Decentralization | Miles Jennings

Transcript

[Music] thank you okay so uh we I have a few presentations that I'm going to give we're going to start out talking about decentralization um then we're going to go into token launches and then we're going to talk about corporate structure and really I think that the key theme across these three right is is going to be kind of what is different about building in web 3 as compared to building a typical startup ultimately you know building in in web 3 building a company that's going to ultimately decentralize takes you down a very different path uh than what your typical Tech startup Goes Down And there are a lot of pitfalls and a lot of dangers that they come with that there's a lot of regulatory uncertainty and so you know a key part of my job is

basically helping our portfolio kind of navigate those those challenges and I think even if you're not building a you know decentralized protocol and are ultimately going to decentralize these presentations should be valuable for you because ultimately if you're if you're building within web 3 you're going to be interacting with companies that are decentralizing and so you're going to want to understand the context for what that means for them what direction it means they're headed in and then also it'll help you kind of figure out you know which companies are kind of and which ones aren't my background here before I came to Andresen I was a partner at Latham Watkins which is one of the bigger International firms in the world I was head of its blockchain and cryptocurrency task force which is about 80 lawyers around the globe I got started in 2017 working with consensus

then started working with Paradigm and a lot of VC funds I ended up working on uniswap's token launch back in 2020 and and kind of had represented a you know ton of companies across the space and since I've been at Andrews and I've basically been working with our entire portfolio basically helping them navigate a lot of the challenges that come up in this space so I promise I will I will keep it as light as possible on the actual legal stuff and we'll talk about more of the Practical realities of kind of navigating regulatory uncertainty all right so the first presentation is going to be generally on decentralization right um we're going to cover why it matters we're going to cover some Frameworks for how to think about it is a very voluminous topic and so we'll kind of try and break it down to make it a

little bit easier for you to to approach uh then we'll talk about some tools that can help you achieve decentralization some timing how you roll it out I think you guys probably all heard of progressive decentralization so we'll talk about that a little bit and then we'll go into some models and basically show how this stuff kind of has been used in practice thus far as well as as kind of where it's headed all right so why it matters um ultimately right A lot of people can give you a lot of different answers in my mind I think decentralization matters it's the key tool that's going to allow us to democratize the internet because it turns web 3 systems into public infrastructure right that is a a key difference from from what we have in the in the current world right where we have a bunch of centralizing networks that

are all kind of accruing benefits to the companies that run them so with decentralization with with kind of operating uh internet infrastructure as public infrastructure what are the the key benefits one is it promotes decentralized or it promotes competition right if if ultimately if you're able to create credibly neutral and composable internet infrastructure then people will be able to build on top of it and as a result of that it will lower the barrier to entry for people building internet businesses and services and as a result of that right we should see greater consumer Choice more diversity in kind of the offerings that people can bring to to Market um the next is safeguarding freedoms right so um you know there's a lot of kind of this is more of a hot political

topic but really when it comes down to it one of the key challenges that we have right now is with with you know giant centralizing networks that are run by corporates that gives them enormous power to censor people to gatekeep and to infringe on their personal Liberty and so one of the ideas is behind you know decentralized uh internet infrastructure is that that power will be removed away from accruing to uh centralized corporate entities and as a result of that we should see you know greater kind of Liberties um you know great example of this is just Twitter and you know everything that we've seen since Elon has taken over and it turns out that the censorship that's gone on is probably not as as fair or transparent as as people would uh expect it to be finally uh the third benefit is that you actually can Design Systems

that more equitably reward stakeholders right currently with with corporate networks um like Twitter and Facebook and Instagram you know those Those comp those companies run those Network and as a result of that all value that is generated by those networks those companies are bound to make sure that that value maximizes and accrues to their stockholders rather than kind of people within the ecosystem and so as a result of that that's why we see enormous take rates um kind of applied to um you know on on corporate networks on web 2 networks so as a result of those three right you get this dichotomy and I think Chris Dixon probably would have gone into this in his presentation with you guys right you get these corporate networks versus blockchain networks um you know the corporate networks look

more like Disneyland and Disneyland's great it but it is a very curated experience it's not vibrant there's a lot of homogeneity there's there's not kind of um a you know a lot of different experiences and no one would kind of go to Disneyland to set up shop to you know build a business if you tried to protest in Disneyland you'd probably be thrown out and ultimately you know the all value is ultimately accruing to Disney stockholders you compare that to blockchain networks right and with the you know promotion of competition um the the uh basically allowing people to reward stockholders or I mean any any stakeholders right and and the ability to safeguard freedoms right New York is a more vibrant place it is not a curated experience and ultimately

um we think that that is what the internet was originally intended to be like and and what we want it to return to all right so let's go into some Frameworks so I think you know decentralization is a is a very large problem um and one that is is incredibly difficult for for Builders to kind of tackle so if we break it down into kind of three different Frameworks I think it'll help people think about it so the first one is is Technical decentralization and what we mean by this right is really just the ability of blockchains and smart contract protocols to you know they are our computers and they can make firm commitments and so they can allow for a trustless and autonomous uh uh ecosystem in which in which you know people can compute and transact in a disintermediated fashion

and so when you think about this right the the kind of Baseline that like the the hurdle for achieving technical decentralization really is going to be dependent on kind of what protocol you are developing if you know for a a blockchain like ethereum right the technical decentralization is is very challenging because ultimately it's quite dependent on the assets you know if being distributed to validators and them not being able to attack the system and basically everything kind of maintaining itself whereas if you look at a you know simple smart contract protocol if you were to just deploy a smart contract and make it immutable then you could very well argue that it was decentralized technically decentralized right away because ultimately no one can actually then step in and change that that smart contract or take funds out of it so as a result

