Ben Sharafian & Stefan Thomas

Coil – A New Business Model for the Web

The difficulties of compensating content creators for their work has been a driving force of the web for decades. It gave rise to advertising-driven tech giants like Google and Facebook and contributed to the decline of industries like music and publishing. Coil aims to change this and leverage blockchain and the Interledger Protocol to build a new business model for the web.

We were joined by Coil Founder and former Ripple CTO Stefan Thomas and Coil Co-Founder Ben Sharifian to discuss their ambitions to change how content is consumed and paid for.

Topics we discussed in this episode
  • What web monetization is
  • The difference between web monetization and web payments
  • The negative effects of advertising-driven business models
  • What the Interledger Protocol (ILP) is
  • The current state of ILP
  • Why Stefan and Ben decided to leave Ripple and start Coil
  • The case for web monetization as an initial use case for Coil
  • The idea of users paying a fixed bandwidth per second to content creators
  • Possible user experiences that could be built on top of Coil
  • How to bootstrap Coil and get adoption
Sponsored by
  • DutchX: The open, decentralized trading protocol for ERC20 tokens using the Dutch auction mechanism. More at epicenter.tv/dutchx.
  • Microsoft Azure: Deploy enterprise-ready consortium blockchain networks that scale in just a few clicks. More at aka.ms/epicenter.
Transcript

Brian Crain: Today, we’re going to speak with Stefan Thomas and Ben Sharafian about Coil, an exciting new project built on the Interledger Protocol.

Sunny Aggarwal: Yeah, Interledger is something that Brian and I have been pretty familiar with for a while, actually. Stefan has been on Epicenter in the past, twice actually. One time, representing Ripple, but the second time actually presenting Interledger Protocol, and this is back when Stefan first created ILP along with Evan Schwartz at Ripple.

And since then, Stefan, as well as a number of his coworkers at Ripple have left Ripple and have created a new company called Coil in order to decentralize the Interledger Protocol-like ecosystem and work on some cool stuff, which we talk about in today’s episode.

Brian: So, Sunny, you mentioned you were also kind of involved in a company doing something with Interledger.

Sunny: Yeah, so as many of you obviously know, I’m very involved with Cosmos, which is like an interoperability project. So, we have our own interoperability protocol of types, which we call Inter-Blockchain Communication, or IBC. And IBC is really used for moving assets between Blockchain, so moving a Bitcoin from the Bitcoin Blockchain onto another Blockchain. Well, ILP is sort of more similar to like atomic swaps and stuff, moving value across Blockchains. So, I think there’s a really cool interplay of work between ILP and IBC that goes on, so I actually advise a company called Cava, who’s doing a lot of this ILP development within the Cosmos and other ecosystems like Ethereum.

Brian: Cool. Then, let’s get to the interview. Hi, so we’re here with Stefan Thomas and Ben Sharafian. They’re two key people at Coil. So, Stefan has been on the podcast a bunch of times before, actually his third time. I think the first time, we spoke about Web Pool, the last time maybe about Interledger. So, this is kind of a continuation of that work, and Ben is here for the first time. So, thanks so much for joining us, guys.

Stefan Thomas: Thank you for having us.

Ben Sharafian: Yeah, thank you.

Brian: So, probably many people don’t remember the previous podcast or they didn’t listen to Epicenter back then, and Stefan, you have an interesting and long history in the crypto-space. Do you just mind giving us a very brief background about how you became involved and what your journey has been through this world?

Stefan: Yeah, for sure. So, I’ve been a serial entrepreneur for about as long as I can remember. I started my first company when I was 14 years old. So, if you’re a small entrepreneur, you know the pain that there isn’t payments. So, what attracted me to Bitcoin, when I found out about it, was that it was sort of the first time that you could just, as a technical person, help write some code, and suddenly, it would change how payments work.

Before that, it was just always the financial mystery, this arcane beast and you can’t really touch it. And I think, then, after joining Bitcoin, it was like 2010, I started to get pretty involved in making animated video about it, starting to give a lot of talks about it, and then I also wrote a JavaScript implementation of Bitcoin called Bitcoin JS. And all of that was kind of to try to get Bitcoin more into the mainstream and try to get mainstream adoption.

Then, in late 2012, I got approached by Jed McCaleb of Mt. Gox fame. And he had just started this new company called Opencoin, better known now as Ripple. And so I joined them and, after Jed left, quickly became CTO, and I’ve been there for six years until May of this year.

For most of the last four years, my focus has been Interledger, and Interledger is essentially like an interoperability protocol for payments. It’s not necessarily exclusive to Blockchain, like you can interoperate any kind of payment system. You could even use something like handing cash to someone and do Interledger over that. So, it’s pretty flexible, and we got to a point where we’re like, “Interledger is a really big idea. It’s a really important idea, so we should really bring it out to the world somehow.”

And what happened early this year, it was basically, we get to a point where we would go to conferences, we’d tell people about Interledger, they’re super-excited, they’d be like, “What can I do with it? Can I connect to it right now?” and we’d be like, “Well, there aren’t really any connectors for it yet, there aren’t really any companies that you can interact with on it.”

And so the impetus for starting Coil was really to be the first company that’s actually doing stuff with Interledger.

Sunny: And what about you, Ben? I know that I think me and you have a little bit similar stories. I believe we both dropped out to join a Blockchain startup. So, can you tell us a little bit more about your story, how you got involved with Ripple and how you jumped over to Coil, followed Stefan?

Ben: Yeah, so I used to be an intern at Ripple, and that was around the time when Interledger was first being worked on. And I don’t know. I found it really interesting and I kind of — I felt like it was one of these things that was going to become a big deal. So, there’s obviously a lot of time pressure on those sorts of things, and so I thought it was best to just leave the school thing and work on this Interledger stuff. And now, yeah, a couple years later, Interledger’s pretty much around, and I think it falls on us to lead the way as a company building on Interledger.

Stefan: And Sunny, you said, “Follow me.” I don’t know who’s following who. I would say this was his idea, and that’s definitely what I’m going to say if it doesn’t go well.

Brian: Okay, so we mentioned Interledger, right? We did a podcast about this before. Of course, we’ll link to it in the show notes if people want to learn about it more in depth. But, can you give a very high-level description about what is the Interledger Protocol?

Stefan: For sure. So, as the name implies, it’s a protocol to connect different ledgers together. So, imagine you will have one person who likes Bitcoin, and so they have a Bitcoin account, they have all their money in there. Someone else, they are not into crypto at all. They have an Alipay account, and that’s how they want to receive money. Interledger is intended to solve the problem of how does the money get from Bitcoin into Alipay.

And so, in general, it’s largely inspired by internet protocols. So, we can now observe that the internet was this really efficient, constantly getting faster and cheaper system, and it seems to somehow bridge the gap between totally different kinds of networking technology, faster networking technology like fiber, slower networking technology like mobile phones or satellite internet, and it can just make that a totally seamless experience, and we start thinking, “Why doesn’t payment work like that? Why is it so different in terms of trying to cross different payment networks?”

And as we started working on it, there are definitely differences and there are some things we have to do differently, technically. But, at the end of the day, a lot of the same principles that apply to data also apply to money. So, we ended up doing things like sending tiny packets of money around. So, if you’re sending a larger payment, maybe that’s split up into lots of small packets. And we do things like routing to kind of find between any source account, any destination account, some path that the money can take where there is actually liquidity available, all these kinds of things.

So, Interledger’s basically just an open protocol that defines what some of the data formats are for doing that.

Sunny: So, how does this compare to something like atomic swaps or cross-chain Lightning Networks?

Ben: So, atom swaps, I think, are if you have one sort of large payment, I think you want to get it through between two systems. That’s kind of what I think of as atomic swaps, and with Interledger, it’s a lot more about having a lightweight network that you can abstract away the actual details of the payment system. So, if I’m sending an atomic swap between Bitcoin and Ethereum, then the time it’s going to take is at least a transaction on one of those.

But, with Interledger, if you want to pay several micropayments while you load a webpage, then you can’t wait very long. So, you need something that sits on top. And I think Lightning is something that does do that as well by using payment channels to make it instant. But, it does also rely on some pretty Bitcoin-specific features. And while you could implement payment channels and Lightning-style payment channels in other Blockchains, it’s not quite as general as saying, “Interledger works over any system where value can be sent.”

