• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Applying Particle Physics to Solving Terrorism

Well, don't you use datamining to gather, organize, and make sense of the data?
 
Not when you are plotting news events (deaths) against time on a logarithmic scale. Unless, by data mining you meant that very general and unintrusive act -- if you did, my apologies, I read it as leaning toward invasion of privacy issues.
 
Marplots,

So the data gathered solely from news events? Do you think these curves will eventually be able to be used to plot traditional crime trends as well as terrorist-actions?
 
Marplots,

Do you have the answer to the question I posted in reply #23?
 
INRM said:
Marplots,

So the data gathered solely from news events? Do you think these curves will eventually be able to be used to plot traditional crime trends as well as terrorist-actions?
Click to expand...

From what I read, they were commenting on a type of pattern (power law) common in nature and seen in human affairs as well. The first time I ran across it was in earthquakes -- the bigger the quake, the fewer.

It's interesting, but it isn't predictive, other than in the most general way. So, for instance, when they talk about an earthquake being 'due' they mean it in the same sense that a roulette player would say 'black is due' because there has been a long series of reds. In no way does it specify that any particular event will happen in any particular location, just that historically, the pattern emerges.

So, for instance, Tim McVeigh comes out of the woodwork and does his deed. The act could not have been predicted by a logarithmic relationship but it fits the curve we are talking about.

Because this is not valuable for prediction, historical data works fine. You could plot crime statistics as well from public records or even news reports.

The only important thing comes up when you can dig deeper into the mix and see if there is an underlying cause driving the rule. That's tougher. It is, however, just what we do with the stock market when we try to predict stock prices. We know there is a random element but we believe that enough information will allow us to pick winners. The difference with terrorism and the stock market is that in the latter, you only have to choose better than the other guy who is investing. In terrorism, you want to predict well enough to influence the outcome -- much harder.

If you think about the level of knowledge needed to predict and prevent a criminal act (and the very short time scales involved) you'd have to agree that the task is nearly impossible.

In my opinion, what works better than observation alone is getting your guys involved and pushing things along until you identify a propensity towards a criminal act. That means undercover folks and snitches mainly.

Although means and methods are often secret, I think you would find many preventative prosecutions are done exactly that way -- someone from law enforcement joins up and pushes the group toward a concrete act that can be charged.

I hope, somewhere in this screed, I've actually addressed your question.

Oh, and by the way, if you haven't yet checked it out: look at "Top Secret America" over at the Washington Post site. A whole network of government agencies and contractors busy as beavers on the hush-hush. Amazing, really.
 
Countdown until someone suggests nuclear physics has a solution for terrorism in 5...4...3
 
Marplots,

Because this is not valuable for prediction, historical data works fine. You could plot crime statistics as well from public records or even news reports.
Click to expand...

And you could do this accurately enough to prevent terrorist attacks without using a lot of computer processing power, and using computers to automatically siphon up a lot of data?

In my opinion, what works better than observation alone is getting your guys involved and pushing things along until you identify a propensity towards a criminal act. That means undercover folks and snitches mainly.
Click to expand...

To catch terrorists, or all forms of crime?

Although means and methods are often secret, I think you would find many preventative prosecutions are done exactly that way -- someone from law enforcement joins up and pushes the group toward a concrete act that can be charged.
Click to expand...

Unless that group was ready and willing to commit the act, that's entrapment. Regardless, there are a number of cases where the government appeared to do just this.
 
INRM said:
Marplots,

And you could do this accurately enough to prevent terrorist attacks without using a lot of computer processing power, and using computers to automatically siphon up a lot of data?
Click to expand...

Yes. You could even do it by hand if you like. But it will not be helpful in prediction. My best analogy was with earthquakes. You can create the same type of power-law graph by plotting Richter scale numbers vs time on a logarithmic plot.

You would get a graph that showed the same pattern. The graph wouldn't help you predict earthquakes. So, we aren't talking about prevention at all. In fact, if the pattern holds across different historical periods (and different ways of fighting terrorism) it might be telling us that terrorism cannot be prevented. That would be an interesting result and would hold even if some individual acts of terrorism are stopped -- the pattern might hold anyhow, so that overall, terrorist attacks as a class of social phenomena might be as 'natural' as earthquakes.

This latter result wouldn't be a happy one at all.
 
Marplots,

So you're saying that this wouldn't be accurate without having to collect a lot of information?

Because, if that's the case, it would either require the government to collect large amounts of information on people without warrants; or it could be used to justify some sort of false-claim of "probable cause" on excessively weak grounds, and would, in turn, result in the government getting the license to go on massive fishing expeditions to preventively enforce the law.

You would have to be a fool to believe that this would not eventually be used for routine law-enforcement, even if it's original intention would be to stop-terrorists only.
 
Last edited:
I don't know how to say it any different than I already have. It's a historical technique that isn't predictive. It wouldn't be a justification to expand capabilities, but I've seen weak justifications used before -- remember when they shut down air flights because of the volcano in Iceland?

Another weak justification was the identification taggant issue with high explosives. To my recollection, they were only ever used in one prosecution. That program was eventually dropped. But weak justifications are used all the time. This part of the situation isn't science, it's politics.

Another example that occurs to me is the mammogram guidelines. So, yes. Government does use inflated science or scientific ideas without thinking things through. We all do. I think this is why we support skepticism and critical thinking.
 
INRM said:
And rather than discuss the merits of the argument, we instead decide to attack the author...
Click to expand...

Probably wouldn't have if it were not palpably the same argument made for each new development you bring up. And, Occasionally, I try to remind you that most large and most small governments could do the exact equivalents in their time and many did - some to an extent that was far worse for their citizens than anything you have come up with so far. Nothing you or I or anyone on the forum can do is going to stop any of these developments and short of insanity, or G. Bushism(a form of insanity - oddly, also paranoia) (ie, I'll start worrying when/if the republickers get back in)none are likely to go the way you are worried they will.
 
You must log in or register to reply here.

Back
Top Bottom