Bar-coded Currency

[The idea]

Bills already have serial numbers. How difficult would it be to give them bar-code serial numbers that can be read by bar-code readers?

Benefit #1: If a given batch of counterfeit money uses a single bar code, then counterfeit money is more likely to be detected at the point of sale when the bar-code info gets into a system that has been notified of the offending code.

Benefit #2: It would be easier to trace money when people are trying to launder the proceeds of crime.

 

[Earthborn]

Old hat.

There are plans to microchip euro banknotes. Although the mentioned date seems a bit unrealistic now.

 

[evildave]

They already have OCR that'll read the serial numbers off existing bills, so why put barcodes on money that will get all those religious and privacy rights people up in arms out about "computers comming to get me"?

They can all remain nice and complacent about how their banks (and the federal reserve) are recording where money goes and comes from.

After all, they record what $20 bills they put in in what ATMs (in case ATMs are robbed), and cash is collected from stores in seperate bags for the individual retailers (and scanned again for conterfeits).

Simple matter of databasing and connecting the dots, and the bank knows pretty much where you spend your ATM cash, and then they can monitor where you habitually spend your cash, and (for instance) put up ATM ads about similar businesses, or sell the list.


To make this simple, imagine they scan and capture the numbers, and load up an ATM. It has the following bills in it:

1,3,2,4,5,9

You are the fifth person to get a $40, so you got bill #5 and #9. You are uniquely identified by your ATM card.

Three days later, they scan the incomming armored car drops from Vons Supermarket #41293. There's #9 in it.

The next day, there's Aarco #34765 that has #5 in it.

Over the year, all the places your money has ended up at (that they collect from) has been added to a database indexed to your ATM card.

Patterns emerge. That same Vons turns up 40 times, and Aarco turns up a dozen times. As well as certain restaraunts and such.

The bank sells the database containing 100,000 names, with individual shopping habits to various "interested parties".

Like Health Insurance companies that would be interested in your shopping and eating habits. They also buy the "Vons Card" data and cross-reference, the UPC codes that were recorded with your card. You're a "junk food junky" and thus a "bad risk".

Not to mention all the extra targeted advertising that comes your way.

Of course, all of this is even easier if you just swipe that credit or debit card. The point is, there ain't no privacy, even if you use cash.

 

[Grammatron]

They already have OCR that'll read the serial numbers off existing bills, so why put barcodes on money that will get all those religious and privacy rights people up in arms out about "computers comming to get me"?

They can all remain nice and complacent about how their banks (and the federal reserve) are recording where money goes and comes from.

After all, they record what $20 bills they put in in what ATMs (in case ATMs are robbed), and cash is collected from stores in seperate bags for the individual retailers (and scanned again for conterfeits).

Simple matter of databasing and connecting the dots, and the bank knows pretty much where you spend your ATM cash, and then they can monitor where you habitually spend your cash, and (for instance) put up ATM ads about similar businesses, or sell the list.


To make this simple, imagine they scan and capture the numbers, and load up an ATM. It has the following bills in it:

1,3,2,4,5,9

You are the fifth person to get a $40, so you got bill #5 and #9. You are uniquely identified by your ATM card.

Three days later, they scan the incomming armored car drops from Vons Supermarket #41293. There's #9 in it.

The next day, there's Aarco #34765 that has #5 in it.

Over the year, all the places your money has ended up at (that they collect from) has been added to a database indexed to your ATM card.

Patterns emerge. That same Vons turns up 40 times, and Aarco turns up a dozen times. As well as certain restaraunts and such.

The bank sells the database containing 100,000 names, with individual shopping habits to various "interested parties".

Like Health Insurance companies that would be interested in your shopping and eating habits. They also buy the "Vons Card" data and cross-reference, the UPC codes that were recorded with your card. You're a "junk food junky" and thus a "bad risk".

Not to mention all the extra targeted advertising that comes your way.

Of course, all of this is even easier if you just swipe that credit or debit card. The point is, there ain't no privacy, even if you use cash.

There are plenty of flaws in your example. The most important of all being that money circulates way more than that. People at Vons and Arco give out change to other people and that #9 and #5 might end up in somebody's pocket and/or a place that would make it in no way possible to trace it back to if the person who withdrew the #x actually spent it.

 

[The idea]

cash is collected from stores in separate bags for the individual retailers (and scanned again for counterfeits).

My suggestion is that they scan it when they take the money, just the way they scan products. Otherwise your $10 bill goes into the till and then it goes to another customer who then spends it, etc. The scanning you are describing only gives occasional snapshots.

