Episode #86: 🎻 Decentralized Playlists w/ Roneil Rumburg from Audius

Audius is building a decentralized platform for music inspired both by Soundcloud and private torrent trackers.

Coin Talk
Coin Talk
Jul 25 · 41 min read

Listen to this story



CoinTalk™️ is produced in partnership with Medium and hosted by Aaron Lammer and Jay Caspian Kang. Press “Listen to the story” above to play the episode. (You can also subscribe on Apple Podcasts, Google Play, download the MP3, or email us at hi@cointalk.show)

Show Notes

  • The advantages of decentralizing communities like Soundcloud’s
  • Building on VC money vs. token money
  • Why Oink was the greatest music service ever


Aaron Lammer: I’m going to be totally honest, Roneil. This show is all about Bitcoin honesty. I get about 100 pitches a week, 99% of which I delete immediately. Not because I don’t think people are doing interesting stuff, but because I don’t think they’re doing stuff that I have any particular insight into or connect with. But you run a, would you say, a blockchain startup? What are we calling these, a crypto startup?

Roneil Rumburg: Yeah. I would say decentralized technology startup. I don’t know what-

Aaron Lammer: Decentralized technology startup in the music space.

Roneil Rumburg: Yep.

Aaron Lammer: Which I’m in a musical project and publish music. So this is an area that’s of particular interest to me and that I’ve followed pretty closely, I guess, from I feel like Imogen Heap was selling music on the Ethereum network. Was that kind of the first big crypto music story?

Roneil Rumburg: Yeah. I think so. Then there was Ujo, which was ConsenSys’s project. Then there have been a number of other folks going after different parts of this as well.

Aaron Lammer: Is Ujo defunct? You sounded like past tense when you said that.

Roneil Rumburg: I didn’t mean for it to come off that way. I know the folks there pretty well. I think ConsenSys is having a bit of issues that are well reported and well known, right? I think that’s making it a bit harder for them too.

Aaron Lammer: Yeah. There’s a whole nother story that I’d love to get into with you about what happens when these sort of nascent businesses are exposed to the volatile market dynamics of something like Ethereum. I wonder how much of Ujo’s fate was sealed by the price of Ethereum more so than their own product. But for you in doing Audius, give me… You’re sitting next to someone on the plane, but maybe that person has a little bit of Bitcoin. How do you describe what you do?

Roneil Rumburg: Yeah, sure. So we help artists connect directly with their fans and distribute music to them. It just so happens that crypto is the best way for us to do that, but we really see ourselves solving this problem that right now content creators don’t have a way to build a direct relationship with their fans, own that relationship, and distribute whatever they want on whatever terms they want.

Roneil Rumburg: So what that looks like more in practice, the Audius product looks and feels very similar to SoundCloud or Spotify or other music players that you’re familiar with, but it’s running on fully decentralized rails. So what we’ve created is a community of artists, listeners, and service and infrastructure providers working together to provide this sort of traditional end user experience that people have come to expect. Yeah, that’s sort of it in a nutshell.

Aaron Lammer: Okay. So I’m a Spotify customer. I’m not sure I’m proud of it, but feels like a necessary evil for someone who listens to a lot of music right now. I have to say for all of my quibbles, I’m reasonably happy with their product, and I also, I guess I’m on both ends of the Spotify equation, because I’m both submitting to the Spotify content library, and I’m simultaneously a Spotify customer. So I guess I’m curious from both of those perspectives, from the creator perspective and from the consumer perspective, how does someone interface with Audius? How do I upload music, and then as a consumer, what does it mean to be a customer of that uploaded music?

Roneil Rumburg: So we see ourselves as being much more similar to SoundCloud, I would say, than Spotify.

Aaron Lammer: Okay.

Roneil Rumburg: So taking a step back, Spotify doesn’t let artists directly upload anything. They upload via distributors who then put it on this Spotify.

Aaron Lammer: I use TuneCore. TuneCore is my service provider of choice, or not even choice. I feel like at the time I started using TuneCore, I’m not sure I was aware that there were TuneCore competitors even.

Roneil Rumburg: Mm-hmm (affirmative). Yeah, there’s a whole ecosystem of these. There’s TuneCore, DistroKid, many others. Spotify actually owns a piece of DistroKid as well, but anyhow, I would almost compare it to Netflix versus YouTube. Netflix is professional content distributed via these sort of well-established rails, whereas YouTube is like anyone can show up, upload whatever they want. That’s the kind of divide that’s emerged on the audio DSP side as well, but much of the industry, driven by Spotify largely, is shifting towards the more professional model. There’s not really a space anymore for folks who want to either do things like share half-finished things, get feedback from their fans and from the greater artists and music community.

Roneil Rumburg: SoundCloud used to fulfill that role, to some extent, but they’ve fallen on some hard times, and happy to get into that more as well. But we feel like by doing this in a decentralized way, we can create a permanent home for this type of music that can’t go away or fall prey to business model issues in the way that SoundCloud did, and copyright issues and many other things. That’s how I would kind of characterize it.

Aaron Lammer: If I upload a song, someone can say, “Go on the web and listen to it.” This isn’t going to require you run an Ethereum node in order to be able to listen to a song or something like that.

Roneil Rumburg: That’s exactly right. We feel like so many crypto products today are really just, frankly, hard to use. It shouldn’t take installing a chrome extension, getting a wallet seed, writing it down, putting it in your safety deposit box, and then signing up for a service to be able to play a song, right?

Aaron Lammer: Yeah.

Roneil Rumburg: That’s a lot of the work that we had to do was really innovating on that onboarding experience and then just the general user experience of using Audius. So there are a number of things that we’ve kind of done around that to enable us to have users sign up with just a username and password. Using that, we essentially secure a wallet that we generate on the client’s machine and save an encrypted version of that to a server we operate, so that if you sign in on other devices, your device can pull down that encrypted wallet, decrypt it locally, and then access the same account.

Roneil Rumburg: So from a user perspective, Audius looks and feels just like any other music player. The difference here from that consumer side is that you get to see a catalog and have a kind of interaction model that you don’t get with existing products, just because… You sharing your podcast on Spotify, do you have any insight or relationship with the folks who are listening to your content?

