ZkPodcast: Gnosis & Full Node with Friederike Ernst

Transcript

[Music] welcome to zero knowledge a podcast where we explore the latest in blockchain technology and the decentralized web the show is hosted by me Anna and me Fredrik [Music] in this episode and the chats with Federica from gnosis they cover prediction markets wallet security exchanges and more so before we start we want to say thank you to this week's

sponsor trail of bits trail of bits recently released their smart contract audit executive summary they aggregated the work of 23 smart contract audits and they found that 78% of high impact easily exploitable findings are discoverable with automated analysis tools 50% of all findings will never be found with automated tools and according to their research unit testing has no impact on security for more about this audit have a look at one of their recent blog posts at blog trail of its com we'll add the link in the show notes so thank you again trail of bits and now here's our interview with Frederika from gnosis so today I'm sitting with Federica from gnosis and full node to

talk about the journey that gnosis took and some of the work that they're doing and hopefully more so welcome to the show thanks for having me so besides you know gnosis in full node you're also known for being one of the co-hosts of epi Center yes I thought we might actually kick that off given that you're now on the zero knowledge podcast mmm-hmm how did you come to be on epicenter how did that start out for you um the existing house asked me so basically maiya approached me and asked whether I think I'd enjoy being a host on the show and I was somewhat taken aback because I've been a long-standing fan so I've listened to almost all

episodes and to me it was just such an institution that the idea that I should be one of the hosts just seemed sacrilegious the way and then and then you know I quickly said yes and it all started so epicenter for those of you who don't know is a fantastic blockchain Casas and I think we're often like zero knowledge and epicenter tend to be like put in the same category yes then it's funny when you joined it I can totally imagine it's almost like crossing over to the other side how do you how do you like it like what what have you taken from being on this other side from doing the interviewing oh it's so

fantastically interesting I mean basically you get to sit down and take a deep dive into someone else's project that's something that being an entrepreneur most of the time you don't get to you just sue in your own soup a lot yeah yeah basically just making the time to look at something else in depth and thinking about how are the parts function together and how the intricacies work and basically coming up with your own hypotheses and your questions and kind of then speaking with the person who is put their live into this it is such a good experience it's a it's very humbling so it's funny cause like in this episode I'm actually gonna

be digging into your world maybe we get to come back to the FE Center stuff as we go on but I think for those who may not know you why don't you tell us a little bit about your background where you come from sure so I've been in the in the I've been interested in adoption technology for a long time so I think probably since 2013 and at the time I was still I'm a physicist by training so at the time I had just finished my PhD in low dimensional complex quantum systems and I was about to start my postdoc I did a string of postdocs and then left academia for good because I thought I loved building new things and I love setting up systems and they're love

things that have the potential to change the world and I could think of no reason why I wouldn't want to do this full-time cool did you feel like you couldn't do that in academia because that's it that's a big shift like it's a total shift in mentality to get out of that cycle of academic research or doing PhDs having courses having teachers yeah and all of a sudden you're sort of thrown out into the world and like judge like how you gauge what you're doing is very very different yeah that's true so in in one way it's very different in some in some respects it's very similar it's I mean it's basically building things in modular ways so that you can improve upon the modules and systematically thinking about the best

ways to actually put together system architecture thinking about the ways that you can put together and and things that you've already understood in novel ways that's a lot like physics and then there's the additional bonus that basically all physicists can hold so it wasn't complete it wasn't foreign to me so it was very much the way that I was already thinking I all right I also had a degree in math so basically cryptography is also it also wasn't terrifying wasn't terrifying to me at all I actually quite liked numbers so what's very different than in academia and academia there are these boxes you

have to tick and buttons you have to hit and building things and finding things out has always given me immense pleasure and I'm I'm not good at being micromanage or directed so I actually quite like doing my own thing and in academia to a certain extent that's very possible but you're still within these confines of you know the things that the boxes you need to take the buttons you need to push sort of thing and entering this blockchain word at the time it was still quite easy to raise money and thus get the freedom to also do your own thing without having other people breathing down your neck to me that was kind of an ideal situation an ideal outcome so I mean obviously there are

advantages of being strapped for cash now also advantages of not being strapped for cash as a startup and I feel like I don't look back so I I still I'm still interested in physics and I still follow you know the latest developments in in the mainstream mainstream media but I've never regretted leaving physics tell me a little bit so this is a little bit about your background but let's hear a little bit about gnosis am i pronouncing that correctly yeah so basically there's different method approach and implanting noses I also say it without the G so for those out there and spelt G n OS is yeah and I have I have I think in the past

