Years ago I saw a poster that asked the question:

"Why is there always enough time/money to do it over, but never enough time/money to do it right the first time?"

Sadly it seems to be a major factor of human nature that we extremely poor planners (soothsayers?). Even in the most obvious of situations we seem to be unable to anticipate the mess and act to prevent it rather than clean it up later.

In the U.S. we have a program that puts police officers in schools. I suspect the motivation was less than noble, but in practice, in my community anyway, it has had the positive effect of reducing the gulf between law enforcement and citizens...especially underage citizens...and, I believe, helps keep at least a few kids out of trouble.

Looks like the NFL won't be much fun this weekend. I just had a look at the schedule, and realized I'm rooting for the Road Team in all four games. Considering the edge home field is, I may go 0 for 4. Maybe this is a good weekend to keep the tube off.

I heard a Bob & Ray sketch the other day in which Wally Ballou is interviewing a policeman, with the gag being that it's like he's never heard of one before. The cop explains that he goes out and catches criminals. Wally asks him if he gets a bounty for every criminal caught, and the cop explains no, they pay him a straight salary. And is there some kind of quota on how many you can catch? No, there's no game laws or anything, you just go out and get them. Well, what happens then? Well then, they go to jail, and the jail eventually lets them out. Oh, so then you can just hunt them down again, right? Well yeah, but of course we have to wait for them to commit a crime again, because otherwise they're not criminals any more. Well, I guess that slows things down a little. And so on.

Of course I cannot find it now, but within the past 6 weeks I read an article about computer models and crime prevention.

Someone developed a model that predicts criminal activity for specific geographical regions in a large city. The police respond to the model by paying more attention to the region with more patrols and things like that.

The preliminary results have been encouraging with expected spikes in criminal activity not materializing in the predicted region, nor does criminal activity appear migrate somewhere else.

Still to early to know for sure, but I'm all in favor of crimes that never happen. It would be great if this were in fact successful at preventing crimes in the first place.

It sounds good, but I'm curious as to what it means. Does prevention mean to have police stationed near areas that might be hit, to prevent the attempt from being made? Because if the attempt is never made, the computer model will soon be outdated. The crooks will go to a place that isn't being guarded. Especially since criminals can use computers too, and can start to target areas that aren't often hit.

As far as I could tell patrols were increased in targeted areas at the cost of overtime rather than reducing patrols elsewhere. The assumption is that the crimes that didn't happen were prevented by the increased presence of officers.

As you point out there are some problems--the most threatening being the "arms race" between law enforcement and criminals that results in a stalemate, or worse, more successful criminals.

I also wonder/worry if these models will end up like some of the wall street stock trading models that can throw the entire financial system into ruin in a matter of minutes.

Oh, that's a good point. Who knows what scope there might be for unintended consequences? It also reminded me of the film / story 'Minority Report', which extends this idea so far that you get arrested and sent to prison for life if there's a prediction that you will commit a crime. It's a slippery slope, you know...

Happy Republic Day to the Indian Ninjas (assuming there are any left).

If anyone's interested, I've just written a new blog post. I got hit squarely between the eyes the other day by something I read about how Christians can use new theological ideas as a defence to stop us actually making progress on the transformational stuff that is, you know, the heart of Christianity:

Romans 12:1-2

This one may be a record. Saw a guy on Playchess today with a 2004 Blitz Rating and a 1402 Bullet Rating. How is that possible? Is he sandbagging? Or does he have really bad reflexes (in which case why would you play Bullet at all?)

User was SlavoljubSerbia, and he had over 2500 Bullet games and 3700 Blitz games, so it's not a question of a small sample.