 Okay, we're just gonna get started and today the talk or the panel is gonna be about the technical obstacles to why there's no widespread adoption of crypto today. And it's pretty obvious that over the last year or year and a half we went from very few people knowing about what blockchain is or cryptocurrency or Bitcoin or Ethereum to basically the whole world figuring out what this thing is. So we've gotten to definitely widespread awareness but no one's actually using it day-to-day. And so we're gonna be talking about some of the things that are keeping us from using it day-to-day and it starts from like the technology layer that's actually preventing a lot of the adoption today. But before we go on to that I want to do a round of intros and I'll start with you Richard. Thank you. Hi everybody. So I am an engineer who was never good enough to be a developer trained in the UK and Cambridge and then I got into building software companies and did a couple of those quite on the kind of plumbing side of things. One IPO and then one trade sale and then 10 years ago crossed over to the other side of the table and for a couple years was really looking for something that really got me excited that there was something transformative happening to quote Peter Thiel to move beyond the 140 characters of Twitter and get to flying cars and I think that the combination of crypto and AI and Internet of Things is actually possibly getting us there and Fabric is investing in that new way of computing. Vinay? So I'm Vinay Gupta, CEO of Materium. Like Richard also a developer. I said in software then went into energy policy for a while, came back to software in 2014 when I joined the Ethereum team, led launch of Ethereum and now run a company which basically exists to handle the legal side of blockchain adoption. Awesome. Joe? I'm Joey Krug. I'm co-CIO at Pantera Capital. We're an investment firm that solely focuses on blockchain tech and then before that how I originally got involved in the space is I created a project called Augur which is a decentralized prediction market on top of Ethereum. Awesome. And I'm Preeti Kastreti. I'm the founder and CEO of a company called True Story. We're building a social network for experts to identify what's true and what's not, starting with information and crypto and before this I was an engineer at Coinbase and a partner at Andrews and Harvitz. So I'm just going to get started and like I said one of the biggest things that we're facing today is there's lack of adoption in crypto so we all know about it but no one's actually using it and we have these like big dreams for what this thing can do for us but the technology that we have to get us to this big dream basically is at an infancy state. It's like Elon Musk trying to go to space but all he has is a horse and buggy. That's kind of the state we are in crypto and we need to get this horse and buggy to a spaceship to really get to the space. So I want to kind of start off by asking you what do you think are the what do you feel are the biggest technical obstacles that are preventing widespread adoption from happening today? So in a sense I'd say I don't think there are any technical obstacles because I'd really like to kind of reframe the conversation a little bit. Partly because I think we can get dragged into quite a lot of detail about how to solve things like scaling or different consensus mechanisms and that's really work that is probably best discussed in another forum in some senses but in as much as user experience is actually you know technically very challenging to get right I think a lot of that the challenges exist in that space whether it be the fact that you don't want to be trying to remember your seed phrase or remember your private key or have to manually do a test transaction all of that should kind of really be handled by your wallet for example and there are companies like Ajahn and Casar and Keyless and so forth sort of trying to address that but I just one sort of additional point is that if we step back and I think about some of the experiences I've had in trying to build companies and failing for long periods of time to get them to take off the challenge of actually building a product that is you know delicious enough to be adopted by users and then cracking the code of how you take that to market is really tough you have to get consumers to get over some kind of discomfort and create a new habit and in the enterprise space you've got to get to the top of the list of priorities preferably of the CFO not of the head of some of the department so this takes real serious effort and concentration and it's many of the reasons that venture capital exists and that even when all of the tech is baked you still you know get a few victories out of several tens if you see what I mean so I think that's that's my feeling about it yeah no I totally agree like I think if you look at the experiences that Facebook or Instagram or Pinterest or all these consumer products have created for us they've done such a phenomenal job of creating really really good products that the crypto crypto products today are not competing with other crypto products they're competing with centralized counterparts which is a really really really high barrier to meet so yeah so I think that some of this is also a question of time I remember installing the first version of the software that runs the web CERN HGTV on university machines in 93 and I don't think the web became generally useful until about 2008 you know the point where it was just de facto it was a generational process yeah and I think crypto is a generational process and when the kids that are currently in high school trading Bitcoin are you know middle management in airlines that's when you'll begin to see talk of like well you know why aren't we selling tickets for Bitcoin you know why aren't we putting these tickets on a blockchain why is there a secondary market the modes of thinking take a long time to soak into the culture and but we've already seen mass adoption of crypto