mispronounced it as well so I think noses is kind of the Americanized pronunciation I think the real pronunciation is kenosis with with a soft G and the original Greek pronunciation is kenosis which is the one financiación that we will not let slide so good gnosis and gnosis is both fine to us but it also is just I don't know the word does not cover this word mean um it's the Greek word for knowledge oh that's amazing so okay what is gnosis and I'm gonna go with gnosis if that's okay yeah that's completely fine and so norsu started off as the very first spoke within consensus and set out to build prediction markets on e

theorem we have that's still very much our core mission we have found that there are building blocks that we need to build in order to get there so now we still have very much this Precure market focus and focus on conditional tokens that we can talk about later as well but we are also building smart wallets so smart contract-based wallets that let you interact with taps more efficiently and have distinct advantages in terms of usability and user experience and security and we also build novel market mechanisms in order to trade outcome tokens for our prediction markets and other things but is is gnosis then is it

like a because now it's it sounds like there's a number of different projects that gnosis is building to reach a like a larger goal but would you almost think of it like a is it a research company is it a dev house is it a is it is it in itself like a miniature and this may be a weird way to think about it like a miniature version of consensus where it's like spinning out kind of newer ideas yeah so I think you know you're you're hitting on a sore point for us I think basically all the things that we've built and that we're building super interesting in their own right if I had to do it again I think I would put more focus on focus

so basically thinking up a new project and thinking about how much sense it would make if we if we had that and how wonderful that would be that's not good enough so basically I mean we're we're only human so basically you can't divide your yourself up between a number of projects and still be as efficient as you could be if you only had one that being said so we have we have taken great care for the last year or so not to start any more projects despite the fact that we have a gazillion ideas so we will concentrate on the three that we have now because they also fit together really nicely in this triptych of create trade and hold so create in this

conditioner token framework trade within these novel market mechanisms that we're building and hold within the smart wallets okay let's go let's kind of step through that because I actually don't think we've ever had anyone on to talk about prediction markets like we have to kind of go back a step and define like what is a what is the gnosis prediction market what was the first initial idea of gnosis the initial idea was that in principle a smart contract based blockchain makes it possible to create tokens that hold worth only under specific conditions so things like one euro under the condition that Bracks it

happens October 31st or one euro under the condition that it doesn't so in principle if you put these two together they're always was worth one euro because either one happens or the other but if you actually have some sort of external input that tells you whether brexit happens or not you could you can make them resolve automatically and you can trade them individually so you can have the yes token or the no token and the value at which they trade so say 30 cents on the dollar ninety-five cents on the dollar gives you the probability and that the market assigns to that event happening not happening so in a way it's a lot like a stock market on you're not

trading parts of companies you're trading information could it also be I mean it sounds like it comes a little bit out of like betting and gambling possibly is it like or what is I mean this this system without the blockchain I believe has existed for a long time right like this is like even digitized but not blockchain necessarily so basically the answers yes so basically betting transfers to this one to one so basically any betting market you can actually display as a prediction market it's not true the other way around so basically there's a lot of contracts that we enter into in an everyday setting that kind of have the same payout structure so for instance say you

have fire insurance for your house then basically in betting terms you could say it's a bet on your house burning down because there's a payout if your house burns down and there's no payout if it doesn't so that's kind of exactly the same payout structure and so yeah bets to have that payout structure but also lots of other things do to also if you enter into a competition for a tender for instance that's also got the same payout structure if if you have an option on something you know on the stock market that's also about the same payout structure so yeah it's it's it's easy to see how betting kind of falls into this pattern but a lot of other

things fall into this pattern as well and if you look at typical betting markets they are they are fairly constrained in what you can actually bet on typically its sports so next next go to school time sort of thing that's very much what we as a company are not interested in so we're interested in making information that was previously not tradable tradable and thereby eliminating information is symmetries and enabling people to take better decisions based on better information do you know any do you know any projects more in like the web to space or like regular startups that are kind of trying

to do something similar without blockchain absolutely so they're the one that most comes to mind that many listeners may actually know it's in trade so in trade was this political prediction market that existed ten years ago or so and you could you could place predictions on you know US American primaries and presidential elections and geopolitical events around the world and there was a decent amount of activity and also the focus turned out to be year eerily accurate and it was actually one of those I mean if you look at the total

