ep 1 – Methods to the Madness

Transcript

welcome to web 3 with a6 and z a show about building the next generation of the internet from the team at a6 and z crypto that includes me your host sonal toxie former showrunner and long-time host of the asics and c podcast network and now editor-in-chief at asics and c-crypto this new show is for understanding and going deeper on all things crypto and web3 including by offering occasional data readouts and insights directly from the leading scientists and makers in the space but in these initial episodes we'll start by setting some quick context before we dive deep the rest of the season on topics ranging from auction design and mechanics nft security zero knowledge gaming decentralized media to economics history infrastructure

roadmaps and much much much more in the form of everything from interviews to oral essays and will always as is a signature respect your time and attention so stay tuned for today's episode and to kick things off we did an asics decrypt a hallway style conversation on the markets mental models for thinking about crypto and briefly touching on methods and metrics as well as sharing key trends that are top of mind our guests include chris dixon founding general and managing partner at asics and z crypto and eddie lazarin head of engineering at asics and z crypto both were also co-authors along with darren matsuoka who led the data and robert hackett who led the editing on our recently released state of crypto report which we refer to in this episode

and that you can find at asicscrypto.com state of crypto for those that are interested we also cover some faqs and methodology that we didn't share in that report at the end of this episode after the credits along with some behind the scenes finally just a reminder that none of the following should be taken as business legal tax or investment advice nor be used to evaluate any security furthermore companies mentioned are not representative of all asics and z investments nor are there any guarantees of their performance or future results please see a60.com disclosures for more important information including a link to our investments and with that let's go

all right so let's talk about where we are so it's may 2022. crypto has been around for something the order of 13 years we just got through what seems like a big bull market a lot of investment a ton of new companies formed it seems like assets throughout the world have declined particularly risk-on assets as they say so high-flying tech stocks things like that although the overall you know markets seem a little spooked people are worried about inflation you might be entering a recession so that's kind of the broad backdrop and anyways what i've been thinking about a lot lately is now as we enter this new what appears to be a new financial environment what's going on in reality in the layer that really matters which is the kind of technology product

layered so in the 10 years i've been in crypto there have been two this might be the third major downturn we'll see the third winter the first one 2014 1516 the second one 2018 19-ish a few years ago we asked the question these cycles seem kind of random to outsiders but as insiders who live and breed this we felt like there was kind of a pattern and specifically the theory was that what would happen is that prices would go up that would generate more interest more press more social media activity leading to more entrepreneurs developers creators entering the space new products being created planting seeds that would then

develop over the next few years at some point people got a little bit over excited about some of these genuine kind of tech breakthroughs but maybe they were the speed at which they would mature was overestimated the markets dropped again but very importantly you had the siege planted new software new products new infrastructure new companies formed which then kind of created the next wave and i think we saw that you did all the data analysis an interesting thing that came out of it we did see and we published this in the blog post and we published it again recently i said crypto report we believe that we saw like a real causal pattern there during the last winter it was a bit over two years ago yeah yeah two years ago and the analysis was like basically if we're right all of these seeds planted in 2018 specifically a lot

of sort of infrastructure solana avalanche near this big wave of l1 l2s optimus etc a bunch of applications fifa unit swap compound etc nfcs openc dapper those would eventually mature get people excited it's kind of what happened right yeah yeah it's interesting is that when you're seeing everything in the flow it's pretty clear to me how many new seeds have been planted over the last couple years and it makes a lot of sense why it should take some time for those seeds to germinate so to speak take for example the scaling of solutions that we're seeing it's just starting to grow now right like optimism barbituram zk sync like they're at their earliest stages but you can see publicly on chain

they're just starting now to take up about one percent of the on-chain block space in ethereum they're just starting to be real and these ideas like optimistic roll-ups and zero-knowledge roll-ups and so on we started hearing about them years ago yeah eddie quickly explained optimistic roll-ups and zero knowledge or zk roll-ups so optimistic roll-ups and zk roll-ups both scale their underlying blockchains like ethereum an optimistic roll-up the basic idea is that you can aggregate all the transactions together into a single checkpoint and put that onto the underlying layer one blockchain and create a period where anybody can permissionlessly come along and check

that the work was done properly if the work was done properly there's the problem after some period those changes are accepted but if there is a problem someone can construct a proof and it's called a fraud proof it's like a cryptographic object that shows definitively that something was calculated incorrectly and if they post that fraud proof on to layer 1 they actually get a reward for doing so so the way that the game theoretics met out is that if there's a problem someone out there is incentivized to find it because it's free money and if there's no problem well as long as the sufficient amount of time passes we can all kind of trust that yeah there probably wasn't a problem because someone would have wanted to get that free mind right hence

the optimistic i love it by the way this is a terrible analogy this makes me think of like one of those lame questions people would ask each other do you give trust by default or do you wait for people to earn your trust optimistic roll-ups remind me of the former model where you give the trust up front until proven wrong it's like how humans engage is this model essentially that's exactly right i love the navy crypto me too and i also like it because i think people forget when we talk about systems of code that they're very human systems like at the end of the day code is just a way for people to create and transact and share with each other oh exactly exactly right and that's literally the purpose of all of this okay zk roll ups so zero knowledge groups are about proving that a fact is

true without having to reveal the fact in this case you're revealing that the computation occurred in a valid way without actually revealing all the computations so a ck roll-up also aggregates a bunch of transactions into one checkpoint or one piece of data the difference is that it comes along with it a special kind of cryptographic instrument a specific cryptographic proof some call a validity proof which is a proof that actually all the calculations that are alleged to be in this piece of data we're done correctly and that's a very kind of counterintuitive idea like if someone says they added two big numbers how do you know that they're right without having to do the addition

yourself of course you could check their work by doing the work for yourself but it's much slower to do that especially there's a lot of complex calculations in doing this it could take you a very long time to replay someone's work especially if for example it's many thousands of transactions that occurred all over the blockchain from all over the world so instead of having to calculate it all yourself if there was a way to construct a proof that it was done correctly so a very short quick and dirty way to verify that everything was done and you could probably accept right away what was alleged and that's how zero analysis are used to scale an underlying blockchain it just takes this long for them to start to work and it will probably take some time for them to

