by Y Combinator5/2/2018
Linda Xie is the cofounder of Scalar Capital, a cryptoasset management firm. Before that she was a Product Manager at Coinbase (YC S12).
Avichal Garg is a Managing Partner at Electric Capital, a digital asset management firm. He’s also an Expert at YC. Prior to that he was the Director of Product Management at Facebook.
Craig Cannon [00:00] – Hey, how’s it going? This is Craig Cannon and you’re listening to Y Combinator’s podcast. Today’s episode is with Linda Xie and Avichal Garg. Linda is the co-founder of Scalar Capital, which manages crypto assets, and before that, she was a Product Manager at Coinbase which was part of the Summer 2012 YC batch. Avichal, whose been on the podcast before, is a managing partner at Electric Capital, which is a digital asset management firm. He’s also an expert here at YC and prior to that he was the Director of Product Management at Facebook. All right, here we go. Let’s just start with a quick intro, so Linda, after you.
Linda Xie [00:36] – Hi, I’m Linda. I am co-founder of a crypto hedge fund called Scalar Capital. We focus on long-term investing in the space, with a strong emphasis on privacy coins, and then before that, I was a Product Manager at Coinbase. I joined pretty early on, and just been passionate about crypto currencies for a very long time.
Craig Cannon [00:52] – Cool.
Avichal Garg [00:54] – Hey, I’m Avichal, serial entrepreneur, part-time at YC. I’ve been crypto since 2013, actively invest in currencies and various companies. This is where I spend a lot of my time, thinking about the way things are going and what the future holds, and where we’ll end up in the next 10 years, because this is a pretty fascinating area.
Craig Cannon [01:14] – Cool. Well, it sounds like you guys are the right people, because we have a ton of crypto questions. Let’s start with definitions. I know a lot of the listeners probably know, or have a loose understanding of what blockchain, what Bitcoin, all of this stuff is, but just so we establish kind of a ground truth, Linda, why don’t we start with you. What is a blockchain?
Linda Xie [01:34] – A blockchain is essentially a decentralized public ledger, where you can have a recording of all the transactions that have happened on it, without having a centralized entity that’s kind of dictating what happens, or of someone that can manipulate the data. All of this is just done in a decentralized fashion, so you don’t have to trust who’s actually controlling this data.
Craig Cannon [01:51] – What are the other things that people ought to know?
Linda Xie [01:54] – It’s important to separate a cryptocurrency from new coins that have come out that are essentially, people refer to them as crypto assets, because it’s essentially more than just money at this point. Ethereum’s a smart contracts platform, and smart contracts are essentially, you can think of it just like a contract that is essentially just programmable logic, where you essentially have a decentralized network of computers that’s executing this logic, without having to rely on a centralized source that can get shut down, or manipulate the information or data. A smart contract, in itself, is just really important, and you can kind of, if you want to compare it to Bitcoin, just think of it as a bit more powerful than Bitcoin, in that you can be a little bit more programmable.
Craig Cannon [02:34] – Anything to add?
Avichal Garg [02:35] – I agree with everything Linda said so far. The way I think about it is like, there’s four core concepts. There’s blockchain, which I think Linda covered well as distributed ledger, and it has a bunch of properties around who can read it, and how you write to the ledger, and so on. There’s this notion of consensus, which is how do we all agree on what gets written to the blockchain, and what universal truth is, and there are different mechanisms to agree on the truth. There’s the idea of a token, or token economics, which is how do we align the incentives of all the actors in the space, and there are different mechanisms to aligning the incentives, and then there’s the smart contracting layer, which is assuming you have a ledger, and you have tokens, and people who are agreeing on the state of the universe, what degree of control and programmability do we give on top of that to the users of the blockchain or the programmers to do interesting things on top of that, which is where things like programmable money, or distributed compute get really interesting. Those are kind of like the four things that I think about in this universe, and you can kind of mash them together in different ways, and different chains mash them together in different ways.
Craig Cannon [03:34] – Yeah, absolutely. We should explain those. There are many use cases for distributed ledgers, for programmable money, all of these things. Let’s break those apart, and we had a ton of questions about it, but just from broad strokes, what are the use cases you guys see?
Linda Xie [03:51] – Essentially, whenever there’s a middleman that’s doing programming logic and saying if/then condition, and is charging fees for that, you can replace them with a smart contract. There’s plenty of use cases for that, in the financial system, health care, legal system. I view that as really powerful, just automating things, and allowing there to not be a centralized point of failure. That’s one of the most important parts of what this produces.
Avichal Garg [04:17] – At a high level, the way I think about it is there are kind of three buckets. There’s essentially programmable money, and there’s a bunch of different facets to that. There’s privacy, or there’s smart contracts that can be used as escrow, and you could argue ICOs effectively are an escrow. You’re locking up some ether, and then you can do something with that, and I guess so that’s just kind of like programmable money is one bucket of use cases. Distributed compute is another bucket of use cases. You can use all these computers, all around the world to execute code. And then the third bucket is distributed apps, so things like prediction markets, or distributed VPN networks, and that’s kind of the third bucket of use cases, and I think pretty much everything I’ve seen so far, kind of falls into one of those three, and sometimes there’s a little bit of an overlap. You could be kind of in two buckets, but I think those are the three buckets I think about.
Craig Cannon [05:05] – Cool, and so to jump into a question from Twitter, J.P. asked, “What are the top use cases you guys think are going to go mainstream in three years?” And mainstream may be caveated by within this community.
Linda Xie [05:17] – Tthe two ones that will go mainstream are decentralized exchange and probably, gaming, something along the lines of collectibles.
Craig Cannon [05:26] – Wait, can you just define collectibles, like what’s an example of that, that’s been out in the world right now.
