EthCC 2: Liraz Siri

hello everyone today I'm going to be talking about a sustainable model for developing open source decentralized networks specifically layer 2 apps and apps we just heard a presentation from a company trying to build adapts and they're running into are boarding problems and they want their dabs to interface with any browser without having to the users to deal with gas fee so I'm going to be giving that as an example this would be building on the work that you have vice my CTO has been presenting this conference on the gas station work network the 1-800 aetherium so abusing an extension of that as an example of how we could fund the development of a decentralized in fira that is an interesting use case it's pretty simple so this is not what we do this is a gift we're giving to the community it's a template we have to come up to solve our own problem what my company does it's a cyber insurance platform for use cases such as crypto custody or controlling fleets of autonomous cars anything that's very high risk and you can afford to trade off time for security and you're doing risk transfer to these Guardians the posting collateral and they're on the hook for letting anything that's not faithful to the intentions of the owner of the asset go through I'm not gonna really talk about that just that when we started working on bringing all this into production realized that there are a lot of missing pieces in the theorem ecosystem and we're having to solve some of these infrastructure level issues ourselves one of them is that there was no sane way to fund the development of a decentralized open source network and this is what my talk is going to be about so what's wrong with current funding models why do we have to fix them so right now we have three options well the last two years we tried I cos we if you want to now you know develop this open-source system you can try engineering it so it has some sort of unnecessary utility token in the middle and that is first a bad bad economic design because good money doesn't make for for a good investment it's it's conflicting interests and it's equivalent to saying that we're going to build a world in which if you want to buy a service or good from a start-up you have to go to a stock exchange you have to buy that startup stock and you have to you have to use that stock as the means of exchange with a start-up it doesn't make any sense but doesn't even make sense in terms of actually working to provide return for investors because after you develop the system there's a very easy way to make a better version that's just removed the necessary utility token so the network will be forked and there won't be totally token so you won't be capturing value for investors so that doesn't work so the other model is grants but grants is is this game that's self extinguishing because the generous grant providers are basically indirectly donating their wealth to the ungenerous free riders that are not participating in the granting process and if you play this game for long enough you end up with a an ecosystem that doesn't have any generous grant providers it only has Free Riders and that gets you to the tragedy of the Commons where they're all these pieces of infrastructure we wish we had but we don't have any way to develop them because you were using relying on on just the grant model so we've been talking about maybe doing things with securities instead because you know we've done that we did that with the internet before and we've that's how we funded innovation but that doesn't really work for the centralized systems or system that have a want to have a path towards the centralization and the problem is if you're creating a financial security it's inherently centralized around the legal entity in a particular jurisdiction and you can't get around that there it will never pass the walkaway test you won't be able to do what Satoshi did and say all right I'm done you guys take it forward from now on so and yet the other thing that's problematic about this is if you are creating a financial security creating a legal entity to protect the value created for investors you have fiduciary duties towards your investors you have to do things like maximize shareholder value file patents try to protect your IP and that's doesn't align very well with the spirit of decentralized open source networks so what are our goals what are we trying to achieve with the this this new model that I'll be proposing well one we want open source to be sustainable so all is these missing pieces of infrastructure like layer 2 networks whether it's for storage or for private computations or fast payment systems anything you could do in layer 2 network or or at the level adapt if it's a public good we want better models for funding it that are sustainable ideally what we would want the same the same type of process that happened in the mid-90s where all this private investment went to fund all the innovation that gave us the modern internet that could not have been done with broken funding models the other thing we it's really important for us to achieve is is a incentive compatibility which which is a technical term that means that we want to align the interests of various stakeholders so we're putting the bootstrappers of the network people who develop the network and then their investors in the same boat as the users of the network now if you want the system to be decentralized we also have to care about well you know how what type of relationship are our investors going to have with the real world with regulators will we'll be able to list whatever we give them and it's important for us because we still live in the world where the regulator has a lot of power to have a path towards regulatory compliance and right now securities regulations means you can't benefit passively the work of others if you want something to not be deemed security also the ultimate test is the Hinman test the walkaway tests can you do what Satoshi did can you leave the network and it still works or is it centralized around the founding team if it is centralized around the founding team then you may make sense to ask the founding team to disclose the financials and and and continue to act as as a central point of failure and another thing that would be really nice as a bonus if we managed to figure out a way to do this that didn't involve creating useless