of it it is kind of a moving hurdle depending on kind of the the protocol that you're you're building the second piece is is economic decentralization and here we're really talking about when you start to introduce uh digital assets to web3 ecosystems uh all of a sudden you have a much more complex product around whether or not the thing is decentralized so for instance in the case of ethereum again returning to that right ultimately the the ability of the system to use its own digital assets to reward people providing services to the system uh and then you know you transact and Eve all of that creates this decentralized economy around ethereum and as a result of that you know that economy your goal is to try and keep

that economically decentralized otherwise if although the value kind of accrued to one person or was able to be manipulated by an individual that would jeopardize the security over the overall system and then the final kind of category is is legal decentralization and so what we mean here right is that decentralization ultimately has the ability to obviate some of the risks that are associated with transacting in assets right or or in in using computers and as a result of that right what we would say is that where decentralization actually obviates risk you don't need to apply laws that are meant to accomplish the same thing to men to mitigate that risk so to give you a good example right in in you know in finance we have a lot of laws our whole legal system is

basically designed around trying to mitigate the risk of one individual having to trust another with money right and so as a result of that we always kind of we have a lot of laws around intermediaries and and kind of rules around how they're able to act and and protecting kind of consumers from conflicts of interest that those intermediaries might have but in a peer-to-peer system like what you can have in blockchain right you don't need intermediaries and as a result of that it's not necessary to apply intermediary based legal regimes that are meant to address the risks of intermediaries to to a system where there are no intermediaries so what that comes down to specifically with respect to digital assets is is Securities laws right so Securities laws are and again this will be the the most

legal part of the presentation but Securities laws are basically meant to restrict kind of how people engage with Securities right Securities are a subclass of assets um that exist and typically what they do to basically address risks around those assets is they apply disclosure requirements right and what they're trying to do there is basically create a Level Playing Field where anyone can invest in Securities and be on the same kind of footing as anyone else that's investing in those right and that's a key thing because otherwise you would end up with Market manipulation people would be taking advantage of other people and as a result of that you wouldn't have efficient Capital markets and the way that our system generally looks at it right is that it's most

likely to have significant information asymmetries when you have managerial efforts that are driving the value of a given asset right so as a result Securities laws are really kind of meant to address kind of situations where assets are based on you know the value of assets are based on the managerial efforts of others so if we just think about kind of the difference between uh Apple stock and and oil right this will make this point clear so stocks right you know shares and companies are the most common Securities that most people have kind of interacted with over the course of their lives and if you think about you know Apple stock right there is a reason why Apple's stock is is treated like a security ultimately Tim Cook right has access to a lot of information that might affect the value

of Apple stock right they know how many iPhones they've sold they know you know how many they've produced they have all of this information that is otherwise not available to the general public and that's also true for some of their you know largest Distributors as well as for for some of their suppliers and so as result of that there's a significant information asymmetry that arises from the management decisions that Tim Cook and others have made and so as a result in order for Apple's stock to publicly trade on markets right they're obligated to disclose to the public all of the material information that might be relevant to the price of that asset right and so as a result that's why we have they're called 10ks and 8ks and all of this disclosure regime that basically Shares are subject to right now compare that to the price of oil right oil is

not a security it is a commodity and as a result of that it is not regulated in the sense of whether or not you can buy or sell it right and so ultimately the way that we look at oil is that they're all the information about oil and what the price of oil might be is publicly available now there might be organizations like OPEC that change the amount of oil that they're going to produce and that might have impacts on the price but you're not talking about the fundamental kind of information asymmetry that exists with respect to shares and stocks and so as a result of that what we kind of conclude is that if a web 3 system can essentially eliminate significant information asymmetries and it can eliminate Reliance on managerial efforts then we say that the system is sufficiently decentralized such that Securities laws are no longer necessary

to apply so the idea is there right is that if you can ensure that both of those things are the case then the risks that Securities laws are meant to address no longer apply and as a result you don't need to apply them so that's pretty complicated and and when it comes to Legal decentralization Founders are often very confused and and you know the rules are not clear because the SEC hasn't set them out in a manner that's very clear but ultimately this is all just kind of a spectrum of decentralization in all kind of one component of it right to be technically decentralized you can't have a management team that can manipulate the ecosystem right if if vitalik could basically you know reverse transactions on ethereum we would all say that that's not really technically

decentralized that the system's not decentralized at all and and so that type of like managerial power right would be something that would also be problematic under um legal decentralization same thing's true as economic decentralization if all of a sudden you could change the way that ethereum works where all of the value accrued to me right then we would say that it's not economically decentralized and we would also say it's not legally decentralized because at the end of the day if I could just change where all of the value is accruing then that would significantly impact the price of eth and my ability to change where the value accrues would ultimately be you know asymmetric information and managerial efforts that are dependent on or that that relate to the price of the underlying asset so you know ultimately legal decentralization is is technically

necessary to actually achieve decentralization in total it's not just kind of a A New Concept that's being layered in that that just relates to uh you know U.S Securities laws at the end of the day fundamentally in order to build decentralized systems um you need to have kind of this this layer of legal decentralization built in okay so now I'll talk about some tools of decentralization so this is a very kind of simple model that I put together of kind of just comparing a web 2 system and a web 3 system um and and really what we're we're looking at the the key kind of difference here right is that in a web 2 system you have a lot of proprietary intellectual property whereas when web3 systems you have very little and as a result of that it makes it inherently