Linda Xie [05:31] – Out in the world in the crypto community, there’s been crypto kiddies, so you can collect a unique digital cat on the blockchain, so that can never be destroyed. You own that permanently and so that has gotten really popular. People are just kind of collecting it, like digital beeny babies in a way. You can breed them together and produce new unique cats. People just love collecting things in general. That’s been very common with stamps and coins and cars, people just love collecting and now, you have a society where you can just collect everything in just a digital manner and you can freely trade this. And so, that pairs well, to me, with decentralized exchange because, so for context, I worked at centralized exchange where you essentially have this exchange controlling user funds. Coinbase is really trustworthy and I’m storing my funds on Coinbase. But, there are a lot of centralized exchanges out there that have risk of getting hacked or getting shut down by regulators, running with user funds, so there’s risk on the custody side. There’s also some barriers to entry, if you live in a jurisdiction that something like Coinbase doesn’t support or you don’t have the right documentations. I’m really excited about decentralized exchange because you can now have a bunch of people that can participate in these markets and trade anything they want.
Craig Cannon [06:42] – Yeah.
Linda Xie [06:43] – Trade digital cats if they want and it’s really cool. More people can just participate in the network at this point.
Craig Cannon [06:51] – Do you think the same thing is true? Like, are those that use cases that happened in the next couple years?
Avichal Garg [06:55] – Yeah, I think both of those will happen. The two others I’d had, or the big one, the other big one, I’d add is basically is programmable money. I think, like, the payment rails will all get built out, things like lightning networking plasma or zero fee payment networks that are coming up and I think the merchant side of those things will continue to get better, so I think payments as a category of stuff will happen. And then privacy is a subset of that too. Privacy tokens are actually a thing. If you talk to a lot of people, hey if XYZ thing went away, would you care about it? They’re like, yeah, it would be kind of unfortunate if that thing went away or I’d lose a lot of money because I was in the ICO. Then you talk to people and you say, well, would you miss it if Monero went away and they’re like, “Oh man, that would really be terrible because I actually use Monero to do something.” The payment side of things is what’s actually going to merge and then yeah, I agree, decentralized exchange in collectibles are too early, like rarely use cases where people actually using.
Craig Cannon [07:48] – What about about the decentralized apps that people are talking about and kind of dreaming about now? What’s the timeline on something like that?
Avichal Garg [07:56] – This is probably a place where we might differ. It’s probably seven to ten years out for most these. We’re really, really early in most of these cases. There might be one-off use cases, like Orchid with VPN, where there’s a real problem and being a distributed network in a decentralized network can be censorship-resistant and so that’s a core feature of the network. Beyond that, it’ll be a while. It’ll be some early cases, but by and large, it’s going to take several years.
Linda Xie [08:25] – Yeah, I generally agree. My attention is just a little bit shortened, maybe three to five. But I think there’s a lot of scaling issues that need to be solved before anything can go mainstream. But, I’m generally pretty optimistic that there’s really a lot of smart people working on that problem right now.
Craig Cannon [08:41] – Be more specific on the technical side, what are the scaling issues?
Linda Xie [08:44] – Yes, with the Ethereum, crypto kiddies itself was so popular, so many people were trying to use it, Bitcoin itself, so many people have been actually trying to use Bitcoin, that the transaction fees have actually gone pretty high. At some points, it was like average $30 per transaction fee, which is like parallel to a wire at that point. You have so many people… It’s a good problem, so many people want to use it, but it’s too expensive and it starts getting really slow. In Ethereum’s case, I know there’s a lot of scaling work being done. L4 is an organization that was funded by the Ethereum community grant to essentially work on some scaling solutions. There’s a lot of different ones that they are tackling. They’re specifically working on state channels, which essentially you can kind of, if I were to compare this, it’s like a bar tab in a way, where you essentially have all these like off-line transactions, where people are just moving back and forth between each other and you can update state and essentially you can update logic and then only when you want to finalize something, you can move it back to the main chain. It’s just a bunch of off chain transactions. There’s also Plasma, which is Joseph Poon and Vitalik Buterin, which essentially you have blockchains within blockchains, and so you just have all this work being done on these blockchains within a main blockchain. And if you ever want to, because the idea is that you really don’t care about
Linda Xie [10:05] – what everyone else is doing, you only really care about what your transactions are all about. You’re isolated in your own little blockchain and only when something goes wrong, you ever have to like leave that blockchain and go back to the main one and report something’s wrong. There’s all kinds of things going on with this.
Craig Cannon [10:20] – Just to understand that, by your own blockchain, are you meaning individual users or individual apps on the blockchain?
Linda Xie [10:28] – Both, essentially you can just be isolated in your own world, in whatever application, or even individual use cases. This is very theoretical right now, but there’s work being done on it. There’s also TrueBit working on doing off-chain computations, so essentially you don’t want really complex computation on Ethereum blockchain, you essentially can maybe move that off and pay people for computational, to be done, and essentially, the promise that you have to verify that the work was done correctly. They have a really cool concept, where you essentially have forced errors in the network, so you have essentially every once in a while, the network will do some wrong computation and something that is incorrect and people who catch that error, actually get compensated for that. You have this whole network where people are just constantly making sure that things are running correctly, because if they catch something, they get a pay off for it. There’s just a bunch of really cool work being done, so I’m generally pretty optimistic on the ability to actually go mainstream.
Craig Cannon [11:29] – Do you think that many of these apps that we’re going to see are going to resemble things that are popular right now? Or is thisvery much like the beginning of the internet and we had no idea what was coming and it’ll look more like that than a dupe of what we have right now?
Avichal Garg [11:45] – Yeah, that’s a great question. The early ideas will look like ports of stuff that we’re used to, just like the early internet was kind of like the New York Times took the newspaper and literally like NYTimes.com.
Craig Cannon [11:57] – The headlines for images.
Avichal Garg [12:00] – It looks like the newspaper right? It’s the easiest way to transport to a new platform and we just couldn’t predict that social media, like actually Facebook and Twitter would be the actual media winners, or YouTube would be the actual media winner. It wouldn’t be CNN. And so, I think it would have been really challenging to predict something like Airbnb. If you could have called that, then yeah, well, you’d have done quite well as an investor, but also, Airbnb, if you look at their early funding history, it was really challenging for them ’cause it was such an out there idea. But it was a native idea, it was the thing that made sense on the internet. The first set of ideas will be essentially ports and there’ll be people trying to sort of, cargo cult map the ideas over and then, a couple of years into it, people who are sort of thinking about these thing natively, will start to play with ideas and say, “Oh, well now you can do this thing that just wasn’t possible before and here’s an idea, and what do you know, if I push a button, a car can come to me, instead of me going to someplace,” right? Those ideas, I think it just takes a while for people to get used to it and for developers to play with the infrastructure and new concepts and just sort of poke around and play around and see what’s possible.
Craig Cannon [13:03] – Do you think it’ll leaped frog by location in any ways?For example, you know how L.A. was perfectly situated to have cars, but medieval European cities weren’t? And so they kind of than get leaped frog and have the train infrastructure. Do you think that will happen with a decentralized app? Say for instance, in Africa somewhere where like, money is not the same thing as it is, in terms of payment accessablity and most things, will they take off in different places first? Or, will that not be the case?
Linda Xie [13:34] – Personally, I never thought cryptocurrencies were going to take of in the US or anything at first. It’s going to be in countries where people don’t have bank accounts primarily and now need a method of actually storing their own funds. Or countries where the currency is just getting inflated away, so the currency is not buyable, so I find that in certain countries, their use cases of cryptocurrencies actually matter more than in countries where there’s really strong financial infrastructure. It definitely varies in my case, so there’s also something like, in China, where you have all this censorship of using different applications, I could really see decentralized applications taking off in China.
Avichal Garg [14:12] – Yeah.
Craig Cannon [14:13] – Yeah, I just always wondered is the Valley a good barometer of the popularity of cryptocurrency in decentralized apps or is it not, that’s what I’ve been wondering?
Avichal Garg [14:23] – That’s a good question. In some sense, geography is just one specific way to say that there’s an underserved market. You will probably see things, like the inspiration for new ideas does tend to take hold in these underserved markets and so geography is certainly one dimension. Or, you could of argued that, college students were an underserved market and they came to social networks, and it was just they were all in one place, there was an opportunity for a bring new type of media to emerge and so there will be pockets, even in developed countries, there are pockets that are underserved and so the question is really, where are the underserved markets, where the new infrastructure gives you some fundamental advantage or some sort of new utility. There’s kind of a related question which is, even if that’s worth the inspiration for where the idea it comes from, how do the companies that build those products actually scale, which actually is a different dimension altogether. And so, if you think about where the bottleneck is for actually building companies are, I think a lot of that is tribal knowledge that’s in people’s heads. Like, what does it mean to have a fast growing startup? Coinbase for example, is a YC company. It’s actually not a coincidence to me. It’s just like what it takes to build a fast growing startup. It turns out that there’s a lot of traditional best practices about how you hire people and how you structure your teams, and how you raise capital and like a bunch of that stuff is like,
Avichal Garg [15:42] – that tribal knowledge that’s in people’s heads in Silicon Valley and so, I think you get to have the inspiration for ideas coming from underserved. But I wouldn’t be surprised if the really big winners in crypto are actually sort of re-centralized around Silicon Valley.
Craig Cannon [15:54] – Feel the same way?
Linda Xie [15:55] – Yeah, generally feel the same way, I mean there is a lot of developers that are, that I hear about working on Ethereum applications in their spare time. And so you’re seeing all of the flock of developer talent from Silicon Valley into working in this space and that’s where you’re going to see a lot of innovation. Although, it is pretty nice to see that there are like different hubs around the world of watching development so there’s like, Berlin and I know places in Argentina and Switzerland is another area. It’s pretty cool to see more hubs pop up in the space.
Craig Cannon [16:23] – Yeah and how much do you think governance and regulation is going to affect those hubs? Does it ultimately mean that there’s a certain density of people that can build this stuff, therefore I’m moving there? Or are these tax incentives, different regulations actually going to move the market?
Linda Xie [16:39] – Yeah, it’s really going to move the market. Right now, I think the US is being smart and not tryna be overly restrictive because they’ll know that it’ll push out of the development. Especially, let’s say if you were just really be harsh on people that were developing on these projects and that would easily just cause people to move elsewhere. You’re already seeing the tax incentives play out here, so people are registering in Cayman Islands,
Avichal Garg [17:02] – Or Puerto Rico.
Linda Xie [17:03] – Or Puerto Rico. Yeah, people are like moving to Puerto Rico and starting a little hub over there.
Craig Cannon [17:07] – I’ve heard of people buying houses in Puerto Rico.
Linda Xie [17:09] – Yeah, yeah, so I actually know of people that are trying to move out of California. Especially, because California has really high taxes.
Craig Cannon [17:17] – Well, you see Nevada plates here all the time. People have been doing it for a while.
Avichal Garg [17:21] – Yeah, the town’s not that far.
Linda Xie [17:24] – Yeah, even like when you have scenarios where people are like mining in China and then all of a sudden China starts cracking down on mining. These miners have to move elsewhere and you’re seeing people move over to Washington, because electricity is cheap. So there’s this movement and it’s just based off of cheap electricity as well. It’s kind of crazy.
Avichal Garg [17:40] – It’s pretty fascinating. The whole regulatory angle to this stuff is pretty fascinating. It’s just how our government’s, the game theory around how the governments want to regulate this stuff is pretty interesting. I actually think the US government has been really, really smart about this and pretty measured and sophisticated, because I think they understand there’s a lot of potential for innovation here and a lot of value to be created and they don’t want to crush that. I’ve been actually pretty impressed with how measured they’ve been. You look at something like the BTC, the Bitcoin ETF, paperwork that went in and all these people filing for it, I thought it was so clever the way that they essentially pushed back on that. They didn’t actually reject the ETF proposals, they essentially convinced the people who put in the ETF proposals to withdraw the proposals. Yeah, which I thought was super clever, so they basically back channeled, I think, I don’t have any insider information, just to be clear, but my read on it was, what they did was, they went to the teams and said, “Hey look, there are a lot of reasons we can’t approve this right now,” and maybe it was like, “Hey we don’t want to inflate the bubble anymore or whatever else, why don’t your withdraw them and let’s keep talking.” And then that way, it wasn’t like the government rejected the ETF proposals,
Avichal Garg [18:50] – it was the ETF proposals being withdrawn. Which is the absolutely right way to do it, right? If you’re like, “Hey look, we’re going to do this eventually, we just can’t do this right now but we don’t want to crush the market, we also don’t want to pump up the market,” really smart way to do it. They’ve actually I think been pretty sophisticated, like the regulators and inside the SEC and the CFTC, and so on. I think they’ve been pretty sophisticated in the way they’re handling this.
Craig Cannon [19:08] – And what about on the investor’s side? How do you guys feel about accredited investors getting involved in ICOs versus non-accredited?
Linda Xie [19:16] – That one is hard because I really believe in decentralized systems and everyone should have access to this, but at the same time, if something straight up looks like a security token and you’re issuing it in the US, I think that rightfully falls under the SEC regulations. I do believe in compliance and that’s actually something I really like about this new protocol, Harbor protocol, where they essentially try to make sure that the trading of tokens is compliant. You can have some wide list of accredited investors that plug into the smart contract and essentially say that you can only issue tokens to these investors. And there’s maybe some holding requirement, you have to hold it for 90 days or something, so this smart contract itself will have the logic that you can actually trade these tokens. It’s important to be compliant with this, as long as it’s very clear it’s a security token. Although, there are plenty of cases where like a governance token doesn’t fall under that jurisdiction, so it’s really important to make the distinction that not all tokens are the same.
Avichal Garg [20:15] – Yeah. Agree with all that. The risk really is that… There’s lot a whole philosophical angle to this, right, which is to what degree should the government be involved and being somewhat paternal about protecting investors from themselves in a sense. And these distinctions are arbitrary… If you have a certain amount of money, you get to sort of do whatever you want, if you don’t than you don’t, right. It’s starting to change with crowdfunding regulations that have changed recently. There is some movement there. Setting aside the kind of philosophical parts of it, generally speaking, trying to prevent scammers in the ecosystem and trying to prevent people from kind of getting and losing their money. It’s and important good and to the extent that the ecosystem can do that itself, things like Harbor, which is pretty awesome project. I think that to the extent that we can kind of make that easy for people, to not lose their money or people inside the ecosystem calling out.
Craig Cannon [21:12] – Well, this is what I’ve been wondering because the ICO market has been rife with these, not necessarily scams but sometimes straight up scams, right? It’s not actually good for anyone. Maybe it’s good for the one scammer who gets a bunch of money but in the long run, it’s not good for anyone.
Avichal Garg [21:30] – Yeah, I agree.
Craig Cannon [21:31] – What other methods are being put in place to avoid these scams?
Avichal Garg [21:36] – Yeah, I’m really hoping that people who are in ecosystem are very aggressive about calling these things out and not supporting them and ultimately, if there are ways for people to call out these projects and then those projects don’t receive any funding, that’s actually the economic incentives will kind of play out in the right way. People will get smarter about it. People will get a little more sophisticated about diligence. And investors can kind of help, the people who are actually sophisticated and understand the technology and have done diligence and understand how product development works and understand all of these nuisances, actually, getting out there and saying, “Hey, this is the real good stuff,” like, TrueBit, super interesting project. You know, like, there’s real tech there and the team is really smart and like so on and just talking about it and saying by the way, this stuff over here, be careful, right? And I don’t want to call anybody out but like, you know, there’s stuff out there that if you just talk to people who are deep in it, then you can sort of poke and prod and realize it’s a little bit more vaporware and people just need to be a little bit more practical about that.
Linda Xie [22:37] – Yeah, there was an interesting case where there was a project called, Crypto All Stars, where essentially you had different collectibles of different crypto celebrities, so there was like one for Naval and Charlie and so if you verified on Twitter that you were actually them, the card owner, you essentially got 4% of the proceeds of the card and then, the creator of the project got 4%. But anyways, this was really popular and people are trading this and it pretty much looked like a pyramid scheme in a way. Someone actually bought the website and bought it from the project owners and claiming they wanted to improve it and they actually shut it down. Someone bought it, I think it was like $10,000 or something like that and then, just shut down the website and just said like, “Hey, this is like straight up a scam and people are pumping the prices on Twitter and you people are going to lose a bunch of money, so I’m not okay with this, I’m shutting it down.” I thought that was kind of crazy. Things like that are happening as well.
Craig Cannon [23:37] – Right. It probably like would behoove me to do some research on the beginning of the stock market. Because I imagine in the beginning, things were getting pumped in the same way.
Avichal Garg [23:45] – Yeah, exactly. That’s the root of a lot of this regulation, is a lot of people lost a lot of money, you know, 60, 70, 80 years ago and so, collectively we all decided, “Hey, it’s probably better if we don’t let snake oil salesmen come in just like take everybody’s money.” It’s like better for everybody. Though I think it’s fair to say, maybe, I think it’s always good to revisit things that have been around for like 50 years and say, which one of these things actually apply today and how we should think about them and how we should evolve them? Even before it gets to that point, hopefully, I think there are people in the community that are good about calling these things out and stamping out the fraud before it really happens. It’s too big and too many people will lose too much money.
Craig Cannon [24:24] – For the people in the community, the folks interested in starting companies, where do you think the opportunities are now? Maybe the real question is, where are the opportunities four years from now that they should start working on right now, to have a company?
Avichal Garg [24:40] – That’s a great question. What do you think?