new forms of currency that we forced into the system then we could standardize on the best form of money all we could create a network effect where all projects are using let's say dye as as their money and they're using something else another financial instrument the same way that equity is used separately for money equity is something that investors use and money is used by users to consume goods and services so if we did that network effects would not only improve the user experience dramatically because you wouldn't have always liked to get the the specific type of token and try to like figure out price discovery on it all these problems and pay for transaction fees but because the underlying good money would be backed by ether we could drive a sustainable increase in the value of ether so the example I'm going to give of a problem we can solve with this model is based on an extension of you always work on the 1-800 aetherium gas stations network and our goal is to get to the point where like you know the previous talk mentioned they ran into this issue they would like all these decentralized systems are creating to be as accessible as centralized servers like you want any browser to be able to access any decentralized network without having to go through a crazy onboarding process without even eating a wallet hopefully and certainly not having to somehow get ethan advance to pay for gas so the gas station network solved this problem and we're giving that as a no gifts no no strings attached gift to the community but what we realized recently is we could use the same architecture to solve another important problem this is a problem that cystic works to the ecosystem this is the problem of decentralizing in fira so if we want browsers to be able to interact trust lessly or any like client with with with these decentralized networks we have to figure out a way to do that in a way that doesn't rely on a third party like in Fuhrer in the middle and it turns out we can use the same network architecture and won't explain how that works in principle so you have a system of relays and the relays they have to post the stake in order to participate in this scheme and they also post their their address where they can be contacted this is the same as the gas-station network but instead of writing to the blockchain what we're asking these relays to do is give us truthful information about the state of the blockchain right and they do that in a way that commits them to being truthful because they're signing sort of like a pledge this is true what I'm giving you right now and if they are untruthful we will find out because we then contact another relay in the system and we asked them can you please confirm that what the previous relay told me is true and if the relay confirms it adds its signature it would be received a like a small percentage of the subscription fee in the system and we can continue doing that until we reach certainty or I really could say wait I'm not adding my signature to that because it's a lie and then we know something is up we'd be able to penalize the relay that lied to us and take all their stake so we're enforcing standards standards of service and we're enforcing them we're bonding them to be truthful and because this this model the costs would be for running these relays would be similar to the cost of running a VPN network mostly bandwidth server fees like you don't have you don't actually have transaction fees if you're reading for the blockchain so it would probably make sense to run this this economy of decentralized and fura in a stable coin so your your cost your staking costs are fixed to your server costs and bandwidth costs and in exchange like the apps that want to provide frictionless access to themselves they would be subscribing to the system they've had like a subscription fee and then all the relays would share this subscription fee among among themselves the one that the Ruby they provide service and you they would be posting a stake and die in order to attract customers to use their their relays but yeah we can get this to work but that's just like the key principle there's a lot of work to actually get this into production and replace in Fura so how would we fund decentralized in Fura all this this development needs to happen how do we how do we get the through the bootstrapping process other than just grants or utility tokens so we need to solve the same problem that equity solves for a corporation but in a decentralized network and what I'll be proposing is a no utility token model for for bootstrapping layer 2 and apps that is sustainable so introduce a few of the pieces at the heart we have decentralized equity tokens it's like a bit of an inside joke because that we all we hold we have a debt towards the bootstrappers of this infrastructure that we need to repay them somehow so if you participated in the bootstrapping you would get debt and then you would be a debt holder now what you do with the debt is you will be able to opal I open these collateralized debt positions this contract where your your your staking debt and you get something in exchange for it so the this economy the example economy would be running on a using die or using die as staking token but you could also play this game with credits so these are these are denominated and die but there are special-purpose loans that are non-transferable you can only use for staking within the system and there's a ratio of credits to die that's controlled by a governance mechanism but just the the go through the simple flow of how this would work so alice is a debt holder and she wants to participate in this game so she stakes her debt as collateral and she receives credits so then she takes the credit well the seed I in this case because the denomination died and she she stakes them in the relay hub contract so she registers a relay and she posted the her collateral and she died now in exchange if she does if she that really does its job well she gets a percentage of subscription fees but if the relay doesn't do its job well then Alice's is on the hook for returning the lost funds and if he doesn't do that like she goes into a delinquency process she doesn't do that then her debt is auctioned off to the highest bidder now when she wants to stop playing this game let's say she wants to sell off her debt to a secondary market like you she would maybe