Aaron Lammer: No. I would sort of argue that the content, the audio content that I have created for the web lives in three different shades of gray, right? So there’s the totally, totally free podcasts, right? I distribute podcasts through Apple. Similarly, it’s not actually directly through Apple. I use Simplecast, which is a podcast host, which creates an RSS feed. Apple ingests that RSS feed. But the resulting product is a free basically unrestricted piece of audio. Then that content is mirrored on Spotify. That’s simply, for me, a choice that I don’t really want to not be there where anyone might want to listen. If somehow Spotify does win this podcast war, I’m like, “All right, fine. As long as you listen to my shows, that’s what my business is is you listening to my shows.” As far as I understand, our analytics are pinged by those Spotify plays, but you’re correct in that we have none of the insights that you get in some of these more open platforms.

Aaron Lammer: Then the third realm of content I’m involved in is music. I have recordings of my own music. I write songs with a band called Francis and the Lights. That music is purely on Spotify and is not available on the open web. Actually, that’s not really true. I think there’s YouTube uploads of most of it. But for argument’s sake, let’s say that that music is paywalled and it’s professionally published, it’s owned by multiple people, there are various legal structures that already dictate its use.

Aaron Lammer: So my understanding is that you are sort of focusing on those first two use cases, the totally free open-market web and people who want to supplement their open web presence with distributing content to multiple outlets. If I had music on Audius, that doesn’t mean I can’t have it anywhere else, right?

Roneil Rumburg: Yeah, that’s exactly right.

Aaron Lammer: Got it.

Roneil Rumburg: To your point, I think that’s really the point here, right? Is music and audio content should be openly addressable and openly shared, regardless of whether you choose to monetize it or not. There should be open mechanisms for it to be ingested and consumed by other applications, right? It’s kind of a question of should the DSP, if you will, the digital service provider in Audio parlance is basically a Spotify or SoundCloud or something, should that be vertically integrated, or should it be in pieces, if you will, right? Should the distribution layer be separate from the consumption layer?

Aaron Lammer: My distribution layer is actually TuneCore, which is interesting, because you get to this screen in TuneCore, and you could actually tell me that Audius was already on this list of outlets, because I don’t ever look at it. I click the box that says, “Distribute in all stores.” Right?

Roneil Rumburg: Mm-hmm (affirmative).

Aaron Lammer: For a penny and for a dollar, you know what I mean? It’s the same price, and you do get weird royalties from Deezer Australia, three cents. You know what I mean? So it’s kind of like why shoot yourself in the foot or put yourself out of those catalogs? But the interesting things to me is always that none of the free catalogs are serviced by TuneCore. So you can populate quickly in the paywalled web while there’s no way to populate into places like SoundCloud, except manually.

Roneil Rumburg: Yeah, I think that’s true, although some distributors do let you upload to SoundCloud or YouTube or places like that.

Aaron Lammer: Yeah. I think what I actually said is actually dated now, because there’s also that YouTube Music service and I think even if you do it, it still appears in regular YouTube, but with an ad or something. Everyone’s encroached on this game, and it’s increasingly like the same catalog, just mirrored at many, many different players.

Roneil Rumburg: That’s exactly it. That’s really what motivated us to build Audius as well was to try to recreate that user generated catalog that used to be on SoundCloud but has now kind of, for the most part, gone away, right? There were so many artists who used to upload half-baked, half-finished things to get feedback, to hear what people think of it, to learn what to tweak and change. There’s not really a place for that anymore.

Aaron Lammer: Well, there’s this interesting way that a lot of modern music works. I will use Old Town Road as the example, as it is the, I’m assuming, still the number one song in the world as we speak.

Roneil Rumburg: Yep. It is, as of this week. It once again topped the Hot 100.

Aaron Lammer: So Old Town Road by Lil Nas X is recorded over a beat he found on YouTube, which is a common way that kids make music today. I think most people over here will be familiar with that it’s not actually that hard to rip audio out of YouTube, and a lot of the people who are uploading beats will include a link for the download and want you to license it, whatever. I don’t know exactly how Lil Nas X found the beat, but the beat has a sample of Nine Inch Nails in it. I can almost guarantee that that sample was not cleared when the song was uploaded to SoundCloud. That’s that greater level of artistic freedom that is available when you’re kind of semi gray area distributing things not for profit. There’s a tremendous amount of SoundCloud material that has uncleared samples in it.

Aaron Lammer: Then when the song actually gets that critical velocity, well, at that point it’s worth going to Nine Inch Nails, and clearing the sample, and doing the official release. Then it eventually appears in Spotify. But the result, at least in my opinion, and this is, I guess, one of the reasons I’m, this being a crypto show, bullish on Audius, is I think that there’s a tremendous power in being the place where the stuff arrives first, not where it’s later polished and put into the paid system.

Roneil Rumburg: Yeah. I couldn’t agree more with that. The case of Old Town Road is a fascinating one too, because I think this is pretty reflective of how so many artists in the last few years have come up. Are you familiar with Juice WRLD at all, the rapper?

Aaron Lammer: I am. I am familiar with Juice WRLD.

Roneil Rumburg: Yeah. So his biggest track, arguably, Lucid Dreams had this Sting guitar riff in it. If you read into the story, it’s pretty interesting, because it wasn’t actually a sample. It’s unclear whether how that guitar riff made it into this track. It’s kind of become this he said, she said thing. But now Sting captures something like over 80% of the revenue that that track generates. That was largely because this existing copyright regime, if you will, enables folks to monetize that type of derivative content in a way that’s somewhat extortionate, we would argue at least.

Roneil Rumburg: With Audius, one of the things we’re really excited about is what comes with artist freedom and artist sovereignty is giving them the freedom to share and upload whatever they want, but to potentially own the responsibility for that too, right? We feel that as a company or operator, if you will, you should not have to dictate what types of content do and don’t live there, which is what happens with Spotify and SoundCloud and YouTube today. But the law forces that upon them, right?

Roneil Rumburg: With Audius, we don’t operate this. We’re putting out tools and giving those away to the community to share what they want to share on their own terms, and it’s up to them to make decisions around what they want to share and how they want to share it.

Aaron Lammer: In terms of applying the sort of anti-censor ability of Bitcoin metaphorically to music, you actually do have these real life conflicts, like at one point, Spotify, I believe, was going to pull XXXTentacion’s songs not from the service, but out of bounds for any Spotify-generated playlists, which you could say, “Who cares? They’re a business. They can do whatever they want.” The other side of that going would be looking into the future of Spotify, the future of what music will succeed economically within Spotify is going to be almost entirely decided by Spotify playlists.

Aaron Lammer: So some of the same anti-censorship, malleability of the catalog issues are sort of echoed between music and because, which is basically this is the most important point, in my view, of the design of Bitcoin is there’s no wheeling back the chain. No one can edit it. You can’t really ban it. It just continues to exist as its own autonomous organism.

Roneil Rumburg: We’ve really been inspired by that kind of ethos, if you will, that philosophy with Audius too. That’s how we’ve designed the protocol, such that our company could go away. We can turn off our AWS survey. We can do whatever we want, and this thing will still continue to live on, so long as there are artists who want to share content on it, right? It’s a really fascinating set of issues to unpack around how music is shared today and curated and all of that, because I agree with you, right? I don’t think Spotify should be in a position to get to dictate what gets featured and what doesn’t. This is just payola all over again, right? We’re replacing one cartel with a new one. Yes, it’s better in some ways, but still the same in most of them, we would argue at least, right? If you look at where all of the revenue in music is going, when you look at who controls discovery, all of these things, it looks very centralized. We’re going to kind of repeat the same mistakes that history taught us or should have taught us not to, right?

Aaron Lammer: So in thinking about the arc of a project like this, and this is always interesting to me in any sort of a project that has a token or has a workforce of people working for you, you’re burning money. People work for you for Audius. What’s kind of a timeline for something like this to succeed? How quickly does something like this need to take off? How long can it run as a niche blockchain, this is cool in that world? Do you want it to break into the mainstream? Do you see it as following crypto itself as mainstream’s adoption?

Roneil Rumburg: We see music being almost the most mainstream use case you could go after. So yeah, we definitely want Audius to break into the mainstream. The way we see ourselves doing that is by having a catalog that’s differentiated. If we can distribute and give users access to music that they can’t get access to anywhere else, we think that’ll be a compelling reason to come try out Audius and check things out.

Roneil Rumburg: So really the types of artists we’ve been going after are hiphop folks and EDM folks for the most part where, as you brought up earlier, they most acutely face these types of problems around sample clearance, around getting featured, getting discovered, all of those types of things. But yeah, I think it’s really important to figure out how, as a crypto community in general, how we can make these products approachable and usable for folks who are non-technical, folks who don’t already have a bunch of Bitcoin or have a bunch of Ethereum or what have you.

Aaron Lammer: Tell me about your own crypto story. How did you get interested in this stuff? Is this your first business foray, or do you have other projects behind you?

Roneil Rumburg: So I got into because in 2012, 2013. I was studying distributed systems within computer science at Stanford, and I really was just fascinated by this idea of having this uncensorable thing that could be worked on and coordinated on by anyone, right? It was permissionless. I think that combination of things was really powerful and interesting to me.

Roneil Rumburg: So I started mining at that time in my dorm room and-

Aaron Lammer: How much of the mining hash of 2012 do we think was being operated from Stanford dorms? It seems like a significant amount of mining was happening on the Stanford Campus.

Roneil Rumburg: That’s a good question. My co-founder, Forrest and I, on Audius now, we actually got started mining, and I know of at lest 10 or 15 other people at that time who were mining as well. So we figured out there was a way to buy… At that time, it was the AMD 7950, or I guess it was ATI at that time. That GPU was kind of the dominant force in alt coin mining. That’s what we were mining. So we were mining Dogecoin, Litecoin, and other scrypt-based alt coins. So we found a way to buy those in Paris and get this vat refund. We could buy them for about half the price we could in the US.

Roneil Rumburg: Forrest was really fortunate that his dad’s a pilot, so he could get free flights to and from Europe, and he would go and bring back some, and we could use those to mine. But yeah, we used to trip the circuit breakers in our house frequently. We actually ended up finding a way to sort of pick the lock for the circuit breaker so we could go turn them on ourselves. There was a very substantial amount of… If you put together the 10 to 20 of us who were doing this, we were probably consuming 30, 40, 50,000 watts of electricity somewhere thereabouts at peak. But I think that was still pretty small.

Aaron Lammer: I’m very interested in the history of crypto culture. It’s probably as interesting to me as the technology itself. What was the Stanford Bitcoin scene circa 2012 like? What were people talking about? You’re mining stuff like Doge and Litecoin. Where did people think things were going then?

Roneil Rumburg: Yeah, it’s interesting to think about. This was seven years ago, so it’s been a long time.

Aaron Lammer: A lifetime.

Roneil Rumburg: Yeah.

Aaron Lammer: That’s 100 years in crypto.

Roneil Rumburg: It really is. We were all just really… We just thought this technology was cool. To be honest, in late 2013, early 2014, a bunch of us sold most of our crypto to fund a spring break vacation. It was not like a-

Aaron Lammer: Have you calculated the present day price of that vacation?

Roneil Rumburg: We have.

Aaron Lammer: It’s probably like a… Okay, I’m going to predict it’s at least seven figures, maybe eight figures.

Roneil Rumburg: Yeah. I think it was a seven figure amount. So one of the most expensive, if you look at the quality of our vacation, certainly it was not a seven figure kind of vacation. But anyhow-

Aaron Lammer: Where did you go?

Roneil Rumburg: We went to Honduras.

Aaron Lammer: Wow, you guys spent-

Roneil Rumburg: We did some scuba diving.

Aaron Lammer: … over one million dollars to go to Honduras.

Roneil Rumburg: Yep.

Aaron Lammer: If you would have just held, you probably could have bought an island in Honduras for that much.

Roneil Rumburg: Yeah, but we were 20, 21 years old. We were like, “Hey, let’s… We never had money before. Now we have a little bit of money. Let’s go spend it and hang out.” So that’s what we did. I think most people at that time just thought the technology was cool but didn’t see this as a thing that could financially change their lives, right? We all just thought it was really cool. That’s actually what led to me working on my first company straight after school. It was a Bitcoin peer-to-peer payment product. We described it as Venmo for Bitcoin. So you could type in someone’s phone number, someone’s name, and pay them some Bitcoin and vice versa. You could receive Bitcoin via your social presence.