once which was e-commerce right you know the little green padlock is an indicator of a crypto legal underlying system and that's two and a half trillion dollars of transactions a year right now so it's not that we're not using crypto we're all using it all the time it's just how long will it take people to get more crypto and I think the web browser is the critical battleground I think brave is super important but then who's using it yeah I would say one thing if you look at the user experience issues in this space about half of them I think are things that you can fix today like those wallet problems that you mentioned are solvable today I think the other half though do actually go back to the technical skill ability issues and the reason is if you think of a regular web startup you can kind of get some traction and then run into a skill ability wall wall or barrier so good example this is Twitter for the first like five years of Twitter the fail whale was up on the site all the time because the site kept crashing and they just kept spinning up more servers and eventually they improve their infrastructure and now it doesn't crash all the time but in crypto currency a lot of the interesting applications are actually marketplaces in particular their financial marketplaces and so for those it's really hard to even get to the point where you hit that first skill ability barrier because to market make in marketplaces requires massive amounts of scale so if you think of a trading firm on Wall Street and they're providing liquidity to some financial market they're placing and canceling hundreds of thousands of orders in that market on the scale of a few seconds and so if you don't have the ability to actually do things like market make it makes it hard for even the smallest amount of actual usage to start to take off and so I do think some of these these scalability problems on the tech side do need to be solved before we start to see you know even kind of small nascent amounts of adoption for some of these applications so I mean clearly I'm not just doing a respond throw it around a little bit if that's okay I mean clearly I'm not saying that we don't need to solve those problems but an interesting question would be I mean first of all how many people in the audience are developing what you might think of as a decentralized application I don't know if there's a show of hands have been anybody a few hands few hands yeah how many people are actually developing an application okay so we maybe had 50% of people developing application but I would wager that there aren't many of those people out there who are immediately running into the scalability problem so it needs to be solved but you know they're probably running into the problem of building indeed marketplaces which even even if it's just an information arbitrage business you're going into the marketplace it's tough to get the two sides you know going but the opportunity that exists with you know new ways of representing data that have been bought in somebody like you know blockchain and express through crypto the opportunities there are moved beyond information arbitrage it's like you know it's like entirely new information businesses around you know whether it be real estate or indeed around finance and so those are even harder to bootstrap into existence I think I'm sort of maybe complacent I but I think that the kind of the guys that whether be in theorem or other projects will continue to crack the scaling side of things we need to focus on you know what are the right problems to solve right now and and be really smart about that yeah I mean I don't think either of you are wrong I think both both are problems like you can't like I think if anything we've learned over the last couple years it's that the if like decentralized applications are not the only form of advocate applications right ICOs are an application and that the ICOs made us realize that actually we don't even have the tech to scale this thing that we dream of and so now we're back at the infrastructure phase and we're like okay let's fix infrastructure so we can do more apps I think it's just like this recurring cycle where we'll we'll try to build apps like ICOs or other things and then we'll realize the infrastructure is not there and then we'll fix that and we'll build applications so it's both they go hand in hand I had a very funny conversation with Michael James Young who was the scaling engineer for Farmville and he said in those days you know most of the social games had to provision their own servers but Farmville was an early adopter of Amazon web services so what would happen is the game would go viral then they'd hit a capacity stall then the virality would stop and then but for Farmville they just kept phoning Amazon and asking for more servers so I think part of the slowing of growth is that when you do get a viral spike Ethereum does tend to grind to a halt like crypto kitties right crypto kitties might have reached 10 times as many people 50 times as many people if it hadn't downed the network on the way up and it's not like you could just call Amazon to do some more provisioning so there is this possibility that we're going to be in a position where we get stop starts with virality you know up until we hit a critical mass of scaled solution and I you know I have reasonable hopes that the scaling will come to kind of like a thousand transactions a second yeah I think you could get there without having to do a scratch read right from what would he say today from like 15 at 15 a second now is that yeah 50 to give a context for people yeah yeah 50 second now so that I think that you could get linear approaches to getting to about a thousand but if you want to get kind of OLTP style visa throughput 130 thousand transactions a second I think for that it's not clear that anything in the blockchain space has a realistic shot at that while remaining decentralized right kind of moving on to that like what are the current solutions that you're seeing that people are