Linda Xie [24:46] – Okay, there are so many applications that I feel rely on some sort of identity and reputation system, because in the end, even it’s a decentralized system, like if you’re doing a decentralized market place, you still have to trust that the person you’re buying your goods from is actually going to deliver to you. It’s really important to have some reputation and identity system that carries with you as you use all these different applications. We’re still at the very early days of that, so there’s things like Uport and Civic, but they’re really rudimentary compared to the identity systems that exist in like, regular financial systems. I would really like to see people work on that problem and another areas, that a lot of projects talk about doing decentralized governance and they say that they’re going to use the tokens to actually upgrade the protocol over time, but there’s so much research that needs to be done there. There’s massive opportunity to really figure out what’s the best way to handle governance in decentralized manner. That’s going to be at massive potential.
Avichal Garg [25:47] – I’m just taking some notes. I just came up with five that I think are probably worth exploring. So if you’re a founder out there and you’re like, “Hey, what should I be thinking about.”
Craig Cannon [25:54] – One, Cryto puppy.
Linda Xie [25:56] – That exists already, yeah.
Avichal Garg [25:59] – That’s a great idea. The first thing were talked about before was scaling solutions. There’s a lot of infra work to be done just around making this stuff more scalable. Everybody’s pretty much in agreement, that there is just not enough people actually working at that layer of the problem. We could use a lot more great developers poking around there. Even more broadly speaking, there are components of infrastructure missing, and so I think, identity as a component for infrastructure is interesting. I think things like oracles are interesting. I think, it’s just this general idea of like, how do we know what’s true and what’s not true and there are like potentially shared datasets that people would want to cross projects or cross chain projects. I would bucket all of that kind of stuff inside an infrastructure. I think there’s financial infrastructure and financial tooling, like a lot of the early use cases are basically payment-oriented or money-oriented and so there’s much stuff to do there. I think there’s a lot of delve tooling to be built, so just like the actual production of a smart contract, how do you test this stuff? How do you deploy it? There’s a lot of work to do there. And then, like a lot of things, I think the early use cases will probably be somewhat trivial and fun and so I think gaming and collectibles and things like that will be pretty interesting and maybe, it’s truly native stuff like, crypto kitties, maybe it’s markets to buy and sell those kinds of assets.
Avichal Garg [27:17] – Maybe it’s games from traditional gaming companies sitting on top of crypto somehow to do digital assets inside the game or cross game digital assets. So, I think a lot of it really like fun, trivial stuff that can take advantage of the new technology will be a way to bootstrap in. And arguably, I was talking to someone earlier today about, in some sense, ICOs are kind of the gaming use case, right? It’s kind of like…
Linda Xie [27:44] – Yeah, it is.
Avichal Garg [27:45] – It’s a gambling use case, right? That’s actually kind of what’s happening, is people are just gambling and it’s real money. So it kind of makes sense, right, that that would take off. It kind of matches the pattern of what you see with early adoption technology. Yet again, gambling is an early adopter.
Craig Cannon [28:00] – Oh, for sure, yeah.
Linda Xie [28:00] – Yeah.
Craig Cannon [28:01] – Is there a porn crypto thing yet?
Linda Xie [28:00] – Yeah.
Avichal Garg [28:03] – I’m sure there is.
Linda Xie [28:05] – And there’s SpankChain which is actually really popular. It uses really cool technology, so it’s pretty great, yeah.
Craig Cannon [28:10] – What is it doing?
Linda Xie [28:11] – So, I think it’s like webcams where people can just pay Spank Coin or something to them, but they use like stage channel technology. It’s cutting edge technology in Ethereum which is pretty incredible.
Avichal Garg [28:22] – Yeah, one of the more awkward things, with all of the early adopted use cases is too is, you want to understand them but you don’t want to go super deep and like, you don’t want to like, in the ecosystem, you don’t want to be known neccesarily as the SpankChain expert.
Linda Xie [28:38] – Yeah, that’s true. Yeah, I haven’t really seen it.
Avichal Garg [28:42] – Yeah, somebody is doing that. I don’t actually know anybody who’s really dug into it.
Craig Cannon [28:47] – Especially, if you don’t massively cash in as the SpankChain guy, then you’re like, “Oh, I’m just kind of around” Fair enough. What about the actually, ICOs have been gigantic. People have raised hundreds of millions of dollars.
Linda Xie [29:03] – Billions.
Avichal Garg [29:04] – It was six billion last year, is number I read.
Linda Xie [29:06] – Yeah, this year, three billion alone.
Craig Cannon [29:08] – What was the largest one so far?
Linda Xie [29:11] – It’s probably Eos.
Avichal Garg [29:12] – Yeah, it’s probably Eos.
Linda Xie [29:12] – Yeah.
Avichal Garg [29:13] – I think they’re up to a billion. I think probably,
Linda Xie [29:15] – Like two billion now.
Avichal Garg [29:17] – Jeez, unbelievable.
Craig Cannon [29:18] – Yeah, so this fundamentally sifts the paradigm for fundraising. We’ve kind of like gone from zero miles an hour to a hundred and twenty miles an hour, in this conversation but for people who don’t fully understand when someone raises a billion dollars straight up, what are the expectations for this company? What are people buying?
Avichal Garg [29:40] – That’s a great question. It depends on, there’s actually a great amount of variation there too. It’s EOS or Bancor or Filecoin/Protocol Labs, or Tezos I think, across the board, people will try a lot of different models. Everything from you’re buying a utility token which is ultimately going to be used for some sort of functional utility inside the network that’s part of a product. Literally, you’re giving money to a foundation. That’s literally a donation to a foundation and you have no expectation in anything in return is like, if you read the dots, that’s actually what you did, you donated to a foundation. That’s actually a pretty broad spectrum and so there is no universal answer to that. It would be interesting to see what happens on the regulatory side on that in the next 12 months as some of these things come into fruition, where people don’t actually deliver on projects. It’s a big open question of what actually happens and which of these things are okay and which ones are not.
