is skewed from reality, that persons Orientation phase will be extended when dealing with the periphery of the skew. The nexus of the skew is frequently exempt; the center of the skew is usually a dearly held religious, political, or personal belief. At any rate, the Orientation phase can be extended long enough for a person or organization to 'get inside it's own OODA loop'. When that happens, stagnation or fragmentation is often the result.

Now, put those all together. As Polsky explains, a perniciously leaky abstraction is never going to be a realistic model of the world. As Pournelle points out, the Left / Right dichotomy is more leak than model. Finally, Boyd informs us that a worldview skew from reality will eventually cause a person to become incapable of effective decision making and hence effective action.

With that in mind, is it any wonder our government is so unable to get anything done?