Roneil Rumburg: Again, it was the same kind of thing. We just thought because was really cool and we wanted to start paying each other for things with Bitcoin. So we built this thing to try to solve that use case. Turned out no one else wanted it, but it was a good learning experience.

Aaron Lammer: So that didn’t take off at all? That seems useful.

Roneil Rumburg: We got to 25,000 monthly active users at our peak.

Aaron Lammer: That might have been a lot of the people who were actually moving any Bitcoin. It’s easy to forget. We think of the total number of people who own Bitcoin, but almost everyone is just not doing anything with their Bitcoin.

Roneil Rumburg: Yeah, and that’s kind of what we realized at that time too. So this was called the Backslash. It was just a Bitcoin walled product, right? What we realized when we were talking to all of the folks who signed up and were using it, no one wanted to send their Bitcoin to anyone. They were trying to kind of stacking Sats, right? They were trying to just accumulate it to the extent that they could. So it was just like, well, no one really actually cared about that use case, but we were spending a lot of time with the team at Coinbase at that time. So we were going to do this kind of white label thing with them. I think they ended up killing that product [inaudible 00:27:07] afterwards. But they were in the hundreds of thousands of active users at the time. That’s what made us realize, “Okay, if a company with this level of resources and this many people working on it can only get to that kind of usage, there’s not really enough here for a type of product like what we were doing.”

Aaron Lammer: It’s funny, because it’s actually pretty right on. It’s not a bad idea, compared to people who are like, “I have an alt coin that’s going to be used to move energy off the grid.” It’s like, “Come on. Come on.” This is a practical idea with a real user experience. But I have one suspicion about why people don’t do this.

Aaron Lammer: When I imagine myself on your trip, let’s say, to Honduras, everyone’s settling up. I’d be like, “If you want to pay me in Sats, go for it. Great. In fact, I’d love if everyone wants to pay me back in Bitcoin, that would be awesome. I love to get a little Bitcoin that I didn’t have before.”

Aaron Lammer: If you’re asking me do I want to pay you back and you give me the option between actual Venmo and Bitcoin, I’m like, “Ah, I don’t really want to give up any of my Bitcoin.” There’s that weird… You know in a board game people don’t want to do a fair trade? They only want to do a trade if it’s slightly in their advantage?

Roneil Rumburg: Yep.

Aaron Lammer: The trade of sending someone Bitcoin at least feels like at best a fair trade or a slightly worse trade for you, because you always have the feeling that Bitcoin is about to go up at least a little bit.

Roneil Rumburg: That’s absolutely right. We saw that behavior emerge even amongst ourselves too. We were all very reluctant to send Bitcoin, but always wanted to receive it. Yeah.

Aaron Lammer: There’s something to be said about just humans. A lot of Bitcoin’s game theory has taught me not so much about Bitcoin itself, but just about how people hoard their pot of gold and get very irrational as they wrap their arms around it and look suspiciously around.

Roneil Rumburg: That’s absolutely right. The only times I think we were actively sending Bitcoin to each other was when we had no money, no US dollars to send. It was like, “Oh, well.” Near the end of that Honduras trip, we were kind of running out of money. We had booked the flights back and stuff, so that was all okay, but there was nothing else to send, right? Then we were forced to dip into that.

Aaron Lammer: Without prying into your current financial situation, which is your own business, does it ever bother you… I assume if you tell people you were into Bitcoin in 2012 and you were mining, they assume that you’re a mega whale, super, super rich, but actually you put a lot of that money into going to Honduras and living your life and starting companies and stuff like that. Does that disconnect ever come into play in your relationship with the company or Bitcoin or other people?

Roneil Rumburg: I don’t think so. Most of the folks I know in the crypto community have similar kinds of stories, whether it was Mt. Gox or whether it was I was-

Aaron Lammer: Sure, a lot of ways to lose your Bitcoin.

Roneil Rumburg: Yeah, like, “I was CPU mining at home and then I forgot about my Bitcoin, and that was in 2010. Now that would be worth $10 million,” or whatever.

Aaron Lammer: Yeah. Bitcoin is a vessel for tragedy. Eventually all the stories, if they go on long enough, are going to end in tragedy. There’s going to be people who are in three generations, and they’re like, “I’ve still got my grandfather’s Bitcoin… Oh no, I lost it.”

Roneil Rumburg: Yeah, yeah.

Aaron Lammer: There’s going to be these stories about lost and found fortunes are just going to… As the time period stretches longer and longer, they’re just going to be crazier and crazier.

Roneil Rumburg: Totally. Yeah, and I think a funny one that happened to a bunch of us a year or two ago probably was when Stellar launched, they had this [inaudible 00:31:08] with Facebook and you would get an airdrop of some Stellar, right? So a bunch of us did it and then promptly forgot about it, right? I noticed during that run up in late 2017 like, “Oh, Stellar is doing pretty well. I wonder how much of Stellar is in that old wallet.” Turned out to be $7 or 8,000 I was like, “Oh, that’s kind of a nice little gift to just sort of discover randomly.

Roneil Rumburg: But yeah, to your earlier point, I think I’ve always kind of taken a zen approach to these things, if you will. I feel like I’ve got a roof over my head, I’ve got food in my stomach, I can travel where I need to travel and live by all intents and purposes a great life. I’m not too concerned about what could have been, should have been, whatever, when it comes to our historical experience in crypto. I just feel very fortunate to even get to work in this time and in this industry, right? I think we’re in very interesting times in general.

Aaron Lammer: How many years into Audius are you now?

Roneil Rumburg: We’re just over a year in. We got started early last year, like March of 2018. So we’re 12 people now kind of doing our thing. We were fortunate to be able to raise some capital soon after starting, and things have been really good. We’re launching a public beta in September. So just been building up our artist board and other things in preparation for that.

Aaron Lammer: Did you raise money by selling a token or something or just by more of a VC kind of structure?