trying to put in place to solve the scalability problems because there's so many different angles that people are taking at this right now whether that's scaling the chain base chain itself or building on layer two like various layer two scaling solutions are just building whole new consensus engines that are optimized for more scalability but like less security per se yeah yeah I would say you have basically kind of two wide ranges of approaches you have people trying to scale things linearly as the name mentioned and I think he's right we can get a hundred x increase in performance doing that in the other way is is kind of exponentially or quadratically and so the ideas there are if you look at blockchains today they're like a really dumb calculator where if you put in your calculator two plus two everyone else in the world's everyone's calculator all of a sudden does two plus two and shows four on the screen and that's kind of how it works today and so people are working on trying to create solutions where you don't need every calculator to process every calculation somebody inputs into it and those are the ones where you know you could theoretically see hundreds of thousands of transactions per second but they're also much more technically difficult those areas are really still an r&d today yeah and it's fair to say that you know with those approaches you make some compromise with kind of like you know speed to finality of a transaction or the level of security or privacy and there's a trade-off that goes goes on and so the answer in my sense to your question is you know yes all of those things different approaches are going to be play out and then depending on the use cases that hopefully people do focus on as well then you know you'll use the the appropriate backend infrastructure and indeed there'll be an interplay between these different blockchains or ledgers you know so they'll use each other as sidechains if you will yeah yeah I mean I think that with this a lot of this is about understanding what the technology is for like to me the current ethereum main chain looks more or less perfect for DNS type functions it's great for key management it's great for figuring out where an asset is to be found but if you're doing serious transaction throughput you kind of need to move to an l2 solution so because we got it into our head that the main chain is where you want all your transactions to happen we keep kind of banging our head against that rather than using the main chain for namespace management key management you're dereferencing to other solutions and then running the transactions on those systems so I think that this is in some ways a cultural shift where we have to accept that you know if you've got a sort of you know 50 signer subchain that has a whole bunch of proof of authority stuff on it and you've smoothed into our probability with main chain that may not be a temporary solution that may be the solution yeah because a full scale you know full blood heterogeneous parallel computing surface for the entire internet that's a big ass yeah right it's all clear that we can get there from here you might have to go back to the kind of 1980s style thinking about concurrent sequential processes and pod calculus and all that stuff and most of the people that really understand that stuff are 15 years older than me yeah you know it's we're waiting for another revolution but you know it may be a bunch of old guys yeah I think you make a really good point there and that like people culturally are still thinking of blockchain like the fact that people think that everything needs to live on the blockchain and it doesn't like not everything needs to live in this slow censorship resistant immutable blockchain the only things you probably want to keep in there are the ones that require the properties of censorship resistant privacy like those are the like those are the only things that need to live on the blockchain everything else can live off the blockchain and so I think my thesis is that layer two solutions or application specific blockchains make a lot more sense so long as you can always peg to the main blockchain and transact with the main blockchain yeah so personally for example on true story we're building on something called cosmos which allows us to do that it allows us to build an application specific blockchain where we we don't have to be boggled down by the throughput of ethereum because we have our own blockchain but then we can always peg to the ethereum blockchain when we need to go to the main chain and so these are the types of solutions that are being built and I think we're still in early days we're still figuring out what is the right way to do it and what is the wrong way to do it and the only way to figure that out is to keep trying things and testing it out so with that I kind of want to move on to asking you like we're seeing this whole battle happening between like ethereum blockchain and bitcoin blockchain and eos and tezos and so you're starting to see a multi blockchain world and a lot of competition among blockchains do you believe in a single blockchain future or do you think there'll be many many many blockchains so I mean as mentioned I kind of believed in the the blooming of a thousand blockchains or blockchain type you know technologies but that said you know it's clear that you know watching the activity in the developer community and seeing the success of ethereum is is the right set of metrics to be looking at and I think at the moment it's still growing in terms of number of repos the it's the fifth fastest roughly growing on github working of asia would be you know on top in terms of type so I think I think that's you know relevant to look at that but nonetheless I think the whole community within ethereum and we can throw around kind of what size we think it is but it's it's less than a million developers clearly maybe less than half a million whereas you know I would I would say depending on how you define developer you've got to be punching towards a