Designing a Bitcoin system to use in prison
Charlie Shrem is going to prison; he designed a cryptocurrency to use on the inside (if it were not illegal)
Illustration: Lia Kantrowitz
In January 2014 I was coming home from a speech in Amsterdam when I was arrested at Kennedy Airport by federal agents and transported to solitary confinement in the Metropolitan Detention Center in lower Manhattan.
The charges were related to my role as CEO of BitInstant, an early Bitcoin startup that allowed users to buy the digital currency quickly and easily.
I knew of—and to some extent enabled—a single customer buying Bitcoin in order to procure illegal drugs on the digital black market Silk Road.
It had finally caught up with me.
A few months ago, I pleaded guilty to "aiding and abetting an unlicensed money transmitting business" and was sentenced to two years. Today I'll be self-surrendering to Lewisburg Federal Prison Camp in Pennsylvania.
I'm obsessed with Bitcoin, and I'm going to prison. So of course I wanted to know if I could continue to use it on the inside
Bitcoin may have indirectly landed me in prison, but it's still my passion. An alternative global transaction network would give people more control over their finances, allowing them to send money with greater privacy and without having to trust (or pay) a middleman. It would also reduce the time it takes to move money by days. If we could do that, we would be another step closer to bringing millions of people out of poverty by allowing people in local economies to buy and sell to the rest of the world.
I'm obsessed with Bitcoin, and I'm going to prison. So of course I wanted to know if I could continue to use Bitcoin on the inside.
Many different types of prison currency, from stamps to cigarettes, have existed over the years. After prisons began banning smoking a decade ago—it was banned in all federal prisons in 2014—prisoners started exchanging tins of mackerel due to their long shelf live and high protein, which is lacking in prison. Laundry may cost 1 mack, a haircut 2 mack, and a personal trainer can run you 25 mack a month. So why not Bitcoin?
As it turns out, running a Bitcoin system in prison is illegal—the Bureau of Prisons strictly forbids "possession of money or currency, unless specifically authorized"—but it's still a fun thought experiment. Bitcoin is primarily used over the internet, but does it need to be?
By understanding how feasible this is, I can see what, if any, infrastructure is needed to connect local communities to the global financial system, even if they don't have computers or an internet connection.
GUESS WHAT? YOU DON'T NEED THE INTERNET TO USE BITCOIN
You've probably heard people talking about how Bitcoin is revolutionary. While Bitcoin may be the first all-digital currency that is feasible on a global scale, the fact that it is digital isn't why it's revolutionary.
The current financial system relies on banks to move money around and make sure it gets where it's going. Bitcoin's real innovation was engineering a system that does not require banks in order to keep money flowing. Instead, tens of thousands of independent volunteers share the work of tracking transactions and preventing fraud.
These volunteers maintain what are called the nodes of the Bitcoin network. Nodes are computers that keep track of all the Bitcoin transactions and update users' account balances accordingly. Instead of tracking this information internally, as a bank might, Bitcoin's nodes broadcast these transactions as they happen. All the nodes talk to each other in real time and update the blockchain, which is a shared history or ledger of all Bitcoin transactions ever.
This transparency keeps everyone honest and the ledger decentralized.
This information can be transmitted over the internet, local LAN networks, TV and radio signals, even Morse code. A farmer in Kenya can connect to the network with only a cell phone and local SMS capability.
But what if technology was even more limited, or virtually non existent? Without computers or cellphones, could Bitcoin still work?
ENTER THE MAK
The answer is yes. All you really need for a Bitcoin-like currency to work is the ability to keep track of transactions in a central ledger, which is double-checked frequently by users.
Without a computer or cell phone, prisoners would need to run a ledger completely offline from behind the prison walls. In other words, we'd need a physical notebook to keep track of all the transactions. Instead of swapping physical cans of mackerel, we'd simply write down the amounts in the notebook.
Let's call our prison-version of Bitcoin Mackerelcoin, or MAK.