Stefan: Yeah, I think my favorite example is this what I mentioned earlier, which is just if Ben and I want to settle in cash, we can still set up two Interledger connectors, we can send Interledger packets all day, and then at the end of the week, Ben hands me a bundle of cash, and that’s a totally legitimate way to settle Interledger connection

Sunny: So, Evan Schwartz, who was one of the other co-founders or authors of Interledger. He published a blog post a few weeks ago comparing Interledger to a Layer 3 solution on top of like Layer 2 solutions. Do you guys see this in a similar way?

Stefan: Yeah, I think that’s a really good way of thinking about it because Interledger’s very much not focused on solving a lot of the problems involved with actually moving the money, or what the underlying units mean, or what the economics are, or how things like mining or consensus work.

And so, even some of the things that Lightning addresses where it’s about scalability and how do we make more transactions through the system and so on is not really what Interledger’s designed to solve. What Interledger’s designed to solve is interoperability. It’s like, “How do I get from one tech stack to a totally different tech stack that might have been built with something totally different in mind?”

We actually think that Lightning is a really good protocol to do Interledger on top of because some of the limitations of Lightning, like the limits on the transaction size don’t really come into play with Interledger because we split payments into smaller packets anyway. So, let’s say you’re sending a $20 transaction over Interledger, even if that goes through Lightning and maybe there’s like a 5% failure rate for $20 payments on Lightning for that path, that wouldn’t matter because the payment would be split up anyway. So, I think it kind of fits in nicely with some of the other solutions out there.

Sunny: So, I’ve been following with the ILP road map for a little while now, actually, and I know it’s actually changed a lot especially since the last time you were on the episode. Can you give a brief rundown of how the ILP protocol maybe adapted over time and maybe how the ILP ecosystem has adapted? Are there like, “Two years ago, really the only player in the ILP ecosystem was Ripple,” but has that really expanded as well?

Stefan: I think, on the ecosystem side, the biggest thing is we got to a point early in the year where we were going out to conferences, we were telling people about ILP, they thought it was great, but there wasn’t anything real, anything live to tie into. And when we would tell someone to, “Go start your first Interledger connector company,” or something like that, they’d be faced with this problem of like, “Well, who’s going to be my customer? There’s nobody using this yet.”

So, what’s happened since May, which is only a few months, but what’s happened since then is that we now have a company called Strata that is like a professional Interledger service provider and Coil uses them for all of our uplink traffic. And there are other companies that have built Ethereum plugins, and Lightning plugins, and there’s a lot of other tokens that are interested in using it for interoperability.

So, I think their community has grown a lot. There’s a lot more people now in the community group, Coil. So, I think it’s taken it from this academic project that people were theoretically interested in to something that, now, Coil and others are actually using in production, and that really changes a lot of the conversation and the field’s very different. We’re now moving like 10 million packets in a day, so we have to think about scalability, we have to think about how do you process that much, and keeping track of everything, and so on.

On the technical side, I think the number one biggest change since I was on the show was around two things. One is how we thought about the system. So, I think before, we kind of thought of it as you have ledgers as these payment networks, and Interledger connects them together. And if you’re drawing Interledger as a picture, what does the Interledger network look like, probably, you would have drawn the ledgers as the nodes and then they would be connected by what we call connectors that sit in between each network.

And since then, we’ve had this what the community has termed the “Interledger enlightenment”, which is basically flipping it on its head. So, rather than looking at the ledgers out there as the nodes and then Interledger connects them together, it’s actually the participants, the users, the connectors, they are the nodes, and then they settle their relationships over some ledger.

So, it really puts a lot less emphasis on the ledger, and without going too deep into the weeds, but it came from a realization that the internet community had, which was basically that the internet is, by tying together networks, it’s duplicating some functionality. Because, what does a switch do? It routes packets from one place to another. Well, what does a router do on Layer 3? It does the same thing. It also sends packets from one place to another.

So, you can actually get rid of switches entirely and just have links between different nodes, and then some nodes also route packets, right? So, we basically took that same lesson that the internet community has come to eventually and applied it to Interledger. We simplified the picture a lot. Again, it gets a little technical, but I guess some of the people listening might get a kick out of this realization.

The other thing was this concept of what we call “penny switching”. So, that’s basically the idea that if you have a larger payment, you can split it into smaller chunks. And the reason that was significant is because, before that, we had a lot of difficulty with the variability of payment size. And I’ll explain why.

So, one example, one issue that people who have been following Lightning might be familiar with is the issue of liquidity. So, you have some path and you want to send a payment through that path. Now, if that path doesn’t have enough liquidity to do that payment all at once, you’re kind of out of luck. You have to come up with some really complicated scheme, and we actually went down really deep rat holes trying to come up with a scheme like that to try to split up the payment and then recombine it. But, we just think that it makes the whole thing way more complicated.

So, the other problem was when you’re talking about exchange rates and you’re trying to find the cheapest path in terms of exchange rates, also the exchange rates might be dependent on the amount. So, you might have a different exchange rate for a very large payment, or a better or worse exchange rate for a very small payment, which then means that, for some paths, or for some different sizes, you might have different paths. So, small payments go this way, large payments go this way. So, your routing table gets one dimension more complicated.

So, what I’m trying to get across is like this variability in the payment size gets really freaking complicated. So, once we started to think about, “Well, what if we just have a pretty consistently small packet size? A lot of these problems go away. You now have one exchange rate. You can just assume it’s like a fixed number because all the packets are small, and you just have a rate for that.

And liquidity becomes much less of an issue because if you exhaust the liquidity on some path, the next packet will just take a different path, and also people that see a payment in progress, or you see the money streaming through, you slowly see the liquidity getting exhausted, you can actually go and provision additional liquidity in real time to make sure that that link doesn’t go down.

So, it just became this much simpler, much more powerful way of thinking about it. And once again, it’s taking a lot of inspiration from the internet where, similarly, you don’t want to send a big file download as a single packet, because what if that packet gets lost or ports and so on. So, those are the two main technical changes.

Sunny: I see. And so you mentioned that these packets are getting very small. You said that, currently, ILP’s handling about 10 million packets per day. Roughly, how much is that in like U.S. dollar volume?

Ben: It’s hard to say. Maybe not very many dollars. Probably on the order of 10 to 100.

Sunny: Wait, so each packet is like a millionth of a –?

Ben: Yeah, the average packet size is about 40 microdollars. And so, you’re actually sending an Interledger packet for every single second, sometimes more often than that that you’re on a site. And of course, the rate at which you send could be tuned for performance, but basically what you’re getting at is that, in real time, the site is getting paid.

Stefan: Sometimes, people ask, “What happens if you lose a packet or something like that?” and when we designed the protocol, we designed it in such a way that the standard receiver are never exposed to any risk, number one. Number two, that there is the right incentive structure. So, no one has an incentive to fail. But, then number three that failures are almost expected and tolerated, and I think that’s, again, something that you can learn from the internet is like don’t try to make failure a non-issue at every layer. Just handle it on end-to-end level and say, “Hey, if this packet doesn’t go through, we’ll resend it, and just make it so that the failures don’t matter commercially, like they’re not expensive enough to matter.”

Brian: This is very interesting. I’m curious because you made this comparison to Lightning as well. Is this something that could also make sense for Lightning Network to basically say that there’s this standard size of a tiny amount and basically things are always on kind of like packets of that amount?

Stefan: Well, so the way I would say it from a technical perspective, that makes a lot of sense because I think the problems that you’re seeing right now with Lightning, and I know that they have their own road map and their own solutions for it. But, I think they’re fairly elegantly solved by the Interledger Protocol stack, and we have a plugin for Lightning. So, if someone out there is trying to build an app on top of Lightning, they’re running into these issues, they should definitely take a look at Interledger and see that actually solves the issues that they’re having.

Whether the Lightning community, to the extent that there is an official decision-making body, will officially use Interledger for those use cases, I think that’s more of a political question. I don’t know if that’s going to happen. But, just from a technical perspective, I think it’s a really good solution for those problems.