become the predominant computing layer for ethereum like and that's kind of been the design all along just like for example the migration to proof of stake for ethereum that's not just some recent trendy idea it's been in the works for years probably five or more years at this point it just takes that long for the incremental technology and like research innovations to take place and it takes a long time because it's complex and also because these are decentralized systems you need a lot of people to come together and agree on them exactly right so from my view when i look at how everything's going it feels like despite financial turbulence things are proceeding according to plan in the loosest sense of course there's no specific plan yeah these are decentralized systems

with tons of different participants with tons of different motivations and interests and incentives and it's chaotic definitely when you're deep in it it's chaotic but when you kind of zoom out you can see that each incremental piece of tech that we've needed to make this kind of thing work or make that type of thing wherever even to be able to experiment with this new type of application like they're all coming in like one at a time and it feels like sometimes the excitement can outpace the level of tech development but i actually think there's a reason for this too the fact that people can buy these things and can own a part of them and that they can trade freely it contributes to the ability to talk about it the prices can change like all these effects are things that weren't even possible

in previous generations of the web because the only way that people could get access to these things before web 3 is either by working at the company being a privileged investor or starting your own company and the fact that anybody can buy these assets now means you get a little bit of the chaos one expects from democratization yeah from just broadening access to wider groups of people that creates turbulence it's also part of what makes web 3 is that everybody can get skin in the game everybody can be an owner everybody can participate crypto has always been retail led and institutional followed like the institutions are always way behind my mental model is retail is very good at macro and bad at micro and institutional is very good at micro batted macro

that's interesting say more part of the point of this episode is mental models i want to hear more about that and how it fits here sure retail gets the big idea so retail gets that like crypto is a big thing retail meaning like you're smart software developer in ohio who like lives and breathes code all day gets why a blockchain's a big deal right in a way that a banker in new york doesn't and so they get the macro they get self-driving cars they get ai they get the macro micro is very hard because they don't have the diligence teams they don't have the access they can't go and say okay here are 10 layer one block chain i did deep work and i can tell you which ones are better from a distributed systems point of view how could they they're like behind the

keyboard playing a game i think much more advanced on the kind of zeitgeist side and just understanding where the future is going than the banker in new york but doesn't have the tools right the banker in new york essentially graph lines through charts don't really have vision don't really understand technology almost none of them have knowledge of computer science for the internet but they have teams of people and they're able to say this lidar company is better than that light our company this ai company is better than that and so i think one thing with crypto is because it's been retail led they i think are directionally correct get the macro get the big picture get excited by new concepts but they also can't distinguish and frankly the regulators have decided instead of going after the bad projects they're gonna have good projects and as a result you

get people who are directly correct they're excited but some of the projects they choose are bad some of those things blow up and then you get this kind of chaotic up and down cycle that we see but one that i think is directionally long-term correct but chaotic in the short term the analogy i've used uh is sort of flying a plane with instruments versus through sight i've never actually flown a plane so this is very much a metaphor but if you fly when it's foggy or something else you know using your instruments and when you use your instruments you have to kind of trust the system right and you can't kind of rely on your senses in my experience if you just sort of trust your senses and by senses i mean the news you read the people you talk to your kind of instincts it's very easy to

get caught in this up and down herd mentality cycle of markets this is not by the way a new idea but only ipad it's a common idea if you read people like howard marks and others who talk about market cycles and sentiment what happens in these kinds of cycles is in a good cycle in a positive cycle every piece of news gets interpreted positively everything is kind of self-reinforcing it's sort of availability cascade where you know more and more good news it's more and more good news in a negative cycle it's the opposite and so a very important thing to do i think when you think about innovation entrepreneurship the future seeing where the world is going is developing frameworks to parse information in the world look at patterns look at the evolution of

products i always talk about sort of computing cycles a very important kind of reimportant feedback loop between infrastructure and applications so you know the first iphone came out in 2007 the app store 2008 kind of the golden period of app development snap instagram venmo uber were all created with 2009-11 that created a whole bunch of new use cases for phones which in turn gave apple and android samsung etc you know they sold more phones they had more money to invest they built better phones better phones led to more apps you had this reinforcing cycle understanding things like where are we in that flywheel cycle of the computing cycle is very very important and so that's the instruments and what you see around you is this show of people freaking out and panicking or in the

flip side everything's wonderful and we're all geniuses you know keep focused on the fundamentals on the technology on the products on the entrepreneurs and the where's the talent going that's the kind of instruments in my mind and that's for me has been a kind of mental model that has served me very well is developing those frameworks and then sticking to them and ignoring the stuff you see around you i love that mental model this is not like a normal technology cycle because it's not just a matter of new technology leads to new types of things that people can build and it takes some time for them to find that new tech and build on it and incorporate it and learn it and add to it and then that creates new things that kind of serial process the fact that you can put price into it

because these are assets that you can own and that you can give to people and that you can receive through doing interesting things when something is permissionless and open and exciting you can get hordes of people who are super interested in the thing barging down the door and that's a powerful super interesting effect you also allow unusual and new types of talents to be attracted to it you know like chris was saying the engineer in ohio who just thinks about code all day maybe was never going to work for a big investment bank or for a major silicon valley company but is a brilliant engineer nonetheless they may be attracted to this and contribute and benefit and have an impact that they

never could have just because of how open and permissionless it is so those two super powerful forces are at play in a new way and that creates chaos in the short run in the long run i think it creates more innovation at a faster pace you know i was going to add one thing to that which is i think it creates more composability and opportunities for composability because when you think about how communities form around lore and like storytelling able to exchange ideas around the frenzy and excitement of an idea like the memes frankly yeah i was even thinking of an example like loot and the activity that is entirely enabled by having a community and that kind of excitement so it's not just an either or something in between two yeah exactly this is like an incredibly

