a16z Podcast | Cryptocurrencies, App Coins, and Investing in Protocols



hi everyone welcome to the asic sincere podcast I'm sonal I'm here today with Chris Dixon and Olaf Carson we the founder of a six in the investment poly chain which is a cryptocurrency hedge fund and we're talking today about what that means both as an asset class and in terms of protocol development and more broadly what happens when you bring capitalism to open source development and services okay that's the intro welcome guys so off if you don't mind people in oh but cryptocurrency hedge fund is maybe you could just play with it absolutely so poly chain capital manages a hedge fund that invests exclusively in crypto currencies and by the way are there a lot of crypto currencies so you've probably only heard of Bitcoin yeah you've heard of a cerium there are hundreds of others like oh I remember hearing about dogecoin and things like that exactly and new ones are launched all the time much more than you might think so in March alone there's been about 15 launches of new holy so instead of investing in companies that are building on top of these protocols we invest directly into the protocols so we take equity ownership of protocols instead of taking equity ownership of companies let's pause on that for a minute because that's so counterintuitive honestly because the whole thing we've been touting here and the committee's been talking about for ages is how important yes that bitcoin and blockchain and everything related to the crypto currencies the protocol is obviously very important but because of what it enables so I'm very confused by this concept of investing in a protocol you guys the way I think of it is it's like imagine you were back in 1993 and the internet was starting there were two ways you could have imagined investing in that one was to invest directly in companies like Amazon and if you were fortunate enough to invest in Amazon of course that would be a phenomenal investment a lot of people didn't have access to that or invested in companies that weren't as successful as Amazon another way to have sort of bet on the internet would have been to do something like buy domain names which a lot of people did actually and they're called them domain errs and there were a bunch of people that saw the value there and a lot of them bought you know whatever pizza calm or something and at the very beginning who'd buy for $8 or whatever the domain registration fee was the aggregate value of like dot-com domain names today you know it's hard to add it up but it's certainly in the many billions of dollars you know so just like you know a typical five letter English word is an example comm is probably five hundred thousand two million dollars so you know you can just sort of go to the dictionary and figure out that having the total value and so I think it's very obvious to that what you have is people menteng these new protocols that have you know things like they they could be for doing distributed storage or distributed payments or a bunch of different use cases the difference now is that you can invest for example indirectly in a company like coinbase which we here did and actually Olaf used to work at Quinn base or you can invest directly in a protocol and so what Olaf is doing is essentially think of it as a bunch of new things like the internet getting started today and they have the feature where you can just go buy domain names and you're buying domain names in a sense right yeah in a sense it's a more of a pure play is the way I like to think about it so instead of investing on the application layer which you're taking a bet on a specific team a specific business model a specific so that's like Amazon Yahoo application exactly and we're betting a layer below that I think in this space disproportionate value accrues to that protocol level so even if you look empirically at say Bitcoin the seed investors in coinbase have done very well on that investment but when they made that investment Bitcoin was at about five dollars yeah and they would have done effectively just as well by simply buying Bitcoin you look at the etherium ecosystem there's been very little venture funding into aetherium companies aetherium here being an alternative blockchain to Bitcoin but ether itself has grown in value by over 100 X since it was created in addition the crowd sales where people are actually raising money for application specific crypto currencies like app coins really gap coins are app tokens those those token launches on top of aetherium have raised over 300 million dollars can you talk about that for a minute because I'm actually just trying to understand what the hell in-app Queen is in the first place when a lot of people think about Bitcoin they think about money or currency they think about this broad-based way to generate value and transfer value really what's happening now is people are creating I like to call them application-specific tokens and these are basically a monetary layer added on top of a specific type of I'll give you a concrete example there's a project called golem named after the Lord of the Rings character I don't know maybe did so Gollum is a peer-to-peer marketplace for computation this means that instead of shutting my computer at night I can rent out the CPU and GPU cycles to some developer in South Africa or Japan or something and all those payments happen in these column Network tokens so the that golem network token is not really money in a sense it's really meant to just at a monetary layer to this one specific peer-to-peer network and so when you're when you're buying that token you are buying either the ability to rent out CPU or GPU cycles but you can also look at it as owning equity in this peer-to-peer protocol that's fascinating so when we you know hold something like Gollum tokens we're really betting that this peer-to-peer protocol for computation will grow now the interesting thing about these projects in this ecosystem is that they're literally aren't companies so suppose you want you like to this Gollum protocol and you wanted to invest like andreessen horowitz investing companies there's really no vehicle we can't invest in Gollum because it's a project not a company it literally doesn't exist so the only means to gain exposure to this new ecosystem is to buy cryptographic tokens or if you rewind like it would have been awesome you know like Linux started in whatever the early 90s you know there's a foundation there's no company at the time it was a hobbyist project today it's 99% of the computers in the world I mean it's Android it's everything in the data center it's you know etc so like so you're saying if there was a way to invest in like the company would have been invested in something that that that had some correlation to the adoption of Linux it would have been one of the all-time great of estimates yeah are there any other benefits to people investing in these at the protocol layer the way you're describing because I mean in don't take this the wrong way it's just benefiting off it monetarily like is there any other benefit to the ecosystem in the flip side of it is you go look at like Linux noise other open-source projects they had to run around raising money or look at Linux has a huge get the heart bleed air bug right so this is a fundamental flaw on SSL that was discovered I think a year ago yeah and if you look at it like the the foundation that's this nonprofit foundation that supports SSL which is the core kind of encryption protocol the Internet and I think their budget is like five hundred thousand a year or something they have to go every year and do like what nonprofits do and and so we basically we you know we've got all these funding models in Silicon Valley for companies but these companies are mostly built on these these open protocols and these open protocols are neglected and the developers are forced to kind of go around and and kind of beg for money what this provides is is it now kind of brings capitalism into protocol development yeah exactly so we've already seen the creators of many of these open-source protocols make lots of money on this I mean the creator of Bitcoin whoever he or she or they are has you know over a billion dollars in in Bitcoin so I think we're already seeing this play out a little bit and it creates the same economic incentives for these projects that you might get as a start-up so when you look at you know taking golems crowd sale as an example when they launched this golem token network they raised nine million dollars in 19 minutes they now have nine million dollars cash to hire developers pay for offices things like that and that looks very much like a series a round of fundraising they also kept an 18 percent equity ownership in the network they're building so they have the same kind of economic model that you might see at a start-up like coinbase where as an employee you have equity and you have salary and now you have quote shareholders yeah exactly and so the thing is that it creates this economic incentive model for founders of open-source protocols in a way that has never existed before absolutely unprecedented the flip side of that as you pointed out is that now the users of the Golum network or any of these peer-to-peer networks and and there are so many of these launching it's pretty incredible they are the equity holders right so when you look at centralized web platforms the last 15 years all the major web services have you know Twitter Facebook Oberer Etsy eBay whatever it might be it's basically a company that built a platform all of the value for users on that platform comes from other users yeah you know we peer-to-peer network and these are the networks that nobody owns just like nobody owns the internet nobody owns email these are networks where the value instead of accruing to a central entity actually accrues back to the users and the thing is that this creates a really really strong network effects among the user base so you can see in Bitcoin sort of fanatics of Bitcoin I was once one of these and I even I would admit I was one yeah and and you don't see that sort of fanaticism around users of Facebook or your users have uber even the power users like they like the product but whatever people in Bitcoin are fanatics and it's because they not only gain a value in a more traditional sense of network effects when when the user base grows they also actually gain wealth from that user base growing and so this idea of users owning the networks that they participate in on the Internet I think is extremely powerful I think another broad historical way to look at it is if you look at the protocols we use today so emails SMTP you know the Internet is tcp/ip HTTP you know HTML etc these these were all developed 20 plus years ago right there's very few protocols that we use today that were developed in the last 20 years I can think of I mean maybe BitTorrent or something but for the most part like you know that's sort of niche and so and those were specifically developed by you know academic academic and government-funded projects and so you had kind of this golden era of protocols and that was funded by governments and academics and then just to remind people that was a time when you actually needed those entities only to be able to fund those well there was because there was no there was no a capitalist sort of economic motive it then of course the Internet bubble happened the capitalism moved in and what you've seen since that period is as all off was saying like a dramatic kind of centralization of power in private companies which you know and a dramatic drop-off in the rate of kind of open protocol development yeah there's only been like one or two like like one of the big battles you know like RSS is a good example where that was a that was a more recent protocol which was with sort of an open social protocol which last decades sort of in the in the you know 2000s had made it run and got popular for a while and could have been a real rival to these closed social platforms like Facebook and Twitter it unfortunately lost I mean it still used the fringes here but it basically lost I mean so basically the central feed the way the last decade you blog who might not have been aware of it but but there was a battle between open and closed in social in closed one right I mean there was actually been there the whole way of a companies that we're trying doing the alternative Facebook's and the all closed one and open los unfortunately I think for the benefit of the world anyways whatever your political views on that it's clear the close close one there's a lot of different theories but one obvious one is just they had all the money they had all the inner you know they had all the people they had all the developers like to add that the other reason for the theory is that they also had a centralized model of coordination which meant a better user experience and a bunch of other features that you have a great user experience with email Gmail is a great user experience and it's built on open protocols I disagree with you I think that Open Pro like web browsing is great user experience it's an open protocol like my view people can disagree on this I think it's because that's where all the money and all the and everything else that they had the better business model they mean clothes had the better business model for the last 15 years now that clothes camp okay now open is developing its own kind of business model around this which i think is going to shift a lot more energy I hope and money and funding and a whole bunch of other things back to the forces of open have a new set of new arsenal and it actually happens on multiple levels they have a new funding mechanism which is the app coins they have just a new infrastructure which is things like you know kind of the blockchain mining infrastructure which is the first time we've seen an example of we've had open source software we've never had opens open services so these are services op X running so basically it's effectively it's a public AWS is what you think of as the miner network so miners is this thing which you know if you read the New York Times they're like oh it's a waste of energy like okay but it's actually running this public and it's like saying a public park is a waste of space or something it's a pub it's public infrastructure yeah it can be used for all sorts of things like the coordination of essential authority there's no central authority and a central owner and it's they sent there are owners of the protocol token owners and individuals and no central point of control no choke point no no one can pull the rule no one can say they can't just cut you off the way that like you know certain social networks cut off developers as a developer you can now invest in this you can build things on top of aetherium with the knowledge that no one can you know Vitalik who's the kind of the you know the leader of it theorem can't pull the rug out from under you even if you want it to self-sustaining so clearly the reason that open might be winning besides the fact of your argument that this capital is rushing in and all these other things how do we solve the problem of trust without having a central authority in the middle the brilliant thing about these systems is the premise is that you do not trust any individual party trust no one yeah you trust no one you if you're a very paranoid person these systems make great sense because you assume that there are all sorts of bad people out there trying to steal trying to dupe the network trying to convince the network that they have a valid you know transaction when they don't or things like that but when you have this decentralized sort of swarm incentive structure and this is what's created say through Bitcoin mining you create this protocol that actually incentivizes very precise human action and then without a centralized kind of top-down hierarchy but instead of sort of swarm of naturally incentivized economic actors you can create amazingly precise organization through this kind of decentralized incentives drug let's quickly break down why those people are incentivizes because they're getting something out of this Network yes exactly efforts failed because no one had incentive to contribute resources and you know you know the contributors to Linux had to monetize in all sorts of you know other supplementary ways but they're actually building Linux and growing the Linux user base didn't really make them any money directly right now as a lead developer behind one of these blockchain based protocols you can actually make money every single time a new user signs up and is interacting with this blockchain so this this incentive structure as you as you put it is very important because it leads to this organic growth that's completely decentralized completely trustless it's totally permissionless nobody owns the blockchain just like nobody owns the internet so you cannot shut down anyone anyone can participate in any way that they want and it's for that reason to me an open and free system how do you know which protocol and I think this is important because we all heard about Bitcoin but as you mentioned there's a theorem in apparently there's like 15 more this new new ones this month how do you know which ones matter and and how did it affect what you're trying to do yeah I mean this is what I spend all of my time doing so we think a lot about basically technologies that are enhancing what these protocols are capable etherium is a great example via