user base it wasn't humongous but it got a lot of coverage because journalists like that kind of thing they like reliable crowd source information yeah in this case like adding the incentive layer actually helped probably drive a participation yeah if it was just sort of like what do you think answer this poll for nothing he's not gonna get the same kind of dedicated participation yeah and so basically that also basically that's kind of like you're serving or apollyon so where we where you basically just ask people what do you think is going to happen but typically if people actually have to back up their beliefs with money they are more accurate and also if people have better insights they're typically willing to place more money on this because if if they think the crowd is

wrong or people are wrong about what's going to happen then to them it's basically free money so basically that's it it's all based on the premise that markets are inherently efficient so basically if you don't believe in that then prediction markets are really not for you but if you if you if you would agree to that markets typically work then this is a super interesting way of eliciting information from the crowds and the way and and why it hasn't actually been that much of a thing in the past decade or so was basically in trade fell apart due to a number of unfortunate events so their CEO died on

Everest and they they had a prediction market on the price of gold where basically the CFTC had told them to please take that down and they had and then basically the knowledge that apparently they had had this conversation got lost and they relisted it a couple of years later and got shut down and basically there wasn't enough seminar to start it up again so in principle there was interest because the the payout structure so closely resembles betting there is friction markets move in a tricky regulatory regime basically explaining to

regulators why this is not betting or why it's not gambling or why maybe you can call it betting but then the stock market should also be called betting this is difficult and we have taken tremendous care to actually interact with regulators and explain them our reasoning and yes so we see a lot of resources on our and have gone into that and this also led to us getting an in-principle license to offer friction markets to retail customers at least in Europe so that will be going live soon and if you would like to sign up for that even now before it happens go to site dot p.m. so site as in eyesight dot

p.m. and sign up and we will invite you for a beta that's starting soon so add that to the show notes so that's so this is I I have to ask another question on the prediction markets before we move on though the other big project in the space that is very well known for being a production market marketplace is like what's the difference between auger enosis M so it comes very much from the same place so it also comes from this idea of crowds crowd sourcing information it is a blockchain based British market the way that fog of works is different in a number of ways first and foremost okay actually makes their money as a decentralized oracle so basically the rep talk and the reputation token that is associated with

augur you you need that in order to determine which of the possible outcomes took place and which didn't take place so basically it is a decentralized oracle mechanism which is a super hard thing to build because obviously you need an escalation procedure and they have that and it is it's it's a really good procedure how do you like do you can you work with them yeah so basically we can kind of we can piggyback off that so basically we decided to go the other way so basically the Ark problem is the one problem that we're not solving natively with in gnosis despite the fact that we're building a million project basically this is a Oracles is the one thing that we've always said we are the Oracle agnostic and you can plug into gnosis whichever Oracle you want

also because we anticipated that our providers will spring up and grow out of legacy businesses so basically if you look at how news corporations are going to develop what the business model it's open to them in the future you know which ones will actually be viable then providing good information to smart contracts is actually really good one and so say the writers and bbc's and Bloomberg's of the word we kind of we anticipate that they will become Oracle providers in their own right and and that for most that for most sort of Oracle decisions you don't actually need

a fully distributed or occur because you're okay to trust some trusted instance say the BBC or Reuters so you don't actually need your own you don't need a decentralized oracle for everything and you want an escalation procedure and that's that's that's that's a good thing to have but you don't need to have it fully decentralized it's a lot of overhead on everything so basically we have we have concentrated on building good mac market mechanisms for the prediction markets that you can build on top of our platform that's one of the things that aqua is struggling with so actually having decent liquidity on the market so if you actually check the markets then

there's a couple that actually have a decent trading volume but the spread so basically if you buy something and then sell it again immediately you actually lose quite a lot of money so basically often on the on the order of like ten or twenty percent obviously that doesn't happen if you have enough liquidity in the market so basically what we've done is we have built these liquidity mechanisms that work even for markets that natively have low liquidity like many sorts of information markets I want to just go back for a second to the question of like to the topic of Oracle's in this context because in in previous conversations and episodes we've talked

about Oracle's but always in the context of price discovery on like for exchanges yeah like that how does an Oracle like I'm trying to sort of wrap my head around an Oracle for information yeah what does that even like you just mentioned the BBC might become an Oracle yes so basically you have information that you can put out there and you just need a cryptographic signature to go with it so that you know that it's actually that you can actually use it so basically and then you you point at that new source and yeah basically that it's it's kind of so nothing none of these are production ready yet so there's one system called reality IO it's a madman at cos project it kind of