Brian: Maybe one final question in this, and maybe this is a dumb question, I’m not sure. You mentioned, for these payment amounts between your 10 and 100 dollars, you’re sending 10 million packets. Is that economically-sensible, or does that actually make it an expensive network to run, I guess some costs associated with bandwidth as well.

Stefan: Yeah, I mean, at the end of the day, an ILP packet is just a packet of data, and there’s no real reason why, for a connector, it should be much more expensive than sending an internet packet, for instance. And definitely, it sounds crazy to say when I load this website, I’m sending 100 payments or something. But, to say I’m sending 100 internet packets doesn’t sound that weird.

Brian: Okay, cool. Well, let’s speak a bit about web monetization, because that’s, I think, sort of a threat that runs through a lot of work that you guys are doing today, and I think, Stefan, you have done over time. I mentioned before the episode that you started working on problems of web monetization, like much before Bitcoin. Can you run us through what was that early and what were the main lessons that you learned?

Stefan: Yeah, so back in 2008, I joined a company in Switzerland, which was basically selling e-books. So, e-books, you can think of it like these little 14 to 25-page guide books on various day-to-day tasks, and they cost around like 1 euro, 1.40 euro, something like that. And obviously, it was very difficult to motivate people to actually take out their credit card and a lot of people in a German-speaking area don’t even have credit cards, or like their PayPal accounts, and try to go through this whole process just to buy this 1.40 euro e-book, right?

And we eventually switched to a subscription model which worked a lot better. So, a lot of people were actually willing to say, “Instead of paying 1.50 euro for one e-book, I’ll pay 10 euros for a monthly flat rate where you get access to all 5,000 or whatever it was e-books that are in the catalogue.”

So, I had that sort of experience in the back of my mind for a long time like neither solution was super-elegant because even with a flat rate, if you’re a smaller provider, it’s kind of hard to get people to sign up to your particular platform, your particular flat rate. After that company, we actually built a tool, we built a PDF to an HTML5 converter, so you’re being able to look at PDF documents in your browser, and it was just part of making the e-books easier to read for people.

But, then we realized that that technology was actually independently useful, and so we started selling it and spinning that company off to basically sell it to publishers who had these back catalogues of old magazines, newspapers, etcetera in PDF format so they could convert that into HTML and then put contextual ads on that.

So, that gave me a bit more of an insight into how the publishing industry works, especially that time, it was a very tough industry. It was kind of an industry in decline, their revenues were shrinking, you would go into an office and there would be like white sheets over some of the desks, and all the remaining people were like, “Yeah, a lot of our colleagues just got let go,” and it was a very tough environment.

And ever since then, I’ve been thinking about, “Why isn’t there a native built-in way on the web where you just go to a website, and as you make the request, you also send some money, and then they send back the data? Why isn’t money the same as data in terms of efficiency to transfer it?”

And it turns out that when you look at the HDP standard, and other people have pointed this out, but if you look at the HDP standard, there’s actually an error code 402 that’s reserved for this particular use case. Like, if a website wants you to pay for a given page, there’s an error code reserved for that, and in the description, it just says, “Reserved for future use,” and they just never, after 30 years, got around to figuring out a protocol for that.

So, when we were looking at Interledger, we were like, “Hey, we want to take this from a theoretical idea to a practical protocol that’s really being used. We need another use case,” and this seemed like such a natural fit, because we kind of looked into the different criteria and then tried to figure out what was a good use case for that.

Ben: Yeah, by having it where you send to so many different websites in a day, that’s something you couldn’t do with a manual process for payment. So, if we’re talking about, “How can we apply this Interledger technology that’s basically making it as easy to send payments as it is to send internet packets?” then I think that sort of many-to-many scenarios are sort of the best in the short term while it’s still gaining adoption.

Stefan: Right, and then also if you want to solve this problem for the web, it needs to be an open protocol. So, an example I often give is like we could have applied Interledger to remittance. The problem with remittance is that what users care about is the cost to them. Like, if I’m sending money back to my family, I care how much gets there. I don’t care if the protocol underlying it is like open source. So, with the web, however, if you’re trying to get your protocol to be officially adopted by, let’s say, browser vendors and can be baked into the web standard stack, there really needs to be an open protocol.

So, we thought that Interledger just checks all of those boxes really well. It’s great at many senders to many receivers, it’s really scalable, it’s an open protocol, and it’s great at going across borders as well. So, it just fits that use case so well.

Sunny: So, does web monetization only apply to this sort of micropayment kind of thing where you’re paying for data, or is it also meant as a replacement for payment APIs themselves, so like Stripe, and PayPal, and all those guys?

Stefan: Yeah, so we’ve already done a lot of work over the last three years with the W3C where a colleague of mine, Adrian Hope-Bailie, who’s now also here at Coil, he’s been the chair, or the co-chair of the web payment’s working group. They’ve come up with two standards so far. One is the web payment request API. So, this would be like a website that’s asking for payment from the browser. This is what Apple Pay uses, for example, when you go to check out on Shopify or something like that, and it pops up with a nice little native UI.

So, that’s something we’ve already been involved with with for a few years, and we’ve done some interesting tests with that. So, for instance, if you look at the Chrome source code, I don’t know if it’s still in there, but they did a test where there was actually Interledger was actually one of the whitelisted payment method identifiers, and that was because we’d been working with them on experimenting with using Interledger for kind of retail payments. If you dig it up, there are demos, and blog posts, and stuff like that about that topic.

I think, for Coil, web monetization, we think, is a new thing. It’s a different standard, and the main differentiator is that, with web payments, you are actively acknowledging each payment. It’s not automated. It’s like, “You would like to check out, which address would you like to use for shipping?” that kind of thing. So, you go through like a checkout flow. Web monetization, the amounts are so small that it’s actually not worth it for the user to think about it at all. So, it should be completely in the background, completely automated.

Now, I think the way that web centers work, it helps a lot if the standards kind of build on each other. So, we’re starting to think about how to make these things more similar and maybe reuse some of the web payment standards for web monetization because that would make it a lot easier to get browser support. But, right now, from a use case perspective, there are distinct use cases.

Brian: I’m curious if you could speak a little bit about what are or what have been the higher order or kind of bigger-picture effects of this lack of web monetization? I mean, I guess that probably ties into advertising being one of the prevalent ways of monetizing the internet. How do you guys look at that?

Ben: It’s just I think the fact that the web doesn’t have any built-in sort of business model for a website is kind of what gets us into these situations because, yeah, it costs money to run a website, and you kind of have to turn to ads, and that, then, kind of goes back and modifies the nature of the site that you’ve enabled for ads to make it ad-friendly or a focus on collecting your users’ data, that kind of thing.

Also, if you want to switch to a subscription model or have your users buy something off their credit card, that also requires quite a big change in how your platform works or how your website, or blog, or anything on the web.

So, I think, with Interledger, you could have this as just, “Here’s my app, and so long as you’re using it, you’re paying me automatically.” So, it doesn’t add friction for the user, but it’s still a sustainable business model for the operator of that website.

Stefan: Yeah, just like if I made a useful app, I made a useful tool, I made a piece of content, I shouldn’t also have to come up with some business model like, “Hey, I’m going to sell your data that you put into this app,” or, “I’m going to inundate you with a bunch of ads,” or whatever.

A lot of those, to me, are workarounds. It’s kind of like the difference between going to a restaurant to have dinner and then you pay, or going to a restaurant to have dinner, and then you do the dishes for three hours in order to compensate them for the dinner you had. It’s just barter is not a very efficient way to pay for things, and so we think that introducing actual payments is actually healthy, and there’s no amount of ads that you can look at that’s going to be worth more than whatever your actual job is and the money you make with that. It’s just going to be more efficient than bartering.

Brian: Yeah, I think that’s a great point. I recently listened to a podcast and somebody was saying, this was in the Sam Harris podcast who’s done a lot of experimentation around basically trying to get people to pay for listening, and donate, and actually podcasting, I think, is a great example where the native ways of podcast distribution through our SaaS makes it very hard for people to, for example, pay directly for listening to a podcast. But, yeah the point he made there was basically that all of the bad things on the internet are due to advertising, and there’s probably a decent amount of truth to that.