KOIN has a kind of programming language that allows you to interact with the protocol it's called a scripting language and Bitcoin scripting language is really limited it was designed this way intentionally so that people couldn't kind of screw up the protocol in the early days yeah now the creator of etherium vitaliy uterine and the great team behind him has created a programming language that is Turing complete what Turing completeness means is you can basically build anything so any computer science language that is Turing complete can build anything that any other Turing complete language can build what does that mean it means that if you can build an application in one language you can also build it in any other Turing complete language what this pragmatically means is that you can write arbitrarily complex software and have it execute in the etherium blockchain much in the way that a Bitcoin transaction would execute in the Bitcoin blockchain so you can build all sorts of more advanced features in aetherium that are simply not possible with Bitcoin and if you're you know Assyrians language it has it has solidity which is like JavaScript and serpent which is like Python whereas Bitcoin is more like bytecode it's almost like assembly very very low-level solidity is very similar to JavaScript yes it's a very accessible language so if you if you're familiar with basic you know web scripting you can kind of write in the theorem contract and so of people may have heard of Bitcoin and aetherium can you talk and you mentioned Gollum can you talk about some of the other yeah so so I can cover you know both application-specific tokens as well as new block chains that are are kind of you know meant for developers to build these applications on top so one underlying blockchain that I'm excited about is called Tasos Tasos is interesting because it uses a proof of stake consensus mechanism instead of mining or proof of work what this means is that the actual stake yes the actual holders of the coin act as the miners so they are the ones actually validating transactions the really interesting thing about Tasos proof of stake system is that the governance is also pushed to the coin holders at the protocol level what this means is that instead of governance and by governance I basically mean decisions about how to upgrade the protocol instead of those happening in in a room full of developers or conversations with minors that are these conversations are very opaque and they don't happen at the protocol level they happen outside the protocol in Tasos all of those governance decisions happen at the protocol level people effectively vote with their coins and the protocol will change based on what those coins like a stockholder of a company appointed by equity early on more equity if you own more stock exactly it's there for more coin is more it's a much more democratic system where the holders of the coin actually have say and how the protocol moves very directly so that's an example of a more underlying blockchain that's really designed for developers to plug into in order to build kind of application specific tokens applications that could be built on top of taser is very similar in some ways to aetherium in that it has this turing-complete scripting language so again it's it's unlimited you know anything that you can imagine is possible okay so thinking about a more application specific token one that I'm interested in is just an are due for a second when you keep referencing application specific tokens that's your version of saying app coin which is a shorthand everyone else uses standardized I like to be very precise in my language is because they're all over the place so that's it's usually a good sign it's an it's a sign of fertile creative if no one knows what to call something that means it's new so that's right so in its fertile days you're saying application-specific quote tokens which can also be up going interchangeable so you know an interesting you know token that I'm that I like is is called maker or MKR so maker is is really a sort of whole governance and insurance system for a for a separate coin that is pegged to the USD dollar value so break that down because well I don't even understand it too because I thought for the whole point of all of this was not to peg it to like things like so this is actually you can view that as a feature or about these coins are traditionally very volatile and it's because there's speculation and trading and things like that now for certain applications this causes problems suppose you're holding money in escrow or you're fundraising on the blockchain well you know you're taking in money that could change in value sometimes aggressively over over days or weeks so I think there is a need for a stable coin and basically the idea is a coin that doesn't change in value now the naive way to do this is to put ten million dollars in a bank account issue 10 million stable coins and say hey it's back then it's a peg to the dollar but that doesn't really work because now there's a bank account and someone has control of that bank account there's a legal entity and it's just not how box chains work there's all sorts of flaws there so you have to have a stable coin that's actually endogenous to the blockchain and actually creates that stability through incentive mechanisms that are on the bar yeah what the maker system is attempting to do and I view it as hugely ambitious with you know likely a high chance of failure but if they can pull it off its massive and it's it's basically a system of smart contracts smart contracts here being the pieces of code in the etherion blockchain that are a bit more complex complicated than just transactions it's kind of a patchwork of smart contracts to create a stable coin and this coin then is pegged to the dollar value and so then you can do