Bitcoin is a Language Best Spoken by Machines
Bitcoin is a beautiful technology with countless facets. It has sprouted so many separate spheres of exploration; it heralds a true blossoming in thought, in which cryptography and blockchains may touch fields previously thought untouchable. The scope of things still unbuilt is vast. There is an inchoate sense that something powerful lurks just under the surface, that the blockchain can mean more than what it means now. Such a novel thing profoundly challenges our presumptions. With Bitcoin unsettling our human intuitions so much already, what unimagined consequences lie in store? It is hard even to wrap one’s mind around without succumbing to what some would call techno-utopian delusions. Yet I believe that we live in auspicious times.
Much work has been put into building the infrastructure to use Bitcoin for payments. It has gone without saying that payments occur between people. It has certainly been satisfying to use Bitcoin personally. Without a doubt, there will be powerful use cases for using Bitcoin between individuals, such as remittances, tipping, an anti-fiat hedge, and many others. There will be many reasons for individuals to hold and use Bitcoin.
And yet people are not Bitcoin’s proper master. Bitcoin is programmatic money. What then could it be better suited to than machines themselves? Bitcoin makes the most sense as money, yes for people, but most especially for devices. Not only should I be able to interact with computer programs with Bitcoin, as I can already to a limited extent, but devices themselves should be able to transact in Bitcoin with one another.
Bitcoin is the perfect medium for devices to exchange value, or even establish consensus on other topics. This is a largely unexplored facet in the Bitcoin world. Certainly the easiest work, though by no means easy in an absolute sense, has been in adapting human-friendly systems to Bitcoin. This is right and reasonable. But Bitcoin’s biggest opportunities lie outside human interaction. A person should at best be an extremely distant end user, on whose behalf transactions are made behind the scenes.
I want machines that post bid/ask orders with one another algorithmically. I want to be able to program my Raspberry Pi to negotiate really intensely with all the other bots. I want to earn small amounts of money for small tasks, such as setting up a webcam above Times Square. Just look at the way Bitcoin miners have been incentivized by small payments, to almost absurd lengths. I want to pay for cheap data on a fiercely competitive, decentralized marketplace. And I want others to profit in wildly open marketplaces, providing small services for small money. The pennies changing hands are too small for people to keep track of. But machines have no such scruples.
Some might ask, “But what do machines have that they need to trade with one another?” This is a worthy question, but it is only difficult because we have never before been in a position to ask it. While I and others may have some ideas for the moment, they must surely constitute a small fraction of extant possibilities. When new tools arise that are generalizable, programmable, and offer a financial incentive, what limits are there in the hands of downstream innovators?
One thing that machines certainly have to trade with one another is data itself. Data has a price, but too often that price has been locked into illiquid, static pools. To my knowledge, there is no free market in data akin to what exists in well-run, liquid marketplaces. Just look at the liquidity of any financial instrument, or even of Bitcoin itself, and you shall see that it has been scrupulously priced, having sampled all the markets’ whims unhindered. Most assets are not traded this way, not because one would rather not enjoy the efficiency gains, but because there was no way to transact trustlessly, cheaply, quickly, and in small amounts.
There ought to be a floating market for Wifi access, or advertising billboards, or webcam access, and many other things besides. Rather than solicit a local monopoly, or interact with a small cadre of suppliers, who resemble medieval guilds, why not empower small actors to compete on an equal footing? Established companies all the way down to individuals, ought to be able to compete transparently. We have seen this in Bitcoin mining, where the playing field is so level. Because anyone at all can mine, it has produced staggeringly efficient hashing power. In other fields, what has been missing thus far is not the will, nor the advantages of open markets, but a mechanism for monetization at small scale.
My goals are twofold. Firstly, I want to help build the tools machines need to handle value and ownership on the Blockchain. This means more than just payments. Devices should be able to store a headers-only wallet which allows them to recognize payments, as well as keep track of Colored Coin tokens denoting their own ownership. Using a meta-protocol on top of Bitcoin, simplified in such a way as to address certain scalability issues, it will be possible to equip devices with cryptographic recognition of who owns them at any one time. It will also be possible to transact in that ownership, as well as issue cryptographically unforgeable, blockchain-evident instructions, such as rental agreements. Devices will own money through Bitcoin. But using the Blockchain, they will also establish consensus, independent of any outside authority, on who is authorized to issue commands and possess access rights.
I hope that, by making machines blockchain-aware, it will be possible to monetize almost anything. Each device could have its own terms and conditions. Different API instructions could be available for different kinds of devices, but all access rights would be governed and transacted via the Blockchain. API commands themselves would need not be on the Blockchain; they would be cryptographically signed messages from owners off-chain. Not only is the money itself programmable, but in the hands of a machine whose ownership is a digital token, the machine itself is a tradeable asset. Secondary markets for ownership tokens could be built, making device access rights themselves liquid, transferrable, and scarce.
If devices may control money, and if devices are controlled by Colored Coin tokens, then devices could even own other devices! The car I own cryptographically might own its own parking space cryptographically, except perhaps it earns a few Satoshi on the side renting it out throughout the day. The parking space hardly knows whether it is owned by Man or Machine; it only knows that it receives properly signed transactions from the Blockchain-evident owner.
With blockchain-aware machines, new forms of economic activity that had previously been unviable, might suddenly become real. My parked car could buy internet from pedestrians as they stroll past. It’s not worth having its own data plan after all. The car itself could be blockchain-aware, responding only to instructions from its cryptographically-obvious owner. When I open the doors, I write an encrypted message from the correct Bitcoin key saying “open the door”. Anyone else writing such a message, from any other address, would be unheeded.
Besides the new economic potential, giving Bitcoin to the machines is only sensible; it was meant for them all along. Machines that compute at the speed of light need a currency to match. Even more sophisticated layers on top of Bitcoin, particularly payment channels (which I am very bullish on), can make transactions highly granular, trustless, and obviate the confirmation time necessity in Bitcoin 1.0. Thus Bitcoin can be made to transact at a computer’s pace.
If bitcoins circulate between devices, it will only be beneficial to the entire ecosystem. The scope of use cases for Bitcoin would grow; a strong baseline of demand stemming from machine-machine interactions would set a lower limit on the price. If demand becomes compartmentalized away from people’s fickle moods, volatility can only diminish.
Bitcoin, as a protocol, has always been a language at heart. It is a language that we humans speak only haltingly and with great practice. Let the machines transact in this new language that they speak so fluently. Trading between each other data, access rights, services, high-frequency negotiation could become commonplace for even the most boring tasks, such as parking a car, or finding Wifi. Perhaps through bids and asks exchanged every millisecond, between devices algorithmically maximizing outcomes, Bitcoin can make the most difference.
Follow me on Twitter at @abarisser
1GgwA7c2ovgWDBoVYsHT5VYXw2QBey1EdF