fine edge yeah you know like this is a incredibly powerful instrument this ability to create growth by like making assets that people can speculate on and you can get as rewards that is an incredibly powerful tool that we don't even know how to use yet i firmly believe that the next wave is going to be people thinking much more carefully about designing sustainable types of growth and using the incredible growth tool yeah but thinking very deeply about retention and keeping people in the system and yeah i think of token rewards as like mentally i picture it like you blow up a balloon and you let it go and it flies around the room and it zigzags and yeah yeah and it's like i'm not a rocket

scientist either nor am i a pilot but i imagine people that originally built rockets and planes had the same problem right and then eventually i don't know if they pass around the room but i'm sure they didn't go in the direction they wanted to and the whole thing was like how do you take this incredibly powerful new force and kind of control it and make it go in the direction you want and i think we're in that phase of grip there right now yeah yeah like it is a nuclear level force the power of taking people who have been disenfranchised who sat on the sidelines who basically effectively you know 98 of the people on the internet are left out of the internet economy everyone who uses facebook is not really part of that economy but the question now to me the most important question

that we'll be discussing the next couple of years is how do you take that new kind of rocket fuel that's right now in some cases flying out of control and doing crazy things and leading esteemed publications to constantly say that crypto's a bunch of ponzi schemes and everything else and how do we take that powerful force and harness it to fulfill the web through vision to me the web 3 vision is how do we build new networks the killer app of the internet's networks this is what you do with the internet the internet is a system upon which you build new networks you can build a network like facebook or twitter or email smpp the web and i believe there's going to be 10 000 new networks created in the next hundred years how do we use tokens to create new networks where the people that do the work of

building the network own and control the network instead of the company and the people that fund it and the founders of that company in a way that builds better networks that are more inclusive and that incentivize long-term positive behavior and then help a bunch of networks that previously there's this very hard thing when you're building a networking beginning the cold start problem the chicken egg problem the initial hum and one of the very powerful of the tokens is we can now use those to incentivize users to help those networks get to critical mass yeah i mean a question i'd ask someone who is skeptical of this is like how do you design a network so that enforcing and encouraging pro-social behavior yep is native to the network

there is no way to do that in a network that is just owned by someone i'll do the other side let me just defend it like the joel scroll ski at stack overflow would be the closest on the other side you use games you know badges maybe wikipedia so there have been a few cases where it's worked i think in the kingdom of a benevolent dictator like it's just fake let's just admit it like the web 2 narrative uh people feel like they're part of these companies and it's a community it's a fake narrative they use these words like community and it's all like no one actually feels outside of the venture capitalists like us and the owners and a few other people feel like they're owners of those companies right has anyone ever shown up to a convention and said you use facebook i use facebook oh my god for

both facebook users that's awesome i feel a sense of kinship to you the way that they just showed up to garyvee garyvee had a convention in minnesota 10 thousand people show up at the football stadium to talk about the francis and the t project to my knowledge i think the first time in the history of the internet that's ever happened that consumers not developers not fa not salesforce stream force consumers have flown across the country 10 000 of them talked about the consumer internet product they did it because they have skin in the game because it's like it's a new thing that people finally feel these people they felt left out and now they feel like they're part of it and it's a very powerful force the other thing i think is that how many scale networks do i mean right now like 20 30 there aren't

that many yeah and there's like a million things you can imagine doing on the internet including like just an example healing the crypto protocol that is trying to build kind of a grassroots telecom an alternative to verizon and 18t and wouldn't it be cool if you could build an incentive system so that people would put up the transmission hardware on their house in other places they've done half of it they've built the supply side like it was interesting and that's a network that very likely could not exist without tokens and there are many others that are probably out there that haven't been created yet actually going back to this idea of networks it's not just networks as a networking and networking with each other and going back to eddie's point about what makes something inherently pro-social and creates kind of these multilateral just like really like not

like hub and spoke interactions where you have like a central player it's all mediated through a platform like facebook groups people can certainly interact with each other in those groups but it is still in as eddie said the kingdom of the benevolent dictator yeah and on top of it they don't control the moderation policies they don't get to set that like maybe at whatever levels that facebook decides and pre-determines they get to pick down from some pre-designed configuration they may be able to tier it slightly the permissions and settings for their group but that's about the max yeah they can't decide like a group can't decide on facebook hey we want to reinvent the breastfeeding groups or whatever groups there are they can't reinvent that from scratch and so i just want to put in context what you mean about the ability to do that and that goes to your point

about community because like this weekend you tweeted this line which i thought was really profound that a community isn't just when people have an affinity for the leaders that's celebrity it's when they also have an affinity for their peers yeah well that's what i was saying about like the facebook line like you don't go to a bar and like oh my god use facebook i use facebook where you would go to a bar and say oh my god do you have a court eight five at board eight oh my god do you experience your the friend i'm a b friend and that's i think directly correlated to feeling a sense of ownership in the community and nfc is an example is the first time that people on the internet have felt that way and that's why no one in 30 years of the internet who shows up to conventions for

facebook google some of the youtubers do because they're making money again skin in the game right if you have skin in the game if you make some money if you feel like you're part of it you'll show up so far the consumers have been left out of the party and you know they're just the products right they're the ones getting surveilled getting shown ads now they're getting to participate this is the whole promise of web 3. i mean i think the only other communities that feel this that weren't mentioned as like game communities like video games yeah but that's also clearly because gamers for really interesting reasons they feel sense of ownership of games even though they don't right they don't have any ownership but they feel a sense of ownership is it status status it has something to do with status something to do with just being devoted

to the thing and like spending all this time like incredible amounts of time and having in-game assets isn't it kind of sad that in the end it's like this fake love affair and then the game winds it down it gets unpopular and they're like one day i used to be really great at some game well what they do is they go fight with the game publisher for years to get some kind of carve out so they can maybe spin up their community server that's like underfunded or they go off market and to like secondary markets to make trade and do more and build community like everything's happening outside of the game because there's such a need for that exactly they do either one of those things my friend eugene way he wrote that famous post about status as a service he has this whole interesting theory about what happens