transactions on the blockchain that you know will hold the value as well as at least they don't just tell you you can't invest in these things now without a mechanism like poly chain so this is the the great thing to me anyone in the world can go by maker and these coins often have venture style returns if you pick the right ones but they're available to anyone in the world Amazon as Chris pointed out earlier was available to a very elite group of venture capitalists it you know it's kind of Club of who you know right and now intelligent twelve-year-olds can go invest so you're saying that's where the democratization comes in there's a bunch of exchanges that where they call crypto to crypto exchanges so usually what you have to do is you go to Quinn base and convert your dollars or fiat into something like Bitcoin and then you can go to these other exchanges and and now you know the probably the most liquid exchange for maker is a smart contract based exchange so we're getting even more abstracted from centralized server exchanges so let's do Central yeah the exchange itself lives in the earthy room that's really really interesting just to take a real world example it'd be like completely removing NASDAQ and Dow Jones and letting people just it's like a decentralized Nasdaq exactly right so this is a whole new asset class you're describing if you can invest in it directly or go through a hedge fund or why does I mean I understand the broad argument that we're bringing capitalism to open source which has been as you guys have observed traditionally very under-resourced so I buy that argument I don't get why there's a hedge fund in this space yeah that's a great question so as you've probably picked up based on this conversation this space is very esoteric yeah and it's hard to understand a lot of these underlying protocols yes it would actually be a 12 year old just throwing a dart in the dark well we don't recommend unless you really know what you're doing you should not go buy these things Bitcoin have gone up a lot but there's a lot of these things that have gone down in value and it's very likely that an average person who's not spending all day on these things will lose their money so if you're listening to this we are definitely not recommend you get that message loud and clear I just I make a point though that that's also true of the regular stock market most people should not be investing in individual stocks you buy ATF's I mean and someday hopefully they'll be a crypto ETF they could buy too but for now they should stay away unless you really know what you're doing I completely agree so yeah I think the as an asset class the thing that's so interesting is that you have units of protocols that you can invest in you there's equity ownership in open source projects so to go back to this maker project when we wanted to do a deal with maker and purchase some of these tokens there was no legal entity behind them it's a pure open-source project and there was no bank account they had so there was no way we could send them money so we actually had to send them etherium this cryptocurrency in order to receive mkr this other kind of crypto token and that's how the deal was done there was no bank account anywhere yeah I involved in this process so it really is a brand-new asset class in that also just when when you look historically this is uncorrelated with oil gold emerging market currencies the equity markets bonds it's it's really just uncorrelated with everything else and it's because it's a totally new phenomena that's so fascinating ok so going back to the the protocols you had your scribe column you describe maker what are some of the other interesting ones and how do you tell what's interesting I mean I know we're saying don't try this at home yeah just how you tell so what we do is we read white papers religiously so these white papers are formal specifications of how the protocol should work yeah I mean it's a cushy nakamoto wrote the original white page there's an italics famous etherion white paper y'all start off as beautiful white paper they specify how the protocol should work and then we want to talk to the founding team so we'll talk to the developers and make sure that they have a good sense of actually how to build this but what we're interested in more broadly is a stack of technologies that can recreate a lot of the centralized services we see on the web today but in a decentralized manner so when you look at we talked about earlier Lee's centralized platforms that connect to different people they're built on a massive stack of web technologies and so they have like the the DNS system which Maps comm domain names into IP addresses and they have a server client architecture that they're built on so we need to recreate all of those slices of that centralized web stack in order to build Twitter the protocol and uber the protocol that can actually compete with these centralized companies so right now what a lot of what we're interested in is those kind of low levels of the decentralized internet staff they're the fundamental building blocks exactly like AWS everything that you can do stock launches spawns the next generation of so there are there are projects related to etherium you know IP FS or the interplanetary file system which has kind of a silly name but is actually a really sophisticated piece of technology to build a decentralized server architecture yeah there's also swarm which is a kind of a similar project on aetherium that helps you build decentralized apps like the domain name system or DNS there's ENS the etherium name system that Maps dot 80h aetherium domain names to small contract cryptographic addresses you know this whole underlying stack of technologies needs to be in place before a lot of these application specific tokens that are more user facing like normal person facing