she had a stock then she returns all the credits she gets to keep the profits that unlocks her debt and now she can do whatever she wants with them so the question is how does having decentralize equity leverage growth why not just wait for this system to exist and then play the game with die why would you want to participate earlier why why would you want to be exposed to earlier risk and the question and the answer to that is you come in earlier you get exposed to more risk but in exchange you get exposed to the growth of the network over time so you're leveraging more than if you just came in later this is how it works so the critical ratio is the ratio between C die and an and I soaked red O's to die now if it's 100% nothing interesting happens as far as the bootstrappers are concerned because they can't play the game but if we set the ratio of C die to die late ninety percent now if we have a million stick guy die in the system we can the debt holders can receive up to 100,000 died as the special purpose loan now as this the economy grows decentralizing Fuhrer becomes very popular the subscription fees make it more economical to run more nodes and and and stake more then people are staking more died into the system and if we keep even if you keep the ratio the same we get more more seed ie as loans to play the game with so this is the centralized in fira after after credo or using credal then you could just use either die or c die for the staking and then there's this mechanism that that makes sure that the the debt holders are accountable for losing a seed I so they're playing the same game now in terms of regulatory compliance that's also important because you're not benefitting passively from the work of others the only way you can benefit is by actually taking your your fair share of risk and profit the other the other maybe even more important reason we think this is going to eventually have a path toward regulatory compliance is that if you set up a system in this way there is a future not not too far down the road where the system doesn't have any centralization around the bootstrappers the bootstrappers they could all disappear they could pull Satoshi and the system still works so what are the use cases for this well pretty much any layer 2 network or Ord app that has staking and we we have a paper that gives a detailed example of a layer 2 network we're calling cash chain which is like a visa scale network that would use this it would be expensive to develop but then there's there's a return so as expensive as that is it would would be worthwhile we would have to fund that with just philanthropic or grants so all these projects use taking and they could all potentially use credo and probably have a better user experience and better alignment between the the interests of different participants especially the bootstrappers and the users so what's the current status we have a draft of the paper explaining how credit works I just explained the key insight but there are a lot of nuances to getting this to work in a way that's not vulnerable to abuse and the paper goes into the mechanism design the various layers of Defense we also have a simulation a computer simulation and net logo which is an open-source agent modeling system and we are seeking comments from anyone interested giving them access to the paper or the computational model and we are interested in receiving feedback before we like you publish this officially so if we have time three questions from the crowd great talk what can you walk through the user experience of what it would look like forget coin for example to adopt it right well actually Alber experience so you would need to he would need to change the currency of the system instead of having our utility token you would say okay so we're doing everything we're doing all this taking you to get coin we're doing it and die and and then you would have this you would inherit from a template where you can also play this game of staking not just with die but also with with C die then there's this governance mechanism that determines what the ratio it is but you get hardwired that is not at ninety percent for example right and then you just have two classes you have everybody that's participating who is in the bootstrapper and they just play the game with die the get coin game and everybody like the debt holders they if they want to participate in the game they're playing the same game but they're doing it with like a derivative they're staking their their debt they're decentralized equity tokens they're getting this long basically we're using the mechanism of credit expansion right but the loan is collateralized by the debt by the decentralized equity tokens and yeah I mean it would it would user experience it would just be oh now get coin uses die that would that would just be it but you would have to think about having utility token or just be transparent so usually when you stake you want to have a reward for staking that is often implemented through inflation but in this situation you would need to find another way actually earn money and that process already right absolutely I mean the the mechanism what you get in return for staking is is specific to your economy I mean the general case is you're taking a risk and in exchange for taking a risk or bonding yourself to some to a duty a to perform a service then you are receiving transaction fees or a percentage of subscription fees or premiums if it's an insurance platform okay go to here comes my second question so you said there's this governance parameter and you had an example of 90 percent so that means that there is kind of 10 percent extra in circulation or something that can you just play through another like percentage number and compare them in terms of risk for the investors right so it's not in circulation this the credits are non-transferable they can't be withdrawn from the system they can only be used to stake within the the local economy so they're not they're not circulating but but yes now you're playing the game you've added another 10% worth of staking coin like inside this economy you've created the credit economy then now you have like 90 tocut and if you have 100 100 100 die in total 90 of that is is died that maker doubt created and 10 of those is see die which you created using the credit staking mechanism alright thank you everybody [Applause]