 

[evildave]

Your $20 bills come out of the ATM, just like everyone else's $20 bills. Most people use the ATM. Period. Any smaller denomination is pure white noise.

People who have denominations LARGER than $20 are rare curiosities. Think about the last time you actually laid your hands on a $100 bill. Unless you go into the bank and ASK for bigger bills, you will get a wad of $20 bills. It's the only currency ATM machines give you (and everyone else). The larger the bill, the more unique, trackable and obvious its journey.

How likely are you to CONSISTENTLY receive change from purchases made with $20 bills, in anything but bills SMALLER than $20 bills? I'll tell you. Darned near zero. The cashier will almost always hand you back bills "too much" over the amount on the register. Most retailers put most $20 and above bills in a SEPERATE place. Watch 'em. Under the cash drawer. Into a slot in the counter. They don't expect to give out $20 bills as change, because it's extremely rare to get anything larger than a $20 bill. If someone makes a grab&dash, they get lots of little bills, but not much money. The $20 notes you get as change from a larger note comes from a bin in the cash drawer. The $20 notes you give go under. Most people who don't use $20 bills are using ATM or credit, which is far more tracable.

All retailers REQUIRE a constant supply of monies to replace the smaller bills they give out in change, and most of the $20 and above bills collected go right back to a bank the very same day. They ALSO happen to end up stacked earliest-to-latest, as they're received. Useful.

There is plenty of "noise" in the system, but buying patterns stand out. Most people keep going to the same stores. If it only hits 40% on your family's buying habits with this techniqe (and it will probably do better than 70% - unless an individual routinely pays other individuals with a LOT of cash ~ something "homeland security" would be interested in), your most commonly visited retail, restaraunt, etc. habits will be highlighted clearly with a quick sort, and by looking at the larger counts on the end of the list.

 

[The idea]

Your $20 bills come out of the ATM, just like everyone else's $20 bills. Most people use the ATM. Period. Any smaller denomination is pure white noise.
[...]
People who have denominations LARGER than $20 are rare curiosities. [...] Unless you go into the bank and ASK for bigger bills, you will get a wad of $20 bills. It's the only currency ATM machines give you (and everyone else). [...]

They don't expect to give out $20 bills as change, because it's extremely rare to get anything larger than a $20 bill. [...]
All retailers REQUIRE a constant supply of monies to replace the smaller bills they give out in change, and most of the $20 and above bills collected go right back to a bank the very same day. They ALSO happen to end up stacked earliest-to-latest, as they're received. Useful.

Those are interesting points.

 

[waitew]

Why have money....Bills,notes..etc...after all?Why not simply bar code everyone..say on their wrist or forehead & have a cashless society??..oh..wait ..that sounds to much like the mark of the beast!!Better not go there just yet..But I believe it's coming.

 

[evildave]

A transponder chip. Just a tiny sliver of a thing. No "mark" at all.

The technology's just about perfected for pets. Wave a wand over it, and "boop", name of pet, owner, address, etc. Very small. Implant just about anywhere.

Just need to encode enough details and trivial bits so nobody digs out someone else's chip and wears it in themselves. A picture, name, some biometric details, etc. that could pop up on cheap scanners, and an ID to lookup for anything else they need to know. Of course, with this much, it doesn't need to be implanted. If someone else has your card, it's useless to them.

Anyway, the chips are theoretically so cheap they're seriously working on embedding them with UPC codes. So you walk out the door of the grocery store, it goes "boop" and rings up everything in the cart, and you too. It all shows up on a bill to settle. It would make shoplifting a LOT harder. You just end up paying for it on your way out the door. Basically, this is accomplished by encoding the identity of the product and a unique serial number. The ID says "eggs", the serial number uniquely identifies one set of eggs. So you walk out the door and it uniquely identifies however many cartons of eggs you take.


The only thing preventing a totally cashless society is the notion of "privacy" (that I've effectively demolished), and some form of inexpensive, pervasive and secure system for performing exchanges. Having bullet-proof identification would be a key requirement of the system.

It will certainly make criminal proceeds a bit more difficult to collect on. After all, if every transaction is recorded and electronically monitored, it puts the criminal trade thoroughly on a "barter" economy. I can see why many religious leaders would be in absolute terror of it. Once you have all that privacy to collect and mismanage funds to your heart's contentment, free of any oversight, all of a sudden your church get tracable funds and accountability. Horrors!

 

[ktesibios]

Dave, it smells very strongly as if you're advocating a total surveillance society- in other words, the sort of police state that even the antheap experts of Red China never dreamed possible.

Think of this- while such a society might well make life very difficult for people you don't approve of, what guarantee do you have that you yourself won't become one of the people who some utterly unaccountable bureaucrat (one of the breeds who would flourish in a system of total social control) doesn't approve of?

Welcome to my ignore list.

 

[evildave]

You'd 'ignore' someone for discussing how it can be done? Sorry, but that sounds pretty stupid to me.

A wise person would pay attention to how it can be done, rather than bury their heads and pretend nothing 'bad' will ever happen if they refuse to understand any of it at all and try to recapture their blissful igborance.

How would you even RECOGNIZE the construction of such a system, if you don't understand how it would work? Easy. You couldn't possibly recognize it.

Research and development would need to be solicited and funded. The feds would approach various colleges and banks to build a prototype involving bank note OCR and tracking. This action could be identified, publicized and stopped.

Someone would have to pass an act that required banks to scan money and report numbers. This action could be identified, publicized and stopped.

Someone would have to pass an act that funded a big, fat database and secure banking network to it. This action could be identified, publicized and stopped.

Someone would have to provide law enforcement access to it, with or without court ordered search warrants. Believe me, by the time they're talking about how law enforcement can access it, it's already built, and too late for you to cry over your self-afflicted ignorance and inaction.

SOMEONE would have to PAY ATTENTION, rather than try to shut out all the bad things they don't like in life.

But what do you care, ktesibios? I'm on your 'ignore' list. You don't want to know, so go cry me a river when you wake up one morning and realize the 'total-surveilance' system is already in place. But what do you care now? They're only BUILDING it.

We already have the USA-Patriot Act to make such research and development a state 'security' secret. So, really you SHOULD ignore it. It's none of your business.

Be complacent.

 

[Earthborn]

Dave, it smells very strongly as if you're advocating a total surveillance society- in other words, the sort of police state that even the antheap experts of Red China never dreamed possible.

I don't think he is advocating it. It seems more likely to me that he is arguing that the technology for such a society is already here, and will be pretty near inevitable that it will become a reality. There are still a few problems with it though...

Of course, with this much, it doesn't need to be implanted. If someone else has your card, it's useless to them.

Unless a hacker hacks the chip and changes the data.

So you walk out the door of the grocery store, it goes "boop" and rings up everything in the cart, and you too. It all shows up on a bill to settle. It would make shoplifting a LOT harder.

Not that much harder. Already many products have little transponders on them and there are scanners near the door. Still there are people who are inventive enough to get the products through the scanner without setting it off. How? Bags that are lined with aluminium foil on the inside. It is a dirty little secret that shopkeepers have known this for years and have gradually cranked up the power of these scanners. They are now a lot more powerful than when they were introduced, so people must use steel plating in their bags instead of foil. And some people do. (And nobody is protesting this increase in radiation, although it is much more than antenna's for mobile phones everybody is complaining about, and people get a lot closer!)

If such technology is really introduced, you are creating a strong incentive for people to try to circumvent it. And nature will find a way.

After all, if every transaction is recorded and electronically monitored, it puts the criminal trade thoroughly on a "barter" economy.

True, but it already largely is. You don't think criminals send suitcases full of dollar bills across the globe, do you? They use very different currency: stolen art, antiquities, fossils, drugs, weapons, people... What is the primary business for one criminal, is exchange for the other.

 

[evildave]

You ever try to change a ROM chip that's been burned to contain data? (Not a PROM chip, or EEPROM chip.)

I suppose with a very good high-resolution microscopic 'xray'* device and some sort of ultra-precise (sub-micrometer) tool drilling, that fails to contaminate the chip with surrounding debris, you could begin setting bits (but not clearing them - for that you would need to bridge gaps). By the time you had barely begun, the CRC would be bad, and a lot of "extra" bits would be set. You'd need to change millions of bits this way.

Hardly the sort of hardware most people would have, and you're caught the first time someone routinely runs the serial number and compares the results.

Then there's always masking your own custom chips. Dreadfully expensive, and you're "caught" the first time someone runs the serial number against the database.

Yes, there are still ways a shoplifter could find and disable a UPC chip. Of course, that requires privacy and time, neither of which is really available in a store. For instance, placing a scanner at the restroom and dressing room entrances, and entrances to parts of a store with expensive and concealable items, you could reasonably expect all of what checked in, checked out again. An automated system could do that. "Oops! You forgot something? Well, we can expect to find it intact on the bench where you left it, can't we?"

