More on Metadata

This has been passed around for a while now, and it’s fascinating.

A certain cyber spook who shall remain nameless at this juncture told me about this sort of data analysis many years ago. It was just one or two little sentences along the lines of, “You would be AMAZED at the sort of things “they” can find out about you.” As I say; that was many years ago, and orders of magnitude in increased capability have been achieved since. As a merchant having done installment sales for more than a decade at that time, I had a database of my own, consisting of thousands of accounts, and immediately could see the power of what on the surface would seem like rather mundane and mostly useless points of information.

BUT, when you ask the right questions, and can sort by multiple data fields, WOW! We used it to make buying decisions, which were extremely important and had to done months in advance of peak selling seasons, and so it involved recent history and long term trends. It was also extremely useful in predicting patterns and indicators of contract violations. If I told you half of that stuff I’d be branded a scumbag, a bigot and a general son of a bitch. Its one and only purpose though was to avoid going bankrupt– You make the wrong decisions and you’re stuck with major inventory you can’t pay for, or you get it out the door and don’t get paid for it, and so you can’t pay for it.

So yes; we profiled in a very big way, you could say. At one time the default rate was so bad that we required a credit check of every prospective customer, but with analysis of a large enough database were able to do far better by asking for only a few simple bits of information.

Learning important things doesn’t need to come from positive data either. That is to say, what isn’t there can be as useful as what is there, when forming the kinds of association tables in the link, or plotting behavior patterns. For example; knowing those times when someone is unaccounted for (enter the Smart Grid, to know when someone is not at home), when compared to certain events, over time, can tell you a whole lot.

So I now wonder what a look at the metadata in the time and place of Jesus (arguably one of the most influential people in human civilization) would show. How about Bedford Falls of “It’s a Wonderful Life” (thoughbeit a fictional town)? In that story, you could say that Clarence showed George Bailey where he fit in the metadata analysis.

The Soviets, using infiltrators, both of the overt and the covert kind, were able to come up with their various arrest and kill lists in much the same manner.

That brings us to another issue, that being the concept of the inception i.e. a planted idea– The tiny seed of an idea that can germinate in the minds of people, multiply and grow into something large. The looks into associations in the link are all very interesting, but being able to predict or pinpoint inception sources is also very interesting.

Reading about the various groups and group associations in the link got me to thinking about the modern version of the 1700s pub. You’re looking at it. Sort of, but it’s much more open to the world.

I was also struck by the modern lack of such clubs and associations. We’re a much more fragmented society. We don’t meet with our neighbors much at all anymore, whereas it used to be common. We often don’t know anything about them.

But your government sure as hell can find out.

And I submit that this is by design. It’s either by the design of mere mindsets, which will reflexively respond in predictable ways, or by some mastermind scheme. I say it’s a combination of both, each nourished by the other.

Call it a leap into Crazyland if you like (I’ll be quick to say I told you so, later) but looking at the data points over my lifetime, things make a lot more sense if one assumes that our government is in a full-on war against the ideals and principles, the foundation of the U.S. and has been for decades. When other people are saying “W. T. F.!???” and getting all exasperated and confused, I’m saying, “Well sure. It fits perfectly don’t you know.”

Try it. Ask yourself what institutions, traditions and associations would be under attack or already under control if our own government were at war against the American founding principles and the American economy. What groups would be funded worldwide, which ones given a pass, and what groups would be targeted for degradation, intimidation or destruction. What industries would be targeted for nationalization? Yup. Gotcha. On all counts.

So accept it. Then you won’t be surprised or confused anymore. The specific organization of this war isn’t so important, not so much as that you realize it’s a war. And we’re losing, mainly because we’ve not been willing to face up to it’s full import. It’s too crazy. Too horrible. We can’t go there. We’re cowards. Look and listen to the pundits. Any of them, no matter how great you think they are. They’re giving you data points that they themselves are unwilling to connect. We’re all afraid and so we’re very easy to distract. Oh look! Gays!


3 thoughts on “More on Metadata

  1. The omissions are just as telling as the affirmative data. That means that when you change something, you have to change everything. If you are moving to encryption, you have to move to encrypting everything. If you encrypt just your email to a handful of people, that tells them everything they need to know. Encrypt everything all the time.

  2. “Encrypt everything all the time.”

    Not enough: “they” can still do traffic analysis. Seems we need to make the traffic metadata itself just a bunch of “noise”.

  3. Noise won’t work unless it’s on a scale of >50%.

    I worked for a company who did shopping data analysis and then printed coupons based on what you bought, what you didn’t buy, and what changed in your buying habits. No shopper card? That’s fine, you probably paid with credit card / check (hint, if you’re afraid of being tracked and thus don’t have a shopper card but use anything but cash, you’re losing money because the card works).

    Now, I’m sure you all see that cashier who has a card handy and just scans it anyway. That’s the ‘noise’ your talking about. It gets filtered toute suite.

    The other thing you need to do to make noise harmful is to make it random. You’re a human, you are not good at random. Try this – have someone write on a board / paper “heads” or “tails” 50 times or so, then have them flip a coin 50 times and write the result. You’ll be able to pick out their seq. because it will be MUCH less random than the coin toss (nobody wants to see 0,1,0,1,0,1,0,1 because THAT’S NOT RANDOM!!!!)

    None of this would bother me if the data was used for its intended purpose – tracking the anomalies and beating the bad guys, but it isn’t going to be. It’s going to be used to squash dissent and crush the American citizen.

Comments are closed.