easier to decentralize um because there are less things that you are are actually protecting there are less things that people are reliant on you continuing to bring to the ecosystem so if we break this down we can talk about how you know the various kind of components of web3 systems influence the overall decentralization as I was talking about earlier right blockchains and smart contracts can help achieve technical decentralization just based on the fact that they allow for trustless computing right and so as a result of that they kind of form a a foundation for technical trans decentralization but they also affect the other elements of of decentralization so one of the things that we're talking about right compare the you know ethereum to Apple stock right Apple knows exactly how many units

it's sold and how much money it's made over a given quarter right and as a result of that that's significant information that would affect the price of Apple stock and and as a result of that they have to disclose that information to the public in order for their shares to continue um uh trading but you know ethereum is very different right every single transaction that happens on ethereum happens on chain and so at the end of the day you can you can anyone can just see what the actual revenue is of ethereum at any given time over the course of a year so you don't have that potential for significant information asymmetry around the economics that underlie the asset value that actually exist and so the the nature of blockchains and smart contracts being very transparent right ultimately means that that enhances the overall legal

decentralization um and then things like you know them being open source uh data portability and composability all kind of lead to more economic decentralization because ultimately what it means is those are all tools by which that take away from the ability of any single you know centralized actor to manipulate the decentralized economy that forms around an ecosystem so you know just for instance right if if you create a a system where uh you know everyone has is in control of their own user data right then those people can just move to another network if they want to they don't have to remain within your ecosystem because they actually own that's very different from what we see in corporate networks now where if you want to take your followers from Twitter to another platform that is obviously

not possible so ultimately all of those things help to accrue economic benefits to the you know to the users of a network as opposed to a person that owns a network and as a result they do enhance uh economic decentralization so digital assets then are the other tool that you basically can really use to to create and form your own uh uh decentralized economy right the real difference between a web 2 Network and a web 3 network is web 3 networks are open rather than closed and so in a closed Network all value and network effects are basically accruing to the people that run that Network whereas in an open network all of the value is accruing to the people that are using that Network and so that's what digital assets really

allow you to do right and as a result of that you're basically enhancing the overall economic decentralization of the system and and that again is is kind of you know a key piece of the overall decentralization and then decentralized governance is is kind of a final tool that ecosystems can use to enhance their decentralization and this really kind of crosses all three uh of the the various uh Frameworks of decentralization so you know decentralized governance can increase the security of a system by ensuring that no in single individual actually controls the system it can increase the economic decentralization of a system by ensuring greater incentive alignment across users and and you know contributors to an ecosystem and then it can um and then it can help with legal

decentralization because the very Act of kind of pushing decision making to a Distributing group of people means that there's ultimately going to be less Reliance on managerial efforts um by a small Team all right so decentralization timing um now here the the the kind of key thing is that decentralization is is ultimately not a uh binary state right it is it is a spectrum and as a result of that it can take a significant amount of time to uh to to achieve right and that creates a fundamental problem right you start with basically a paradox where if you need to decentralize in order to use digital assets but you need digital assets in order to decentralize you have a chicken and egg problem right and and

that is a key challenge that a lot of companies face in this space and it's one that our you know Securities laws do not do a very good job of providing for um as you know they don't really kind of take into account that if you're ultimately trying to decentralize that that you then get you know more flexibility around kind of your ability to to stay within the rules so what that means is that Progressive decentralization is a concept that people have developed around basically how do you kind of achieve ultimate decentralization when you're trying to kind of navigate the uncertainty that's created by Securities laws uh and the uncertainty that's created just around um you know not having a robust and and secondary market for your assets so really what what Progressive

decentralization comes down to is it's a balancing act where you're basically balancing the decentralization of the system uh the product development and the community building right and and we'll talk about this more in the context of token launches and when to time your token launch with respect to uh you know those three kind of uh elements um it then you know Progressive decentralization in like a key component of involves incentive design right one of the the key areas that that we look at when we're talking about decentralized systems is how many other people are building within that ecosystem how many other people are contributing and ultimately what we haven't kind of seen yet but what will happen um you know in the future right is is kind of autonomous and automatic incentive mechanisms to to basically

enhance the overall decentralization of systems so right now what you see in a lot of uh ecosystems right so for instance um like a lot of layer ones like Aptos Avalanche Solana right A lot of them have built up ecosystems by basically incentivizing developers to come into those ecosystems and and develop smart contract protocols on top of them and and mostly those have been like have been uh structured as Grant programs that they give out to those people but fundamentally that that kind of structure is a little bit um the antithesis of decentralization right because it does mean that you're then having someone that's in control of kind of um you know funding the overall development of the ecosystem and so what

we we should start to see going forward right is more kind of autonomous automatic um uh uh incentivization of developers and people you know operating within ecosystems as that way you can remove that kind of layer of centralization of Grant programs and stuff like that and just have it happen automatically um and then finally risk mitigation there are various things that we help a lot of companies work with to try and kind of mitigate the overall risk of of navigating this uncertainty right of this chicken and egg problem around you know decentralization and um and and securities laws so if it works if if done poorly right the progressive decentralization uh leads to stagnation and the death of

your project so it's a pretty important uh topic right and the idea is is that you know ultimately as you decentralize it becomes harder and harder for that core management you know the core team the core the initial development team to actually continue developing and as a result of that if you haven't set things up properly right you're going to have this significant drag and securities laws risks around kind of your continuing uh development within the protocol and so as a result right you know you you see this a lot where companies kind of launch their tokens a little bit early uh and then they struggle to kind of develop and iterate because they didn't have product Market fit and and then projects you know slowly die over time now I've done well right the exact opposite happens if you