Stefan: Yeah, I think that’s sort of something where like instead of having a one-on-one relationship where the website owner is trying to provide a service to me, I’m trying to reward them, and it’s a very simple, straightforward relationship, with advertising, you’re bringing in a third party.

So, you get things on YouTube when they demonetized a lot of content because advertisers weren’t sure about the types of content that their ads were being placed on. So, now you have content creators that have to modify, maybe they have to use less coarse language or whatever, which actually their audience doesn’t mind, but it’s like some advertiser that doesn’t want it associated with their brands. So, it’s just like an extra complication, and I think that by having that direct reward, you kind of align incentives a lot better.

Sunny: Yeah, Stefan, I think I remember you once telling me about how the advertisements that are shipped actually make up a sizable percentage of mobile data costs for users.

Stefan: Yeah, so there was this statistic where I just took it from the New York Times where they basically looked at what is actually the cost to the user, given a certain mobile bandwidth cost to ship all these ads onto your mobile phone. And in some of the cases, it  actually was more cost to the user to deliver the ads to their mobile, again, depending on their data plan, whatnot, than it was revenued to the website, and it just shows like, “Well, clearly there’s something pretty weird going on.”

Brian: Oh, that is insane. That’s super-interesting. That sounds like a deeply-broken thing. Well, let’s move a bit to Coil and get a bit more in depth here. I mean, first of all, you were both at Ripple. Stefan, you were at Ripple for a very long time. What was the reason why you decided to start off in a different company as opposed to maybe continuing to work on Interledger, Interledger use cases from within Ripple?

Stefan: I’m sure your audience is familiar with Ripple. I think they’re very focused on use cases around the digital asset, XRP, as well as enterprise use cases. So, kind of making payments more efficient for banks, payment service providers, their customers, corporates, and so on.

So, what I found with working with some of those kinds of customers, and this is sort of something that I think was more of a mutual realization among the Ripple team and myself, a research team, I mean, they are kind of interested in whatever their customers are interested in, and their customers, the customers that matter the most to them are probably some of the larger corporates and that kind of thing.

So, if you think about Interledger as this thing that we talked earlier about what are the things that Interledger is best at, if you think of it as a thing that enables a new class of use cases, kind of how the internet enabled a new class of communication use cases, it’s unlikely that large corporations are going to necessarily be the very first to adopt it. They’re going to be looking for cost savings under existing use cases, they’re going to be looking for ways to maybe scale their existing use cases better or things like that. They’re not going to be looking for, “Okay, how can I send a fraction of a cent to halfway around the world?”

So, if you want to go out and you want to build totally new products, I think everyone felt that it was much better to do that as part of a new company. So, I don’t think it would have been possible to build Interledger as Coil because you kind of need enough funding and momentum to put in all this research and pay for a whole team to do that. But, as far as productizing it, you kind of need something like a startup that’s just not doing anything else, that’s just going after this one opportunity. So, that’s kind of what Coil is intended to be. I don’t know if you want to add anything to that.

Ben: Yeah, we just really need to create the demand, because somebody has to do it, you know?

Sunny: I mean, one thing I noticed about the Interledger ecosystem is that there’s still a lot of work to be done on the protocol layer side of things. So, for example, right now, the only finished plugin is to XRP. And there are some like Carver is building some on Ethereum, and Bitcoin, and stuff. But, what made you guys decide to focus on going out to start building end user applications instead of working on the core protocol? Because, I don’t know, in my opinion, I see that, actually, as having a lot of work left to do.

Stefan: I think, actually, there’s a little bit of leapfrogging going on. The last community group call, I’ve heard that some of the Ethereum plugins are actually starting to look pretty good, and we need to go back and maybe update some of the XRP plugins. I haven’t looked at it myself, but that’s what I heard. So, I think those plugins are actually getting close to mature, and I don’t know if you have a —

Ben: Yeah, and I think on whether we should wait until there’s more on the core protocol level, I think you really can’t get the core protocol to be ready or to be very good unless you’re actively doing things on top of it, like in production because now that we’re actually accountable to keeping the network up, like our customers aren’t going to be messaging us if the Interledger network’s down. Those are much higher stakes than it was purely for research.

Stefan: Yeah, exactly. I think the healthiest protocol is one that evolves with actual usage. I think, as far as some of the core data formats, like for instance, the ILP packet header, which is the most important thing you really need to standardize, I feel pretty strongly that those are actually done. And so, it’s more some of the protocols below, like you said, the ledger integrations, some of the protocols above like transport layer protocols where there’s still like, especially on the implementation side, not even just on the protocol side, but the implementation side, there’s a lot of room for improvement.

But, there’s enough there that you can go out, and you can build an app, and have it be rock-solid, and very usable, and very performing, and so on. I don’t know if it’s like super-scalable yet. I think it’s the amount of throughput that we receive per node. I mean, you kind of heard the volumes that we’re doing. I think, certainly, by Blockchain standards, it’s pretty good.

Sunny: So, you must read the Union Square Ventures’ blog post from last month about the myth of the infrastructure phase. And so, I guess you guys have a career with that model and believe that you need this curvature path where the demand and the infrastructure are kind of building off of each other?

Stefan: Yeah, I mean, that’s exactly what we found with this idea of having a connector. I don’t think that Strata would be a commercial connector company if it wasn’t for Coil being their customer. And I think it’s you’re kind of mutually creating the demand, you’re kind of solving the chicken-egg problem by co-evolving together, and you’re also kind of leading the way for other companies to get involved.

Because, I think there’s so many other people that I’ve talked to before starting Coil where they kind of look at it like, “Yeah, it sounds nice in theory, but you’ve got this cushy job at Ripple, you can do this all day. I need to actually make money, build my company,” and so they don’t feel comfortable actually jumping in. But, now, I’m 100% dependent on it, like my personal fortunes are all tied up in Coil, so if this doesn’t work, you know.

So, if I talk to entrepreneurs now, it’s like, “Oh, you know, you’re really going to stand behind this, you’re really going to make this work no matter what.” So, I think, totally, like Ben said, it really totally changes the tone of how you work on it and what people react to it.

Brian: Are there other ideas that you guys were thinking about besides this web monetization and micropayments that you felt like, “Okay, this could be an interesting company to build around Interledger”?

Stefan: Yeah, we had a board of 49 ideas. I’ll let Ben talk about that.

Ben: Yeah, I remember there was a day when we were trying to think of, I think it was what to do at the next conference we were attending. So, we just started listing all the ideas we could think of on top of Interledger. So, now, in the context of starting Coil, we’ve zoomed in more on the web monetization one. But, even within there, I think we could definitely list another 50 things to do with web monetization, and probably we’re going to have to zoom in somewhat there too.

Stefan: Right. I think, once you’re in a startup, suddenly it’s like, “Okay, we got to focus, and once we focus, we got to focus more.” I think one of the ideas that was a frontrunner for a long time was Codius, so that might be a name that some of your viewers recognize. Evan and I worked on this project a long time ago at Ripple, and we kind of got it to a point where the only thing missing was the payments interoperability protocol because Interledger didn’t exist back then. So, when we started Coil, we actually went back and built a prototype of Codius because it was literally like — how long did it take us? Like three weeks?

Ben: I think that was our first couple of weeks, yeah.

Stefan: Yeah, two or three weeks. So, that actually, kind of, in and of itself, was exciting to me because I think, before, you could go and you could have an idea for decentralized application, and it would take years to kind of put together an ICO, and you want to create a token, and you need to have a Blockchain, you need to have a consensus mechanism, you need all these different components.

Whereas, you can literally just like, for Codius, which was hosting using Interledger. So, we basically just took a piece of software that could already do hosting, and it would run different containers in their own virtual machines, and then we just added an API, and we made it like a paid API with Interledger, and then we released it, and then a couple weeks later, we had 400 people running Codius hosts.

Unfortunately, Codius isn’t that core to what we’re doing right now, so we can’t devote too much time to it. But, it’s just one of those ideas. It’s just like it’s so obviously useful that we just felt like we had to give it at least a bit more the light of day, you know? So, I don’t know, what I took away from is just how easy it is to build stuff with Interledger.