it's got this it's a decentralized oracle in the sense that you can have the Agua escalation process but in principle supplying real world information into the blockchain is going to be a major business and a major challenge and a major challenge so basically because it you rely on your reputation a lot because if you think about how how much is it worth potentially to falsify one of these claims then the deterrent obviously you know your reputation loss has to be bigger than whatever you stand to gain so this is why we think that already existing companies that have a lot of reputation to lose um are

predestined to grow into the space interesting so from what I got gnosis and auger are sort of in the same space could we work together but are focusing in a different direction yeah so basically no it's good even so basically because Agua actually has their business model centered around New Yorker we could in principle even piggyback off of them so basically you could actually specify org or as an Oracle if you really wanted to so the way that that would work is you would set up the same friction market in Agua and then just say I use whatever whatever outcome our NGO decides has actually happened hmm cool okay so the platform the gnosis platform what's the other part like so I now understand the

Oracle part quite well what's the other part okay so the other part is what we what we call the conditional token framework so basically it is in the most general sense it is the transfer of value the tokenized transfer value under a certain condition and this this you can use this for prediction market markets you can actually also use it for a number of other things so basically for completely diverse use cases so you could use it for instance for access rights you could use it in games you could use visa like

insurance payouts in principle you can use it for insurance payouts you can use it for all kinds of financial products you can use it even as conditional payment say for instance I hire you for a job or you put in a tender for a job that I want to hire someone for and I could give you a conditional token as payment that only becomes worth something if you actually get the tender so that would actually enable you to already give these tokens that you already received to your subcontractors and your subcontractor contractors already know that they'll be paid if

only you win this tender so basically the the use cases are endless and yes bacing it's it's a kundus touken it's a conditional token Standage are you is it a plat like is it a product is it sort of an interface that you've built or what is the actual thing is just the mechanisms fine it is the mechanism behind it so basically it's my contract based and it's very much against common wisdom of how you should build smart contracts it is a what we call a monolith so it is one giant big smart contract to which anyone can append events and you can then have tokens that payout under whatever

combination you want out of all the events that have already listed in that in that monolith contract and so this is in principle I mean this is infrastructure anyone can use it we put it out there and then we as a company have built an application on top of it to showcase it that's site that we talked about earlier so it's the information discovery use case it'll start with very simple with very very blockchain centered questions like what's the maker stability if you going to be are there going to be more than a hundred thousand one hundred million died out there when we might equilateral die go live and then basically you can

you can trade all of these singly so if you have strong beliefs for any one of these but you can also create them you can also trade them in a correlated manner so you can say I want to place a prediction on the on the stability fee under the condition that might equilateral die go live by a certain date so basically you can kind of extract all kinds of information around the way that things happen or the the causalities or even you know just correlations so it'll we think it's a tremendous tool to actually make information tradable and actually for

people to incentivize to be incentivized to trade on these markets in order to reveal what they know and what they believe and in order for the people to make better decisions off of that Wow so this but what you're talking about there is almost like the like not a single prediction market but rather like these linking of other it's like you are building your own contract somehow your own like unique cases like you'll say this sort of and if statements like you've basically said I want to predict that this will happen if this and this and this and this also happens exactly and if none of those are done if none of these happen it is refunded you're funded yet exactly so baby you can you can make very specific predictions and yes in a way it maps out the possibility of space of the future

which basically to any single thing that could happen under whatever circumstance it can in principle assign a probability and obviously for now we expect our first users to be humans but in an age of big data and artificial intelligence I fully expect that in five years time almost all of our users are going to be a is that trial data and basically people are engineers who have good algorithms to extract this kind of data and they're basically being paid by society or by the people who actually won this information to provide this information leading to a wealth of information out there and and enabling

people to make better decisions would you say that this is still the core business yes this very much is the core business this is what you wanted to achieve him and basically everything else that we're doing is is kind of informed by wanting that to work so we actually built the very first version of the smart contract in order to hold our own ico funds and that sit around so basically that's the mighty sick that held the largest amount of value in the space so it was you know in the couple of billions just because you know more or less many many ICO projects used it and private people also use it in that kind of segue into this project the safe

and which is a personal smart wallet that is mobile first and you can add multiple keys in order to increase security I realize we haven't actually talked about your team we should probably talk a little bit about your co-founders and how the company is a little bit distributed and what those other projects are so you are one of the cofounders of gnosis um no technically I'm not a co-founder okay you're not no I joined very early and I was there from the get-go but in the beginning I had I had a real job so I joined Martin Jeff and after they had already founded Martin and chiffon they are two fantastic guys they met at university fifteen years ago or so and they found that they can work

together marvelously so they started working together in 2015 2014 2015 when Martin met Joe Lubin Joe obviously is the founder of consensus and one of the cofounders of etherium back in the day when consensus was literally him and his assistant Ashley and Martin and chiffon kind of became the first consensus employees and started working on this basically they had had this prediction market project on Bitcoin previously and Joe convincin that this would be way better suited to have this on'y theorem despite the fact that if your impact then wasn't actually what wasn't live so there was the white paper there but that was about it and they started putting

this on e theorem yeah so basically they started as a spoke within consensus and we're spun out into our own company now in early 2017 we conducted a token sale afterwards had you already joined by this point yeah I had joined then yeah and then we conducted token sale and that was two-and-a-half years ago and we've been building stuff ever since okay and you were like an earlier sale early 2017 you said yeah so we were in early 2017 so yeah we were one of the earlier ones yeah we lucked out in that respect yeah so obviously there's there's always a big element of luck right okay so you erase that those funds

started to build the prediction market yeah we had actually something life by then so basically we were the one of the very first projects that were actually live so we actually had a friction market platform life by the entire time that we conducted the token sale and then basically getting regulators to approve was took a long time so so we started building the infrastructure around it that we anticipated we'd need okay I got it and so what are those other pieces of infrastructure so and like who's maintaining what in the company that's a good question so basically mocking Jeff and I we have kind of this rough division that I am responsible for prediction markets Stefan is responsible

for wallets and Martin is responsible for the exchanges we talk a lot so basically everything is kind of consensus based between us but those are like the rough responsibilities then we have a fantastic team so we we are just over 50 people now we are headquartered in Gibraltar we have an office there all our finance and compliance functions are there we have a second office in the north of Spain and then we have a deaf hub in Berlin he had four node and then we have another ten fifteen people who are fully remote hmm so that's the team that's a team and so you just mentioned these two other projects we kind of hinted at it

but it was like the wallet and the exchanges yeah um I think you've already kind of explained what the wallet is yeah it's a multi sick wallet and this actually what it was safe exactly and it's called so easy it's it's the second version of this mighty sync which is a personal smart wallet quite the safe and there's a team version out there as well which is basically the exact continuation of the gnosis multisig only now basically back in the day when Stefan programmed it more or less in the better part of a week and that's what it looked like as well as the basics Jeff and the fantastic program and he's not a designer and it was never hacked so we have to I mean it was it was a good

product it was just very much not the thing that we anticipated to take off like it did yeah so basically back then obviously gas costs when such an issue and basically if you wear to deploy a new nose it's mighty sick these days it'll cost you like 20 to 50 dollars depending on the depending on the exact gas price on that day and the new version is much much more efficient I see and it's also been formally verified so as far as I know it's currently the only smart wallet out there that's formally verified that's interesting so that that's sort of this has come up a few times with a number of guests this idea that like you build something with no idea how much value

it's actually gonna hold it's um yeah I mean how do you how do you deal with that actually I guess in this like did you immediately realize that you needed to kind of build a a better version were you keeping like a very close eye on this contract once the value accumulated it started to grow yeah I mean it all happened so quickly right I mean basically if you if you think back to 2017 I mean it was mostly that ether just appreciated so much in such a short timeframe and all of a sudden this thing was holding billions of dollars and yeah I had this recurring nightmare where basically someone would call me in the middle of the night and say like yes there's something wrong with the mic you think we've had a hack and I found knock on wood it never happened

but yeah of course you think these thoughts I mean that is that is very much like the nightmare scenario and we know people who've gone through something like that how did you like watching that sort of from did like did you see the hacks happen to other projects and other multi six for example and immediately look back at your project like did that force is that why you did formal verification like was there like a lot of security fear yeah of course there is I mean for one our own funds when that wallet but also if you think about everyone else who stores their funds in that wallet yes so basically when when the parity

happened it hit very very close to home and we've we've been we've left out so far but yeah with the sort of value stored in these things security really you have to put that first and foremost and we have very strange and practices as to when code can be changed when audits happened basically we went through the formal verification despite the fact that we are actually not making any money off of the safe we did that kind of as in it's kind of an infrastructure project and we still spend a lot of money on paying for formal verification just because we