Roneil Rumburg: We followed a fairly traditional venture style structure. There was essentially a corporate entity that owns a pool of tokens. At some point in the future, our company will either distribute those tokens and cease to exist, or some other thing will happen around that. We felt like that gave us a lot of flexibility to iterate on both the token model and everything else without having to do so with public scrutiny. But we’re pretty excited now about being able to open up what we’ve been working on and starting to actually go after larger numbers of users. I feel like we’ve learned what we can in this private testing period.

Aaron Lammer: So you’re venture-backed, which means the money you have is not from some shit-coin trader who pumped your ICO. But you do have a token that is going to be used within Audius? Or is it actually two tokens?

Roneil Rumburg: Yes, it’s two tokens, although I should clarify, one of those is our native token or we call it kind of our governance token, follows a traditional work token style model with a staking system around it. Then the actual transactional economy would be conducted in third-party stable tokens. So we feel like we’d support any stable token that an artist wants to be able to receive, but we didn’t feel it would make sense for us to try to build our own transactional token.

Roneil Rumburg: There are a lot of folks working on value stable, low volatility, whatever you want to call it, mediums of exchange that could apply really well in this use case.

Aaron Lammer: Are you open to Libra?

Roneil Rumburg: Yeah. We actually think Libra’s going to be net good for this whole crypto ecosystem. There’s a lot to unpack around that too, but we’re actually pretty excited by it.

Aaron Lammer: I’m actually I’m a contrarian pro Libra thinker also, or at least I take it seriously. I don’t think it’s my place to be for or against it, but I don’t think it’s irrelevant or a joke.

Roneil Rumburg: Yeah. This is two billion people, right?

Aaron Lammer: Yeah.

Roneil Rumburg: You can just disregard that. Not only that, but I think if Facebook can do this right or if let’s call the Libra Foundation can do this right, they can actually enable interesting use cases around censorship resistance and other things. It’s just too early to say. But I do think there’s an opportunity there that could be really exciting. We would be more than open to accepting Libra as well. We kind of wanted to be agnostic, right? To any of these transactional mediums.

Aaron Lammer: Is Bitcoin an option also?

Roneil Rumburg: Yeah. We certainly could support it. I think it would be wrapped Bitcoin of some kind, right?

Aaron Lammer: Right, right, right.

Roneil Rumburg: I think the scripting system in Bitcoin is a bit too limiting and whatnot for that to apply directly. But yeah, to your point, I think it’s kind of people should be able to pay in whatever currency they want, and people should be able to be paid out in whatever currency they want. You can use 0x relayers and other things in the middle to make that work, right? If I want to pay in Dai and you want to be paid out in Libra, there should be a permissionless way to atomically swap and have you be paid out in a single transaction.

Aaron Lammer: Okay, I’ve got a question. So my friend Ledger Status, who comes on this show a lot, he’s a crypto Twitter trader kind of guy, but really great guy, really smart guy. He started playing this crypto game called Hero in beta. You’re basically shorting or longing each candle using this game token. He was playing it really early and basically built up to where he’s one of the biggest holders of that token.

Aaron Lammer: So I’m curious for artists who come into you system early, is there a first actor advantage insofar as the catalog won’t be very big and there will be some sort of amount of free-floating token that tip heavily towards people who go in on it early?

Roneil Rumburg: There will be, yeah.

Aaron Lammer: I’m not asking for myself, but maybe I am asking for myself.

Roneil Rumburg: There certainly will be, and we can talk about-

Aaron Lammer: I’m not good at shorting and longing Bitcoin. This is closer to my skillset.

Roneil Rumburg: Oh, we would love to have you in Audius. I’ll follow up after this with an invite and everything. But yeah, I think we certainly want to incentivize folks to come in early. If you look at how previous content platforms have played out, it was the folks who were there earliest and understood how to build a following there, how to engage with users and all of that that ended up doing the best on those platforms. So there’s that kind of soft incentive.

Roneil Rumburg: But the harder incentives we want to create are around we’re going to have a bunch of incentive programs to directly incentivize early artists, everything from, say, the top 10 artists being listen to any given week or within any given genre, whatever could get paid out bonuses and other things. We’re still figuring out from a legal standpoint how best to execute these types of things. So yeah, we haven’t published a lot of detail yet around plans there, but to your point, yeah, we would definitely want anyone to feel incentivized to share content here, interact with their listeners, all of that, right?

Roneil Rumburg: I think that’s one of the most powerful things about these crypto protocols that you have essentially this currency flowing around within your system. You can use that to incentivize behavior, right? If you think about Dropbox’s signup model, for example, if you referred five friends, you’d get 100 megabytes more of space or things like that. I think the really cool opportunity is to explore those direct incentivization models around crypto economic networks as well.

Aaron Lammer: Do you foresee that this token, not the necessarily what’s being paid for, the music, but the governance token, is that something that will trade on exchanges at some point?

Roneil Rumburg: At some point, it likely will. That’s not something that our company will play a role in.

Aaron Lammer: Got it.

Roneil Rumburg: It will need to be traded at some point in order for the service provider economy to work, but from our perspective, the longer we can not have it be traded, not have to worry about price and all these other things, the more time we have to build long-term value here, the better.

Aaron Lammer: It seems like there’s a divide in most… I come from a product design user experience background, and there’s basically these, “Pay us a flat rate subscription, and then past that, what you do is basically irrelevant. We’re just taking the money.” Then sort of direct sales, “Let me sell you an album.” Bandcamp, basically kind of on a software, like pay-for-download model basically. Are you on either side of that user experience, or is your user experience something different than either of those things?

Roneil Rumburg: Yeah, I would say we’re on both sides of that, in that we want to enable both types of models, and artists can choose whether they want to participate in one or the other or both. But from the protocol standpoint, we essentially support two ways of getting artists paid. One is every time you play a song, your client is making a small payment to the artist. The other is you pay a flat rate subscription, you get all you can eat listening. However, your listening behavior is logged on chain in a way that payments out can be done transparently, right? Versus the existing subscription models are typically these kind of black box things.

Roneil Rumburg: So on the pay-per-stream model, we actually envision there being a sort of free market around this. As an artist, you could say, “I want to get paid X or Y per stream.” If you’re below a certain threshold, maybe the client doesn’t prompt a listener before they click play to opt into the payment, but if you say, “Hey, I want $10 every time someone plays my track, maybe the listener does get prompted and the listener could choose.”