It’s just like a way to solve incentive problems, and you can almost forget about how any of that works, you don’t have to worry about token economics and stuff like that. You can just go and say, “Well, if you want to access my mesh network, this node pays this node. If you want to upload something to the server, you pay. Whatever you’re trying to build, whether it’s an incentive issue, if you download from me, you pay.” It’s just like you just rely on Interledger. It feels a lot more like building things on the internet.

Ben: I think the cool thing about web monetization, as well, is that it gives you that Interledger capability just in your browser. So, that means, really, anybody with a Coil subscription could use any of the applications we build on Interledger. Like, you can upload a smart contract to Codius from a browser.

Stefan: And pay for it from the browser.

Ben: That’s right.

Brian: And so is the plan to get back, at some point down the future road map, into the Codius sign thing? Because, I heard from someone, I’m not sure if this is true, that the name of the company, Coil, was actually like an amalgamation of Codius and Interledger.

Ben: Yeah, that’s true. It’s kind of a smushed those together.

Brian: Okay.

Ben: It also fits in with some of the other naming schemes that are floating around like, for example, Xpring at Ripple, that’s funding us, basically. So, I think, yeah, those are definitely the two pieces that are very important.

Stefan: Yeah, we had a lot of names, like we have Council, which is a service selection protocol. We could talk a little bit about that. We had something called Cog, which is like how to pay for APIs. So, the CO kind of meme was around a lot, and it came from Codius, mostly.

Brian: So, let’s move to Coil and dive a bit more in depth there. Maybe one good way of explaining Coil to our listeners is if you can briefly run through, “What does my user experience look like?” Ask somebody who’s going to use Coil as an end user, and what about the user experience of using Coil as a creator?

Stefan: Right. So, I mean, our goal has been to make it as easy as humanly possible. That’s sort of been the primary goal. The other goal, I’ve always been after mainstream adoption. So, I’m looking at it from the perspective of a mainstream user, not somebody who’s already way knee-deep into crypto themselves, right?

So, for the end user perspective, the way it works is you go to Coil.com. Right now, we’re in closed beta, so we actually have a waiting list to let people on slowly onto the platform because we want to make sure that when they get on, they have a decent experience and they can scale fast enough and so on. So, right now there’s a waiting list, but eventually, you’ll be able to just sign up, put in your credit card, you pay $5 a month, and that gives you access to the entire platform.

One thing that’s, I think, unique is that we’re not taking that $5 and splitting it up in some way and then paying it out. It’s a flat rate. It’s a true flat rate. So, you can use more, you can use less, and Coil makes up the difference. The reason we think that’s so important is because the ideas gives the user peace of mind.

Like, I used to use this site called Flattr, and basically what would happen is Flattr takes your monthly budget and they kind of split it up between all the websites that you go to. So, what that means is that if you go to some website, they initially get all your budget, and then as you go to other websites, that gets subtracted from the first website you went to.

So, I remember I would go and think about, “Well, if I go to these other sites now, then the website I really like the most will get less,” and so suddenly I was thinking about it, and that’s really not a good way of doing it. So, we’re trying to get more of an experience like you’d have on Spotify, or Netflix, or something like that where I have access to all this content, to all these websites, and I can use as much as I want.

So, if I leave my Spotify running overnight, sure it costs Spotify in like licensing money, but I, as a user, don’t have to think about that. I don’t have to worry about it. In fact, I have a strong incentive to use it as much as possible to get the value out of it. And we think that that’s a more engaging way of building an application.

Anyway, just to wrap up that discussion of what’s the experience, as a user, I sign up, I pay $5 a month, and now I can go to different websites, and right now, it’s mostly about supporting creators that I care about. So, right now, there are a lot of people from the XRP community that are either blogging, or making tools, and so on that are web monetized. So, if I sign up to Coil, I can support those creators if I care about their particular content just by using it, reading it, following them, and so on.

We think that that same model could apply to other communities as well. So, let’s say you’d apply it to the Fortnite community, and so suddenly, people that are Fortnite streamers, websites, forums, etcetera, they can become web monetized, and now users that want to use those websites, they can be web monetized, and they can get a cloud subscription and go to those sites.

So, that’s kind of the idea. And then for the creator, it’s also a pretty simple experience. Although, on that site, there’s still some crypto involved. So, if I’m a creator, I have to sign up for an ILP-enabled wallet, and as of recording this episode, hopefully there’ll be more options soon. But, as of recording this episode, the only option is XRP Tip Bot. And so, I have to go there, I have to make an account, and then I receive XRP.

But, what’s cool is that the underlying technology is Interledger. So, if someone out there has an Ethereum wallet and they want to integrate, they can make it so that creators can earn their money in Ether or in some ERC-20 token. So, it’s like the technology allows you to receive in whatever you want, as long as there’s, provided, that gives access to that particular currency.

Brian: So, just a question briefly on that. Let’s say now, I add my website, I want to receive Ether, someone else, Sunny, wants XRP, he’s his own creator, the idea is then that you, as a company, Coil, will basically buy whatever crypto-assets for whatever the creators want and pay them in that token?

Stefan: So, we’ll actually send it out on Interledger. Ben, you want to explain how that works?

Ben: Yeah, and Stefan briefly mentioned Strata earlier. They’re our Interledger service provider. So, essentially we just have an uplink to them where we send ILP packets, and then later we settle them using XRP. But, due to the multi-hop nature of the Interledger network, Strata is then passing it on to the next hop and so on until eventually the creator themselves gets whatever currency they wanted. So, Coil doesn’t need to worry about what payment methods exist in the Interledger network because that’s sort of what their routing does for us.

Brian: So, who when actually does that exchange, let’s say, from XRP into Ether?

Ben: So, that would be, potentially, Strata or a peer of Strata. Some connector in the Interledger network will hold both XRP and Ether. So, to one of their customers, they’ll be sending packets, Interledger packets, which will be settled in Ether, and they’ve got money incoming that’s being settled in some other asset.

Stefan: And there could be multiple currency exchanges on a path as well. It doesn’t have to be direct.

Sunny: So, you mentioned about Netflix and Spotify, right? Do Netflix and Spotify pay their content producers? So, whether it’s the movie studios or the artists, do they pay them based off of number of views or number of listens, or is it like a fixed fee that they pay the producers, the licensing fees up front?

Stefan: I think that content licensing is a very deep and complex topic that I don’t think I can do justice. But, I would say it’s fair to say that, by and large, there’s generally a relationship between how many times a song is listened to or how many times a video’s viewed, and how much the content creator receives. There’s just a lot of other variables too, like the negotiating position of like a large major label versus a smaller independent artist or something like that.

I think what’s interesting about Coil is that we have this fixed amount of money that the user’s paying us. And sure, somewhere we need to make a profit. I think, initially, we’ll probably operate at a loss just to help get the network going. But, eventually, let’s say user pays us $5 and we pay out $4.95 on average, like across all of our users. So, now we have some profit that we’re making, 1% or whatever it is, and our goal has to be to make sure that those $4.95 are spent as efficiently as humanly possible.

And when I say efficiently, I mean our users probably don’t care how much money that some clickbait website that they accidentally clicked on make, but they might care about getting that HD video that they saw on that web monetized website where you can only watch it with a certain amount of bandwidth, or that one article that you can only read after you’ve paid a certain amount of money. They don’t want to wait for a long time to read that, things like that.

So, I think that we would optimize, and we try to get really good at figuring out, based on things like user feedback, user voting, user behavior like, “How do we pay just the right amount to give the user a good experience without overpaying, which then gives them a worse experience on some other site.

Brian: One of the things that I struggle with with this model is that, let’s make an example. Let’s say I have a video, I do some livestreaming video, and this is like a high activity that takes a lot of effort, and now I want to offer that, “Okay, people can get the HD stream, and then they should pay more, and then they can have a lower resolution, and they should pay less.” So, is there some way to kind of differentiate between your maybe high-value services and low-value services that are being provided?

