'Layer2.finance: DeFi City's First Public Transportation System' - Mo Dong

up we have mo dong uh leading the seller project uh seller is actually another team that's been working rigorously through the bear market uh and so they were focusing on state channels they also uh you know created a really revolutionary state guardian network and what i most remember seller for is the gomoku game that you guys did that got a lot of traction uh so welcome mo yeah thank you thank you for the invite and it's great to be here so uh today i guess i would just first share my screen here and uh you know talk about the uh uh some of the new things that we're working on um so as uh oh let me just see if this works all right all right uh as uh eva had like pointed out we are a scaling platform and we started out with uh uh state channel uh solution and we actually are the first one launching on ethereum with generalized channel and you know the generalization outside of things are running very smooth right now the the initial gomoko game right now is has evolved into a whole platform called zerox gaming platform and the games build on top of that is processing payments using uh state channel and now has approaching uh one million users in the u.s market actually um but most recently we're also kind of looking at different kind of scaling technology and we always see layer 2 skating as a holistic landscape that we need to have combined different kind of technology pieces together to offer a holistic view for the uh for the market and for developers so this is like one of the motivations that uh leads to uh the things that i'm going to talk about today which is layer 2 finance and what this really is is the d5 cities public transportation system and i'll get back i'll get into how that is the case what we all know d5 is awesome like basically uh this this summarizes it all uh the most stable and safe d5 protocol provide 10 to 50 50 of apy and if you want to go to uh olympic djing world it's really no no no upper bound on this while we look at the entire traditional financial world uh you know center along like uh banks traditional banks even if we talk about think tank banks new banks chinese banks these guys are always 50 times 100 times away from the kind of yield that can be offered by the d5 space you know what this tells us is that this huge amount of a different difference between the kind of yield we should see a huge influx of demand and exponential growth of user user numbers well not so much so this is these are the data we pulled from doing analytics and you know what this the graph is telling us is that you know after the last d5 summer that we're seeing like a very booming trend for the entirety of spice space the internet effect grows the key protocols on them i'm just showing some of that is actually slowing down this graph the y-axis here is the data day-by-day user growth new addressing and the access access here is actually the time so you can see that after the huge spike around the last summer everything started to quiet down a little bit and another interesting way to look at this is that we look at the total amount of the daily active user for the entire d5 space this is also data we pulled from dual analytics and so basically based on launching data we're seeing about 40 000 daily active user in this entire device space okay so we have forty thousand dollar dau in the entirety kind of work what does this mean exactly okay so it basically means a u.s apple app store based app i'm just talking about u.s market and it ranked about 500 in cairo game category which is like one example is a cut rule the this particular games the daily active user exceeds the total daily f user of the entire divide world so d5 is still small and it is not exponentially growing it backs the question of why really why that is case it just doesn't make sense so um you know one one way to kind of look at this problem is to really visualize the device landscape as a city right so like you have your home which is your metamask what it and then you have all these uh shops which is d5 protocol you can park your money on at and kind of a earn reward now the problem happens at the point where you want to travel from your home to these shops to actually drop off your money and the the problem about this particular city is that this particular city only have one type of transportation system which is expensive limo service like super expensive highly expensive taxi this works fine for alice who has like a million dollars to play with so alex goes uh you know to come a shop called compound which is offering ten percent annualize the game paying about 250 dollar for the demo fee and what's gonna happen is that uh uh you know after one day sitting in this particular protocol with one minute dollar minus that limo fee um she can turn into profit in just day one now let's look at a different scenario who is bob early majority uh he wants to try this out but only have like one thousand dollars two some dollars to just try the water and test out what what's this d5 all about to actually get hands dirty so same thing but just as uh the uh uh the limo fee is is fixed no matter how much like uh money you're actually taking to these shops and you know what's gonna happen to bob is that after even after one year bob still lost 175 dollars after parking his money uh to compound while paying that the super expensive limo fee now we probably all know that what this limo phase referring to is basically the layer one uh transaction fee we need to pay in these uh you know these days like to interact with uh these kind of protocols now this is why we believe that d5 hasn't really started to grow exponentially because for many of the early majority adopters in the space it is actually very hard to get to the promised apy while you know eating all this kind of up from the gas feed cost and the kind of learning frictions of using define so you know any kind of a reasonable city designer and city planner would at this point say that hey this totally doesn't make sense we should have like a public transportation system in this entire city where you know people can just gather around and take the train all together and split the uh the vehicle costs or split the gas cost of this entire train and everyone will pay like one dollar 250 limo fee well this is exactly what layer two finance is trying to do layers of finance is trying to build a uh layer to roll up system that essentially act as another layer on top of the existing d5 landscape without actually asking the d5 protocol to move to a different layer but just adding another layer of liquidity on top of that to allow user to use defy system as if they are taking a public transportation system in the default world so this is what layer to finances looks like let's uh walk through a very simple user flow here and uh you know it's right now live on robson this is actually the new ui we're launching in midnight next week um so you know in this particular uh uh you know uh ui the first step is really uh just move user's fund from layer one to layer two well you know using the same kind of a subway analogy you're basically buying a subway ticket