Total Information Awareness?!?

Here's an example of the BS that is being shoveled to protect this harmful boondoggle:

"Let's say lots comes in and we don't see anything interesting," said a source who helped develop the technology. "Tomorrow we find out someone is communicating with a known terrorist. When you go back and look at the past data, there may be information that you missed. A pattern that was meaningless suddenly makes sense."

This is patently idiotic. IF we knew who the terrorist is, they would be under surveillance and all the calls they made would be tracked. This program is being built because we DON'T know who the terrorist is. It is attempting to tease potential terrorists out of the data on patterns of behavior where no link to a known entity exists. In short: a very harmful boondoggle.

My money is on the cop, trained in counter-terrorism (like NYC and LA is going to), that works the beat.

Comments

[Andy]

Investing in human solutions over giant technologically based data dumps? Never happen John...

[John Robb]

Too bad, it is the only thing that works in both counter-terrorism and counter-insurgency.

[Andy]

Like I said last week, the situation is either not serious enough for radical change or the system simply doesn't care (or "get it.")

[Don McArthur]

John, with your background in finance, I'm sure you see the similarity to computer programs designed to digest years of equity and commodity price/volume data and spew out buy/sell signals -- none of which work. LOL.

[Dimitar Vesselinov]

The NSA’s New New Phone Database
http://www.radioopensource.org/the-nsas-new-new-phone-database/

[Michael]

I would be interested to see the press reaction to the first Representative or Senator who puts forth legislation asking for enough cops to triple the forces in every major metro area. Most certainly a good idea, but equally certain to bring calls of “See, I told you they want a police state!” The only saving grace is that like the technical programs now being lambasted, most average folks would probably welcome such an approach. Now, to find the right lawmaker . . .

That the worst-part of politics rules the process by which we live is unfortunate, but it is what we have to face up to.

[a z]

Michael: I doubt people would be upset by tripling the police force. I suspect most would welcome it. Only if the police literally searched the population in there home on a daily basis would a majority complain.

Doesn't look like Americans give a care.
http://www.washingtonpost.com/wp-dyn/content/article/2006/05/12/AR2006051200375_pf.html

Although since the incredibly small sample was done via phone, one wonders if doing a poll about phone tapping via the phone is a rational methodology.

It seems to me the most ridiculous claims sold by the administration (merely taking about the program helps terrorists, Bush's actions are legal despite no legal citations etc etc) are just eaten up by the American people like spicy Cheetos(TM).

Truly our government does represent the people.

[Michael]

"one wonders if doing a poll about phone tapping via the phone is a rational methodology."

Like sending someone details about a Word virus in a Word attachment? ;-)

[Larry]

John, you seem to be assuming the Administration actually has an interest in catching terrorists.

Given that the most popular worst case scenarios involve the deaths of millions of predominantely Democratic voters in "Blue" states, what you or I might see as a catastrophic system failure would be seen by the Powers that Be as a brilliant success.

[Gina]

By monitering phones calls one is assuming that all terrorists are stupid. Monitering phone conservations only drive them to find better ways of communicating making any effort to find them even more difficult. Taking this action is not in hopes of finding something but is only a device to indicate to the American Public that our government is doing something.

[Walter]

John,

While I think this program is detestable, and I agree with your earlier post that suggests the unbridled power of our own government is a far greater threat to our way of life than Islamist terrorists could ever be, I believe that the reasoning of that unnamed source is not quite as idiotic as you make it out to be.

There are at least three scenarios that you're not accounting for.

(1) For a known terrorist, you're probably correct that all the calls they made would already be tracked, but, strictly speaking, the previously revealed NSA eavesdropping only applied to the calls of a known terrorist when one of the endpoints of the call was outside of the US -- so not *all" calls would have been tracked, only domestic calls. I doubt that it was ever actually true that we weren't tracking the domestic calls of known terrorists, but that has not previously been publicly acknowledged, and in any event that tracking would not have extended to the terrorist's callees' calls, and so on, so network analysis would be truncated at just one level deep.

(2) Someone calls a known terrorist for the first time. Prior to their call, the person might not have been on anyone's radar. With a database of all calls ever made, the NSA can see who else this person has been calling, and who those people have been calling, and so on, and do analysis on that network of calls. (God help the sap who dials a wrong number.)

(3) Someone becomes a known (or suspected) terrorist for the first time, not through any analysis of call patterns but through some out-of-band discovery. We could then analyze their domestic calling networks, and it is entirely conceivable that, given some out-of-band evidence to point us in a certain person's direction, such analysis could strengthen our confidence that they're a potential terrorist and reveal new suspects (well, it would definitely reveal new suspects -- whether they would be *quality* suspects is the rub).

I think all of these are non-idiotic use cases for a domestic call database.