Ben: Yeah, I think with Coil, it’s slightly different from how you would think of buying different tiers of content today because with the web monetization standard, the user is kind of, you can think of it as like the user’s almost opening the negotiation on price because they have a certain amount of money that they can send per second. So, it’s kind of up to the site, then, to say, “This is good enough and I’ll give you the content,” or to say, “This is not good enough, and you have to watch at a lower quality version of the stream,” or something like that.

With payment methods right now, the site tells you how much the content costs, and then that forces the user to interact with it in order to pay, which means most of the time, the user’s not going to pay at all. So, yeah the user can still differentiate into tiers, but the user would need to then basically buy a more expensive web monetization subscription or turn up.

Stefan: Yeah, they’d have to pay more, essentially.

Sunny: So, is there like a slider right now on the Coil extension where I can like increase my payment bandwidth or decrease my payment bandwidth?

Stefan: So, we thought about a bunch of different models. Right now, it’s a single fixed fee with a single fixed bandwidth. But, obviously that’s our minimum viable product. That could get more sophisticated in the future. I think one of the things that we thought about is allowing the user to, on medium, there’s like this cloud feature where you can kind of say, “Hey, I really enjoy this article.”

So, you could have something like that where the user can kind of signal that, “This is a creator that I particularly care about,” or, “This is a website that I keep going to.” So, we might be willing to pay more for a website that the user has previously given us feedback that they liked.

Another thing could be like the opposite of that, where it’s like, “I think this website is clickbait. I don’t want you to pay them anymore,” and it’s funny because it doesn’t cost the user anything, but there’s still websites where we get the feedback that, “No, I actually actively don’t want you to pay them,” right? Because, I don’t want to reward that kind of behavior.

So, if we get a lot of feedback like that about a website, then maybe we’ll default to a lower bandwidth, or even no bandwidth at all if it’s literally an abusive website or whatever it is, like they’re just clickbait.

Another, I think, differentiator in the future could be that we could have different tiers of subscriptions, right? So, we have an entry-level subscription where it gives you access to certain types of content, and then there’s like a premium subscription which, if you really never want to worry about bandwidth ever again, you can get that, and then everything will be maximum tier like HD and so on.

I think what’s going to happen is there’s going to be a certain amount of interplay between the publishers who are trying to set their price levels, and then the users, as well as their providers, like Coil, who are setting their bandwidth levels. So, we might look out there and say, “Okay, publishers have these kinds of features at this bandwidth level, and then these kinds of features of this bandwidth level,” and so we’re going to introduce a new subscription that has a higher bandwidth level or that lower bandwidth level. So, both sides would kind of react to each other. That’s how we imagine it could happen.

Ben: And because this is just built on web monetization, which is just a JavaScript API that we had to found, an Interledger which anybody could join, you also would have the interoperability routine different web monetization providers where maybe users don’t like a tiered system. Maybe they just want to have a different provider where they just pay as they go, and they’re okay with that. Then, that web monetization ecosystem could fit that in as well.

Brian: Now, if Coil succeeds and if this works at a large scale, I mean, you guys alluded to Bit, right? So, you could expect that I, as a website developer, as a creator, I now say I customize my user experience and I give a better user experience, some premium content, or some other preferential treatment to people who have this activated and are paying me, basically, for the data I’m providing, and to some other guys, I do this.

At the same time, from the user perspective, I may say, “I want to now join this web monetization thing and I want to pay my subscription not just because I agree with some of the ideals, or maybe I don’t like the advertising business model, but because I get a better user experience, my websites work better now.” So, is that kind of where you see things going? And can you give some examples of things that you hope people will build out if Coil actually sees widespread adoption?

Stefan: Yeah, I think, Ben, you actually built some apps that kind of use web monetization. Maybe you can explain some.

Ben: Yeah, so I would almost think of it as the low-hanging fruit of some pretty useful applications you could build. One of them was file upload, which I think is interesting because the way I did it testing out was that what if you upload a file, and then the downloader pays to the uploader. So, that actually means you could throw some files up there, and then embed maybe that content in your website, like a video or something, and then even if you don’t have any programming knowledge, you could do that as a way to make premium content that web monetization users can access.

Another one was a livestreaming site where basically you can tune into a stream and so long as you’re streaming payment, you’re also getting streamed video. So, that obviously has some obvious uses of — maybe you have just a Twitch streamer or something or other kinds of streaming that want to make sure all of their users are paid.

Stefan: One thing I would kind of highlight is that, right now, there’s sort of a set of user experiences that we’re used to. There’s the premium experience where I have to sign up, I have to put in my credit card, I have to get on board it, and then I can finally enjoy the premium experience. Then, there’s the ad-supported experience where it’s like, “You know, it’s free, and maybe I don’t even have to sign up, but it’s kind of like there’s ads in my face all the time. I don’t really know if their privacy policy is going to be very much in my favor, if it’s more in the favor of the advertisers,” and that’s just kind of like, “Take it or leave it,” that’s how it is.

And this almost creates a third class of website where it’s like once I am web monetized, once I’ve set this up once, I can go to a website I’ve never been to, I don’t have to sign up, I don’t have to pay, even, if it’s part of my Coil subscription. I don’t have to look at ads. It’s just an experience that’s designed to be as frictionless as possible.

So, imagine for a file upload website today, they have these really horrible ads with the fake download button that you click and then it installs malware in your computer. Or, you can sign up for an easy payment of $7.99, and I’m like, “I don’t use the site that often. I don’t upload files that often.”

So, we envision that Coil subscription would just be something that you kind of get, and then it supports a bunch of creators that you like. It also gives you a bunch of services like that, like a file upload service. You just go to it, it lets you upload infinite-sized files because the larger the files, the longer you spend on the uploader. So, it’s kind of paid for. But, it lets you download at whatever speed you want because the larger the file, the longer you’re going to spend on the download.

And then it doesn’t show any ads. It’s a clean, simple user interface, and that’s it. And just imagine you just have this less BS web experience, and that’s what Coil gives you for $5.95\. That’s my pitch.

Ben: I think it has some implications for privacy as well, because if you imagine sites right now, if they want to charge payment, like you said, you need to sign up first, they need to have an idea of who you are so that they can remember that you’re a paid user. With web monetization, you’re just streaming payments while you’re actually on the site. So, they can see, “Oh, this user, who’s completely anonymous, is sending payment. So, give them the premium experience, but then you can forget about them once they leave.”

Brian: I do think that is a very desirable thing. One of the things that stands out to me here is the way we’ve been speaking about this is web monetization and this almost fundamental new way of economic relationship on the internet. And at the same time, when we speak about the role of Coil, it very much feels like the role of traditional startup, like we made the analogy of Spotify, and I as the user, I pay. I give my credit card information to Coil, I pay you the monthly amount, you guys have some discretion on, I guess, the rate at which you pay out to the different websites.

So, it feels like there’s some almost misalignment here in terms of this ambition of this big, open, fundamental revolutionary protocol, but then at the same time, also of building kind of a traditional web startup. How do you look at — do you see a tension there?

Ben: Yeah, I think it’s important that, along the way, we are offering an alternative. Obviously, I think for people to actually sign up for Coil, if they don’t understand what the web monetization standard is or if they’re not familiar with Interledger and these kind of things, then it’s not a huge selling point to them. But, I think it’s really important for just the evolution of how this part of the web evolves that we have a non-Coil way to do web monetization.

So, we actually have some software that you can download on your machine, which basically connects to Interledger, and in your browser, you can go to web monetized sites, and this software you’re running will pay this site instead of Coil. And so that kind of gives you actual credibility that it’s not just Coil, because it’s easy to say, “Yeah, anybody could come in and do this.”

Stefan: Yeah, you can actually do it, like right now. Go ahead.

Sunny: So, this web monetization stuff, so, the Coil product is specifically this subscription-based idea where I pay you $5, and you guys do all of it for me. If I wanted to, you said minute in order to load my own wallet or something of some sort and I can do everything by myself. That’s kind of where the difference is?

Stefan: Figure it as like the difference between Gmail or G Suite versus running your own email server. And I think one of the reasons that that’s significant is not just because you could run your own server. I think a lot of people are not going to do that. But, someone could start a competitor to Coil, and we’re actually totally okay with that because we think that we can provide a good enough user experience that people will want to use Coil, and we’re totally comfortable competing with someone else on that.

