The Bitcoin Fluid Dispenser is true to its name. The device accepts Bitcoin and could potentially be used to spew any of a range of fluids, like gas, fuel, H2O, and alcohol. The possibilities are vast.
Andy Schroder, the machine's creator, released several videos describing how the dispenser works. As we've reported before, the intriguing device has evolved quickly over the past year and ease of use remains a crucial element of the machine. He recently unveiled an update that includes several new features, including Bluetooth, NFC, and the Bitcoin Payment Protocol:
His latest video demonstrates its smooth compatibility with various wallets, like Breadwallet and Mycelium:
Cointelegraph was lucky enough to catch up with Schroder, to learn about ideas behind the machine and the surprisingly far-reaching use cases.
Cointelegraph: Why a fluid dispenser?
Andy Schroder: As a mechanical engineer I wanted to figure out how I could best apply my skills to help Bitcoin become mainstream. Most people involved with Bitcoin are computer programmers or cryptographers—or some sort of software-related expertise. There have been a few people who've made mechanical devices that interact with Bitcoin.
There's some Bitcoin ATMs out there, a couple of people have made prototype vending machines. ATMs are interesting, but they're not taking it to the next level as far as how do we make a Bitcoin-only economy. They're kind of just a bridge to that next step, but I think we need to look towards the next step in order for the world to see that vision.
Fluid dispensing is very complicated because of the high metering accuracy requirements that the industry has progressed towards. I have a lot of expertise in fluid mechanics and thermodynamics—very much so that the fluid aspects of the fluid dispenser are fairly basic from my perspective. I knew that I could do a very good job putting something together that would accurately and reliably measure fluids—particularly fuels, which are a dangerous fluid.
CT: Can you explain briefly how it works? How has the dispenser evolved in the past year?
AS: The first presentation was really focused on being a proof of concept, so the control system wasn't as well-integrated, the software wasn't as seamless of a user-experience. Since then, I've evolved to integrating to a more traditional fuel dispensing cabinetry with more industrial hardware. With this latest revision (2.3), I've added new software features: Near-Field Communication, Bluetooth and the Bitcoin payment protocol.
With version 2 with the more industrial cabinet, I also added a receipt printer because fuel stations do require printed receipts based on the regulations here in the states. I don't know about the rest of the world in much detail. Version 2 was also more focused in as a use case. It was focused on fixed applications with a dedicated screen whereas the first version was mobile, it was cellular, and it had a preview of a web-based interface, which is cool because you don't really need a screen to make a purchase.
Unfortunately, fuel station dispenser regulations require screens right now. Even though I have this pretty cool phone-only interface, it's not something I could drop right in the service. When the regulations were created, no one thought you could have your own personal screen you could carry around with you. So I'm focusing on how this could best be integrated into the marketplace.
Version 2.3, with the NFC, Bluetooth, and Bitcoin Payment Protocol, was a big update. NFC's a lot easier than scanning a QR code, which is what most other payment workflows do. You just touch the phone on a certain area of the dispenser and it starts the process. It's less steps for the customer and it also works with intense sunlight. I don't know if you've tried to scan a QR code on an outdoor screen at noon in July. It's not too easy.
The Bluetooth allows you to make transactions if your cell phone doesn't have cellular Internet or WiFi—if your wallet supports Bluetooth. So that's a huge benefit because not everybody has a phone or tablet with cellular Internet on it. Lots of people talk about bringing Bitcoin to the unbanked of the world. If they're not set up with a bank account there's a high chance they're not going to have a cellular Internet plan.
The Bitcoin Payment Protocol basically allows for explicit refund address rather than heuristics to detect a refund address based on where the funds were originally sent from. It also facilitates that Bluetooth offline paying process. It allows the use of multiple addresses so you could have improved privacy.
It allows for digitally signed payment requests, which are really important for things in public because machines that are exposed in public can be hacked. That's a concern. Digital signing allows some authenticity to the payment request to be presented and verified by the customer.
CT: What piqued your interest in Bitcoin in the first place?
AS: Well, I'm very frustrated with inflation. I think it's a fairly reckless thing. Particularly that it's causing interest rates to be artificially low, making it extremely difficult for people to work hard, save their money, and make a large purchase—whether it's a house or a car or starting a business or whatever that may be the current financial system with an uncontrolled money supply—it's pretty much pressuring people to go into debt in order to get where they want to go. I think that's a very bad concept.
There's that, and then there's financial privacy. Bitcoin does have the potential for improved financial privacy. Right now, it's a little bit difficult to harness all these privacy options just because some of the tools for achieving maximum privacy aren't easy to use for the average person. But I think there's definitely a lot of progress being made in the privacy area of Bitcoin. Those are my big concerns: inflation, artificially low interest rates, and financial privacy.
CT: Do you have an interest in economics?
AS: Economics affects everyone's lifestyle decisions. To me, I try to lead a life that's very rational in my decision-making process. It's always been an interest to me because the business world is interesting—every engineering decision at its core is built on economics.
Maybe you want to build the most fuel efficient engine in the world, but if it's not economical than it's truly not the best option. I'm no expert in economics but I try to stay fairly well-versed from a layperson's perspective.
CT: What were some of the challenges you faced when designing and building the fluid dispenser?
