Friday, July 13, 2007

Poll

Google just added a new feature to its blogs: polling. Just as an experiment, yesterday I put up a poll about a simple science question: why is our sky blue? I'll be putting up more polls, on whatever subject I feel like – please keep an eye out for them...

And vote!

Cognitive Bias

Cognitive bias is just a fancy term for a readily observable phenomenon: we humans aren't necessarily logical in our thinking. Often these cognitive biases can be traced back to some evolutionary advantage they conveyed (which may or may not still be an advantage in our modern world).

For example, we humans have a special horror for risks of death by unusual means. Around these parts, many people are quite frightened by our abundant rattlesnakes – and it is possible for a rattlesnake to kill you. Last year, in fact, seven people in the U.S did die from a rattlesnake bite (two of these were small children; four were adults who did not get treatment for various reasons, and one was an elderly man who apparently had a heart attack just before being bitten).

The same people who are frightened by rattlesnakes will climb into their car for a trip without the slightest care – despite knowing that over 100 people die every day in traffic “accidents” (39,189 in 2005, the most recent year for which data is available)! By any logical analysis, we should all be much more frightened by driving to the grocery store than we should be rattlesnakes...

A similar cognitive bias affects the lamestream media's reporting on the war on terror, and most people's perceptions of it. I've noted this in quite a few conversations – in general, people seem to very badly overestimate our war casualties. Of course this affects their perception of the war and how it's progressing; they see the war on terror as being unreasonably costly compared to it's benefits. The sad thing is that this perception doesn't stand up to any rational analysis, much like our fear of rattlesnakes versus our fear of driving.

For starters, ponder this: in over four years of our war on terror, the total military casualties are about the same as the traffic fatalities from a single month. Even that is an overestimate, because it neglects something rather important: our military personnel are at risk even when they're not at war. From 1980 to 1999 (a relatively quiet period), our military suffered (on average) about 1,600 deaths per year from non-hostile causes. Adjusting for changes in the size of our military and for the fraction of our military directly engaged in the war on terror, a reasonable estimate is that those forces would have suffered about 2,400 deaths in the same four years, even if the war on terror had never occurred. So the incremental casualties directly caused by the war are more like 1,500 – about the same as the number of traffic fatalities in just 10 days.

Ponder that.

The men and women of our military engaged in the war on terror are heroes to me, and I mean that in the most literal sense. I think about them and admire them often; I read everything I can get my hands on about them. I mourn those who have been killed or injured. But I am also very conscious of something else: that our military's competence, combined with the amazing technology we possess, has lowered the human cost (to us) of warfare to levels far below many other risks that face us – and at the same time has increased the lethality to our enemies. By any rational measure, our military heroes are kicking al Qaeda's ass, hard; they are dying in large numbers. That's not a pretty fact, or a politically correct fact – but it is a fact. It's also the objective in a war, as General Patton famously said:
No bastard ever won a war by dying for his country. He won it by making the other poor dumb bastard die for his country.
The only way that al Qaeda is going to win this war is if we give up. Their poor dumb bastards are dying in droves (especially in Iraq), while our astonishingly expert military is suffering very light war casualties. Partly because of this cognitive bias, a significant fraction of Americans (and virtually the entire Democratic party, except Joe Lieberman) are ready to give up, and hand al Qaeda a tremendous victory. Nancy Pelosi and Harry Reid make almost daily calls for withdrawal from Iraq – they want us to unilaterally surrender.

Such “leaders” shouldn't be allowed outside to play without adult supervision...

Fun Project

It's a rare and special treat when I can combine my hobby with my professional skills and experience. Over the past few weeks I've been spending some of my spare time doing exactly that – creating a web site to display slide rule information. The information is a compendium of all the models of slide rules ever made by Hemmi, who was the world's largest slide rule manufacturer. The data on this site was collected by Paul Ross, with help from several other slide rule collectors. It is now the “gold standard” reference for everything Hemmi.

The original web site was created (by Paul Ross) using Microsoft Excel. This was a tool familiar to Paul, and therefore very easy for him to use. But the resulting web site only works well with Internet Exploder (Microsoft pays little attention to compatibility), is very slow to download (because the HTML file is huge), and is a purely static page (you can't interact with the web page in any way). Paul is a slide rule god, not a geek. I'm a geek. I volunteered to help, and Paul eagerly accepted.

Last night I posted the first beta test version of the new site, and sent a message to slide rule collectors to have them start banging on it. I got my first bug report this morning, and I'm sure I'll get more. The new site “understands” the Hemmi slide rule data – you can do things like sort the slide rules in various ways, and limit the slide rules that you're looking at. From a technical perspective, one of the most challenging things was figuring out how to display the large amount of data in a way that was both compact and easy to read. This depends on the way the web browsers lay out tables, which is a very tricky thing. To help with this problem, I made many of the user interface details changeable on the fly – something not often seen on web sites.

Bottom line: it was lots of fun!