if you progressively decentralize in a way where you're incentivizing third-party development then the more decentralized you you become right the more development that's actually going to take place by Third parties and as a result of you can kind of see exponential growth and that's what you would see right if you look at kind of the way that that ethereum works now um right there's way more development happening on ethereum than it could ever be possible if it was just a single centralized company that was running everything that was being built on top and we see that not just in kind of right that's not just ethereum but if you look at the App Store the same thing can be said there right I mean ultimately that platform has enabled thousands and thousands and thousands of people to develop applications and run their own businesses on it now the

problem that we have with that kind of model is that Apple takes an enormous take rate because they own that network but at the end of the day the the idea is is that as a centralized company you're never going to be able to develop more than you would be if you built the platform on top of which other people could develop and so the whole point of progressive decentralization is basically unlocking a pathway where um you know you're incentivizing decentralized development and and building out the ecosystem so that doesn't just rely on you okay models of decentralization so here is a a model of decentralization um for a blockchain network and this is obviously looks looks pretty simplistic um so the the thing I would say though is that there's a lot that goes on here

right so when we look at whether or not a blockchain is is actually decentralized there are a ton of things that go into the analysis and you know these are a series of questions that one could ask I've labeled all of these uh in a way that kind of look at you know whether or not we're talking about it as technical uh economic or legal decentralization but if you look through these right you can see so how diverse is the validator set so this is something that you're going to care about because ultimately right if the validator set is not decentralized then someone would be able to come in and and basically take over the network by you know getting a number of of different validators to uh to go along with them and as a result of that

the entire you know security of the system is jeopardized um and you would say that it's not decentralized um then if you look at like who is contributing to the protocol well so this it really matters because fundamentally if the protocol is you know developing over time um and it is you know dependent on kind of uh you know new versions being released or upgrades being released over time then it really matters right who is actually contributing to those things because the economic value of the asset of the system is going to be dependent on on whether or not those those targets are met and so as a result of that right you want to have kind of a decentralized group of contributors very much like what ethereum has now um and and what you would say right is that

like you know obviously you have to look at kind of how meaningful the contributions are I think if you looked at you know I think if you looked at the change from proof of work to proof of stake um for ethereum uh you know everyone like that was obviously a Monumental achievement but at the same time it's not like it led to a significant rise in the price of ethereum and so ultimately like these increase you know the the uh upgrades that are are brought to Market over time um you know and their impact on pricing is all something that factors into whether or not uh the system is is decentralized um so just a couple other like technical ones right how many nodes are are accessible right that just kind of relates to how people interact with the chain um how advanced is the product roadmap

again this is similar to who is contributing to the protocol um and and what you know what direction is it is it ultimately going in um and and is there any kind of point of centralization there um who controls decentralized governance is obviously a very key one right at the end of the day if a single individual or group of people acting together we're able to you know kind of uh co-opt uh governance to do what they wanted then it would be hard to say that the system was um actually decentralized and so you can see that with you know with things like optimism which has on-chained governance um you know if if uh token holders went rogue or some token holder was able to acquire enough tokens to then attack governance right they could essentially

you know let's say take the entire Treasury and award it to themselves which would obviously be very detrimental to the price of the token as well as the security of the network so as a result of that right we would say that that kind of both factors into the technical and legal decentralization of the system now um I think that the one of the key questions here that I would focus on is how many Builders are deploying protocols to the network right and I think that that is ultimately a fundamental component of the decentralization of any given blockchain so with ethereum right there are thousands of developers that are all deploying uh protocols to um to ethereum and as a result of that the the amount of developer activity really is a strong

signal that that ethereum is decentralized particularly because right the price of eth is is significantly based off of uh you know the the overall economic model behind the token and the more transactions that go up the more Heath is burned and as a result of that right that has a significant impact on the overall economic model for ethereum so as a result of that right the number of builders that are actually you know participating in this ecosystem um is is kind of critical so if we go from this model to the next right we can see a little bit more complicated structure right so here we have a smart contract protocol right that has been deployed to a blockchain and then we have you know it runs a an application on top of that

and then that interacts with users and so we'll add in kind of a dow as well that that oversees kind of a treasure so right here now this model itself does not look that decentralized right and as a result of that so let's say that you had just a single client application built on top and that was the only way that people could interact with the protocol and it was the only way that that actual value was driven to the digital assets well that point of failure with the single client or app right would end up being a point of sensualization that could undermine the the decentralization of the overall ecosystem so we see this model right was was kind of a lot what was what people used in D5 really in a lot of the kind

of first D5 protocols and applications that came out so if you think about you know uniswap or compound right the primary application and client that people use to access those protocols unifwap.org or compound dot Finance right and those are run by the you know initial development teams that develop those protocols now the reason that we ultimately get comfortable that that kind of you know application being run by the development team doesn't create problems from a decentralization perspective is because those applications are quite simple they're not um they're not like very robust and Rich uh experiences basically they're just kind of explorers that allow you to interact with the underlying smart contracts of the protocol and and as a result of that

they can be replicated very easily and and so what we saw with you know uniswap and with compound with a lot of other D5 applications right is that despite the fact that that the team was you know at first the only team that was running a a front-end to those protocols hundreds of other um you know front-ends kind of uh popped up so with respect to you know swap right there are tons of uh decks aggregators there are tons of other you know websites that basically enable you to interact with the uniswap protocol um and then you know things like metamask right has a direct swap feature that routes to uniswap's protocol so as a result of that right you basically have a very rich system of of applications and clients that popped up and as you know if if

uniswap the company was to shut down right their application it wouldn't really change what is going on with the protocol because ultimately people would still be able to interact with it through all of these other interfaces now this model right really kind of um is based on the premise of of a Visa that's called the fat protocol thesis and what that idea is right is that in in these kind of ecosystems all of the value really accrues to the protocol itself and not to the client application layer and and what that means right is that that ultimately um the application and client layer is very commoditized and that's exactly what has happened with compound and with with uh Ave and with uniswap and a bunch