this time like you're entering this entire uh you know layer two space and then the next step is uh among all these protocols that you see like basically we're actually aggregating all these protocols together like the compound the wire alpha ape and all these other future protocols and strategies the user can express their intent to actually put their money in so user can actually click to deposit and specify a certain amount of phone to actually deposit in and this is like they're now getting at a station to wait for a train to leave okay so what actually happened under the hood is that they're sending a layer to transaction in this layer to load chain indicating that oh i want to allocate my phone to compound by how much and what's happened after that is that they will have a aggregation pending phase where this draw option is is essentially aggregating and then waiting for a sufficient amount of user who are all going to compound basically you know having the same kind of intent and then what's going to happen is that it's going to aggregate all these demands together and form a kind of a layer 1 call basically layer 2 to layer 1 call in the form of kind of a commitment and saying that hey now let's allocate the fund pool in layer one to basically direct you know three users combine the fund all going to compound at once and therefore you know if we have like 100 users all wanting to go to compound in this trade they will be able to split the cost by 100 times so basically everyone will get the cost reduced by 100 times so this is like the basic idea of layer 2 finance and using row up to solve the d5 kind of a problem and then after that the user will be able to just directly uh get their gain on layer two uh while their fund is actually moved to that oracle the good thing about this is that it doesn't cost any kind of liquidity fragmentation so basically you know you we don't need to push different protocols into different chains we're merely moving users and their liquidity in layer two and aggregate their demands in layer two and the combiner demands back into layer 1 and allocates the entire poor fund all together at once and split the single transaction cost instead of asking each and individual user to make one single layer one transaction cost all by themselves so this is basically the uh the kind of a high level idea of uh uh you know layer two finances this is what we are we're building and actually um you know it's there is a live test net running today and we're also releasing a new um you know ming nut with this new shiny ui uh next week so the immediate question is hey wait wait so this this sounds like really something like you you know i'm putting my money all together and someone seems to be managing it so like either possible for the uh layer two or whatever the system you're calling it actually be malicious and actually take the money and just wrong um well that's a very very legitimate question the good uh the answer here is that no really uh because this is actually built as a in-house optimistic role up uh solution right now so uh you know how the process works is that a user starts from layer two layer one and they all have fonts the first step is uh you know they will put their font into a row up front smart contract just like any row app they will just put the phone there and then they're moved to layer one and layer two essentially they have balances on this layer to finance the roll-up chain and then they will specify their uh you know uh fund allocation intent basically to this layer two finance chain so this it's very critical note to note that this type of fund allocation intent now becomes the transaction on this special layer too and what's going to happen then is that a see for example these three intents are all sharing the same kind of a strategy goal and what's going to happen is that uh this rollout chain is going to oh i'll of course pull post all the other call data and stay ruled and very importantly a fund allocation proposal in this case it's basically telling the layer one fund for contract and say hey i can prove with the call data and state route that inc in combination you should allocate 300 to this strategy zero which is let's say compound and then there will be challenge periods passed um and what's going to happen is that the fund allocation confirmation will happen and the phone call will actually allocate fund into the corresponding strategy adapter and then the strategy adapter themselves will actually interact with d5 protocols to actually allocate font to the default protocols it's essentially kind of a you know combining many user's fund allocation strategies and the transaction into one so this is how this is done now immediate question would be okay does this actually cause a lot of kind of delay in the in the sense well not so much there of course will be the aggregation delay but the challenge delay is definitely not going to be something like multi weeks it's actually going to be in the hours the reason for that is that the roll up state in this particular row up is so much simpler because it's not an arbitrary evm it's just like a kind of uh you can think of this as kind of pretty simple ledger that is i'm putting my money into which protocol or which strategy and the the the type of transaction is also pretty simple so it's very easy to verify it's very easy to catch any kind of malicious intent it's extremely cheap to challenge in the case of any kind of a malicious attempt so let's talk a little bit more about the strategy side the strategy can be very simple direct the wrapper to the existing uh strategies like compound where you know we basically wrap the compound with a auto harvest of the liquidity mining token com token function uh but you know and curve and make you know all these other protocols can be very simply wrapped around and now you know uh what can also be done is we provide this thing called layer to finance strategy sdk to allow users to basically develop a strategy as long as it's conforming to a certain interface that is connecting to the font pool um you know it can be plugged into this uh layer 2 finance system dynamically what this means is that you can create a strategy for example put ease into maker and you know land of die at compound in the early interest rate and you know we can combine multiple kind of strategies all together that's all perfectly fine the the good thing about this is that you know you can really start to expand the complexity of the underlying strategy while maintaining an extremely simple user interface for the user to use so in a way you know let's finance become a very nice to use and easy to use aggregator and at the same time it avoid any kind of you know the key challenge people face in today's layer one aggregator where you know in today's layout aggregator people are actually paying higher fee because this additional layer is actually causing higher amount of fee but like actually in the uh layer to finance um the retention of user using this aggregator is going to be pretty strong because like you know they really they will they really will be better off using layer2finance and at the same time uh you know uh uh uh in