hundred million developers around the world and maybe that you know that's growing quite fast so it's still a small you know it's the foothills of its kind of potential growth so there's a lot of possibilities for that that that position of you know primacy to change over time so so I wouldn't for that reason be dismissing other approaches I think in some ways it's more important to think about the wallets and the underlying chains and you know you could easily see a situation in which you wind up with a near monopoly wallet where you just wind up with one front end and pretty much all of the crypto stuff is in that front end I don't think the users particularly care because you know if you ask them to differentiate between bitcoin ethereum and a bunch of other stuff you know for most users it's numbers that go up and numbers that go down and either your friend takes it or your friend doesn't so I'm I think I'm much more interested in the front ends and the wallets and the development of user interface concepts like I feel like we still haven't had the breakthrough on interface all the blockchain stuff still looks like kind of third-rate banking applications rather than like whoa look it's super computer and we don't even like the first-rate banking applications so I mean Monzo is not bad right but some of the new it's only not bad right yeah it's only not bad like why isn't finance fun yeah yeah and like I think when you think about a blockchain a blockchain is a distributed system in a distributed system you can't get all the properties you can never have a perfect distributed system so you're always making trade-offs whether you want more security less security more scalability less scalability and so I think every blockchain will probably optimize for a certain thing and build for that and so you can't ever have a blockchain that does everything this that's just my thesis and so even though people give for example projects like EOS a lot of shit for being centralized I think it does serve some use cases because maybe gaming doesn't actually require a sense doesn't need censorship resistance and that's enough for gaming right sure I mean I was gonna say that there's definitely of course and this happens if you you know see these generational shifts driven by new technologies you know you definitely get the some of the earlier protagonists and adopters being pretty religious about the use of those technologies in the approach and clearly people have religion about decentralization yeah and definitely what we've started to see and reflect in the way we're looking at opportunities is people who think about not just minimal viable product but when they're getting going but also minimal viable decentralization a minimal viable you know use of crypto you know and and be much more pragmatic about that even to the degree that people might think about from a fresh start I want to disrupt this marketplace I want it to be a very information data intensive marketplace that I'm gonna build but I'm not ready I don't need to use crypto and decentralization day one there's other problems for me to for me to tackle and so you might say for example in a health care use case you might say the most critical thing thing is just the privacy of the data and and I need to use you know some of these fresh technologies for example zero knowledge proofs to to address that particular problem but the level of decentralization that's something that can perhaps evolve over time yeah yeah I think that's a good point was your knowledge proofs like I sort of feel like we're about to enter this kind of Cambrian explosion of new algorithms yeah you know there are there are all kinds of new things kind of crawling out of the mathematical research with the ring signatures of various kinds of zero knowledge proofs and ZK Starks and Starks and all the rest of that so it may be that in two or three years blockchain is one of half a dozen crazy things that happen in cryptography that are now running around on the internet yeah I think that the stark thing is so shatteringly powerful yeah I mean it's a jaw-dropping set of mathematical innovations I think it's going to tear reality wide open um I mean on my front the the angle that we're working on is not so much the trade-offs in in the pure digital domain but how do we break out into the real world so that you could get hold of assets and it's kind of like namespace management right we've got all this work which is happening on digital identity for people a unified digital identity standard blockchain or not would be a game changer same thing for assets like houses and cars and all the rest of that stuff just being able to internet route your way to physical assets to move them around or internet route your way to somebody's legal identity to serve them with papers this kind of stuff that's also transformative for the blockchain space it's scaling but it's not scaling in the pure technical it's reaching out into different parts of people's lives into their identity into their billing into their banking you know into their health records so maybe that it's as much as scaling outwards as just roll more transactions just different areas of life that finally become kind of digital because we did shopping right we did shopping and we did hanging out with your friends but those are that was the last revolution right we haven't had the next jump but i mean on that point of what we've done done though i mean if you look at the the stats and actually try and work out where we are in the potential digitalization of the world and even if you take kind of e-commerce and retail consumer spending it's still it's some you know single digit percent tentage of the of the world's consumer spending two two and a half two and a half percent world trade digitized yeah and that that's because on a lot of those other categories you know a much more complex to digitize and and that's some of the opportunity but obviously you know also the challenge even advertising you know if you narrowly define advertising