AS: Well, the latest version as far as Bluetooth goes—that is only supported by Andreas Schildbach's wallet [Bitcoin Wallet]. Just since I made that announcement another wallet has announced a different type of Bluetooth functionality. Bluetooth is very new and something that I did with very detailed advice from Andreas Schildbach.
I drafted a post and set of specifications describing how the communication works. Most people in the technical areas know them as Bitcoin Improvement Proposals, so I have two proposed BIPs that I've drafted up along with Andreas Schildbach's contributions. Understanding how this wallet's bluetooth functionality worked when it wasn't formally documented (outside of his source code) and creating a formal documentation that, hopefully, other people can adopt in other devices to make this a standard was a fairly involved process for someone who is new to such things.
That's just how Bitcoin goes, I guess. You create a new technology, document it, and encourage others to use your standard. I hope all those people creating bluetooth functionality can come together and agree on some common standards, even if it doesn't end up being exactly what my original proposal entails.
As far as version 2, there was a lot of research that went into place as far as selecting the right hardware as far as the cabinet goes and what goes inside the cabinet. Making sure it's up to the standards of the existing fuel industry. The controls were completely developed from scratch. I have a fairly custom control system with inputs and outputs that are much more primitive than version 1, but that was basically a necessity to bring it to a platform that was much more in tune with the way Bitcoin works, on a computing level as well as making something that could be mass produced more cost effectively.
CT: How much time have you invested in the project?
AS: I've been working on this for over a year and a half now. I'm also doing some smaller consulting on the side as well as working on a doctorate in aerospace engineering—focusing on thermodynamics—so it's not 100% of my time, but I'm definitely a hard worker, so it's probably 100% for somebody else.
CT: I'm super excited to buy fuel at a dispenser that accepts Bitcoin. Are you looking into future partnerships? Do you have plans to roll it out?
AS: I'm definitely trying to commercialize this. I'm looking for someone in my area geographically who would be interested in a pilot installation. I definitely need to keep it close so I can monitor this and it needs to be robust.
I think it's in a fairly good state right now, but I don't want to put it half way around the world and not be able to check on it every week and make sure everything's working okay, talk with the operator or retailer and see how they're doing – stuff like that.
CT: Where do you live?
AS: I'm in the greater Cincinnati area.
CT: Let's step away from the fuel dispenser for a minute. What are you looking forward to in the broader Bitcoin space?
AS: I've definitely been excited to see many of the mobile wallet creators working on the deterministic wallets. Schildbach's wallet just launched with deterministic addresses recently. Breadwallet has deterministic addresses. Those are both peer to peer wallets—which I think is very important. Then there's Mycelium. They're pretty close to having a production-ready deterministic wallet. Theirs is not peer to peer. It uses some dedicated servers.
Deterministic wallets are a big thing as far as privacy goes. I tell everyone I know “You should use Bitcoin,” but without having a wallet that's trying to be private it's hard for me to feel comfortable setting them up. Now there's options for both Android and Apple which is kind of cool.
Another cool wallet for desktops is Armory. I think that's really powerful for security as well as using deterministic wallets. It allows you to set up multiple accounts, whether personal or business or savings. It's really useful for that. I'd like to see Armory be less resource intensive so the average person could use it because it takes a powerful computer to really use.
The Bitcoin Payment Protocol is really important because it encourages and embraces the use of multiple addresses. I'd like to see more people use that as far as wallets and vendors, whether online or in-person retailers.
There's definitely other privacy things that I see on the horizon. There's stealth address and CoinJoin. Those two have a lot of potential, but they have a little bit of a scalability issue as far as – I think stealth addresses may add a lot of data to the blockchain.
CoinJoin seems to be great for proving privacy, but right now we don't have a lot of options for using it natively in a wallet. For CoinJoin, you need some sort of a server that acts as a meeting place. I'm just a little concerned it's not distributed enough or integrated into the Bitcoin network itself for more tightly.
CT: What is in the future for the fluid dispenser?
AS: There are some things on the backend of the fluid dispenser that I can improve to scale this up to mass use at stations. Most stations have six or 12 units—or whatever number—and they're integrated into their payment terminal in the station. That's a big thing to make a marketable technology. It needs to be beneficial for people who don't care about Bitcoin in ways that they can see directly and indirectly by just making it more scalable to the fuel station itself and more products.
Normally you have 87 octane, 89, 90 or 91 or whatever it is—and diesel fuel. That's some improvement there. There's also the potential for Bluetooth low energy in addition to classic Bluetooth. That's something that I'm looking into.
There are so many things I could do. There's other dispensing applications. A lot of people are interested in dispensing at bars. There's also slower dispensing like water in an apartment complex or some other rental property. Many times landlords just lump water costs into the rent because they don't feel like metering each individual tenant and billing them.
The same goes for electricity for rental property. This is a great technology to morph into more of a longer term type of a sale—maybe over a month period. It would allow people to have better incentives to conserve resources and also have a more fair pricing strategy because then individuals will pay directly instead of things being averaged out across all the units.
CT: Thank you for your time, Andy.
AS: I appreciate it. The more people that can learn about this, the better. I think it's a pretty practical application and hopefully it resonates with some people as far as what Bitcoin can really do and how it can be simpler.
Did you enjoy this article? You may also be interested in reading these ones: