Could the sniper have been caught earlier?

Orin Kerr has a link to an NYT article [1] that apparently says that the sniper could have been caught earlier with more and better mega databases. He thinks "that it took the police lots of uses of databases, information sharing, and citizen-informant work to catch the sniper." I'm going to comment on the database part. I haven't read the NYT article, but anyway:

Poindexter's dream DB (a dream now shared by IBM and perhaps Dumpster Divin' Larry "call me Lawrence" Ellison) could have even caught John Muhammad before he went to DC. It would have tracked gas purchases, stolen credit cards, gun purchases, etc. After Muhammad shot the friend of his ex-wife and shot a couple people on golf courses and started travelling across the country committing other crimes, a query might have put all those things together and come up with a short list of possible suspects.

Of course, that dream DB could be put to other productive uses as well. Such as electronically ruining the lives of political enemies, checking up on all the people who've ever read, say, Desert Solitaire, and other good things. We'd be an infinitely safer society, and many crimes would be prevented in the envisioning stage. The DB could even record cases where suspects were trying to avoid the DB, such as by using cash as much as possible. And, even easier, it could record those who say bad things about it. Then, those suspects could be dealt with.

As for the sniper case, all that was needed to solve it earlier was a) better police work, b) better cooperation between humans, and c) less PC.

Cops have made statements along the line of "we were looking for an angry white man in a white van." Moose knew from the first letter that there was a good chance the perps had some Jamaican connection. He stated the perps could be any race. Which is a good thing to say publicly, but he should at the same time have informed police that they might be black.

Then, there were the five or so times that calls from the snipers were ignored.

And, most important of all, two independent witnesses to the shooting of Pascal Charlot spotted a dark Chevy Caprice. There was some question over the color, and a mini-truck was spotted at the same time. However, that was a very key piece of evidence that was more or less ignored. It's no surprise that that was the only shooting that occurred in D.C. (insert Chandra Levy case here). Perhaps the D.C. police didn't inform the rest about the Caprice spotting, or it was ignored for one reason or another. The D.C. police have said that they were looking for a Caprice.

You put all that together and you see that, in addition to talking about the white van and the white truck, they should also have talked about the Caprice. There are many fewer old Caprices driving around than white trucks.

Any one of the dozen or so cop stoppings of the Caprice could have turned into either an immediate arrest, or further investigation. That is, if the cops who had stopped the Caprice had realized that a Caprice was being looked for. No ultra databases needed there, just telling the various agencies to look out for Caprices in addition to white vans and trucks.

And, note that the witness reports of the Caprice are in an entirely different category than all those tips that were being called in. There were thousands and thousands of tips being called in. That's too much for a human to deal with. The cops already have software to record and sort leads like this. That's a good use of DB technology.

However, there were at most a few hundred data points from all the shootings. By data points I mean witness reports, shell casings, tarot cards, surveillance tapes, and all the other evidence. A human can easily handle this amount of data. One human should have seen how important the Caprice sighting was and should have made sure that it was being searched for as well.

Bear in mind that the snipers were spotted by a citizen who heard on his scanner that the cops were looking for a Caprice. If that report had been promulgated earlier, they probably would have been caught earlier.

Parenthetically, you can read my theorizing in the archives. I, Jim Henley, Freerepublic, and many others spent time trying to figure out why the white truck couldn't be found. Muhammand and Malvo were to a certain extent living under the radar; a mega DB would have to include, say, spy cams at rest areas to record entries and exits and be hooked into the YMCA.

ObImmigration: Did you know that 58% of Mexicans AGREED with this question: "the territory of the United States' Southwest rightfully belongs to Mexico."

For the full poll results, click here. At what point in time do we start putting quotation marks around "country" and "citizen"? For more on, say, Jorge Castaneda's threat to being "propagating militant activities," see the entries below or the comments here [2].

What if we polled American citizens with this hypothetical question:

"58% of French citizens believe the Louisiana Purchase to be invalid, and believe that the state of Louisiana rightfully belongs to France. The French government does not discourage French citizens who emigrate to Louisiana. Should we allow those 10 million French citizens currently in Louisiana to become U.S. citizens?"

I would hope the numbers would be about 95% against.

UPDATE: The Pascal Charlot witnesses might have been confused over the make/model of the car, but even if that wasn't resolved by showing them various makes and models, it would still indicate an old Chevy.

A $250 Chevy is an urban ride; I'm sure it more or less stood out like a sore thumb in suburban Washington SUVland or pickupland further south.

The guy(s) who caught the sniper at the rest area had the car's license plate number; the cops had not released that info, they'd got it from their scanners. However, even if the cops didn't have a license plate number, like I said Caprices aren't that common.

-------
[1] volokh.blogspot . com/2002_11_24_volokh_archive.html#85728876
[2] mattwelch . com/archives/week_2002_11_24.html#1529