yeah can can really flourish so we're investing in a lot of these things that are part of the low levels of that internet stack as well as what I might call the middleware so something like Gollum for example is a computing platform on top of aetherium but even Gollum is a product for developers it's not really meant for normal people but using say Gollum and and safe I'll coin which powers the ipfs system developers can build more complicated emergent applications right I mean in golems case you're essentially reallocating GPU resources that are unused yes exactly so you can do all kinds of things yeah and so then you know it's too early still in my opinion for really viral end-user applicants or app tokens so I think that's a lot of what we're thinking about is is what technologies will enable that and that's what we want to invest in right okay so assuming that this becomes a real asset class that we really are bringing capitalism to open source and all the wonderful mechanisms that come with it what are some of the challenges obstacles have to be overcome these are brand new technologies so these aren't facing just a market challenge these are hard technology challenges as in no one knew that something like aetherium was possible it's not that it was just a matter of time right we're actually creating things that previously did not exist and some of the things will fail because they are literally impossible so these are hard technology challenges where you're actually sometimes making a bet on something that it could turn out is not even really feasible I think we need literally thousands of brilliant technologists around the world to come together in order to build this decentralized web stack that's a massive challenge luckily I do think that the space because of groups like poly chain is seeing more and more capital available to build these great projects so I do view it just as a matter of time until these things very quickly become it kind of are you know or rather are building the next era of the Internet I have to ask a tough question which is you know there's so many stories in the media and in in the developer communities around Forks and conflicts and frankly open-source is a bunch of people working together at the end of the day we can talk about technology but it's humans coordinating to work and develop code and for build something building something are there any challenges that are unique to this world that aren't because I think of like there are no politics or disagreements of media companies large groups of people have politics and disagreements yeah and I think the only difference is that the arguments happen in public corporate board might have just as large an argument and one side fails but the shareholders never hear about that you're right very right in this case these are literally on public forums on mailing lists that anyone has access to IRC channels that anyone can join that does one more question though around like resolution conflict resolution because when I think of a centralized how does conflict resolution like if you listen you have a group of people trying to design their friend it's all built into the protocols so like in Bitcoin there's a there's the miners through their hashing power vote essentially vote on what solution they believe in mm-hmm and then and you know some of the things that we're investing in or alternative governance mechanisms like that project tasers I mentioned allows one coin one vote kind of democracy style voting on how the protocol should upgrade so what comes next if this becomes an asset class there should be more players like this in this space and a lot more activity what does that mean like what comes after this I think we'll see a lot of competition as other people kind of catch on that this is a new and growing asset class don't get mad at this podcast makes that happen yes that's totally funded by the time there's lots of competition that means that the poly chain fund has done well but a long-term vision for me that really matters is a web that is owned by the users of the web yeah so you know I think there's massive problems around privacy and data mining and sort of pseudo monopolies that exist on the web today and I think right now we're seeing a wave of technologies that could swing the pendulum all the way in the other direction you really own your data you decide what to share with services and those services are ultimately peer-to-peer protocols that are owned by the users so all of the wealth that's been generated by venture capitalists and entrepreneurs of centralized companies I think will get pushed back to everyday investors who have access to all of these things and that all the value that accrues will go to the early adopters and users of those platforms and protocols instead of to the people that kind of owned these these web platforms part of this is that's gonna have emergent behaviors that show up that we have no way to predict is there anything that you can anticipate though that might happen after that reality hits us yeah it's very easy to compare what's happening right now to the centralized web because it's what we know right but just in the way that the internet was not simply a digital library right you know it actually enabled much much more than that I anticipate that these protocols and economic incentive structures that create sort of swarm intelligence will generate things that we could never have imagined yeah just like the internet generated behaviors we never could have imagined it might sound a little hand wavy but I do think the biggest things that come out of this right now no one has ever conceived that's amazing well Olaf thank you for joining the a 6nz podcast thanks guys yeah thanks for having me yeah thank you

Leave a Reply

Your email address will not be published. Required fields are marked *