I think, also, there could be a lot of variety in terms of the kinds of models that people could try. So, we talked to somebody the other day who was thinking about doing a token-based model where you actually have a dedicated token, and then that’s what you’re spending through Interledger to pay for web monetized sites. So, there’s a lot of token economics involved with that. We talked to someone else who wants to incorporate this into their ad network. So, you might look at that network’s ads in order to pay for your web monetization usage.

So, are those competitors? In a certain sense, sure, but a lot of their users wouldn’t be the right users to pay a $5 subscription, right? Or, you could think of different kind of content bundles, right? The provider can choose what they want to pay for. So, one example I have for that is imagine if you have a company and they want to provide web monetization enabled browsers for all of their employees.

They probably don’t want to pay for their employees to go to Facebook all day or watch Twitch streamers, but they might be okay with paying for things like educational content, news, things that are going to make their staff more productive. So, someone could come along, they could say, “Our web monetization subscription is specifically for paying for a corporate use case.”

And it’s like, again, not really competitive with what we’re doing. I mean, some of these things, we may want to do in the future, but my point is that I think the system’s a lot bigger than just one platform.

Brian: We spoke before a little bit about in the end, if we have a success of Coil or a success of web monetization, you’d have websites that build around it and users that you specifically sign up for it. But, of course, in the short term, like right now, there are very few Coil users. So, as a website, it’s probably not worth to integrate it, and then at the same time, as a user, there are no or probably hardly any websites, or maybe none at all that provide a better user experience if I’m using Coil and web monetization.

So, it feels like this is a very tricky problem where you need to have significant scale on both ends until you really have a kind of economics at work. What is your strategy for getting to that point?

Stefan: I feel I’ve gone through this a bunch of times now. When we started out with Bitcoin in the early days, it was like I would go to people and people would be like, “Well, what websites can I pay on with this?” and I’d be like, “Well, right now you can pay for Alpaca Socks at this family farm in the U.S.,” and people were like, “Oh, that’s cool.” And I think what you do is you just find the people that are excited by the technology and they’re excited by this vision.

And yeah, maybe the first couple people that are embedded, it doesn’t make that much commercial sense, but at the same time, it’s like you put a piece of JavaScript on your website. It’s not like it takes you weeks and weeks to do this. And you have pretty much nothing to lose. Like, worst case, you don’t make any money with it, but you’re also not really losing anything.

And then, as you get some initial users, and we have about 5,000 people on the wait list right now, that’s enough to kind of get some initial feedback and get a better sense of what are people looking for. I think one of the biggest pieces of feedback we’re interested right now is, which websites would people actually want to see web monetization enabled so we can actually go to those websites and say, “Look, our users really want your content. Our users are really interested in your website”?

And then also, like Ben alluded to, we signed a financing deal with Ripple, so we do have a little bit of XRP that we can use to incentivize websites to support web monetization, that kind of stuff. So, once we get a good idea of what our users would like to see, we could go to those websites and we can make it happen. So, I think we have enough. We have all the pieces in place, but it is going to take a while to really take off.

Sunny: I saw on your website that the Coil plugin currently supports YouTube and Twitch. As far as I’m aware, YouTube and Twitch haven’t enabled web monetization on their platforms yet. So, how do you guys do this?

Stefan: So, I think that’s an interesting one because there are certain platforms that have such a dominating position in the market that A, they’re not going to be able to,  or it’s not going to be relevant for them for a very long time. I think that’s a good way to think about it. But, at the same time, they have a lot of creators that are sort of captive on their platform that really would like to use web monetization.

So, I think a lot of similar systems have done similar types of integrations where we just use a browser extension to say, “Okay, well if somebody goes to a certain YouTube channel and that YouTube creator verifies with Coil, we can actually make that transaction happen.” I don’t like it as much because it’s not an open standard. Like, unless YouTube actually goes and becomes web monetization enabled, the open standard users are not going to be able to pay unless they talk to the Coil server, etcetera.

But, that’s just a limitation of using these closed platforms. Like, if it was up to me, I would definitely say people should move off of YouTube onto a more open platform where they could actually put a piece of JavaScript themselves and have control over that. But, you got to deal with the world that you live in, right?

Sunny: I know YouTube doesn’t have a donate button already, but I believe Twitch does, right?

Ben: That’s right.

Sunny: And so is there maybe thoughts of maybe integrating Twitch donate as an ILP plugin?

Ben: We actually have already, because that’s actually how the integration to Twitch works is that, basically, you can go to any channel, and if they’re a Twitch partner, then our receiver on Interledger is going to be accumulating these packets, which they then settle out, essentially, to that Twitch streamer by donating in their channel with however many Bits it’s accumulated.

Stefan: Bits is Twitch’s currency.

Ben: Yeah, that’s the Twitch —

Sunny: That’s really cool. I guess that’s a little bit confusing with the Bits and Bitcoin.

Ben: Yeah. And the interesting thing is that because it’s done all over the top, so it’s not actually through Twitch adding anything. We’ve just basically done this over what they exposed to regular users.

Stefan: So, what happens from the streamer’s perspective is like let’s say a bunch of Coil users go watch their stream, after a while, this bot will join the chat. It will be called, “Coil_Twitch_Bot,” and it will just donate bits and then pop out again.

Sunny: Oh, that’s cool.

Stefan: Yeah, so it’s almost like it’s acting a little bit as a marketing tool as well because people are like, “What the hell is a Coil Twitch Bot and why does it keep donating?

Ben: We’ve definitely got a minimum limit to prevent spam as well.

Stefan: Yeah, we don’t want it to be a nuisance.

Ben: Yeah, you could be dropping in one cent every couple of seconds, which maybe some people wouldn’t appreciate.

Brian: Just one point very briefly on this. I did actually activate this on the Epicenter YouTube channel. So, if you go to Coil.com, I guess there’s a waitlist at the moment, but if you do get the Chrome Browser extension and then go to any of our Epicenter videos that you should see it in the extension that it’s donating tiny little bits of money to Epicenter.

Stefan: Yeah, and I know that some of you viewers might already be on the waitlist, so we actually just started working through the wait list and starting to invite people off of the waitlist actively. So, if you’ve been on the waitlist, just give it a bit more time. We are trying to get through it as quickly as we can.

Sunny: So, I guess one more question before we start to close up. In the web motivation, like the interception of web monetization with crypto, I think the other big name that most people are very familiar with is the Basic Attention Token, and they seem to be kind of going for a similar product to you guys, and Brendan Eich and everything, everyone knows these people. So, how do you see your platform comparing with Basic Attention Token? Have you ever talked to them about integrating with the web monetization work and ILP work that you guys have already been working on?

Stefan: Obviously, I can’t talk about the specifics, but I have reached out to them. I think that they also care a lot about the same things that we care about like privacy, openness, and so on. So, I have some hope that, eventually, both will use the same standard. I don’t think that’s going to happen anytime soon. I think they have a lot more traction, at least in terms of browser users. So, I don’t know they’re open to working with us on a standard. I think, ultimately, long-term, I think our standards are going to be more powerful.

I think, with Interledger, you’re not tied to one particular token. I think that’s going to open it up to a lot more communities to get interested in this. If you think of Ethereum community, Bitcoin community, our proposal is something that’s much more aligned with like, “Hey, if I just want to use Bitcoin, I can just use Bitcoin, and I can still use all of the technology that the Coil guys have built.

So, I think that’s one aspect that’s the difference. Another aspect is that, which we really try to make it a browser API so it’s a direct communication between the website and the browser. So with Brave right now at least, you verify with Brave.com, so it’s like it’s actually talking to their server and also, of course, it only works in the Brave browser as opposed to any browser. So, we think that it’s a little bit more of just a neutral standard, a little lightweight of a standard I think is the best way to say it. And I’m trying to be nice because I hope that one day, we can work together.

Brian: Great. So, very final question. What does the road map look like? At what point do you hope to be 12 months from now and maybe two years from now?

Ben: So, I think sort of next up is premium content, because that’s the things that’s obviously a very crucial part of making a Coil subscription work, and that’s something we haven’t actually explored that much yet.