of other D5 protocols now the problem with that thesis though right is that if all value is accruing to the application or to the to the protocol layer then how do you actually you know generate real businesses to be built on top of those protocols right and that fundamentally becomes a very important question particularly in D5 because ultimately we are very likely to have Regulatory Compliance obligations at the application and client layer right and if you have Regulatory Compliance obligations at the client and app layer right then you need to be able to run a business when you're running that client and app layer otherwise if you have no ability to accrue Revenue then you're going to have no ability to actually you

know engage in Regulatory Compliance which costs money and as a result of that the fat protocol thesis just doesn't work in a system where you actually need to have a you know Regulatory Compliance or right when I you know first started in Andreessen D5 understood this model understood why D5 was decentralized but the the thing that I couldn't wrap my head around was like well how do you go from this type of structure to a much richer ecosystem where the app client layer looks more like the applications that we have in web 2 right you know uniswap's front end is very different from Facebook's app or Instagram or Twitter right and as a result of that it's very unlikely that you would kind of find client and applications that

were basically replicating um what uh is very unlikely that you would be able to find clients and applications that we're going to be able to be built and replicate those types of Rich experiences right without an economic model that then supports them uh to be built and continue to operate so as a result of this this model just doesn't really work for like the grander vision of what web3 is it doesn't really work for what's going to happen when Regulatory Compliance obligations uh are are start getting applied to these protocols and so what we need is is kind of a a better model that that is going to allow for those things to happen and and the model that that is right is is

just this right so it's diver it's basically incentivizing diversity at the client application layer right so that you can decentralize that whole layer right so now if you look at this right you have a smart contract protocol that's deployed to a blockchain the blockchain is decentralized the Smart contract protocol let's just for easy of use let's say that it's it's totally immutable right now if you only had one client that was a rich you know Facebook like or Instagram like um client right again that would create a point of of centralization that would undermine the decentralization of the entire ecosystem but if you're able to incentivize multiple different clients to kind of operate within this ecosystem then all of a sudden you do have a decentralized

layer as I was saying with D5 that kind of happened organically and naturally because it was very cheap and easy for people to build their own front ends but with these more rich and robust ecosystems right that's not going to be the case and so as a result of that right what we need to see is we need to see actual incentivization of that client layer being built and it needs to happen autonomously so let's think about just a comparison here of let's say openc to uniswap if you go to openc right that is a much more robust uh website for people to navigate and use right there there are you know they have all of the images there's a ton of different um a ton of different assets that are listed and compared to unit swap it's

just very kind of um it's just much more robust and their their operations reflect that as well right openc has a massive trust and safety operation they file SARS reports they do all of these things where it's it's not just it is a business that is built on top of of their protocol the protocol that openc kind of runs their Marketplace on top of is Seaport and so if if you were to think about okay well how do they actually go about decentralizing right it really wouldn't work if let's say that there was a token right if all of the value of that token was then generated through sales of nfts that were coming through opensees marketplace right because now openc could say well we're just going to turn off our Marketplace we're going to change something about our Marketplace and that would create information

asymmetry it recreates Reliance on managerial efforts and as a result of that it wouldn't be decentralized but if openc you know if there's a what 2.5 Commission on every sale let's say that that all went to the token if openc then had other marketplaces that were built on top of Seaport right that would be how you would start to decentralize now that's a very uh counter-intuitive kind of approach right because it means that you are essentially incentivizing people to be your competitors right and as a result of that like it's it's a very kind of odd uh uh business structure to approach because it means again that you're you're incentivizing and basically enabling people to compete with you but fundamentally that is really the only

way to kind of achieve decentralization Within These ecosystems and what I would say is that that fundamentally what we believe right is that that will lead to a bigger pie than if you you just kind of retain it yourself so the idea is right if if you are open C and you're running your own Marketplace well there are a thousand other entrepreneurs that might want to run an nft Marketplace and they might want to do things very different from you right and as a result of that incentivizing them and enabling them to do so ultimately will lead to a greater value accruing to Seaport the infrastructure that you built and as a result of that there's a race and a real kind of economic value to developing that infrastructure right but at the same time you you do need to be able to

incentivize kind of competition to build their own ecosystems on top of it and and and so we'll see that right we're already starting to see this with you know forecaster where people are building their own clients and applications to interact um with the protocol and and we'll see it more and more when it comes to these more robust protocols and and the ecosystems that are being built on top of them and so this was a fundamental kind of piece of where we expect things to go it also has a lot of knock-on benefits um where with some other pain points that we're experiencing within web3 you know one of the things that that Porter will talk about tomorrow is just decentralized governance and what what kind of this abstraction does by creating a multiple client layer

right is it allows you to kind of practice governance minimization at the Smart contract layer because what what what this does right is you can basically push a lot of the decision making that otherwise you might want to have your community make to the client layer so for instance if if you know let's just again using openc as an example right if if you wanted to have multiple different marketplaces and you had a marketplace where people didn't want you know you to show some type of artwork because it offended who you know whatsoever right or they just didn't want that kind of material on there ultimately that is something that they could easily achieve right because someone could just build a client that doesn't have that kind of material in it whereas if you you know try to do that at the Smart contract layer then you

would be introducing censorship at the Smart contract layer and that would fundamentally kind of undermine a lot of the values of of the the technology so this is kind of um you know the direction I think that we're headed and and it's it's key right figuring out kind of that incentive mechanism the only the only place that we really saw this within D5 I think is is a protocol called liquidy where the team actually built their protocol uh and didn't launch a front end whatsoever uh but they did turn on a incentive mechanism where basically anyone that built a front end they would receive a small percentage of of kind of earnings of people that use that front end uh in interacting with the protocol and so that is exactly the type of thing that we expect to now see