terms of like a transaction fee is also much lower comparing to directly using their one so for that we are offering uh seven hundred five thousand dollars to uh as prices to build using layered finance strategy sdk with two prices or three thousand dollars uh to the best layer to finance strategy and uh once on five dollars each for three runner-ups and you know we hope that to see the community start to build some very cool strategies and connect them to layer to finance and allow more users to use there to finance and you can always find us at east global discord space and you can follow us at the center network twitter and we have pinged the tutorial video on our twitter and all the relevant codes and documentation can be found in our github so uh you know we're basically trying to uh uh build a d5 public transportation system so to allow everyone to be able to easily access d5 uh so yeah that's uh that's all from us awesome thank you so much mo and it looks like we have a few minutes for questions if you don't mind sticking around uh sure yeah i can definitely stick around cool the first question i have for you is you know you've been on this journey for a while as you see you know uh gamoku into a gaming platform starting from state channels to where you are today um what has been the biggest learning for you you know as a researcher in how you've seen l2 scaling solutions evolve yeah i think uh you know uh the depicts the learning is that it's always a trade-off game and uh you know uh it's basically defining and uh finding out what he said the right trade-off uh to choose uh you know in the in the early days like if you think about uh think back like in in the plasma days people are thinking we can achieve this kind of quadratic scaling by cascading plasma layers like a one layer on top of one layer two and another layer two and then you know along the way people quickly find out that the you know the trade-off doesn't actually pay off while we kind of continue to cascades and uh you know i think i think the biggest learning for us uh is that uh you know um there are different kind of applications narrow suitable for different kind of underlying scaling solutions and you know our goal is still very much the same as we first start out which is like trying to combine different technology pieces together into a holistic platform and a holistic application to offer to the end users and bring more kind of end user to the platform yeah got it and how are you guys thinking about the future you know so we've got quite a few layer two options you know and my personal view is that every dap every layer one chain you know they'll choose the solution that makes most sense for their use cases um but as we also expand to e2 how are you guys thinking about sellers role in scaling ethereum yeah definitely i mean like you know there are there are a couple of different pieces right so like you know while we're moving to this multi-chain work uh you know center also have this uh uh you know our state channel solution is easily extended to something called sea bridge which is uh able to bridge different kind of uh uh chains uh from layer two layer one and all that stuff so uh you know we're we're launching pretty soon with uh uh all the other uh layer two solutions uh together and uh it will be something pretty interesting that everyone actually can participate and provide liquidity for the for the liquidity bridging between different chains uh it's kind of almost like a liquidity pooling um yeah which is pretty cool um and uh you know uh for the roll-up perspective right so like there are really two uh kind of a stream that is going on one is that let's build a separate chain and move every application possible to that chain and what's going to end up very soon um you know towards the end of this year is that we'll have we'll end up with almost a shortage of exam basically you have like multiple kind of roll ups that is you know spinning up um it's good for scalability but really horrible for ui ux especially for new beginners like they they will have hard time in terms of figuring out where to send the phone to and the ecosystem also need to catch up a lot like these exchanges need to be able to figure out how do they connect with different kind of ethereum does the layer 2 exam actually have the same price as the the underlying serum because there's always this kind of inherent friction there and you know these kind of questions will actually create increasing uh kind of uh user usability hurdles and our approach here is that you know instead of taking this view we're kind of trying to take a unifying view but but but like basically helping all these layers and all these kind of different chains uh be able to aggregate together all of their application to aggregate all of their application together and into a unified interface for the users and into unified pool of liquidity for the users so from a user's perspective they don't really care like where they're traveling as long as they're in this kind of d5 subway system they can kind of a travel intercity or kind of across different city basically um i really like that analogy of like traveling between cities and each layer you know two or smart contracts its own little village um maybe one last question and this one can be sort of a prediction um so you know it's very clear that d5 needs layer two as we can see through gas prices and just you know massive adoption what use case do you think will be next for really needing layer two to solve the ux and and getting more uh users on board yeah so you know um i think i think uh um i i honestly think that we are still like just scratching the surface on the decentralized finance and open open finance space and i'm seeing a lot of the exciting things popping up like uh with a very exciting new uh dax uh kind of design and uh things like that and what slayer two is gonna enable is that it's just like 4g enabled the tick tock basically like you you couldn't have tick-tock in 3g networks in 3d mobile networks and you know with kind of this layer two capabilities the capacity is going to be quickly filled uh in layer two and with a new type of application that that couldn't be built uh with the existing layer one capacity capacities and costs basically um so yeah and uh i think there are a very large class of open finance applications that can be built and couldn't be built today because of this kind of a high gas cost and though you know i'm really just looking forward uh to that and aside from that uh you know i'm um uh you know i'm looking at various different things but like i haven't found out that it's something that is uh extremely interesting except for several very interesting experiments going on with some kind of social tokens and uh social interactions and kind of a self-organizing stuff which is uh kind of dated back to in even the doll doll time right so like the you know it's still social coordination but like i'm very interested in kind of how some of these experiments gonna go awesome yeah looking forward to seeing you guys also support a lot of these apps that are coming to fruition yeah definitely thank you thank you thank you guys so much