when there's a loss of status and that we don't talk about it enough and i think that really applies to the web 2 versus web 3 analogy like we haven't actually spent a lot of time thinking about that as well in these networks that's very interesting by the way just one more line that dixon had this weekend there was a guy who responded to his quote about what's a community which i thought perfectly described what you guys are talking about yeah he said the difference between a community and an audience is which way the chairs are facing and to eddie's point the pro-social version of the chairs face each other but in the web 2 version the chairs are facing one way like it's only bilateral you face towards or against this audience fan versus creator or like look at the tweets about vcon this weekend

there were a few tweets that showed people looking up the stage but the vast majority were a bunch of people that had never met who were internet friends all meeting together and like it's the same idea right it's peer-to-peer not peer to leader exactly one question i want to ask you kind of the segway here like when i think about crypto i think of a lot of analogs to open source and og open source communities which i spent like almost two decades studying and writing about and editing and i have a question for you guys which is what's different now a and b what can we and can't we borrow from like open source to now so good question first let's talk about the history of open source so open source started in the 80s with richard stallman mit it was motivated by extreme anti-copyright

sentence he created these famous sets of unix tools gnu gcc i think a whole bunch of other things and then kind of in the 90s was i think mostly dismissed as a kind of left-wing political movement and then in the 2000s and 2010s morphed into a very major technology movement i think today my guess would be 99 of the software on earth is open source almost every server you run your android phone is linux a lot of your iphone is open source so it's a very interesting story which i don't think has been properly told first of all i think that the main reason it won its couple of things one and more from political movement to the technology movement and to fit strategically in with a lot of big corporate interests so the biggest

funder of linux is intel intel funds linux as a counterweight to windows and other proprietary operating systems that run on top of it it's a complicated kind of strategy question as to why so that really works but i think the most important thing is what we call composability so composability is the idea that you can take a piece of software on github or some other place and fork it copy it and build another piece of software on top of it and assemble the software together the way you would lego bricks and this allows you the community the world to develop software in a much faster way yeah how is web3 how is open source in crypto different from open source before one immediate path that's clear to me is

that in a crypto project when you're working on a crypto project in open source and this isn't true of all crypto projects but it's true of many printed projects that they can eventually become a decentralized protocol that people can have access to those tokens whereas an open source project by default pre web 3 is not actually an investable project it may become a super popular library that people love that spreads all over the internet and becomes like a critical piece of the developer's toolkit but that's a different kind of result then something's becoming economically very meaningful that it creates a lot of value that a lot of people can use just a fundamentally different thing like if you look at for example react to like facebook's open

source front-end framework or look at like requests like the open source library for python for making http requests like those are huge libraries it would have been very interesting to track their activity but as far as i can tell react is only monetized indirectly and as a result of i'm not even sure exactly like how facebook benefits from it and then request hasn't at all just subsists on a very meager diet of donations whereas open source crypto projects they can potentially become something that's self-funding or community owned so on and so forth it's just a totally different sort of evolutionary pathway they can become much more sophisticated much more advanced starting from the same early form just a little bit of

cone will contribute to i'd like to say that composability as the software is compounding interest is to finance the same way in finance you put a dollar in and it grows and it grows and grows and it's kind of seemingly magical way because every year the interesting counts upon the interest in the same way with composability the software compounds upon the software and i think it's the core driver of open source so crypto of course almost everything in crypto is open source it's a ethos it's lined with the movement there's weird cultural tensions that a lot of current prominent open source people actually are anti-crypto i think a lot of it's some sort of weird mix of generational differences and political differences i think crypto's perceived to be slightly right-wing

because of the history of bitcoin it's not that's a non-true narrative but a narrative nonetheless i think some of it's generational crypto is a new thing it's threatening it's different they dismiss it for a bunch of reasons they're culturally very different they're ideologically very aligned in that they believe in open access permissionlessness democratization of knowledge and ideas i think over time they will come back together as generations evolve change retire as the technology kind of rises above the politics you know in the early 2000s you're in technology you read about microsoft i read everything about my sophomore mind use the word linux to my knowledge like in the antitrust trials all the other things never came up linux

was ultimately the thing that dislodged windows and that's 99 of the servers in the world of linux not the windows server it's now like i think a fairly niche server at the time it was like java and all these other kind of corporate competitors it's really remarkable how it went from mixed political movement to irrelevant to a toy to by far the most dominant software in the world it's very underappreciated even by a lot of technology advocates like exactly what was that driver that made it go from this niche political movement to 99 percent of software in the world i think it's an undertold and under-appreciated story i would argue you can't begin to understand software and technology if you don't have a good explanation for how that happened i was just going to add that code like content

all these things it wants to be free it wants to be distributed across the widest set of people people to use it people to understand it people to build on it that's the natural end state of complex and interesting systems web 3 makes it possible for people to do that but then to still maintain some ownership of the underlying fitting one of the things that makes me so excited about web 3 as a movement is that it has been too profitable and too useful as a business tool to make opaque the most important types of code and because of the new kinds of business models and the new types of ownership and ways of thinking in web 3 that it just seems much less likely that the companies of the future will want to

hide their code i agree i'd add a third thing because based on my studies of evolution of open source i feel like i could write this book someday but i would add dows to this as well like decentralize autonomous organizations as mechanisms for organizing the communities and activities and that's very native to crypto and the reason is when you look at the history and culture of open source and how it's evolved a lot of it boils down to how the communities were organized as very big kind of monolithic entities and then they would fork and have like different ways of splitting the communities they'd be very religious wars all very interesting to follow but the other big movement that kind of came around in the last decade and a half was obviously like this move towards microservices more modularization and daos are another way of organizing like smaller groups to

transact and create culture essentially and transact and change governance models around code and how to use that code and what to do with that code and so i would actually add that as a third kind of cultural thing to this as well that's going to change the way open source comes into the age of crypto totally agreed i think we're still in the earliest days just because forget the tech forget the tokens and the blockchains and multi-sigs and dow votes and staking and all this type of stuff like just on a pure cultural level it's a new thing for groups to be able to very quickly and easily align their incentives around like a specific specialized goal and manage like a pool of capital that's like global and transparent like all these types of