didn't want that to happen to us did that did you do that after the parity hack happened yeah so basically we did that on the new version so basically the new and gas efficiency improved one so basically the old one I think it's fair to assume that it is safe just because the you know if you think in terms of if you think in terms of potential payouts it was just a huge honeypot right so basically had there been the vulnerabilities probably would have been exploited by now so yeah we did it for the new improved version and that's come back yeah and you can actually we we make it a habit to to publicize all our security audit so you can actually go

look at that on the Internet okay so now we're kind of getting into the third part of the business so of the three main categories we listed earlier on the last one being exchanges but what does that actually entail like what what is an exchange or what are exchanges in the nosis context um that's a really good question um so exchanges we are no exchanges so decentralized exchange is centralized exchanges and so on and basically to us it seemed that building an exchange on a blockchain actually comes with very specific constraints namely that time on block chains is not continuous and that is very much not

like if you go to a stock market then time is continuous and that's why I am why traders like to be closer to the stock market in order to have lower latencies in terms of their orders but block time is inherently blocky so it's discrete so we thought about we thought about auction mechanisms or we thought about exchange my and that don't see this as a flaw but as a feature and that's basically that that's basically the approach that we have taken to building exchanges so do you build did you build a decks did you build because I also heard about the DX Dow which you've also done so where does

this all fall yeah so yeah so basically we've we've experimented a bit so basically the very first decks that we built was the Dutch Dutch exchange and which was based on the Dutch auction contract that we actually used for our own token sale the idea was that for low liquidity tokens actually having an order book it's very inefficient and if you go to Texas and I mean there's a couple of liquid pairs but if you look at all the other pairs they're illiquid and you have a large spread and you have large slippery so basically the price changes fast just because there's not a lick not enough liquidity and one way of combating this is just batching things together for a while so saying like we

are batching a couple of hours together and then basically everything will be cleared at the same price so what we actually had is we had this mechanism where an option started at an arbitrary price so actually twice the past clearing price so say the last clearing price for ether was 200 I so busy you'd start at 400 I and then basically everyone who wanted to sell ether Ford I could put the ether into a contract then the price for the ether in this contract would drop and people would send died at that point in time where they were they are happy with the price and they would receive that price or less and basically when the total volume of either to be so

it and died to buy this ether would kind of match up at the current price and the auction would stop and the entire thing would be cleared at that price so basically the time at which you send your payment token this indicated your willingness to pay and from a game theoretic perspective this is a really good auction mechanism because it makes sure that well first of all everyone pays the same price and everyone pays what is the fair market price the market has actually defined somehow exactly regardless of what they were actually willing to pay how much it was worth to them so so it can be argued this is the fairest way of

actually conducting a sale for fungible goods and this protects that front-running case exactly so basically you can't front run on these exchanges so basically if you look at decentralized order book exchanges front running is a huge risk and basically they have they have been analyses of the of this so basically maybe we can put the link in the show notes but tilde and he's done this entire page on how much front running costs your indexes and it's like one to two percent so that's how much you lose on all trades and that combats that it also combats the spread so basically if peak because everyone gets the same price and in principle so

basically it's slow so that's the one drawback so basically it's not an instantaneous thing so you send your your tokens and you get your return tokens a couple of hours later this and we try to make this into the tagline so slow slow trades fair trades take time and the the you can actually still find it on the Internet's it's it's at slow door trade and it didn't really take off it's like an amazing experiment it's an it's an amazing experiment and people have used it for auctioning so basically we've we've done experiments ourselves where we said okay we have we have extra die we will action this off for ether

and we were auctioned off a hundred thousand died at once which is something that typically decentralized exchanges they cannot take this kind of you'll get enormous slippage if you if you do if you put say 100 thousand die on idec's you will and you'll lose a large portion of your of your money and we did this 10 times in a row so actually we we auctioned off one million die and in the beginning we had like a little bit of slippage speak up before like market makers well we're looking onto the so we're basically the on the very first option we lost 8% but the subsequent options they closed within 1% of the actual dye price and basically having a

decentralized auction mechanism and where you can actually put a $100,000 worth of anything on there and get more or less fair market value this was an oven but obviously the user experience was a lot different from what people who I used to and yeah with exchanges because people typically want it also gives you a peace of mind right so basically you don't need to think about that you know they're still like something a process open that you need to check on at some point in time and also it doesn't really guarantee so basically despite the fact that we did these experiments and we kind of showed that it kind of closes more or less within market within market price there's no guarantee for that so