Roneil Rumburg: So the way we really see this, with every product decision we make, we make it from the perspective of enabling artist choice. We want to give artists the power to make these decisions rather than forcing those decisions upon them.

Aaron Lammer: That makes sense. There may be emerging markets that don’t even exist yet, like selling high quality stems of songs.

Roneil Rumburg: Totally.

Aaron Lammer: Track separated. There’s already sort of an emerging economy in audio quality, but there’s so many more dimensions of distribution. I also kind of imagine something I’m surprised there isn’t more of a market for sometimes is the flat rate $20, everything forever. Everything that I’ve ever put out and everything going into the future. You know what I mean?

Roneil Rumburg: Mm-hmm (affirmative).

Aaron Lammer: Basically like a pass for an artist. Because the models that people are publishing on now are so rapid that I think something like Patreon I think actually more closely follows the fan experience for a lot of people.

Roneil Rumburg: I couldn’t agree more. What if that artist pass gave you other benefits? Like what if you could participate in an early feedback system with the artist, right? Maybe if you pay $100 or $200, you get early access to content and get to give feedback, or you get backstage passes when the artist visits your city. I think there are so many different avenues for monetization that artists haven’t been able to explore today, just because they don’t own that relationship.

Roneil Rumburg: If you in your band, for example, wanted to say, “Who are all the people who listen to us and who are the people who listen to us most?” You couldn’t answer that question, I imagine, given I don’t know what-

Aaron Lammer: I actually think you can look at that in Spotify, weirdly. Spotify, for all of its shortcomings, Spotify does do some somewhat interesting tracking that I’m sure is somewhat invasive of listeners, but you can see it by city, what cities do you over perform in? You can look at times of day.

Roneil Rumburg: But you can’t look at specific people, right? What if you wanted to give an offer to your super fan?

Aaron Lammer: There is some drilled down… I remember Chance did a show where the audience was made up of the number one SoundCloud listeners maybe. But anyway, I always find the most interesting metric on Spotify is just when you can see someone’s rank in the world. That’s what’s pretty wild is that they’re not only really exposing your analytics, but they’re publicly exposing what artists you do or do not out-perform analytically.

Roneil Rumburg: Yeah, and it’s pretty fascinating. Then you can almost gamify those things too, right? If you-

Aaron Lammer: Well, they are. I think they are pretty heavily gamified.

Roneil Rumburg: Yeah, yeah, yeah.

Aaron Lammer: What would be interesting would be from a data science perspective, to see a graph of artists ranked, including plays generated directly from Spotify playlists, and then totally exclude playlisting and re-rank. But then again, the people who are heavily featured in playlists, you get tail-off subscription, whatever, it’s probably actually similar. But it is hard to tell how much of that is sort of internal algorithm, understanding of the artists in Spotify versus purely organic plays.

Roneil Rumburg: Mm-hmm (affirmative). Even being able to map from being featured in a playlist to what outcome that has for the track, it’s pretty hard to analyze these things today. One of the other frustrating things there I think is that a lot of those playlists kind of have this payola style model where either artists or labels are paying the curators of given playlists to feature their track, and that’s because the folks who operate playlists don’t actually get to monetize the fact that they curate content into this thing and people listen to it.

Roneil Rumburg: So that’s actually one of the other areas that is very near and dear to our hearts at Audius. We feel like if you command an audience in whatever way you do, you should have the opportunity to monetize that. So if you create a playlist in Audius, people are listening to content via your playlist, you actually earn a rev share on any of the listening that you’re enabling.

Aaron Lammer: For a minute, I thought you were going to go like full crypto and be like, “We wanted to create a more friction-less market for payola. So we’re having our own payola token.”

Roneil Rumburg: Yeah, I think it should just be the best content wins. If someone wants to get paid for placements on their playlist, they could do that, but listeners should naturally gravitate to the highest quality curation. I think it’s just unfortunate that the situation we have today is people are forced to do this payola style thing because they can’t make a living otherwise.

Aaron Lammer: Yeah. If there’s one universal consistent in the record industry that extends from sheet music through future blockchain on the moon music industry, it’s payola.

Roneil Rumburg: Yep.

Aaron Lammer: Payola has never gone away. It’s always existed in different forms. People have always been like, “Finally this new medium will stomp out payola forever,” and it always comes back. It’s always quasi-legal. They did ban payola in the ’80s, but now they have modern legal payola.

Roneil Rumburg: Yeah, and I think in the ’80s too a lot of it was organized crime related. So I think part of the reason it was banned from a radio context was to hurt those folks, not necessarily because what was happening was illegal. Don’t get me wrong, I’m not trying to make a moral argument for why we shouldn’t be doing payola. I do think you should be able to see if payola is happening. There’s a utility there. There’s a reason why this happened, because it’s just impossible to get heard between the noise. I think where that becomes problematic is when the only way to get heard amongst the noise is to participate in that.

Aaron Lammer: If your project succeeds beyond wildest belief, this is a thing we like to do on this show, which is basically just like when someone spins a crypto narrative, you’re like, “Okay, what if that actually happens?” In the sense of Bitcoin maximalism, we’re living in a scorched earth and all eating steak all the time. In the case of Audius, if this really succeeds, it becomes the Bitcoin to Spotify’s Libra. Then it becomes a battle of is switching from that to a harder version worth it for the benefits? To me in Bitcoin, it is. I think Bitcoin is better than Libra. I think that when people start thinking about Libra, it exposes lots of the qualities that Bitcoin has. Do you think that those qualities are important enough in music to drive people from a centralized to a decentralized alternative?

Roneil Rumburg: I do, because there are just so many types of content that you can’t really listen to easily today. There are so many times I’ve seen my favorite artist, for example, getting their content pulled, getting de-platformed, getting demonetized, and there are many different reasons why that happens. So if, to your hypothetical there, in 10 years, if Audius is as successful as we’d like it to be or hope it could possibly be, I think what this looks like is a community-owned co-op where artists, listeners, these infrastructure service providers can all collaborate and work together on this shared thing, which is the world’s music. If there’s an incentive amongst this community to curate the world’s catalog of music, I could actually foresee a future where, for example, Spotify pulls content from Audius as well. This is not like a closed-off, vertically integrated thing. This is an open system for anyone to publish content and for anyone else to consume that published content.