things on a personal human level are sort of new yep and we haven't quite developed the cultural technology around like what it means to be a stakeholder and how to get people aligned and how to measure people's participation these are sort of totally non-tech things but they're important prerequisites i think of being able to advance the tech that's one of those things that might just take generational time to resolve but i think is inevitable that it will be resolved yeah i agree 100 for me one of the questions that does come up is when i look at our data we do a lot of talking about developer activity but yet it's not only enough it's open source has had plenty of users is developer activity enough of a proxy to say like this is going to be a real thing because you

know open source has a lot of activity yeah it's a really challenging one actually because you need to classify all the repos on github that are crypto related right not every crypto related repo is just going to say it's crypto repo so we use electric capitals crypto ecosystems repo on github as a mapping we've contributed a little bit to it and extended it made some modifications but we're building a bunch of internal tooling to be able to quantify live developer activity across all these different crypto ecosystems and that'll mean that we can be way more granular about showing how many monthly active developers are there in different portions of crypto the second question i would ask you you did a lot of work with darren on this too but like in this

recent update which is essentially an update to the 2020 post the state of crypto what changed methodologically to account for the new era of this current ford cycle of crypto well one example of a change that was really big but worked out and we spent a lot of time ringing our hands about was in the original blog post the social media activity section was actually reddit comments and the way that that worked was i collected every subreddit that was just clearly a crypto subreddit and there's some that are kind of on the fence i didn't count those just the ones that are just obviously about crypto yep and then just counted all their comments and just like this is just a measure of kind of engagement at super high level not

trying to be cute about parsing each comment just let's count the number of comments and that was really clean the hardest part was just collecting all the subreddits which is actually not that hard like if you can query all of reddit data it was only several hundred subreddits at the time in this case we actually moved away from reddit and now the social media activity is all twitter all tweets not discord not discord and the reason discord would have been good the problem is that discord's api is difficult not just for pulling messages but also for finding all the relevant discord servers and then joining them discord is less public right it's harder to find discords not all are public yeah it's more like a dark forest if you use

the analogy of like the fermi paradox and reddit is more like a public directory exactly and so is twitter it's like they're very public you just go there you see basically everything by default we looked into using discord data but it's just really really gnarly to get that the twitter data is way easier you just need to query the api the hard part in that case is you have to pick the terms that you're going to use and of course you might be tempted to use terms like token these days a lot of the uses of the word token have to do with crypto but it's far from 98 percent as far from precise enough to be really really tight about the data so we curated a list a non-exhaustive list by design of keywords that are just almost certainly crypto related like

ethereum zk rolla optimistic roll up all these different terms and the idea here was not to try to capture every single tweet that was crypto related that's impossible but instead to really tightly quantify some lower bound of tweets that are definitely about krypton the hope there is that you'll actually more carefully capture the waves the ebbs and flows of interest super interesting yeah because otherwise you're just getting a lot of background noise and like the ebbs and flows may be a function of twitter generally so we went back and we segmented different years to find as many keywords as we possibly could from each era it's actually a really funny thing going back to like the

2018 era and seeing sort of the lack of certain terms even when crypto kitties is a thing nobody's really talking about nfts in 2017. one thing that's funny to me is like terms like dapps when they first came out app coins and dapps were a thing decentralized apps now people do use dapps but i think intuitively tell me if this turned out true far less than they did before oh yeah way less also aiko right ikea was like an obsession oh right like initial coin offering right like my god now no one mentions icos at all which is great because it was also like i was about to say it's a regulatory nightmare on that one it was a total regulatory nightmare and a horrible idea yeah frankly even annotated an adverse signal so you see these patterns and of course as the

space has grown there's been a ton of new language so you know for better for worse a lot of the language is of course more contemporary language and that's because there's more new language just because it's bigger like the wagners of the world and stuff we're all gonna make it speaking of are we let's talk about what do we think's gonna happen like let's say there's a downturn what does that mean dad i've been thinking a lot about it share your unfiltered thoughts it's so hard to say like on the one hand i just still viscerally remember what 2018 2019 early 2020 were like and just how there was still all this cool stuff but the space was just so quiet and the normal conversation i was having with other people was like oh isn't crypto

that thing that happened and then like ended like a year ago it was like the normal conversation i can totally see the market cooling off especially there's accompanying sort of macroeconomic changes but i find it just so hard to imagine getting back to a product mindset that's the same as back then i think it was 2019 i remember sitting with you and maybe a lee or something at the end of the year and there were like very few products launched at all if any that year it was particularly like this famous quarter and everything's going to launch it didn't launch it so what happened right ethereum 2015 you had that ico thing and then you had people say hey ethereum's cool let me riff on it let me mod it essentially which is the other layer ones layer twos

so that was 2017-18 and that all kind of came to fruition around i'd say 2020-ish 21. yeah and today it's still happening when you and ali and eddie were at the end of that last cycle and you were looking at the like not anything new had been built were you guys actually bummed like what gave you the conviction not to be depressed oh yeah i was but also we knew it was in the pipeline like we invested a bunch of stuff we were waiting for the launch yeah so nothing was out there nothing was being built that'd be one thing there was nothing had launched like there's two kinds of blades and software delays like it's never going to launch and delays like it's actually delayed luckily it was the case and we believed it was the case of the talking that it was the latter kind

but they had just disappeared away so it turned out to be like these are hard systems to build yeah and it just took longer i remember that time nothing had launched it was like ethereum was the only fully programmable yeah watching at the time and there's all these other ideas for other ones and they just kept getting pushed back we were actually meeting with a lot of people who wanted to build and kept asking it's like oh where should i be building all one of the cool things you see launching it's like oh man it's kind of awkward there's all this cool stuff but it's not out there yet it's like just wait please you know whereas now now there's a hundred things you can point them to yeah this just feels like a totally different time the only thing that was live at the time was

like maker this is literally i think the first time in human history we have a dow that's persisted this long with this kind of governance in practice like i don't think there is any other example this is it totally oh no no absolutely not no this is absolutely the leading example like the difference now is that we have to have minimally viable infrastructure you know it's not perfect but we have solana we have intel one oh yeah like new later ones like avalanche solana solo all these new layer one networks as well as new kinds of applications like dydx was in this category we have layer twos they have avalanche we have a bunch of other kind of infrastructure polygon we've got a bunch of stuff which you know they all have pros and cons but they work in a way that they didn't

three years ago number one number two we have a lot more we have open c we have unit swap we have d5 and these things all really work well and have significant users and usage but then we also have just like a lot of really great entrepreneurs building at the app layer on top with a much faster product cadence like they'll be able to release products some of the games nft projects just sort of social media like they have a three to six month product cycle they don't have to go and figure out these incredibly complex things you have to do when you build a layer one they can just go build an application it's much more like building a mobile app we're kind of getting that phase where it's sort of like building a mobile app i guess and therefore i have no idea what the prices

will do i think that the products will see a lot of activity in the next 12 to 18 months and i think there's a scenario where it's actually good a lot of these products have launched in an environment that is not so financialized yep definitely where the incentives can be much more around behavior and true economics yep so maybe i'm an eternal optimist but i actually feel like this is all a positive thing and that things were getting kind of silly maybe a month or two ago and some really bad projects and again like the regulators only going after good projects never the bad projects exacerbate this and so we had a whole bunch of bad projects and speculation and now maybe we'll get back good fudges yeah i totally agree it's so easy to make a token and that token can represent ownership of the thing that is

just new you can do this really interesting thing where you can distribute that token as a reward to the early adopters of the system or the protocol or the dap or whatever and that can create and when you compare that kind of growth with web 2 growth even like paid acquisition growth where you're just piling money into marketing and throwing users into the top of the funnel when the market is really really hot that web 3 growth case where you can fund user acquisition with the token is set in its most crazy most overblown hottest form which makes it really really hard to disentangle like real authentic growth so something i'm excited about is if the market cools off

hopefully that effect is dampened a bit and also hopefully entrepreneurs and new projects rely less on it and experiment in other more creative more sustainable ways it goes back to your point about retention and focusing more on retaining versus only initially engaging and that actually ties to what dixon just said about how you can actually innovate more behaviorally because you're focused on what people really want to do versus what they think they want to do just because it's a way to like maybe flip something or make more money like moving well past the financialization so i do have to ask you guys both here how do you think about this idea that community precedes product versus the other way around and it's kind of along this line of what you're talking about you have this token as a way to quickly

incentivize people get excited about something etc but even beyond that how do you think about the difference between honestly a bunch of hot air and something that's kind of real like do you have ways of analyzing it looking at it measuring it yeah that's such a great question so about community proceeding product part of the reason for this is creating hype and buy-in of course but there's good reasons for this too one is that the more attention that a project gets before launch the more contributors that it earns and because of the open source ethos and the sort of partnership and sharing decentralization ethos so common in web3 you just benefit from more attention eyeballs whenever i look at a new project and i

want to kind of sniff out whether or not they have a real community of course i go to the discords i go into the githubs i like message people i ask around i look at who's talking about it in other communities and you can always tell that something is a high quality project not because of just the pure hype and the people talking about it but the people thinking about the core ideas and the tech in it and the things that they want to copy from so that's kind of a qualitative measure quantitative measures for better for worse or a little bit easier to gain i think thank goodness we don't have at least yet a huge problem with like fake github bots starring rebos i've heard reports of it and i've looked into it but it seems like the problem is not

nearly as bad as say fake accounts on twitter maybe i'll leave it at that dixon would you add anything to that the whole community preceding product how do you think about it well i think it's also like token product i guess the web 2 model was always build a product with no money in it no monetization make it popular like instagram and then get acquired after or bad advertising later like snap one thing i learned working closely with uniswop is if you don't have a token early on the community kind of rebels they want to feel ownership they want to feel like they're part of the project they want to have skin the game and so this old playbook with first building the product and then building the community the modernization is reversed and it's a broader pattern in web 3 that's one of the reasons that a lot of

entrepreneurs and others don't understand what's going on in web 3 is that i'd like to describe it as an alternative planet where there's different laws of physics and so things like all of these kind of lessons that we've learned over 10 20 years on the traditional internet are really flipped in web three so for example what i just described this idea that you first go build a popular product and then you layer in kind of the community and the skin the game and the modernization that's flipped the idea that in web 2 interoperability is a pain in the ass like why would you interoperate and so like for a long time and like 18 years ago there were these things called mashups and apis and other kinds of things which basically all disappeared there's almost effectively no consumer apis anymore on the internet like there's stripe and business ones

but not consumer ones they all went away and they went away because interactively turned out to be a pmess and a thing you don't want to do is web 2 company you just want to like build your network put a wall around them keep them in there and then like put ads in front of them surveil them and put ads for them that was the model and in the web 3 model it's no you do want to interoperate because it's a way to acquire new users and to build your network and to grow your token and to have more people build on top of you and compose on top of you build software on top of you it's a very different way to think about things you know and you said community community is many things it's users it's developers it's creative people like the

way i think about web3 it's the right way to build networks we finally figured out after 30 years the internet and longer ability networks the right way to build the network and the right way to incentivize network development and once you do figure that out it becomes all about how do you grow that network how do you where the developers how do you grow the users how do you grow the creators how do you get the ecosystem going people to build software around it and continue to grow it so it becomes the only thing defensible really ultimately because all the code and everything else is open source communities only defensible yeah it actually forces everyone to do better by their people too because you can compete so easily yeah the switching costs are very low you're not locked in as a user like i'm locked into twitter i have a bunch of

followers i can't leave i'm locked up they give you the money that you're logged in like that's kind of web 2 today it's like you're just kind of stuck there's a famous scott commoner sent me it was a clubs with entrapment or something it's like this whole famous old economist article anyways but that's what web 2 is yeah it's like this whole you join the club it seems great it's like it's horror movies next thing you know it's like this vampire cult like that's web 2 yeah hotel california funny that you say vampire culture because i was just thinking of like vampire attacks too and like that's kind of a cool thing actually to be able to like suck out actually users and do things just for context for people who don't know scott commoners is a researcher on asics and z crypto and he recently published an

article that he co-authored in harvard business review talking about how it's actually going to spawn a whole new age of competitive theories and intelligence for how business is done yeah because everything is open source and because so many projects in crypto are composable by default and because there's no shame in kind of copying something that's working and borrowing a business model or retooling some code i think that something i really really expect is that over the next year two years if something starts to work i think the speed that it can be incorporated into other projects and then combined and recomposed with other types of existing infrastructure will be so fast that we may end up

actually seeing like progressive waves of extremely rapid mini cycles and lots of little growth as the interest area is a crypto expand right so for example if someone figures out something that really really works in crypto games the speed that that can connect and have consequences in dows connected to that game or in d5 or in nfts related to that game or all the other crypto areas the speed that the ripples can have their effect across these other areas is so fast because of their interconnectedness that it just seems so unlikely to me that we end up in a period of general long stagnation i think that's a really interesting thing in whims free that's fantastic what are the things in

the pipeline that you guys are excited about that you guys know of along those lines i mean one obvious one is like the merge you know things like that or like other things that are public are going to be public what would you say keeps you giving the faith right now oh a lot of good stuff so yes ethereum like the merge and fruit and steak and just generally kind of upgrades l2s and ethereum i think you'll have kind of no trade-off l2s within six months on ethereum like four eight four four for the ipa and like there's a change that's gonna happen to ethereum it's gonna make them 100x cheaper yeah the ip4844 also known as product sharding is a way to create a new type of data in ethereum that's more short-lived they actually call it a blob

a blob of short-lived data that can be used by transactions in the blockchain what that means is that instead of properly storing in the fully robust manner most data in ethereum is stored more or less permanently and oversimplifying a little bit we may introduce some epochs some some valid life spans of ethereum data but 4844 introduces a special area to store data in a special structure that's very cheap very short-lived which means that a whole flurry of transactions can use that cheap and short-lived data store to do complicated things in a way that doesn't require persisting as much data in the long-term storage of the

blockchain it's almost honestly a little bit i think of like the human mind and like the analogy of like short-term working memory and like your long-term memories and like letting you do more like the older you get the more you forget actually you know oh yeah exactly right like the more that you store the more expensive it is the more onerous the requirements of the machines that validate and preserve this type of data if you have a bunch that you want to do in a relatively short period and you really don't need to store all the data for it but you do want to sort of course the long-term consequences of all those transactions you're doing then it sounds like a good idea to store in a temporary let a bunch of people use it and then only persist what needs to be kept okay great thanks for explaining that

and so keep going dixon what else would you add kind of lightning round real quick to this there's so long there's a whole set of blockchains in the sort of next wave that's going to come out there's a whole other wave of kind of things coming on bridges so bridges moving across blockchains layer zeros a bunch of other good stuff better wallets better developer tooling that's the infrastructure side the app side i would say like a bunch of really interesting having the media side of nfts people creating kind of like centralized content creation storytelling a bunch of really smart people in hollywood media have gotten involved a bunch of relation gaming kind of related things social network funded two very high-end top-end entrepreneurs recently who are well known in kind of creating

decentralized social media so i'd say we're really firing both levels right now like firing multiple cylinders in both the infrastructure and the application layer and content yeah i think it's really exciting time would you add anything else to that list chris covered a lot of the big ones for me but a couple i'd say zk math is entering like escape velocity where like everybody everybody i know of segment zero knowledge proofs and zero knowledge math with regard to privacy and scanli that's the going totally parabolic chris alludes to this too with unchained games so this is like the whole family of games where all the game's logic is on the blockchain it's way way way way out actually seeing this in the wild for mainstream players but it's a super interesting area where

the sort of composability and interoperability components of blockchains get to shine in a game context which is super fascinating to me what does it get you by the way really quick when you do that when you go fully on chain with gaming well means you can decentralize the game and we've never had a decentralized game and that sounds like a really awful important thing to build historically speaking games of course are made by like a single team that has total control over the whole experience but when you decentralize a game yeah you'd lose a lot of control over the development experience maybe but you get a game that can also act like a platform something that people can authentically build on build very seriously all and that could be things like mods and it could be things like new clients and new

types of experiences all on something to do with the underlying game being shared i could talk about it forever got it no that's cool then like another thing is that there's so many people well it's still a very small total number actually but there's so many more people programming on blockchains now that we're actually getting people who care a lot about programming languages on blockchains which is really exciting that's a super interesting subject because until recently programming languages for blockchains have typically been very loosely adapted contemporary programming languages like javascript to solidity or rust to solana right these are like programming languages that already have some prior purpose and history in web 2 world and are only

loosely adapted to blockchain programming purposes but now we're getting a lot of people talking about custom-built programming languages for blockchains and that includes special kinds of primitives around ownership around transferring assets and value things like that move is an example that came out of facebook's dm project that's being used by a few layer ones that we're working with there's also new programming languages on ethereum even there's like the fe or iron fe lang project on ethereum and a bunch of others but it's finally something that's getting attention as a first class subject which is totally overdue and super exciting for me especially as a programming languages

nerd i think overall we're in a really exciting time in web 3 crypto blockchains when i look back at past computer movements they tend to be these kind of golden periods i've talked about before and i think we're very likely in that period now where there's been a ton of really interesting talent that's coming in and people building cool stuff and so i'm really excited about it yeah despite the chaos i'm as excited i've ever been because of all the really high quality people that we run into so i'm just so excited to see what everybody builds fantastic wonderful way to end thank you eddie thanks for joining dixon yeah thanks for having me fun to talk to you both all right thank you thanks eddie and that's our first episode of web3 with asics and z there's a little bonus

discussion about methods and metrics that follow these ending credits along with some behind the scenes internal team chit chat this episode was produced and edited by sonal taxi that's me the episode was technically edited by our audio editor justin golden with thanks to long time sound engineer seven morris credit also to moonshot design for the art and special acknowledgments to chris dixon founding and managing partner cmo kim milosevic and several others on our team here for their support to follow more of our work and get updates resources from us and from others be sure to subscribe to our web3 weekly newsletter you can find it on our website at asicscrypto.com you can also find show notes with links to resources books or any papers mentioned or discussed transcripts and

more at asicscrypto.com as well thank you for listening okay so let's cover some other methodology and metrics as well as a couple of faqs that came up with our recent state of crypto report which again people can find at asics and zcrypto.com datacrypto we touched on this already in the episode but was there anything else that changed for you when you did the original price innovation cycle posts from a few years ago and then the work that the team including you did for this recent report like what else changed another change from the blog post was that for the prices section back then we used the bitcoin price just the bitcoin market cap and that was because back even in 2020 i forget the exact number but

bitcoin market cap was the vast majority of the global crypto market cap and that is no longer the case there's tons of other stuff that contributes to the global crypto market so now we just use coin market caps approximation which is not perfect but is much more precise than just wrong yeah it doesn't even matter frankly the precision it's the directionality and the variety that matters actually it's a mix that it's no longer a single dominant thing only that's the narrative there yep exactly right anything more to add like what about addressing some other frequently asked questions that came up yeah yeah a couple days i didn't actually have a bunch when is the estimated creator revenues slide we use

averages which of course i know like any good data scientist a median would be much more descriptive here not even a good data scientist does anybody knows anything anything anyone who knows anything about data you know i am like just for people who don't know like the difference between a median and mean here is really significant because for a lot of creators there's a lot of big outliers and when you're using an average versus a median you're actually counting the outliers when you probably shouldn't be come on way more exactly and that drives me nuts because i would totally rather have a median yeah these kinds of creators you're saying it's totally power law distributed which means it's very prone to a mismatch between the median and the mean that's right more than anything it's like the worst case scenario so i can't even

explain how much i would rather have not just median data but i'd rather just have why not just the full distribution i'd rather have a histogram for each one of these the full quantiles for everything and break it all like i'd rather have all that of course the problem is that that data is just not available for spotify for facebook for instagram for youtube right that kind of distribution data that you'd need to really get it to the descriptive statistics is just not available yeah now ironically it's totally available on the blockchain it's actually very easy to calculate the median earnings of an nft project that's actually totally straightforward totally easy to calculate but we're trying to be as applestables as we can right and of

course none of the comparisons are perfect it's just meant to illustrate on a high level the differences between the ways that people who make content are monetized and participate in monetization across different types of platform what 213 i was just thinking to myself if you actually use ironically correctly you did because it's dramatic irony if you interpret it that way because otherwise it would be paradoxically i was just mentally thinking you know dickson's will not ironic chart yeah yeah yeah well it's ironic in the sense that the very feature of a blockchain is the transparency it's the transparency which we can use yeah exactly

exactly no i got you i was just chuckling to myself i might not we gotta be careful we gotta be careful about misusing irony that drives me crazy too well dixon is so funny about it it's so hilarious like he's so funny he took it off because he's embarrassed by it but i feel like he should put it back on the internet it's like a useful chart yet yeah um what are the other things you would add besides a creator and mean versus median versus the histograms and yeah another is that and that create a revenue slide we chose to describe revenue as opposed to profit just because there's all these other factors that play into the profitability of a specific company whereas revenue directly speaks to the market value of the underlying content selling the attention that corresponds to right so for example in spotify a lot of the

revenue spotify is not actually that profitable and that's because they have to pay upstream licenses for all the contracts extremely generously for all that content yeah and then like youtube and facebook like google spends a ton on research still and that's going to take away from the sort of profitability if you're just talking about pure profit on like an accounting basis same with facebook like they all spend a ton of other unrelated projects which are moved from the overall profitability of the system i think that the revenue is a more precise way to capture the market value of the underlying content and then you had another thing you said the other is that when we talk about like the number of creators nfts it says 22 400 creators

creators is of course a broad word it really corresponds to nft projects but there's a lot of projects that are the work of a single creator and of course even on youtube like when you talk about a youtuber often they have many people working and collaborating with them like one youtuber is not necessarily uh solo show they can be many are right but they can be actually like a very produced tons of people behind the scenes or maybe they don't look that produced but that's actually part of the bit you know they actually have a lot of really clever people behind the scenes making that show work so some people had some reservations around the usage of the word creator i think that just like artists sounds like a single person creator sounds like

a single person channel could be a single person i hope that the nuance is conveyed here that a creator can be potentially a group just like an artist can be a group or a channel could be totally i have a lot of thoughts on creator collectives actually i love that great and one last question on this one thing you guys actually didn't define it in the deck was tvl like total value locked yeah this is a super tough one part of it is that we don't want to lean into tdl because it's an imperfect metric and it's also defining it is also a total nightmare because yeah there's all kinds of ways you could define it that i wish we didn't use tbl actually yep it's totally imprecise it's vulnerable to gaming it describes some types of protocols

better than others unfortunately just for path dependent reasons it's the key metric that a lot of projects measure themselves by the right thing is to use vertical specific metrics got it so like lending like define lending tvl is fine for uh dex decentralized exchange total volume is a far more precise metric for an nft project we're kind of in the early days of still defining what the key metrics are but there's already some really interesting ones like the number of tokens in a collection that are on sale at a given time that's an interesting metric another one is the diffusion of the project so the average number of holders per 10 000 tokens yeah totally i would

add my own i don't know if you guys have this on the list but i care most about the extensibility of that particular nft like how many derivatives have derived from it essentially because that is the true art of nft being something that is composable and something you can build around oh totally now back in the early days of unislaw and compound we actually used to we used to talk about the forkability of a protocol being a leading indicator of its quality if a new protocol drops and a bunch of people are working and copying it yeah and trying to rip it off like that's not a sign that it's weak and should be abandoned then it's just easily copied no that's a sign that the original is incredible yeah it actually undoes a lot of classic dc and business strategy

mindsets frankly about what modes are okay anything more to add because i think we covered all the data methodology stuff that's it thank you so much saul