Linda Xie [30:39] – Yeah, what’s crazy is you can’t even possibly spend that much money on developing applications so,
Craig Cannon [30:44] – I disagree. I guarantee you someone is going to spend the money.
Linda Xie [30:48] – Oh my god, I just can’t believe it. But people now, have so much excess capital that they’re giving it to crypto funds and being like, “Hey, manage this capital,” and now they’re like an LP. It’s so crazy that this,
Avichal Garg [30:58] – Yeah, some of these projects have become funds of funds, basically.
Craig Cannon [31:00] – Really? I didn’t know that.
Linda Xie [31:01] – Yeah, it’s crazy.
Avichal Garg [31:03] – In some cases, you can argue, I think… The Ethereum community fund is doing an interesting job here, where they’re saying, hey look, Ethereum is an actual real ecosystem and we want to encourage development within the ecosystem, let’s invest in teams and companies doing really interesting work that may be under invested in. A lot of this infrastructure stuff is under invested so let’s give developers an incentive to pay attention there. In certain other cases, it’s just like, “Hey, we raised two hundred million dollars and by the way, that and now we have a billion dollars. We’re not exactly equipped to handle a billion dollars, we really only needed 25 million dollars to build the thing. What are we going to do with 975 million dollars?”
Craig Cannon [31:40] – Right, now you also have to be a money manager as a founder. How much is liquid? How much is on crypto? How much do I pay people?
Avichal Garg [31:48] – To Linda’s point is, we probably don’t want to be money managers so let go find a money manager and then handing their money over to people to be money managers.
Craig Cannon [31:55] – Man! That’s insane. On the founder’s side, what are you guys saying to people that are honestly considering an ICO and they passed the threshold. Okay, this actually makes sense. You’re not just generating some pyramid scheme. You’re not just trying to grab money and there’s a real reason to do it. What do tell them?
Linda Xie [32:16] – Very, very few times is it reasonable, in my opinion, to do a token sale. There’s a lot of really great projects that can do decentralized applications or create their own block chain or whatever. They don’t actually need to do a token sale, they can just go traditional equity route. We seen that with some projects like Dharma and dYdX. I actually recommend a lot of projects to go that route, because you don’t necessarily want to tack on a token that you don’t know if it’s actually going to make sense. It creates additional frictions to the network as well. There’s still regulatory uncertainty as how this is supposed to be handled. If someone still wants to really do a token sale, I recommend that they work with lawyers that really understand this space. Don’t try to be cheap with the lawyers. I’ve heard people like, “I’ve found some discount lawyer who’s willing to do this token sale, for really cheap.” I’m like, no, you want to make sure people have expertise around this and can guide you through things. I generally also say, no one’s listened to me on this, but, I think they should raise in a series. There’s no need that you have to raise all that money up front. Let’s say, you really wanted to raise all that money up front, you should probably lock it up for some time and as you hit certain milestones, the funds then release. Because it can also be sitting on 200 million dollars, can also kind of hurt incentives of the team being like, “Oh, we just have a bunch of money, we can just spend like crazy.”
Linda Xie [33:33] – The incentives really changed. And the last thing, which people have listened to me on this, the vesting schedule of tokens. It’s really important to kind of look like a traditional company. In that sense, you want to make sure that employees are not going to just leave and cash out their tokens when the token sale happens. On the flip side, for investors, you want to make sure that they’re aligning with you as well. Early on, a lot of projects didn’t have a vesting schedule or lock-up for these investors and the investors just came in and flipped their coins. They got their discount like 20-30% and then during the actual token sale, they just sold everything and they’re no longer investors. It really changes incentives, so it’s so important to think through how your token model will actually works. What’s the governance of it, essentially.
Craig Cannon [34:18] – Do you recommend four-year vesting for employees? Like same deal?
Linda Xie [34:23] – Yeah, that’s pretty much what I’ve been doing so far, like one-year lock-up,
Craig Cannon [34:25] – One-year lock-up.
Linda Xie [34:26] – One-year lock for your vesting schedule. That’s what I’ve been recommending, but I get, it’s,
Avichal Garg [34:31] – It’s pretty tried and true. I mean, I think a lot of these things like vesting or being thoughtful about when you raise money and how you raise money, they’re tried and true for a reason, right? They actually like underlining it all, there’s a group of humans that get in a room or get in a telegram group and write some code. Those human dynamics haven’t really changed and so things like vesting are important. I think we figured that out over the last 20 years, that some of these things are really good. Now I think that are places where you might want to reconsider them in this new world. Where’s the value accrue, is it equity or tokens? Are your employees vesting equity or are they vesting tokens? There are like some interesting questions there, but I think some of these foundational concepts like vesting, they exist for a reason. They’re best practice for a reason and I think a lot of teams should be adopting these things, if they haven’t already.
Craig Cannon [35:19] – Again, when you get a bunch of humans in a room, they create games and if you raise the billion, I’m going to raise a billion and one and that’s the game, unfortunately. To boil it down, you’re saying to founders, listen, you don’t necessarily have to do a token sale. It’s not required, so if you’re a founder and you’re interested in crypto, you can get in, but you don’t have to do this. When it comes back to that, are you saying to them, apply to an accelerator or go and raise money traditionally, what are you recommending them to do in that path?
Avichal Garg [35:52] – What it comes down to is, just outside of crypto and startup world at this point, there are a lot of places to get money. I know it’s different if you’re in different parts of the world. Some places have more investor depth and investor liquidity than other places, but there are a lot of ways to raise money. You really want to be thoughtful about who you’re raising money from and what value they bring to your company. You can do the ISO and get a bunch of people to send you some eth, but what happens if things don’t go well? Those are the kinds of projects that you know, your retail investor raises up their hand and says, that person stole my money. Meanwhile, there are all these great investors like Linda who have actually been at companies and started companies and helped scale companies. That’s actually who you want involved. That’s actually who can help you over the next three, four, five years as you build your company. What we are going to see is a lot of people returning to that idea of “Oh, I could get money from five different places, who can actually add value to my company for the next three to five years and help me build this company.” I think we will see that sort of re-aggregation towards people who actually add value in the earlier stages. That could be an acceleration. That could be an early stage investor. That could be somebody who’s really early in crypto made a bunch of money but actually is very sophisticated about how they think about crypto. There are a lot of those people out there too.