You've also expressed skepticism regarding the ability to do effective data mining. I share your skepticism, and your fear of the harm that will be done by flawed mining (both the sins of commission from wrongly harassing innocents as well as the sins of omission stemming from a foolish overconfidence that no terrorists can elude our NSA super-brain). However, I am reminded of the game-changing revelatory power of something as simple as the Google search algorithm, imperfect and limited as it may be, and therefore I think it's possible that the NSA can achieve some remarkable results if they get the data and the algorithms right.

I wouldn't be so afraid of the program if I thought it was a boondoggle, doomed to fail, but unfortunately I don't think it is, and it would be much easier to make the case for opposing the program if it was "patently idiotic," but unfortunately it is not. Collecting data on every call made in America *would* make it possible for the government to identify and track relationships between people in very powerful ways. That is exactly the danger and exactly the reason why this program must be opposed, yet it is simultaneously the grounds on which the current administration will defend the program and the reason why a majority of Americans will *not* oppose it.

Regards,
Walter

[John Robb]

Thanks Walter. All the of the items you mentioned are covered via FISA. The first is done through initiating tapping and getting the warrant afterwards. The second two are found in records that are at the phone company and could be accessed with a warrant.

The only reason to do it this way is data-mining for terrorists based on patterns in the data alone. That is impossible. I don't care how smart you are. It is also a gross violation of personal privacy.

Of course, the program is being defended based on the items you mentioned, which are reasonable but didn't require any change from previous behavior.

[Walter]

John,

Those are good points. Much of what they hope to do could already be done, legally, if they were willing to bother with getting warrants, which they aren't. But I'm not sure it *all* could be done currently.

First, while I'm not an expert on FISA, I'm pretty sure that *none* of the items I mentioned are covered via FISA, or at least not completely covered. FISA applies to agents of *foreign* powers. To get authorization to eavesdrop on a U.S. citizen under FISA, you have to demonstrate probable cause that the citizen is colluding with a foreign government or at least a foreign group of some sort, and that the surveillance is a *necessity* for national security.

I don't know (because *nobody* knows outside of the FISA judges and a few others) how strict FISA judges are about "probable cause" -- maybe they freely allow fishing expeditions, or maybe they don't, but if they *do* allow fishing expeditions I think they would be contravening FISA.

I don't think it's possible to use FISA to get authorization to conduct surveillance on suspected domestic terrorists, including, e.g., radical Islamists who might develop from amongst American-born-and-bred Muslims (perhaps Iranian-Americans who just can't understand our God-given right to launch unprovoked airstrikes against whoever we want to), etc.

Of course, warrants could still be obtained from non-FISA courts.

Second, I'm not sure (perhaps someone in law enforcement will chime in) if warrants for phone records, whether obtained from FISA or other courts, extend beyond the individual suspect. If you demonstrate probable cause and get a warrant to view the phone records of one person, does that give you the right to *also* subpoena the phone records of all the people that that person had talked to on the phone? I would be astonished if that were true. I have to think that you would have to demonstrate probable cause for each additional individual/business whose phone records you wanted to see, and I think it would be nearly impossible to do so.

Third, are there any regulations governing how long or in what format phone companies must store their records, or how quickly they must comply with subpoenas? I've heard some companies keep records for 12 months, some for 18 months, probably none of the phone companies keep records all the way back to, say, September 10, 2001. I gather from TV shows that phone companies *can* turn over records in a matter of minutes or hours, but do they have to? Phone companies refuse to comply with some subpoenas, after all, and I imagine their lawyers spend at least a day or two determining that such and such subpoena is one we can and should reject.

Obviously it would be much easier and faster and just generally better if the government had records of all the calls ever made, dating back to the inception of this program, in their own massive database, that they could instantly access and analyze without first obtaining bothersome warrants (assuming they even could get warrants, which for any kind of network analysis extending beyond the root level would probably be impossible).

Finally, as you've already said in your other post, the NSA will be able to add all kinds of other helpful information into the data warehouse, and then it will be even more useful, and then they can use it for monitoring and restricting the free association of citizens who are prone to asking irritating questions, demonstrating in public, and so on. They could never do that under existing law.

You gotta look at the big picture, John -- it's really a *beautiful* thing, as long as you don't give a crap about freedom and democracy. (Of course, any real terrorist is going to be using throwaway cells and anonymized, encrypted VoIP/text messaging, but let's not mar the beautiful vision with ugly reality.)

- Walter

[Walter]

Here's a blistering attack on the phone records database from a fairly conservative guy, Joe Scarborough:

(2 minutes, 30 seconds)

http://movies.crooksandliars.com/Scarborough-Country-Big-Brother.wmv

[John Robb]

That was excellent. He is exactly on target.