um with with these more robust ecosystems now here's a another model that we'll look at where this basically takes what we we just talked about but then adds on a layer of complexity so one of the things that will happen if you if you're talking about let's just again sticking with the openc example if you have an ecosystem like that right how do you actually create or or an environment where competition and clients and applications can arise how do you how do you basically set people up to build their own marketplaces and operate them right well one of the things that you could do is you could build a white label version of a Marketplace and enable anyone to use it right and as a result of that that would

help people do it uh help people set it up but there are other ways that you can also do this right so if if there are significant Regulatory Compliance burdens right you could create a you know decentralized actor within the ecosystem that just provides that so you know every every single Marketplace that's built on Seaport right could have access to a trust and safety uh team that basically you know operated on behalf of all of the clients and so as a result of that you can create these incentive structures where people use you know shared internet actual property and shared products and services by teams to basically you know overall kind of lower the barrier to entry for people to build

Within These ecosystems and if you lower the barrier to entry for for people to build Within These ecosystems you're ultimately going to end up with greater diversity at the client and application layer and and what that does then is is overall enhance the the level of decentralization um and so you know this area I think is is one that is is going to be you know continuously explored where we're seeing this a lot right if you think about in the context of uh video games this is something that that really comes up because you know if you're if you're trying to decentralize a video game right there are various components that you know can't really be decentralized immediately right you have the underlying game engine uh and how that operates and so as a result of that right you can kind of Imagine worlds where

um you know the video game Operator Operator um you know licenses the the game engine and and everything that you need to basically play the game game to the you know Dow that oversees an ecosystem and then people are able to build with inside that game um and so I think that you know ultimately from here right you can extrapolate out to a lot of of different models but the key like premise I think that we're we're trying to focus on is that you know if the economic engine that ties to a digital asset you know has a point of centralization then fundamentally there's going to be problems under Securities laws but also that that is something that jeopardizes the overall decentralization of your system and so what what this like kind of approach to decentralization does or

focuses on is really about building redundancies at every single layer of of the tech stack and and you build those redundancies and you want to have them be autonomously and incentivized to come in into place and so by using on-chain incentive mechanisms like token Rewards or you know by Distributing the fees that are are accruent to the protocol you can basically um you know set up an ecosystem where that decentralization should happen naturally over time and when we talk about Progressive centralization right having those systems in place where people can permissionlessly come in and build on top of what you've already built and have an actual business model that's associated with it it leads to a much more powerful like opportunity for decentralization to exist um than if you you know you don't have those in place and are trying to do it

manually um so that's it um questions um so my question is um as as you mentioned that the progressive designation is very important but then what's a good timing for that like when a company is like um ready to should be ready to like convert to like start to become like a decentralized yeah so I mean I think that that the I think that the the overall guidance that we tend to give our portfolio companies right is is to wait on the defense realization piece as long as you can there are a lot of different pressures that come into it and we'll talk about it more in the in

the token presentation but you know decentralization is a very hard project to um to undertake and it then becomes very difficult to um you know find product Market fit uh post decentralization and so one of the things that we always encourage people to find is to find product Market fit right and only after you've kind of done that to then really like engage in the process of decentralizing now there are various tools and again we'll talk about it in the next session that you can use to try and and basically you know dip your toe in the water towards decentralization start to get some of the benefits of incentive of token incentivization but at the same time not kind of committing you to a path that uh you know you're not going to be able to

kind of reverse from right I think one of the the hardest things is that you know once you have a token out there um you're kind of Trapped into a a paradigm uh that you otherwise can't you know depart from and and that you know there's a bunch of Securities laws pressure that comes in to play but so as a result of that right like the decision of when to issue a token is probably like there are things that you can do to decentralize ahead of that and then the issuance of a token there are a couple things that we do to to kind of lengthen the time Horizon that you can have to to you know have a little bit more control so does this mean the decentralization has to be tied to a like a token launch it doesn't have to be tied to a token

launch but but really I think it it all depends on on the protocol I think it's very hard to decentralize without having some type of token because ultimately um you know if you if you don't have a token like what is the mechanism for control of the ecosystem and so you know tokens are a key component of of decentralizing um thanks for the answer um to add on um like follow-up questions like go do analysis for like openc uh do you think they are decentralized too early or too late um considering the blur kind of yeah well so and and you know we'll get into this in the next presentation as well but I think you know for openc right I think I don't think that they've they've gone

about it the wrong way I think that they built it the right way I think that it it takes a really long time to build a really good product um and you know building a a website and then kind of engaging in like decentralization theater what is what we say right where you say oh we're really decentralized but then you're not actually um you know not having product Market fit still needing to build a lot of different um you know things like doing all of that and kind of pretending that your decentralization is or decentralized is not a good path and I think openc has done a very good job of kind of not decentralizing too early which is is one of the probably the biggest mistakes that people make in this space wanted to talk about the information asymmetry you talked about and how in

like ethereum and maybe other tokens um that doesn't exist but I wanted to get your thoughts on like kind of like the gray Zone where you know there's like a foundation behind ethereum like there's a case where like Elon Musk is like pumping Doge and you know like yeah that's kind of in a way information asymmetry I think um as a Founder like how how can we think about creating the Rights Management Systems so that we do have you know that input but like you know we stay legally like sound yeah so and we'll talk about that more in in the in the corporate structure presentation but but basically right I think there there are uh there's really two different approaches that you