basically in principle it can drop to zero and I mean obviously that's a market failure and it shouldn't happen but there's no guarantee against Ed because in in effect you actually you're placing a market order and I think people are inherently adverse to that it was a super interesting experiment and I think I'm happy that we undertook it I still believe in the market mechanism of touch options for many things I think in principle was a super valid experiment and we actually ended up gifting this protocol em to a dhow that we helped summon together with tower stack it's called the DX now and basically it can make decisions on whatever it's a Dao it's it's own thing but and we have no part of it anymore so we just help

jumpstart it but we gifted the Dao this protocol so basically I mean obviously there's parameters that you can set so basically how much funding doesn't option need to have in order to actually set off a new option how much of the so this is there's the in the the in the on the Dutch exchange you didn't extract value but there was a liquidity fee that went into the next part and basically how much that is that can be changed and what kind of tokens can be whitelisted to the reward talkin and I mean basically there's all kinds of parameters that can be set in principle and we gave that to the DX tau so the DX tau in itself that

that it was also a project that I heard about and I think I've mentioned on a previous episode with Aragon okay so what I heard with DX tau a friend of mine ping means like oh my god there's this tau and it has incredible distribution and participation everybody's really excited about it I mean at this be the way you sort of say you've summoned it and then you sort of like let let it be what has happened with the DX tau and the DX I was running the Dutch exchange so it's still very much out there and it there's proposals in there that you can look at now and there's around I think three to four hundred distinct individuals who hold reputation within this tau and it's

based on the Taoist AK protocol so in principle is about this scalable consensus mechanism that they call a holographic consensus and it is a really nice piece of software did it become but like the story that I heard though is a few a few members start to accumulate more of the there are a couple of people who hold a large amount of the reputation so basically if you say if you got the right 15 people maybe the right 20 people into a room you would get more than 50% but it's it's people we don't know and we are not affiliated with who did this of their own accord and there's 300 more people who are who are also members of the Dow and to us

hold reputation and vote in votes okay I have a last sort of question on the tech I also so basically we're we're building a new exchange that's why I said exchanges so do you want me to go into that now for sure yeah why not okay and so basically the Dutch exchange was in our in my eyes this is like the first version of an exchange that we build that kind of uses the advantages that blockchain actually has over continuous systems we have we're taking this to a new level with a product product whose working name currently is

diffusion exchange in in in so the diffusion exchange is also a batch option based exchange and it's not based on the Dutch option and it's not based on pairwise options so you wouldn't option off ether against dye and dye against ether but you actually throw all all possible orders into one pot and then you find the optimal solution across all of these different orders so this this has a number of nice side effects say you want to say SL token a for B I want to say B for C and the third person wants to sell C for a on a traditional order book exchange we would never be matched so we would rely on

arbitrageurs to actually do a barrage between these different order books but in a one part exchange there would be a ring trade or a higher order ring trade and that can actually still solve this natively within the exchange and because we're assuming that there will be we'll live in a in a mighty token word with a lot of tokens and I mean we can even see that now with the way that stable tokens are going so there's not just one sable talk when they're 10 and probably in the at the end of the year there would be 20 that have like some amount of traction and you can actually trade any token against any of these stable tokens and in principle would totally make sense to prove the liquidity on all of these

order books but basically on an order based model you can't do that so basically in in like the one part exchange and you can find the optimal solution across all of these and if you look at how that's better than a conventional order book exchange it's marginally better if you have one central token against which everything is traded so currently a little bit like ether but if you trade a large number of low liquidity tokens against each other order books become incredibly inefficient and having this one pot actually helps a lot and we're building a proof of concept that is out in the coming months with a working number name of stable X and which does exactly that first

tockman's because that's a lot easier to build because you know every everything is close to one and people can provide liquidity at more or less no risk to themselves it's funny because just going back to earlier in our conversation where you were like no we're getting more focused you should have seen as a year ago but basically it's like we went why I don't know all the different ideas that we had but basically it really fits into the entire day because if you look at if you look at conditional tokens and outcome tokens these are ERC 20 tokens and you can yeah and basically if you look at information tokens that conditional tokens that have a register

of different conditions by the by definition they will be low liquidity and you will need a good trading mechanism in order to trade them efficiently in order to actually it to enable this exchange of information that we envision envisage and this is why we need the diffusion exchange that sounds like that's fair but I have one other question about another project or a side project or I don't know if it's still around but I remember there was a team working on zero knowledge yes and they kind of I came into contact with them because they were organizing a trusted set up I think maybe thinking about doing a universal trusted set up so what happened with that team and where did that live and that lives in the diffusion space so obviously having this