then maybe it's heading towards 50 percent of advertising spends on online something like that but if you include kind of like you know dropping mail into your into your door and so forth then it's actually again probably single digits so it's much less so there's still a long way to go it's early days i mean the b2b space has nothing at all like you know wearing my material you know we're just chiseling the way on what you need to get b2b to work for e-commerce because e-commerce is practically zero b2b there's almost nothing there you know exactly and we know that there's an explosion waiting to happen and the question is blockchain the necessary enabling technology for that explosion or not and then we think it is yeah and we see some people in that e-commerce space or kind of b2b e-commerce you know again starting very pragmatically so thinking like down to kind of the kind of fundamental building blocks of of e-commerce say like an invoice and saying for some set of invoices in some industry let me re-instantiate that as you know a smart contract and and then start from that fundamental building block and then and then work up and i think you could start in many different places rather than saying you know from from the get go we're gonna just take this entire industry and transform the entire supply chain which is just by definition uh tough yeah i want to say something about ico's actually if i is that okay is it okay so yeah indeed that was an application pretty popular so we talked about adoption it got pretty popular at least amongst a certain segment and indeed maybe presented some scaling problems from time to time but it also of course presented some regulatory problems uh it's i mean we i still think that you know alongside venture you know but also in different stages of investment it's not going away the ability to kind of crowdfund like that but obviously you know regulated but i also think that you know what we see is that there's still a huge potential for different types of financial application that you know what 30% of the world is unbanked entirely and probably 30% of of the US is essentially unbanked even or unfairly banked perhaps and and that we see both on the consumer side and then on the retail side a lot of innovation going on now driven by the crypto capital markets when people realize that they can do things differently i think we shouldn't just like you know forget that bit and then be hunting for supply chain or medical use cases we can do that as well but let's not forget sort of the capital markets yeah absolutely i want to jump into a couple questions from the audience and one of the ones is asking when do you think it's too late for mainstream adoption when yeah i mean like over a decade from now it's it's still so early i guess one one thing on the point about you know people are saying what you can make the trade-offs between you know scalability and security and all this stuff i think we're so early that it's almost slightly too early to even discuss those trade-offs it would be like talking about you know trade-offs on the speed side of the internet back before dial-up even existed back during that time you should have actually just focused on laying more wire and solving the the huge infrastructure problems as opposed to making trade-offs and so to route that back to you know when was it too late for mainstream stuff i mean this tech is still so early i i view bitcoin almost as like the arpanet equivalent in this analogy you know big ethereum is maybe the very early internet and ethereum has only been out a few years so it's it's it's still like day zero go for it to build on that i would say that it was about 85 that sysco systems was founded i think a couple of stanford folks and and they and they i think as i understand it if you want to boil it down you know the ultimately sysco sold most of the kit that went into building the internet that then enabled the web as an application running on top of that but at the time the thing that stood out was that they made these relatively cost-effective multi-protocol routers now who gave a damn about the fact that they were really good at building multi-protocol routers you know then let alone you know ten ten years later but we're still at that stage yeah that people are building those kind of those kind of products took ten years then to get to the net scape scape ipo i'm going to take a slightly contrarian view and i'm going to say i don't think it's ever too late because i see i see crypto playing out in two ways either it does get to mainstream adoption or it remains very very niche and i don't think it i just don't think it ever dies though if that's the beauty of this system it just never actually dies to to um badly misquote kind of Arthur C. Clarke or whatever it was maybe another way of thinking about it that all the kind of most powerful technologies are you know most useful once they become invisible and that you want to move into the kind of Carlotta Perez phase of deployment in these technology ways where it's it's just super widely spread so it doesn't die it just kind of just becomes ubiquitous yes kind of like ai yeah like ai has just like creeped us underneath us without even realizing it's not like you you go in and plug into an ai solution every day just it powers your life and maybe that's how crypto takes off too it just creeps into our lives slowly well it raises an interesting point because that that is the case about ai and there's a lot of debate about ethical ai and about whether or not you want to allow ai to pervert our emotions and therefore possibly political outcomes and i think part of this whole discussion is how the combination of these technologies can you know make sure that this new wave of services that we're going to indulge in and develop habits around actually are you know kind of good for us individually and families and countries and whatever awesome we're running out of time so we're going to end it there thank you so much for being on stage with me thank you