Avichal Garg [37:15] – It’s not to say that people who haven’t previously been in the ecosystem are not value add, I think there are actually a lot of people crypto-native in some sense, who really understand this stuff and have been in it for years, who are very smart and very thoughtful and you want them involved. It’s really about how do you find people who can add value and get them involved in your company and being smart and thoughtful about that.
Linda Xie [37:33] – That’s so true because when these projects are doing token sales, they’re distributing their tokens to thousands of people and especially, if you did some heavy marketing around that, you end up getting this pump and dump kind of crowd. All they’re doing is saying, when moon, when… They just want things listed and price to pump. If you look at those Slack groups, it’s just like mayhem. They’re not adding any value, they’re creating tons of distractions for the founders. It’s so important to have value add investors, not just try to make money really, really quick or something.
Craig Cannon [38:04] – You see the micro example in Silicon Valley already, right? If you’re a big quotes hot company, you can raise from anyone. It’s just about finding the smartest money. You see that on a larger scale here, where there’s billions of dollars just floating around.
Avichal Garg [38:16] – Yeah, 100%. I actually think in the last three months in particular, I don’t know if you’re seeing this, but it seems like we have moved back into that world, where the really, really good teams in crypto and a lot of the teams that have a lot of early momentum, are going out and seeking out the high value investors again, and saying, “Hey, I don’t want to do a public sale, I actually want to find people who add value to the company and be thoughtful about it.” We might look at this historically you know, in five years we might look back and say, oh, that was a funny six month window in time. Kind of like, reverted to the mean the basically, so we came back to the basic principles.
Linda Xie [38:50] – I saw some stat on that, it was saying, nowadays 60% of the funds raised in the general token sale, is actually raised in the presale round. It’s really shifted the momentum so far.
Craig Cannon [39:01] – Just in a year.
Avichal Garg [39:03] – The last few months, yeah. Since the beginning of the year. Yeah, things move so fast in this space.
Craig Cannon [39:08] – Alright, so let’s go into some of the questions from Twitter, King Croasis asks, so what are thoughts on optional vs mandatory privacy in transactions?
Linda Xie [39:18] – Yes, I have a lot of thoughts on that, so I essentially think that it needs to be mandatory privacy, if you’re going to actually have a privacy coin. If it’s ever optional, you get looked at if you’re starting to use a privacy feature, like, “Why are you using this privacy feature, do you have to something to hide?” First of all, you paint a target on your back. Second of all, of contacts, I did blockchain changes when I was at Coinbase and so Bitcoin is very easy to trace and a lot of people give information away about other people. Even if you’re being really smart and someone transacts with you, they give away information about you and so that can happen in a scenario, where you only have a small subset that’s actually being private and the rest of the users on the network are being public. They can give that information away, so it’s really important to just have mandatory privacy. That’s something I really love about Monero, all they transactions are private by default and that’s absolutely crucial.
Avichal Garg [40:11] – Yeah, I philosophically agree. Mandatory is the way to go. It creates some interesting challenges though. That’s a harder pill for regulators to swallow, generally speaking. I heard this great story where in the early days of the internet, you know SSL, HTPS is like a standard thing. It’s on by default in a lot of cases. But actually, the Netscape team had to fight to make that a thing that was possible inside the browser because the government was worried, “Why do you need to encrypt your transactions? Are you doing something illegal?” Why shouldn’t we have this be public? It turned out that one of the selling points was commerce. You don’t want to send your credit card number over the wire unencrypted. The fact that they were like real legitimate use cases allowed things like SSL to emerge like, “Oh yeah, we should actually allow that.” That is actually the challenge for the space is philosophically, you agree with Linda which I do, I think then that the onus on us to demonstrate use cases, where default privacy is really important. Things like, if people do start to get paid in crypto, for their job, well you kind of don’t want everybody to know how much money you make and who’s paying you. That’s reasonable. It gets even more interesting if you start talking about, well let’s say, that you have a particular sexual orientation and you live in a community that doesn’t really support that and you’re donating to a charity that does, do you really want people to be able to sniff that out?
Avichal Garg [41:41] – You take it even more extreme when you say, well if you live in a country where the government is not, you might trust the US government but there might be other governments that you don’t trust, even if it’s not your government, do you want that government to be able to sniff out your transactions on a public blockchain?
Craig Cannon [41:55] – Right.
Avichal Garg [41:56] – I think being really concrete about, it’s not just a philosophical thing, although I do agree with the philosophy a hundred percent. There are really practical consequences to this stuff. What would make it possible, I think, for regulators to understand why it’s so important that it be on by default.
Linda Xie [42:10] – Yeah, absolutely, and also, there is this concept of selective transparency and both , so you can essentially have this view key where you can share with someone to now view your transactions. This you can share with auditors, if you want to be compliant or if a charity wants to display their transactions, they can share their view key to everyone in the world. The idea is that if it really mirrors the traditional financial system, we don’t broadcast everything to the world, but we download our bank account statement and give it to someone when they need to see it. I don’t think it’s far off from what we’re used to.
Craig Cannon [42:45] – So just default private and then, if you need to be public on anything. Cool, so transitioning from that into another governmental related question, Claudio asked, how do you see the US government or do you see the US government launching their own crypto currency backed by gold?