Say I'm working out in the gym, and Dave needs to pay John 25 MAK for his monthly workout training. Instead of having to fill his pockets with 25 tins of fish, he'd simply ask me to move the funds from his account to John's. I'd open the ledger and write down "Dave paid John 25 MAK," and the date. Then I'd subtract 25 MAK from Dave's balance and add 25 MAK to John's.
(This isn't quite the same as Bitcoin, because as the single node, I'm acting similar to a banker. If we really wanted to imitate Bitcoin perfectly, many inmates throughout the prison would be keeping track of these transactions and then synchronizing them together, just like nodes in the Bitcoin network. When Dave paid John 25 MAK for a workout, every inmate in the yard would take out their notebook and jot down the transaction. Later, when John pays Tim 2 MAK for a haircut, all the other inmates in the salon would take out their notebooks and each record the transaction. Then at the end of the day, all the inmates would meet together and enter all the transactions into the main ledger. It's sort of like the Island of Yap, where the residents used enormous stone coins that were too heavy to move—when money changed hands, everyone in the village simply spread the word of its new owner.)
Taking the experiment further, think of a prison bookie. Bookies need to be able to make dozens of transactions a day, from betting on a game of Monopoly to guessing whether dinner rolls or cornbread will be served with dinner.
Bookies maintain ledgers themselves but have to struggle with collecting and dispersing funds. If the inmate making the bet had a MAK balance, the funds for the bet could be frozen until the outcome. If he wins, I move funds from the bookie's account to his, and if he loses, vice versa. This way, he does not have to trust that the bookie will honor the agreement. Meanwhile, the bookie doesn't have to break any noses to collect.
Somehow, at various points of the day I broadcast my ledger including balances and transactions to the outside world—say, by standing up in the cafeteria and shouting out the day's transactions.
I would pin the value of 1 MAK to the price of a tin of Mackerel in the commissary, $1.50 USD. Meanwhile, I'd charge a fee—say 1 MAK—to cash in and out of the system with physical Mackerel, just like Bitcoin exchanges do with the US dollar and other currencies. Ideally, the MAK, which could be exchanged for Bitcoin once the balance holder leaves prison.
The ledger is the most important component to the theory. I'd need to keep real-time books of everyone's balance and transactions, all without technology. In order for me to maintain my own honesty and the integrity of the system, I would mail a copy of the books every day to at least two other parties outside the prison who act as external nodes. These parties will publish copies of the past ledgers, and also confirm that each transaction is valid—that John actually had 2 MAK to send to Tim.
I would mail a copy of the books every day to at least two other parties outside the prison
This gives the system a block time—the amount of time it takes a transaction to be confirmed and published—of about three days, the amount of time you'd expect it to take for the books to reach the external nodes. (For larger transactions, you might want to wait until the external nodes confirmed payment before exchanging goods or services, but for smaller transactions, you're probably fine taking the risk.)
The external nodes double-check my math and confirm the transactions. The mail postage acts as a timestamp. My external nodes on the outside publishes the daily ledger to a website or blog, where any inmate can call or email a friend or family member to verify transactions. This helps prevent me from arbitrarily debiting or crediting from someone's balance.
But the real check on integrity is that everyone, including me, has an economic incentive to keep the system honest. If anyone thinks the system is corrupt, the whole experiment collapses. This is what would (hopefully) prevent someone from cornering me in the bathroom and forcing me to change a balance.
Implications beyond the prison walls
In my personal trainer example, carrying around 25 tins of Mackerel to pay a personal trainer is somewhat cumbersome, and inmates are only given a small locker for all personal items. With MAK, an inmate can simply ask me to move a balance of 25 MAK from his account to his trainers.
Of course, this is an experiment and can easily be shut down. Since I am the only node or access point, I can be shut down and sent to a higher security prison. This problem could be solved by recruiting more inmates to be nodes—and prisons, by design, make it difficult for inmates to organize.
But what about the real world outside prison walls? There are tens of millions of underbanked people in the world. They are slowly being brought online, but in places where the internet is not available, the offline blockchain could be a stepping stone to onboarding them into the global financial system. And even if not, it's still an interesting way to explain how Bitcoin actually works.