Roneil Rumburg: We have a [dap 00:50:21] that does that today, but our hope would be that anyone could and many other people will go build user experiences on top of it, targeted at specific use cases or at specific types of geographies, let’s say, right? The Chinese user might have different desires and/or expectations than the western American or western European type of user, right? So that’s what gets us really excited about this. If we’ve done our jobs right, our company should be able to disappear, go off into the wind. I think that would actually be the best thing for this protocol if there were no authoritative central party that was driving the product development, if this could be fully community-driven. So yeah, that’s the hope, man.

Aaron Lammer: So is there the possibility for an incentive for people to build clients and things like that? Like I build a killer iPhone client, is there a token incentive in the way that there’s sort of these blockchain development fund for that kind of stuff?

Roneil Rumburg: So we’ve explored or thought about incentive grants. So those those governance tokens for folks who are building things that are valuable to the protocol. I think there are also other forms of incentives, which are… So I mentioned at the protocol level, we support those two methods of payment, right? The subscription model and this pay-per-stream. But at the client level, pay-per-stream doesn’t necessarily mean a listener spending money. The client could show a listener and ad every so often and make those on-chain payments on behalf of the listener where essentially you are earning a little bit of crypto for seeing an ad, your client spends down that crypto, and then you see another ad, and you’ll earn a little bit more. But the client could actually earn a spread on that, right? If their ad revenue is-

Aaron Lammer: This is starting to get dystopian.

Roneil Rumburg: How so?

Aaron Lammer: Any time people start… I don’t know. It kind of reminds me of those coin-based, “You’ll learn at Khan Academy and earn 0x by learning about 0x.” It just the ads for token to pay for the service just, I don’t know, it goes Black Mirror quickly for me. Maybe I’m just too squeamish.

Roneil Rumburg: I don’t know. We see it just as a simple way to get people onboarded. You mentioned that you pay for Spotify, right?

Aaron Lammer: Yeah.

Roneil Rumburg: Yeah, you probably started out listening to ads, and then you get annoyed, right? You’re like, “Screw this. I don’t want to hear ads when I’m trying to listen to music, so I’ll pay.” But that ad supported at the beginning is very important, right? Because if you just showed up and Spotify said, “Pay us $9.99 to start using this thing,” you’d be like, “Eh, I don’t really know how this is useful to me yet. Let’s try it out and see.”

Aaron Lammer: Were you a BitTorrent person?

Roneil Rumburg: I was.

Aaron Lammer: Was torrenting a part of your life? Okay. Were you a member of private music trackers at all?

Roneil Rumburg: Yeah, so I was part of Oink and What.cd, all of those. That was really what-

Aaron Lammer: Okay, we’re speaking the same language.

Roneil Rumburg: That was really what enabled me to discovery my love for music. That’s what was so exciting to Forrest and I about Audius too was that we had this chance to combine two of our biggest interests in a single project. But we’ve been inspired by a lot of what they did around… I don’t know if you remember the bounty economies that they had, right?

Aaron Lammer: Yes.

Roneil Rumburg: Yeah, it was just this fascinating model. Look, don’t get me wrong, Audius does not exist to enable piracy or promote piracy, but I think in that era, in the mid 2000s, that really served a utility for those of us who were 13, 14, 15 years old. We didn’t have a way to discover music. I couldn’t afford to go down to the record store and pay $25 for every album, so I found different ways to support my artists. I’d go to live shows and things like that, but-

Aaron Lammer: I was 25 at the time. I don’t have the same excuses. I just wanted it because it was the best music service ever built. The catalog of Oink at the time is the greatest collection, most perfectly labeled, most comprehensive, perfect quality, flack releases of everything. I’m glad to hear that that was inspiration for what you’re doing, because I have always thought that the end state of a crypto-based music service would look something like Oink. I don’t actually think piracy is a huge part of the Oink story.

Aaron Lammer: To me, when I think of it as a piece of system design, for starters, you’re given a little bit of seed to begin with. Then you need to upload as much as you download, which forces you to get creative and contribute to the community, either by uploading music that wasn’t previously available or just by serving as a seeder so other people can benefit. I think you could also earn credit for cleaning up listings and moderating forums and that kind of thing. The idea of a co-op is basically right. I wonder what ways other than watching ads people can perform… Playlisting is one of the ones you brought up. Services so that basically if you want to put in a little work, it’s free. Otherwise, a lot of people are going to unlock just like, “I don’t want to deal with it,” and just pay for it.

Aaron Lammer: The cool thing about Oink was that you couldn’t pay for it. You had to be a good seeder. You actually had to be a good member of the community. The only people who were active in the community really could take advantage of it. Otherwise, you got ratioed and they’d turn you off. I remember the day I got ratioed on Oink. I was like, “No.”

Roneil Rumburg: Yeah. Oh, I did too. I got warned many times as well. It was really hard, right? It was actually really hard to maintain a ration, because everyone was-

Aaron Lammer: Very hard.

Roneil Rumburg: … fighting so hard to maintain their ratio.

Aaron Lammer: I had bad internet at the time too, and some people were like, “Oh, we got a bunch of kids in Stanford dorms on T-lines here.” I’m on this crappy Time Warner cable, and they’re crushing me in the ratio. That’s basically mining. All of that seeding is basically competitive mining, because it’s just looking for the fastest nodes, isn’t it? I don’t actually really know how BitTorrent works, to be honest. But it always has been a pretty good metaphor for mining to me.

Roneil Rumburg: I think it really is. It’s philosophically aligned with what mining was meant to be early on, which was that anyone could show up and join it. Now mining has centralized, right? In a way that I think is somewhat antithetical to the ethos behind these systems. But anyhow, yeah, one of the other really cool things we’ve been able to explore is like with Audius, once you’ve listened to a track, that content’s now on your computer, right? So your computer can turn around and go distribute that, redistribute that to more people. That’s what enables us to be able to scale this in a way that wouldn’t otherwise be possible with decentralized infrastructure, right? In a way, the more people listen to a thing, the more replicated it becomes and the more available it becomes on IPFS, which is our content distribution layer that we use.