can kind of take right I think that that um on the one hand you know if you look at things like uniswap right um V2 is still one of the most successful dexes that's out there right um V3 is obviously like the most successful uh deck that's out there and and both of those Protocols are in place they're immutable they can't be changed and as a result of that there's not really any possibility of information asymmetry coming into existence with respect to the how those Protocols are are kind of functioning and I think if you notice right there's there's not been any kind of um there's been no setting of expectations right that that the uh that the team that is behind uniswap is working on you

know other things that are going to drive ultimate token value I think the the key thing with respect to uniswap right and and its token relates to the fee switch being turned on right and as a result of that that is the the kind of key component to what will the economic like model be behind that token but it's much more about that and what that model looks like in people's use of uniswap than it is about the management team trying to building new products for the uniswap ecosystem and so as a result of that they've struck a nice balance right where I think it'd be very unreasonable for someone to say well I expect uniswap to do this and it's going to make me money whereas you know if you think about Apple stock that's exactly the way that people think so so one way right is

to basically be done with a product and and have that product out in the public right and then the other way is trying to kind of uh really Foster decentralized development um post uh you know having decentral realized initially I think one of the hard Parts there that pathway I think is much harder right because what we've seen is that it's very hard to get um you know other actors into your ecosystem to develop particularly if you don't have an economic model that incentivizes them to do so right because like for instance why would I build on top of your ecosystem when I could just copy your code and go deploy it myself on another chain right and as a result of that those incentives to kind of people to build copycats and all of that has really undermined

um like the possibility of decentralized development and so I think what we what we need and like what I'm stressing here right is better incentive models to kind of incentivize third-party actors to build within your ecosystem and if you can do that right the more that you can get third parties to build in your ecosystem then the more flexibility you have to also build because now you're just one kind of piece of the puzzle rather than you know the only piece hi um thanks for the great talk I have a question around um you know like business models for these clients or front-ends does it usually come from like the protocol revenue or maybe end users of this client Spain like SAS fees or other type of fees and what are the some of that are the most successful

sure so yeah I mean it's going to depend on the protocol right but I would say that in the context of like like game let's just use openc right if you think about um so right now you know when you engage in a transaction on openc right you pay a 2.5 commission right so the the fat protocol thesis would say that that basically all of that value should accrue to the protocol right so the protocol gets all 2.5 but as I was saying right that doesn't really lead to a scenario where people are going to be able to build the application layer on top of it and have it be decentralized um you know openc has a ton of employees a lot of those are engaged in Regulatory Compliance like functions and as a result of that if all of the value accrue to the protocol then the only way

that openc would be able to continue to run its business is if like you know they had it had a token and then all of the value came through them through the token which doesn't really work because again you'd have you'd have centralization right so we have to kind of find a happy medium between the fat protocol thesis and and the other alternative which is all 2.5 goes to the client layer right and if you were to do so instead of 2.5 going to the protocol let's say all 2.5 went to the application but if you did it that way right now the protocol has no actual revenue and and there's no network effects that are accruing to it and and that opens up like interesting possibilities of of also centralization right if you have one Marketplace that is then responsible for 75 of the you

know the economics that are coming through the system well now it's going to be sucking away all of the money that is basically you know coming into the system and so as a result of that right the the kind of opposite extreme of having everything go to the app layer doesn't really work either right so you need to find a happy medium where let's say like let's just say you did one percent to the protocol and 1.5 to the application right and then what what that would allow you to do right is a it incentivizes people to build businesses on top of the the protocol and create their own marketplaces but in addition what it allows you to do is it allows you to have a lot of diversity at that client layer because maybe you run a Marketplace and you say well we're getting 1.5 for every transaction we're

going to give our users point five percent of that and maybe I'm running one and I'm gonna say well we're going to give them all 1.5 percent right and so all of a sudden you have this very diverse kind of an opportunity for various different business models to kind of operate at that app layer and as a result of that you know you should you should basically end up with a lot more choice and diversity for people to choose from and then you would also give like native tokens to this client so they can also get some value from the protocol exactly right yeah the whole thing right is is about incentive alignment right you want you want Builders to come into your ecosystem to build and you want them to be aligned with other builders in the ecosystem to grow the overall pie and so

that's what I mean by you know a centralized business is you're you're you're taking a pie and you're growing it yourself and you're keeping the entire thing right and the idea is is that if you do that you you can't kind of you can only your Upward Bound about how big that pie can get whereas if you only are going to take a slice of that pie right that actually might be better than taking the bigger like all of the the smaller pie right if you can grow the pie massively then you you take a slice of it and you end up economically better and and that's really you know one of our our um the guys on our research team uh Scott commoners wrote an article basically about the economics between when it actually makes sense to decentralize um from an economic perspective and like when that will lead

to potentially greater growth that would then um yield a better benefit to the person that built the protocol so I have a question about um if all the revenue accrue to the protocol and then the protocol distribute um let's say like cash or like in token to a foundation or developers that are building on top of the protocol is that considered legal or yeah so so there is a um and and we'll actually speak to this in the next presentation but like in terms of the design space and and what you want um your token models to be right I think what and and what I'll get into this but like ideally

you shouldn't be thinking about tokens as a proxy for Equity right and so you shouldn't be designing systems where the tokens basically function like Equity does in a in a shareholder world so you don't want a system right where basically uh the system accrues revenue and then it dividends it out to token holders right because ultimately it's very hard to argue that that is should be treated from a regulatory's perspective different from um from uh tokens right and so as a result of that we try and stay away from those systems and it really actually limits your overall like the utility that you can drive to an ecosystem if you kind of adopt one of those models and so I'll walk through in the next presentation um some different design choices that

people can make to to really I think think outside of the box of of kind of what Equity does hi Brandon here thanks for the fascinating talk so we have a protocol called share and we have kind of the permissionless smart contract layer in addition to the application layer and we believe in kind of that redundancy model that you talked about by having multiple clients that being said I feel like one and take away from the talk is that a prerequisite to meaningful decentralization is achieving a dominant enough position such that there's economic value that can be distributed to people that want to build on top of that yeah and so I wonder if in this case in the end is the elimination of trust really an anti-pattern because you need Founders to get to a point where there's value and then make a conscious decision to

decentralize even though they actually don't need to yeah well I mean I think it it is a uh I think it's a it's a it's a very critical decision right and I don't think it makes sense to decentralize in every instance right I don't think that um and and honestly that's what a lot of companies in this space um you know struggle with that decision right is is do you decentralization is very hard it involves a bunch of regulatory risks right and ultimately is that pathway worth it um uh you know for the end goal if you built a system that you know has a legit business model right and is earning money like what are the drivers for actually deciding to decentralize and taking on all of this risk right and I

think that the what it comes down to right is you know I I don't think that every single blockchain network is going to be an open one but I think that the way that we think about it is that ultimately open networks are better than closed networks and so as a result of that I I think we we tend to maximally believe right that opening your network is ultimately going to be the thing that drives the motion value and that you know as long as your network is closed it's always going to be somewhat constrained and so as a result of that I think you know and I think it's too early to like know whether or not that that bores out like successfully or not right um but I do think that that fundamentally like if you're running a closed Network you don't have a token

you're not decentralized right there's always going to be the risk right to users of that platform that you ultimately start to extract more value over time just like the corporate networks of web 2 did and so you know it I think we often write like struggle internally right like how much do people like everyday users actually care about decentralization right and I think that users care a lot less about it but I think that people that are building on top of these platforms are going to care more about it and I think that we you know we start to see that with a lot of the web 2 stuff right people are kind of aggravated over the fact that the take rates are so massive you know Apple's app store has been one of the most successful like project projects ever

right but at the same time people you can starting to see cracks in kind of the lack of trust of these institutions and the you know the fact that they're taking enormous take rates um and continuing to kind of flex on on builders yeah I think I think that spot on to me like a big part of the promise of web 3 is that compatibility across the app layer to where you have some of the web 2 players actually saying you know what you can achieve um the same ux using the thing you already know but it just seems tough to get them to actually uh adopt The decentralized decentralization Primitives without first achieving a pretty dominant position in the market yeah look I think I think that's fair I also don't think that it's necessarily bad to build a centralized business that

has product market like I think it's a good idea right to try and find that product Market fit and then struggle with the decision to whether or not to decentralize I think if you if you look at openc right I think that they've obviously built a very successful business um but the longer you go right the harder it is to then change the course of of the direction that you're going in to then decentralize right because if you start to build out internally organizations and or organizational structures that are very centralized and are dependent on centralization have hierarchies and all of these things right to them very difficult to say well oh now we want to actually decentralize and so as a result of that I think like I think it's beneficial to kind of you know build with decentralization in mind but obviously the like the commercial realities and

practicalities of that are very challenging uh I want to ask about the client kind of level and the monetization there there's another good question I just kind of wanted to yeah uh go on top of that like you mentioned there's like for example liquidy and like others that are taking like percentages of their revenue yeah but do you think it's maybe similar to um like royalties in a way and how those are like users like you know if I'm trading like maybe like millions of dollars like I'll go to the front end that you know that doesn't have that percentage and is there a potential like a race to the bottom or is that like more of like a um yeah do you think there's enough defensibility around basically front ends to justify that yeah well so I I think what I would say right is that

that is exactly why I think commoditization of that client layer is problematic right because if um if if the protocol does not reward the client for having a transaction routed through it right then you're always going to end up with a commoditized client layer because fundamentally there's no business model in like oh who gets to Route the most transactions through right now or you put it on those those front ends to charge their own fee right on top of whatever the person has to to use to trade through and so as a result of that like you I do think in that scenario you end up with a race to the bottom where it's like whoever just charges the least fee right and I think that if you but if you flip it and you

have fees accruing directly to the client layer right then those clients can basically figure out what business model works for them to basically how much of that money can they pass on to the user right versus uh how much do they need to keep in order to like um actually you know have a user experience that people are going to want but I think if you if you think about um you know unit swap for example right there hasn't been a lot of like robust Rich uh clients that have been built on top of it right there's no um you know like robust framework for people to to use it's all kind of like uniswap's very simplistic uh application that you know that people can have replicated um across the ecosystem so as a result of that I think I think that

the without having protocol fees going to clients right you're always going to end up with that most simplistic cheapest uh situation and I think that that's incredibly problematic when it comes to like regulated businesses where you know we obviously are going to see that within D5 and then it's going to become a question of like what types of regulations are going to apply to let's say forecaster applications right you know there there are privacy regulations and stuff like that that are going to apply and so it's really hard to see a world in which you know those applications can actually survive if they don't have a revenue stream yeah so then I guess like you're saying that the protocol should incentivize like clients to I guess be built for it and then in that way they yeah they basically can then they make better clients and also

even Mark like incentivize marketing by basically giving giving funds to whoever can route the most users to it right exactly so and so if if you're again like let's say your C Port um if you're taking a you know let's say the protocol is taking a one percent fee and passing 1.5 off to the client layer right I guess the the question is right is is the protocol better off taking 2.5 and the number of transactions it generates or if it takes one percent right gives 1.5 percent to the client layer are you now going to generate more transactions such that like the overall value to the protocol is higher awesome well thank you so much for that all right that was great [Applause] [Music] thank you