Of course, what such a blind, technical solution can cause is when children shoplift, it blindly goes onto the parent's bill. But that might be better, after all, it's FAMILY business when the kids grab stuff off the shelves. The store properly charged for it, and has the video records that virtually all stores already have today.

* It needn't be 'xray', but any device to peer through a solid mass of plastic gunk that they embedded the chip in.


A new "Left Behind" interpretation
Another interpretation for the "Mark Of The Beast" crowd: instead of bar codes on upstanding and law abiding citizens, perhaps it might be scars from the criminally minded (or paranoid, or plain stupid) who dig their chips out. After all, a pinhead-sized device could be implanted without leaving any kind of "mark" at all, but digging around for it with a knife would certainly scar - AND BADLY.

Certainly people who choose to be completely unaccountable might be doing so because they're doing bad things. So, naturally, for any sort of mass evacuation, these people who "don't scan" might very well be "left behind", especially if space or weight is at a premium (and it usually is).

 

[Ed]

But, if I forged a bill using a legit serial number, it would appear in the database, right? So they would need some fancy programming like the fraud protection stuff that Visa and Mastercard use so that "patterns" of appearence of a bill would set off an alarm. Sounds messy to me. I think that the problem will be with us until we have a cashless society.

 

[Tony]

Is a cashless society even possible? Won't there always be a need for hard currency or something of it's equivalent? I see A LOT of black-market activity going down if such a cashless society ever manifests itself.

 

[evildave]

No, a forged bill would (eventually) appear in more than one place at the same time. Instant counterfeit detection. Red flag. Secret servic auto-notified. Someone go dig that one up.

You OCR the SN, date, value, etc. SN earlier than date? Forged. SN mismatches value? Forged. SN not used yet? Forged. The more advanced scanners actually look for common conterfeiting errors. Scan for that little bar, look at the color, etc.

 

[Ed]

No, a forged bill would (eventually) appear in more than one place at the same time. Instant counterfeit detection. Red flag. Secret servic auto-notified. Someone go dig that one up.

You OCR the SN, date, value, etc. SN earlier than date? Forged. SN mismatches value? Forged. And the more advanced scanners actually look for common conterfeiting errors. Scan for that little bar, look at the color, etc.

I think not. How many bills are out there? According to the US Treasury it is about 16 billion. Eventually they would get a hit, then what, recover a $20 bill?

 

[evildave]

Is a cashless society even possible? Won't there always be a need for hard currency or something of it's equivalent? I see A LOT of black-market activity going down if such a cashless society ever manifests itself.

Yes, but now the 'black market' must define a currency for its self. You can't routinely PAY people in that currency, and allow them to have anywhere to live, or eat or shop legitimately.

How many "rare portraits" are there to go around, and how many low-level pimps and drug pushers will be able to convert that to actual money?

Sure, there's PRECIOUS METALS, but that's obviously very tracable, and the people who BUY drugs would have to routinely get their hands on it, and turn it back into money. You think crossing the border with CASH is a problem? Try it with gold bars and coins. "Hey kid, what's in that very heavy looking bag that goes 'jingle-jingle'?"

Not using cash is a big barrier to establishing your drug habit. Especially when access to whatever goods the kids need to use as cash becomes heavily monitored and/or restricted to adults.

Computer parts? Personal electronics? Surely someone could warehouse commodities like this. Of course, getting them back into the retail chain, and all those sellers who would have to explain how they keep getting legitimate money without spending any legitimate money. And once again, the users have to keep buying electronics at detectably consumptive rates. You can only buy RAM chips at Comp*USA so many times before someone says "Hey, aren't you in here buying RAM every single day?"

Retail channels that willingly launder money will simply stand out, just as banks do, now. Various "black holes" form in the accounting, and these can be traced back, because the cash has to enter the system again for it to be valuable to anyone. Having it automatically traced back would make the law enforcement end of it easy. Most retailers will only discover that they're in this position because of employees, and will turn them over without even blinking.

 

[evildave]

I think not. How many bills are out there? According to the US Treasury it is about 16 billion. Eventually they would get a hit, then what, recover a $20 bill?

Is that ANY kind of bill, or just $20 bills? It doesn't really matter....

16 billion sounds like a lot. Lets allow a generous 128 bits per bill serial number, and assume no compression. 256,000,000,000 bytes of data. I could store all of those serial numbers on the computer right in front of me, right now. Add the ATM/store number. Let's give 52 more bits for that. 180 bits. 360,000,000,000 bytes of data. I need to buy one more hard disk for my RAID 0 array. Or wait one year for 400GB drives to appear.

Using one T3 connection (45mbps each), bridged to my network port, all of those bills could be scanned and sent to my personal computer every day, and stored.

86400(seconds in one day) / (16,000,000,000(bills) * 180(bits/bill) / 45,000,000(bits/second)) = 1.35 times per day. Allows for some protocol overhead.

Assign banks a scheduled reporting time, and I personally could receive bank notes and unique accounts associated with each bill for every note in existance, EVERY DAY. If I bought 14 of these desktop machines, I could probably keep up with the storage demands, then processing them through the remaining two weeks. Building a grid with more machines, I could do the comparisons.

Now think what a big, expensive SMP computer with a lot of network and CPU bandwidth and very high speed storage could do, compared to what's on my desktop. For one thing, a Sun Fire E25K Server could fit all of those serial number/account pairs into RAM at the same time, and have 72 processors concurrently sharing that memory and working away at elements of the problem, including where the money is going at the same time. With multiple T3 connections, they could update the image multiple times a day.

Now think what 14 of these could do.

You think the "billion here, billion there, pretty soon we're talking real money" government can't afford 14 such machines? A hundred? And do you think they lack access to better, more interesting hardware?

Heck, even *I* could write the software to process the data into useful results. It's not rocket science. Tally, sort, report. The seemingly time-consuming part would be matching up transactions, but that's just another set of binary searches against data sorted by bills. Produce a sorted list of dollars that were taken by accounts. Produce a sorted list of dollars that were received from accounts. A sorted list of accounts that received dollars. Ship it our for mix&match. The various results comming in could also be pre-sorted, per bank. Keep track of that, too.

Another thing that makes this easier is that most results will be regionally based. You get money from an ATM, and most of it will be spent in the same town. The money that travels is much more "interesting".

 

[evildave]

And, BTW, all 16,000,000,000 bills are not transacted every day. That makes the problem even easier.

For instance, a binary search based on building a binary tree (dictionary, or std::set using STL) will need log₂N iterations for looking up that data (and some extra pointer overhead). So, on average to match a bill in a data set of 16 billion bills, it would take (up to) 36 steps to insert or find something.

Not billions of comparisons to insert or check a new serial number. 36.

And though there is 52~56 bits set aside for account IDs, there aren't that many accounts, either. Call it a generous 200,000,000 accounts, or 28 steps.

Now, to match 16,000,000,000 serial numbers to 200,000,000 accounts so that records can be matched. Assume the algorithm consumes 1000 CPU cycles per iteration. 16,000,000,000 * 1000 * 28 steps = 125,555 seconds, for a single 3GHz CPU to do. That is more seconds than in one day of processing. But we can divide elements of the work down among multiple CPUs.

One CPU is collecting numbers.

One is managing the workload.

Several are doing lookups against accounts, because we split the trees into several 'bins' for different threads (or processes) to operate on.

Technically, it's a fairly trivial process to crunch the serial numbers and account stuff to match them up.

Then comes the database work. That gets a little more interesting, because now the data about what account historically makes transactions to other accounts needs to be maintained, so all the patterns can be mined out of it.

The first thing that can be done is to simply leave the data "anonymous" at this stage. You don't need to know who belongs to a particular account here. Only they a particular account is involved.

What we need is a list of account numbers that received money, and add whatever account numbers have not been spent on, yet, and increment a count on the ones that have been visited before. The precise date/time of transactions isn't very important. The fact that transactions resulted a certain number of times with other accounts is.

This data is stored in "N" database servers with a front-end that receives a flow of inputs from the crunching computers. These inputs are seperated formulaically from the account numbers. It may be desirable to expand the account numbers so regional data can be derived from them.

If you want to know who "102372981" is, you look that up seperately. If you want to know who "102372981" gave money to, look it up, sort by "hit" count, and see the major, clear expenditures down to the "noisiest", one-off hits.

You can then "flag" certain numbers for routine mining. Naturally, if bits were set on accounts for "criminal history", accounts that ended up with unusual concentrations of money comming from these accounts could be flagged for closer scrutiny.

Once you have this data, interesting graphs and reports of who does what with cash can be mined from it. Who saves it. Who spends it. Who drinks it. Who withdraws it and never forms a clear pattern with it.

If someone hit the bank every day, took a bunch of cash, and that money tended to end up "everywhere", then it might be surmised that this is a cash payroll.

If you seldom get cash, but deposit "random" money a lot, it might be surmised that you're selling something (drugs?) to a lot of different people.