Stefan: And actually, just to add to my previous point is we’re actually able to pay out in real time thanks to Interledger, which I think is another kind of big difference to some of the other solutions out there is that, if I’m a website, I don’t have to know who is this provider that the user is using because I’m actually getting money through my provider in real time. So, I can provide premium content based on that, and so we want to obviously unlock that.

Ben: I think that is actually one flaw of the maybe Brave and Flattr approach of paying it and dividing up a subscription is that you don’t actually know who gets how much until the end of the month. So, in the time that you’re actually browsing the site, the site can’t tell how much money it’s getting.

Stefan: Right, because it could be more or less.

Ben: Yeah, so premium content is probably next up, and I’ve also been working a lot on tools for that and developer tooling to make more advanced sites with premium content. But, I think that’s the thing that’s pretty important, and then also just letting more people off the waitlist, getting more creators on there, pull these people from a lot of different communities to see who’s going to be the most enthusiastic about adopting this because it could be maybe it will happen to be like gaming, or music, or YouTube creators.

Stefan: I think that’s a big aspect of this is like we want to have features on Coil.com where users can tell us, vote for websites and suggest websites that should be web monetization enabled because we want to have that feedback of where people’s interests are.

And then also, I’m going to personally be spending a lot of time trying to talk to different people about this. So, I’m in L.A. for the rest of this week. I’m actually taking part of the team with me, and then I’m going to be in New York next week, and then I already had a series of meetings here in SF, and the goal of all of these is to try to talk to as many different kinds of creators, and people in the creative industry, and people involved with monetization as possible to really try to get a sense of where their pain points, what’s working, what’s not working, and just trying to understand that world better.

Sunny: I saw, recently, a partnership announcement or something with the Bill and Melinda Gates Foundation. Is that something, can you tell us briefly about that?

Stefan: Sure. I mean, that’s a little bit of a personal passion project of mine. That was something that started while I was still at Ripple. We started talking to the Gates Foundation, I would say, like three years ago or something like that, and they had to kind of identify a lot of the same things that we identified when we were doing research on Interledger basically around the real deep problem in payments as a lack of interoperability, and the way that it manifested for them was they would go into a country and there would be these different payment providers, and they could help the incumbent, the biggest provider, but then they’re kind of monopolus, as soon as the project ends, they might crank up their prices, things like that.

Or, they could work with some of the smaller providers, but now they’re just helping to fragment the market, and so now you have merchants that have like three phones in their shop just to accept different mobile payment methods and things like that.

So, they started to identify that if you want to have a frictionless, easy user experience, you want to have low cost, and you want to have a high level of competition in terms of payments, you really need interoperability. That’s really what the reason that data transfer has gotten orders and orders of magnitude cheaper while payments have actually gotten more expensive over the last 20 years is because that lack of competition around interoperability, right?

And so, we started doing a project with them called Mojaloop, which is basically people’s eyes might glaze over, but it’s an open source national payment switch. So, it’s something that a country would deploy and them basically allow its banks and other payment providers to use it.

So, long story short, we’ve been involved in this project for three years, it’s gaining a lot of traction, there’s a lot of implementations there in discussion now. Unfortunately, nothing announced yet, so I won’t spoil it yet, but I think that I see a lot of potential in that project, and so one of the things I wanted to do even though I have a new company, new startup, I want to continue to work on that project.

And so, thanks to the Gates Foundation. They made that possible so myself, Adrian, who I mentioned before, and some other people here on the team, Ben, others, are also helping with that project, and we’re continuing to drive that forward. And it’s just something that we like to do.

Brian: Cool. Well, thanks so much for joining us today. Of course, we’ll have links. People can learn more about Coil and check it out, and yeah one of the cool things, it’s working already. So, right? I still have to tip off, and it was on a YouTube channel before, and now there’s a little bit of XRP that was received by the Epicenter gitpod, so yeah that’s pretty cool that it’s actually working. So, thanks so much for joining us today.

Stefan: Thanks so much for having us. It was a lot of fun

Ben: Yeah.

Brian: So, [where do you faults on you? 1:24:18]. How did you find this conversation?

Sunny: Honestly, it was actually pretty interesting. I guess my first thing that I would really say is it’s kind of a pretty brave in a way, I would say, that they’re really jumping into building a product right away. Like I mentioned during the episode, the whole Union Square Ventures’ blog post about the myth of the infrastructure phase where you don’t want to be stuck in a period for years where you’re just building infrastructure without building product, and I feel like Coil is kind of an example of that idea really taken to heart where they’re like, “Yes, we understand that the ILP ecosystem is very immature, but we’re working on that, but at the same time, we really want to just jump in and build a product to show this thing actually has a real use case.”

Brian: Yeah. The thing that just really stood out to me is that this action product, live in a short time, and the user experience was pretty neat and easy. That being said, I do think there’s a little bit of an awkward fit in a way, where on the one hand, you have this idea of almost like base layer internet protocol and infrastructure, so they’re talking about W3C standard, and in the browser, and these fundamental ways of how the internet works, and at the same time, almost this sort of business model approach where it’s more like a Spotify or a Netflix type thing. So, I do think there’s some tension there, and I’m really curious how that’s going to play out once Coil moves ahead and gains a bit more traction and users.

Sunny: Yeah, absolutely. I also did think that ILP is this very broad protocol that has a variety of use cases, like you could use it for interoperability, in a way you could honestly just use ILP as a dex of sorts, you could use it for micropayments, and they really seem to home in on the micropayments aspect, which honestly, to me, seems to be like an interesting choice considering there’s a lot of projects in the space who are already pushing this micropayments idea like Lightning, and Raiden, and a lot of the Layer 2 networks.

And I feel like what really makes ILP unique from most of the Layer 2 networks is its focus on interoperability, and I kind of find it interesting that they chose not to make that the focus of their product.

Brian: Yeah, that’s a great observation. I didn’t even think so much about that, but I think you’re totally right. But, yeah in any case, I think it will be exciting to see what comes out of Coil, and certainly is a very strong team. So, I’m looking forward to seeing what they built in the long term.

Sunny: We mentioned during the episode that when we started, we actually turned on the Coil plugin on our Epicenter YouTube. That was like two days ago. So, you want to give the listeners an update on how that’s going?

Brian: Since hardly anybody uses Coil, I presume not much has happened, right? Because, you need to have a Coil user, and they have all these people on the wait list that goes on the Epicenter YouTube page, and then we would start earning some small amounts of money.

But, what was interesting was that when I turn on Coil and then I went to our Epicenter channel, and you see there’s like tiny, tiny amounts of money going into our account, and then I kind of leave it open, and then two minutes, come back, it’s like, “Oh, we’ve made like two cents.”

And it’s interesting to notice, even instinctively, this almost desire to gain the system that then I realize, “Oh, I’m basically kind of — is this defrauding Coil?” So, for an industry that thinks very much about having a game theoretic strong incentives and stuff like that, I think that isn’t such a — there could be an issue there down the line.

Sunny: And I guess another actually one of my biggest takeaways from this episode that I guess I didn’t really — didn’t register to me, maybe because I just didn’t do enough reading was this difference between web monetization versus web payments. So, they talk about web payments as sort of like a pull where the content creator or the website is requesting money from you and you are saying you accept it and say, “Okay fine, I want to pay you,” while the web monetization is almost like a push payment where the user is the one that has this constant stream of money going.

I was actually super-excited about this web payment stuff, but I’m actually, honestly, a little bit skeptical about web monetization just because — see, what I see is we talked about, what if different content creators want different amounts of payment for different types of content, right? So, what is the user supposed to do? Is there going to be like a slider in the Coil extension that allows you to change the amount that’s being broadcasted.

But then, if you’re having that, then now you’re kind of like — from what I understood from the episode was like one of the main benefits of web monetization was this lack of need for user interaction, but it seems that, now, we’re bringing back the user interaction, and maybe it will be easier if we just stick to a singular web payment standard rather than like monetization plus payments.

Brian: Yeah, I think that’s a good observation. It is a very interesting and unusual concept, this idea of basically streaming money to the web and all the websites you visit. But, yeah anyway, let’s follow the project and see what happens with it.

Sunny: Cool. Thank you guys all for listening.