Avichal Garg [43:01] – Yeah, I have very strong thoughts about this. The specific incarnation I think backed by gold, we can sort of set that aside for a second, in general, I hear a lot of people talking about how governments might regulate crypto or ban crypto and there is some risk of that. The bigger and scarier potential risk is that governments fully embrace crypto in the wrong way. Kind of back to this idea of privacy, right. Do you really want a government that you may not trust to have full visibility into everything that you are spending money on? Everyday and every transaction and every person that you’re transacting with, right? That has very, there’s fundamental human rights privacy sorts of aspects of that and they’re very practical. If your government can do it, odds are other governments are can do it, sorts of downstream consequences. Increasingly, governments are getting sophisticated in a way that they think about technology and adopting technology and using technology. The bigger risk is actually, what Claudio was calling out, is governments being really smart about adopting this in a way that makes, in my opinion, the wrong trade-offs between sort of security versus privacy. There’s not enough people, so many people are worrying about governments banning this stuff, that we’re not looking at the other side nearly as deeply of what happens if the governments fully embraces it but does it in the wrong way. What if a government you may not trust, even if you trust your government, if another government that you don’t trust embraces it in the wrong way.
Craig Cannon [44:35] – Is there a positive outcome in which a government embraces it, creates for example, USCoin, whatever it might be, TrumpCoin in the worst case scenario, where it’s actually not a negative thing but we move from the dollar to USCoin or whatever it is, possible?
Avichal Garg [44:53] – Yeah. That’s one of the great things about crypto is that it is global and there are all sorts of opportunities for governments, especially smaller and more nimble and willing to adopt this stuff. Estonia has done a great job of pioneering some of this stuff. Singapore’s poking around with this. There are definitely places between voting or making it easy to file your taxes to making a banking system more efficient. There are lots of places where large existing institutions could adopt this and that’ll be great. Even making stable coins, maybe some government should just say, hey, we have 500 billion dollars and you trust us and we’re just going to issue stable coins and now crypto can actually be a real thing. There are lots of positive ways governments can play in this space and hopefully some will.
Linda Xie [45:41] – Yeah, Dubai’s been pretty good about that. They said that all visa applications and some other documents will be on the blockchain by year 2020. It’s going to save them over a billion dollars a year and just being more efficient. They’ve been pioneers in this space and I know Consensus, the company, works very closely with them. I completely agree with everything you said. What’s crazy is Venezuela already issued their own
Avichal Garg [46:07] – You’re talking about the Petro Dollar.
Linda Xie [46:08] – Yeah, which they claimed their 5 billion dollars was raising the presale but they haven’t proved it. But what’s crazy is all these sanctions against Venezuela and now there’s this foreign capital coming in apparently, to fund this. You run into all these issues where now you’re supporting a sanctioned country financially. That will be real interesting to watch how that plays out.
Craig Cannon [46:30] – Right, well you see every day how little the average normal person cares about this sanctions between borders, it’s just like, “I don’t know, go buy something in Venezuela or Mexico or it doesn’t matter.”
Avichal Garg [46:41] – Yeah, it’s fascinating.
Craig Cannon [46:42] – Alright, another question. Navin Mischra asked, what would be the best onboarding ramps for adoption and what roles will oracles have in decision making in the future?
Linda Xie [46:52] – Best onboarding ramps for adoption… Honestly, it’s something like Coinbase. They do a really good job of just linking up your bank account and just purchasing cryptocurrencies. Just having a nice user experience there is really important. I think the more interesting thing for me is what role will oracles have in the decision making process. I’m fascinated with this idea of decentralized oracles, so you can essentially have a system like Augur, which is a prediction market platform where anyone can create a prediction market on it. The past issues with prediction markets is they often get shut down by regulators just because it’s kind of closer to gambling. But in prediction markets on top of Augur, no one can shut this down. You can really have these markets operate and so people can start creating markets where essentially you measure the citizen’s happiness, you measure GDP, you measure all these different statistics of a country, and you can have people betting on the outcome of whether or not a different policy is going to increase or decrease these statistics. And from there, policy makers can now start making their decisions based off the outcome of this and that’s called futarchy. I just think that is so cool because people are actually putting their money where their mouth is at this point. They might say one thing, like “Yeah, I really support this policy,” but if you’re actually betting on it, then I think that’s a stronger signal. I think that’s going to be something that is totally new
Linda Xie [48:17] – that these systems are able to produce.
Avichal Garg [48:22] – The only thing I’ll add on the onboarding piece of it is, I suspect that the way most people get onboarded will either be very directly like Coinbase, so it’s like they know what they’re getting into, they’re buying into crypto, or of it’s going to be totally opaque to them, they’re not going to realize at all what they’re doing is interacting with blockchain and crypto, they’re just doing it a thing that you couldn’t do before. It’s going be so sideways, that people just won’t realize that’s actually what’s happening. I suspect those native use cases will be that way. It’s an interesting question of what percentage of people onboarding into the space do it very directly through something like Coinbase or they just happened to do a thing and it happens to sit on top of crypto. It kind of reminds me of several years ago, we talked about mobile companies, so right now we’re talking about crypto companies and at some point in the next five years, it’s not going to be that you’re crypto company. It’s that you offer some value to the people who use your product and it so happens that you sit on top of blockchain or use crypto economics or whatever. That will actually be how a lot of this stuff goes mainstream, in a way that people don’t even associate directly with crypto necessarily.
Craig Cannon [49:28] – I would bet 100% on that, because if you ask the average person, “What language is Facebook in?” What? What’s a language? Why do I care about this? You see it when people are pitching their companies. They’re like, we have the fastest implementation of Python you’ve seen and you’re like, “No one cares. No user cares at all.” Good point. Thanks for coming in guys. This has been really great.
Linda Xie [49:54] – Yeah, thanks for having us.
Avichal Garg [49:54] – Yeah, thanks for having us.
Linda Xie [49:56] – Alright, thanks for listening. As always, you can find the transcript and the video at blog.ycombinator.com and if you have a second, it would be awesome to give us a rating and review wherever you find your podcasts. See you next time.
Y Combinator created a new model for funding early stage startups. Twice a year we invest a small amount of money ($150k) in a large number of startups (recently 200). The startups move to Silicon