one pot exchange is computationally complex and doing this on chain is unfeasible so the idea is that people submit anyone can submit a solution to a to a list of tokens that are in a closed batch and say this the solution how to redistribute them in a more or less optimal way away and and the ideas that people submit this alongside is a zero knowledge proof that this is actually a valid solution obviously the privacy aspect is also great but mostly in terms of the scaling that it gives you namely that you can prove that this is a good

solution without actually having to go through the entire solution on chain and we we are still interested in that the team is still following developments but basically coming up with three different things at once namely how to build this novel market mechanism you know even you know outside of the blockchain space and then how to put it on blockchain and then the third how to get it to scale efficiently we found that there's so much happening in the zero knowledge area that that's kind of it's kind of taking care of itself and basically we feel like we we need to figure out the how to build this marketplace on blockchain first and then we are worried

about the scaling in a second step but yes we are basically integrating their knowledge into this diffusion exchange it's very much a thing that we were doing the future so it sounds and I can follow that too I mean all right I can I can attest to that as well which is that there is a lot of development in ceará knowledge and things that were limitations like six months ago may no longer be limitations with new protocol exactly raids are ideas and these things are kind of they're happening they're underway that that dedicated teams working on them and basically we fear that if we check back in four months when we figured out the other stuff we can just piggyback off of what everyone else did probably and I mean by then hopefully there'll be a universal

trusted set up for a theorem I mean that would be wonderful I think it's been really amazing to hear all of the things that gnosis is working on you know I've worked around you guys for quite a while and I feel like I didn't know really what you guys are working on we're not good at marketing so marketing is not one of the skills we bring to the table we're bitter this sounds like it's I mean it really is you know there's so many ideas and so many tests being done and I'm very I'm definitely excited to see where you guys go next but let's talk sort of to wrap up about Berlin and this Brill in dev office and where we are right now where we're recording out of because we're in full node and full node is a project that you lead on the

side of all of this what is full node where did it come from four node is a wonderful space that is kind of become the heart of the Berlin Blockade community so it very much started with us in that we were working at the East F office in vitamins before it moved that was wonderful because everyone who worked on aetherium was there at the time and as nice as that was it was a horrible office I I hope no one I hope no one blames me for saying this but basically it was it was more less an apartment they were no more there were no ways to actually have meetings there efficiently everyone took important phone calls on the fire

because that was the one place where you could actually talk but basically having this community was such an asset and we we had raised the money we knew that we were going to hire people so it was limited in space and we also knew that we want to pop office so we started looking at office spaces and it just was it was so sad you know just the three of us looking for office spaces when we had you know we and when we had worked in the space with this really vibrant cool community and at the same time Brian who is my co-host at epicenter I mean Brian Brian crane yeah he was briefly the CEO Oh at cosmos and he was very much in the same situation that he was looking for an office space for cosmos and he had

the same experiences namely that you know just leaving leaving the East F office and its community spirit just seemed daunting so basically we teamed up Brian and I and we started looking for spaces and we found this and we had just signed the contract when Brian stop being CEO at cosmos and did all fell on my shoulders clearly still friends yeah I don't hold it against him I die these things happen yeah I hadn't anticipated how much work it would be so basing this place there was nothing in here it looked a little

bit bombed out so we we set up an office and basically now a hundred and five people work from here it's there's regular meetups here it's kind of where people come to if they come to Berlin and visiting blockchain people and that's wonderful it's also got a really good name the full node.js for anyone who hasn't been to Berlin it's definitely a spot to check out and I agree with you I think it really like any time there's a book like the blockchain week or events happening in town you'll come in here and there'll be all these teams from all over the world in the co-working space and sort of the open space and there's these meetups I think you know the time that I've spent here has been really

really great so thank you for putting that together for for all of us and you know helping to make this Berlin scene kind of really flourish in a way it's cool that it's not necessarily only in the it's like the eath dev office also would have been a theory of only but actually this office I feel has also branched out like it's it is more kind of it's a wider ecosystem oh yeah well sure so basically and I mean the number of people who've worked in here and moved out again because they grew those are also in West they're friends with them nestled in the network so basically C or X moved out and refuge moved out but I mean also people move in so recently matlab's moved in and I mean basically there's there's a lot of the

best movement in the space and that's wonderful very cool so listen thank you so much for coming on the podcast oh it's been my pleasure and I've said this to SATs uneasy on the other epicentre guest who I've had on here but I'm always open to another crossover you guys happy Center I don't know what I talked about but like put it out there will be super alright so thanks again and to our listeners thanks for listening you [Music]