Aaron Lammer: So I think you’re about 10 years younger than me, I’m going to guess, or seven or eight years younger than me.

Roneil Rumburg: I’m 26.

Aaron Lammer: Okay. Make that 11 years younger than me. So when I was in college, taking it back, this is before even the Oink era, this was the post-Napster, everyone’s coming out with peer-to-peer clients era. A lot of these peer-to-peer clients would allow you to form a LAN, a local area network within the dorm and just grab any music from it. So the very first crazy free-for-alls in music I was exposed to, even before that Oink era, was people just had these network parties in dorms where there was terabytes of music moving around. It was super fast, because it wasn’t actually going anywhere.

Roneil Rumburg: That’s super cool. So I do remember when I started college, Spotify, etc, were not a thing in the states yet, right? I think Spotify had just started in northern Europe and whatnot. So what we did do was everyone would share their iTunes library.

Aaron Lammer: Yes.

Roneil Rumburg: Have you ever seen this on iTunes? You could essentially browse other people’s libraries on the local network?

Aaron Lammer: Oh yeah.

Roneil Rumburg: It was actually-

Aaron Lammer: That’s how these services worked. That’s how the thing in the dorm worked. The software… It wasn’t Kazaa. It was like a Kazaa-ish thing, but that got popular in dorms. It knew, because everyone was on an Apple computer, where to look for your iTunes library.

Roneil Rumburg: Oh very cool.

Aaron Lammer: It would automatically index your iTunes library into the local area network of everyone, and you couldn’t see whose computer it was on. It just had a search function and a filter function, and it was just a giant set of files sitting there on the local area network. A lot of security concerns there, but different times. Different times.

Roneil Rumburg: Oh, for sure.

Aaron Lammer: Hey, if you’re getting into peer-to-peer, hey, anything could happen.

Roneil Rumburg: Yeah. I think it’s really that ethos that we want to get back to. A lot of platforms today have kind of de-personalized music. There’s very little of a social layer remaining on Spotify at this point. SoundCloud did do pretty well with this, but for other reasons, I think they’ve been losing their audience. I would love to be able to see, for example, what you’re listening to and sharing and be able to discover it and be able to see who has similar music tastes to me and what are they listening to and why are they listening to it and all of that good stuff. It’s just frustrating, right? Yeah, I came up in the same era you did around that.

Aaron Lammer: Yeah. Well, is it possible for you to eventually ingest… For instance, every podcast I’ve ever done, I run three podcasts, the biggest one is long-form podcast, we’re over 300 episodes now. They’re all just sitting out there on the internet. They’re basically a giant RSS feed of mp3 files. Is it possible that something like Audius would eventually want to ingest all these freely available podcasts and then make them available in the system, and then hold whatever they earn in some sort of an escrow waiting for someone to sign in and claim them as their content?

Roneil Rumburg: Yeah, I think that’s very possible. I think that would be a great use case or model for this where it doesn’t have to be us either. Anyone in the community could decide, “Hey, I want more podcasts to be available on Audius. Let’s go find content and put it here.” So yeah, I think those kinds of models are really exciting. You can even create… You could imagine having a bounty-based system around what you described for cleaning up tags to make sure that the correct person is attributed or correct people are attributed for a given piece of content, such that perhaps anyone could upload a given thing, but the network could figure out who deserves to be paid out for it and gets them paid out. Then people could challenge that and whatnot as well.

Roneil Rumburg: But that gets into we’re building out this arbitration system within Audius to help kind of moderate content in a community-driven and transparent way to make sure that the right people are getting paid out for their content, right? If you go upload the Rolling Stones discography or something to Audius, there needs to be a way for the rights holders of that content to come on and say, “Hey, that’s actually ours, and we deserve to get paid for that, not Aaron.” Right?

Aaron Lammer: Right. There probably should be some sort of a safeguard in the system for gray area content, which is to say, “Hey, you uploaded this work that’s not yours. We’re not going to take it down.” It’s some sort of a collage or something, “But you can’t monetize it, because it’s someone else’s intellectual property.” There is that weird zone of a lot of YouTube stuff which is basically not-for-profit piracy that I feel like is generally tolerated on things that are old enough or outside of the purview of anyone looking.

Roneil Rumburg: Yeah. I think in that case, we actually would want to enable a rev share structure where-

Aaron Lammer: Ah, I see.

Roneil Rumburg: … right now, if you… Let’s say you go do what Kygo did recently. If you go sample Whitney Houston and make a track but you don’t clear it by her estate or what have you, there should be a mechanism for her estate to say, “Hey, Aaron’s track is derived from our track. We deserve a revenue share on it,” but have the community decide what would be a fair revenue share there. Right now, the ability to take down that content is used to extort kind of unreasonable rev shares, right? Like Sting getting over 80% of Lucid Dream’s revenue. That’s one of the things we’re really excited about is when you can resolve these disputes in an open, transparent, and cheap way, rather than by throwing a bunch of lawyers into a room and seeing what happens basically, which is how it’s done today, we think a lot of creativity can be unlocked.

Aaron Lammer: Excellent. Well, when can people expect Audius to be something they can play with and check out?

Roneil Rumburg: Yeah, so we’re going to be launching beta in the fall, so late September, early October timeframe. Yeah, follow us on Twitter, on whatever social platform you prefer. We’re on Instagram, Discord, etc, if you want to get involved, we’re always looking for artists to give us feedback. There are plenty of other ways to participate. We’re just really excited to get this out to the world. It’s been a lot of engineering effort to get to where we are today, but we’re here and we’re finally going to have our coming out party soon.

Aaron Lammer: Excellent. Well, thank you so much for doing this interview. This was really interesting. I look forward to taking it for a spin.

Coin Talk

About this PODCAST


Come ride the crypto rollercoaster with hosts Aaron Lammer and Jay Kang (and guests) as they laugh their way through the week in Bitcoin and beyond.

Come ride the crypto rollercoaster with hosts Aaron Lammer and Jay Kang (and guests) as they laugh their way through the week in Bitcoin and beyond.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade