EB139 – Martin Köppelmann: Gnosis - The Ethereum Prediction Market

Transcript

[Music] this episode of epicenter Bitcoin is brought you by Ledger now accepting pre-orders for The allnew Ledger blue Developer Edition a Bluetooth and NFC touchscreen Hardware signing device learn more about The Ledger blue at Ledger wallet.com and use the discount code epicenter to get 10% off your first s and by Jax Jax is a user-friendly wallet that works across all your devices and handles both Bitcoin and ether go to jx. and embrace the future of cryptocurrency

wallets welcome to epison bitcoin the show which talks about the Technologies projects and startups driving decentralization and the global blockchain Revolution my name is Brian faan cran and I am M Roy today we have a very special guest uh Martin copel man uh who is the founder of fair and gnosis which are prediction markets on bitcoin and ethereum respectively Martin is also very involved in the current episode uh the debacle that is the da and so we'll get some updates from him as well so before we start off on these interesting topics let us have an introduction from Martin Martin your intro please yeah hello um yes thanks thanks for for

having me um yeah as you said I'm for for a while in into the space so my big two topics are or big topics are prediction markets um I'm also involved with basic income project and but um yeah prediction markets are kind of biggest topic and just recently uh I mean there's no way to overlook the Dow currently so I would not say I'm heavily involved but I mean I'm I'm just observing it writing about it and maybe trying to help uh solve solve the problem or solve the mess so you made some long threads on on some of the on Reddit about sort of everything that's

happened with the Dow now can you give us a bit of an update from you know we know roughly about the original Heist I think people are vaguely aware of that maybe you can just briefly sort of Rec recapitulate what happened there and what's been going on since then like where are we today yeah I think um my my main motivation to to do those posts are um that I think it's just extremely important for everyone to know exactly what's going on and especially if if we as a community want to do um decisions like um hard Fork um then it is in my opinion extremely important important that

everyone is is educated and and just knows what exactly is going on what uh what are the reasons um it's it's also um I mean it it would also be extremely unfair to those who currently hold Dao tokens and have to make the decision whether or not to sell them or or maybe to buy some um so I I really think transparency is is important that being said so so concrete answers um yeah what what happened um yeah a heck there there was a bucket exploited it's a so-called re-entry um thing that you could um do

basically multiple transactions or multiple uh withdraws or multiple splits within uh one um transaction so may maybe I I I try to to um explain very simple let's imagine um the Dow is an ATM and let's imagine there are two steps so the first step is um uh get out get out 100 euros and the next step would be uh reduce the balance uh by 100 euros so so that that would be the logic of of the ATM uh first pay out then reduce um the balance of of the user um and that is and the problem is

that those are not uh Atomic steps so with atomic I mean it's not done in one step it's it's done in two steps so first it's paid out and then the balance is reduced and that's exactly what what the hacker exploited so he um he set up or he made a setup that it was paid out multiple times um and then the balance was only reduced or or set to zero basically and um so so back to the ATM say your balance is um1 and there's the option uh cash out all um and and the step is cash out the amount you have1 EUR and then set it to

zero then um he called this cash out function like 20 times um and then after 20 times 100 EUR were cashed out then it was set to zero uh 20 times um that was kind of the heck he even combined it with something that in the end before it's set to zeros he could even transfer those 100 EUR to to another account so then um this the account where yeah that was already zero that was then sent to zero and then he could do the same thing uh from from the second account so that that's that's kind of what happened I mean it it seems to me that like whoever

the whoever the hacker is um is like an extreme extremely smart person uh because like when the when the attack came out I I clearly remember that there being a phase where I was kind of looking at at all of the posts that were coming in about the attack and I was like these none of these posts completely explain the whole attack because uh they were they were as you said like there's this one attack in which uh you do the accounting only after after the withdrawals this was well known before the Dow hack and everybody was expecting this to be used somehow but then there was something something strange about this attack as in he was able to execute this attack so

many more times than he should have been able to and you were the one of the first people to come out with a correct explanation of how he did it and it seems like he exploited like two different things and combined them to make them very potent and these two things were and the second thing was also not very well known so this seems like a very very accomp guy it seems maybe few comments on that so so first the the general attack was described like like a few weeks before the hack so so um the hacker just very carefully looked at at the contract and found found the option to exploit this the specific attack that no one else uh found found before and indeed um so he

came up with this um or they or she came up with with a thing that um that that they used this re-entry not only to do multiple withdraws but then in the last step to to also uh um yeah transfer money to to to so so they they had two teching contracts and they send it so in one transaction they use one 20 times and then send it to the other one to to to use it from the other one so yeah that that was a smart uh thing uh but um you you claim that that they were very smart um yes and no so so they also did a big big mistake in my opinion um and they were basically too

greedy so my my opinion is um let's so I mean we will maybe come to this later but I think it's quite likely that we will see a hard for now um but let let's imagine the hacker would would have only um stolen like let's say 100,000 ether maybe maybe 500,000 ether or so so let's say significantly less than they did um I would I would assume um that in this case uh the um there wouldn't have been um uh a majority for for the heart

Fox so so I I think if if the goal of the hacker would be to maximize the their um monetary profit uh it would be smarter to uh to have stolen less yeah that that might well be the case so so talking about hard Fork uh there was a soft Fork first right what happened there I know there was a lot of criticism that somehow the softw wasn't proper and there was some bugs in that okay so what happened there and how did the sort of consensus around the hard Fork now emerge yeah so so the the soft Fork actually I mean in the end it didn't happen so so it was proposed was um implemented and the idea was to

basically um basically block specific kind of transactions or sensor specific kind of transactions those who would reduce um eer from uh from a dow contract uh and it turns out that it's uh quite complicated to or maybe even impossible to uh to censor specific type of transaction transactions with ethereum um because you cannot uh if if you see a transaction you cannot immediately see from the outside from from kind of the input data or from from we call it the static analysis you cannot see from the outside what this transaction will eventually do without

executing it so that means you have to execute it you have to do maybe um expensive computational steps uh to eventually find out well this transaction is doing something it's not allowed to do and then you are you you the transaction uh wouldn't even have to pay for its gas because it should not be included so that means it's not it's maybe not possible or it so that that opens the um that opens the problem that an attacker can just create thousands or millions of those transactions and uh spam the network at no costs and so that that uh that was the reason why um the

softw was not eventually not not activated or eventually not executed is it the case that there's there's a silver lining in in uh in this soft Fork episode so what you're essentially saying is in ethereum even in the future it will be very hard for for the developers or anybody or the mining Community or anybody to to create a software that allows the allows the some of the transactions related to a particular address on ethereum to be censored because if they were to implement this kind of feature anytime in the future for any other contract apart from the da the same denial of

service vulnerability would exist right this in a sense means it's very hard to censor addresses right if if if you want to keep ethereum working which is actually quite different from uh what might happen with Bitcoin in Bitcoin it is in theory possible but the community is hard against it but in ethereum it just might not be possible at all at least not with a with a soft Fork so if you would do hard Fork where you would say um um those transactions they would just be invalid but still would have to pay the gas um and and and again that that is a hard for because it would change the rues set um then those things might

be possible but yeah with soft par or and I the the important thing to know about a softb is um miners in principle miners could do it uh without asking anyone so so it it doesn't need to be the developers who develop a new uh version in principle miners can always do this kind of censorship of transactions clusion um and I mean to some degree it happens in Bitcoin so so some miners would not include transactions that that uh used what was it some gambling sites or I I don't don't remember but but but this happened at Bitcoin and yes it turns out it's

harder for miners to do this in ethereum so talking about the the hard Fork now um is it correct that essentially this will mean that you know the miners come to a consensus and and the money is essentially taken away from the hacker and given back to the original do holders and then they can withdraw it uh yes but I would uh would say it's important to know that um for especially for hard Fork uh it's more about the whole community so yes miners are obviously a part of the community but also um the core developers uh also people who write applications on on top

of um uh on on top of the protocol also obviously the exchanges all those are important in the decision um to make a hard Fork so I mean just just imagine um everyone the miners and so on exchanges they they would support the hard Fork but let's mention all all people who build um stuff on uh ethereum so all dep developers they would come together and say no we don't support this hard we will stick on the other chain and we will only write applications uh for this other chain so so that that that would um that would still have a big influence so so hard Fork is is is bigger

consensus of the whole Community W uh so yeah so hard Force essentially need the consensus of the whole Community but you have also in this in this particular field you have come up with a new set of data and observations that a significant portion of people people who own ether also own Dow tokens and tell us like what data you found and H how did you come to this conclusion and what it means for uh the probability of the hard F going through yeah I mean so in principle those those data are available um on the blockchain I mean and the first very simple data point is that what was it 10 to 15% of all ether uh is in the D so I mean that

that is already hugee and then you can just look at few other numbers so you can look at how how many accounts exist on the ethereum blockchain with more than one ether um and you can have a look at how many accounts um uh are or how many yeah accounts do uh do H um Dow tokens and um and the third or so so um uh yeah a third so so the number of of accounts that hold Dow tokens is a third of the number of accounts that have more than one eer so it's it's quite likely that um that the super majority or yeah

way way more than 50% or all all Dow token holders together will are most likely holding way more than 50% of all ether at least if you if you include people who uh I mean I I guess most people just tried it out or most ether holders spent at least I don't know one ether or something like that just just to try it out let's take a short break so we can go to Paris I stopped into L Bitcoin the house of Bitcoin at The Ledger offices and I met with Ledger CEO Eric so he could tell me all about The Ledger wallet Chrome app The Ledger wallet Chrome app is the perfect companion app for your Ledger h

W1 or Nano we have a very powerful and cool feature you can use multi accounts for instance personal accounts business accounts this is very useful also when you want to make a transaction we use a second Factor verification you can either use a physical security key or cryptographically securely pair your Android or iOS smartphone to your Nano this way when you issue a transaction a payment the transaction will pop up on your Android or iOS phone and you will be able to verify the amount and destination address finally The Ledger Chrome app has an API with which you can easily integrate third party

applications for instance if you want to create a multi signature account with coin guite or cop it will be known using The Ledger wallet Chrome app Ledger is making Hardware wallets easy and convenient without compromising on security if you want to get a secure setup for storing your Bitcoins go to Ledger wallet.com and use the code epicenter to get 10% off your order we'd like to thank Ledger for their support of epicenter Bitcoin you're now working on on a prediction market right but you originally got into uh into Bitcoin quite a while ago and you started working on on a project called fairlay which is a Bitcoin prediction Market actually fairlay also was the the

very first sponsor of this podcast right long long time ago so can you tell us like how did you originally get into the space and what was it about prediction markets and Bitcoin that you saw like a strong fit and you wanted to create fairly so those were kind of two separate events so so how I got into into Bitcoin into the Bitcoin space uh that's quite a funny story to some degree so uh I think in well I'm not 100% sure it's about the year but it should be 20 11 um there was this in in Germany there was this um New Movement the pirate party they they

became quite successful in Germany so so and one of the ideas of the pirate party is this idea of liquid democracy so everyone can directly vote on on um on political decisions so I joined the pirate party just was exciting uh so and and so I was therefore as every everyone every member else ask to vote on decisions like uh those bailout uh packages on on Greece uh so and I I just tried to make made a um um made a to make a good decision so I tried to inform myself and

try to to understand what what's going on on there and that was kind of the point where I started to realize or started to think about what is money uh and I I started to realize that um money if if you use it from day to day and if you use it on a personal level is very different or the properties of of money are very different on on on this personal microeconomic level compared to if you look at it uh if you look at money from a global scale so if you suddenly talking about uh billions trillions uh uh created uh as as in in those bailout packages it's m money has

then really different properties and I guess that was the point where I also discovered Bitcoin and realized well money is in the end just a social construct and if people agree uh to to use something as money um then it it can be used as money and obviously I have a computer science background and Bitcoin is super interesting from from that perspective as well how was your experience running fairl uh tell us like so was was fair like a centralized prediction Market based on bitcoin and how big was the how big uh how many trades did you get at its peak right so uh so after I I I got into Bitcoin I don't know started to buy a

one of those butterfly Miners and stuff like that I I just um want or just wanted to build something on top of it and um and I was yeah quite interested in in um in prediction markets and I saw well uh there there is an opportunity to build a prediction Market that that purely runs uh on bitcoin so I started with um Stefan co-founder uh we we started fet and fet is basically centralized prediction Market um and and the only kind of the only

blockchain component or is that it uses the currency Bitcoin but it doesn't use um it doesn't use the blockchain to to to to run the bets or to to the logic is is not on on the blockchain logic is in a regular server uh server client model and so how did that project turn out did you see a lot of traction or a lot of interest do you feel like it using Bitcoin was a big benefit to the platform um yeah so so right now so I'm not um I'm not longer um uh actively involved in in in F it's some someone

else took it over but uh so quite now right now um the beding volume in terms of um is roughly a million a million use or bitcoin's worth a million US dollar a month um so I mean our initial goal with P was um I I was always fascinated in the idea of you could use the predict Market to aggregate information so you don't you want to do a forecast on something uh then the best tool to aggregate all the and so on in the end turned out sport betting was was made way more successful than than all the markets I found personally interesting

will Bitcoin do a will Bitcoin do a increase the block size limit will they impl Implement segregated witness before blah blah blah and we we we tried all those markets and they had not not too big success but in the end sports betting went quite well so so that that's kind of the story of f well and and then once once you cease to be involved on a day-to-day basis with fail you came up with the idea of gnosis and the big difference between gnosis and fail is of course that one is based on ethereum and the other is on bitcoin so tell us the fundamental reason U you are interested in building a prediction Market on ethereum what's what's

different from Bitcoin yeah I would say um there are um maybe two two big points so so first um with with ethereum I mean we we had this go with with fa as well so so we thought about how can we do it decentralized basically how can we uh set it up set it up in in a way that during predictions uh we do not need to hold some money because in the end that's just liability so um it's just uh um additional costs it's it's it's it took us we we spend so much time to to

make it secure or to try to to um to uh um yeah secure the Bitcoins in the end you you can only lose so so if it works well then okay that's nothing unexpected but if you do mistake then it's big big downside so so obviously uh it would be great to to have um to use a decentralized system where you are not um um yeah responsible for securing a server or securing a cold wallet um so so that that that is one one uh big point and uh um so it's to some degree

cost efficiency argument it's it's also a argument that that it's easier or might be easier to gain the trust of um of um of the people uh but but to me um that that not that's maybe not the really important point the really important point is um is this decentralized platform aspect so uh if if if I um if I look at the Bitcoin uh Sports Bing space there are I would say five six bigger uh sport betting sites that have maybe even or I think a few might might have higher

volumes than than f but but but so so who have a decent volume uh from a from economic or from efficiency perspective it would be so much more efficient to have them all on one platform so to have all the liquidity uh combined on one PL platform so I mean I mentioned the other problem the for for those markets um for those markets uh uh will Bitcoin do this and that where where we could not reach the the necessary liquidity to have serious uh uh markets so liquidity is a big issue so you want to have all the liquidity you want to have all the

markets um in one platform um that's just way more uh efficient but how do you agree on one platform so I mean obviously we want to have LA but obviously others they they want to have um their platform um so there are two two ways in my opinion so the one way would be to raise a lot of capital um and just to make this big land grab uh and and and get 80% uh on on your platform by just outspending uh the competition um but there might be a new model and and this would be a set up a decentralized

um uh platform where where you have some ownership maybe and we can discuss this later but it's it's it's still you can guarantee um good good conditions um or you you can give certain guarantees about those who decide um to join the platform May Ju Ju Just to to make this more clear so so the other example is or if you look outside the crypto space uh another platform is is Uber for example and and there there's lift obviously as a competitor and there is in principle no good reason to have um two competing platforms you don't want to have

liquidity liquidity of drivers split up in in Uber and lift as a user you you want to have all in one place you don't want to have uh eBay or Airbnb you don't want to look at at five different platforms to to find out what you need you want to have all the liquidity all the flats all the drivers all the whatever in in one platform that is way more efficient however you don't want to necessarily have um this one company that that that that controls everything because then you have the disadvantages that come with a mon Monopoly and and um too much um yeah concentrated power so that they

can unreasonable fees or they can they can exclude someone from the platform uh who who should not be excluded and and and stuff like that so here um that that is the big uh thing I I or big application um I see for for blockchain and ethereum in the next years to build those platforms where everyone can agree on because they do are not controlled by single entity wow I think I think I think that's that is a very powerful statement and uh yeah I'm even trying to think through the implications of of a statement like that they seem like very multi M dimensional but yeah it does seem to be the case that with a lot of

blockchain applications not just on ethereum but even with people trying to build U permission networks it seems to be that uh you want as many people to join one platform as possible because that is when the true utility of the blockchain shines through right as being as being one one it system on which all of us could coordinate right and uh and yeah maybe maybe the maybe the future does enable a model in which uh we can not only bring liquidity in prediction markets on one platform or or cars on one platform but all sorts of other things right u i I mean my my statement is Uber Airbnb eBay and so on they will all be replaced within the

next uh 10 to 20 years by by blockchain platforms today's magic word is prognosis that's p r o g n o s i s head over to let's talk bitcoin.com to sign in enter the magic word and claim your part of the listener reward in theories it's possible to see it's possible to kind of argue about it and I I I I'm not aware of any good counterarguments to this yet uh but it does it does also seem to be the case that the engineering side of of this technology is quite immature for or or

something like that yeah I mean I I think that that is um there the internet is a good comparison so in in I think we are now in 91 uh of of of the internet um and of course some people could argue in 91 well this is too slow this will never replace uh uh you can never watch movies over the internet and so on and well I guess you can yeah absolutely and I think it's such an exciting an exciting vision and I totally agree I think this idea that you can have a a sort of a shared platform that then different people build their businesses on is extremely exciting um going a little bit more in depth with

gnosis so we had the the centralized prediction Market using Bitcoin right and then we have gnosis which which becomes decentralized so can you talk through to us what does that architecture look like and how is that different from it a centralized prediction market right I I I guess again there are those two components there is one the component that um that we can guarantee towards the user um that specific things work like they like they are written in code and not like um uh like like we want them so I mean in a centralized predic Market you you send your Bitcoins to someone they they

store the Bitcoins um and in principle I mean in principle they could or technically it would possible that they that they just run away with it um in a decentralized prediction Market um uh during the um during the uh prediction The Ether or Bitcoin tokens I guess sooner or later there will be Bitcoin tokens on ethereum and dollar tokens and so on um the money the collateral will be um will be held in a smart contract and um and that contract defines exactly uh who will get the payout under what what conditions and

usually you have uh an oracle or you have an oracle that will um set this condition so if you do this canonical example uh who's the next president will Hillary Clinton be president yes no you create um you put in one one ether you get two shares um one yes share one no share yes share will will um become one ether if she becomes president no share in the other case and you will have an oracle that will just provide this data point uh is Hillary Clinton president yes or no and everything else will be um will be triggered uh by by the contracts and in a predictable and fair way so the way

we perhaps we could imagine it is um in a particular market like like you took the example of the predict presidential election Market you might imagine of let's say I don't know 1,000 people that are willing to bet on it and there is either one or a set of smart contracts which is like a jackpot in the middle and uh and basically like like these people are kind of depositing money into that jackpot and getting shares in exchange for it and uh once the bet resolves uh people who are right get the get the full full money inside inside the jackpot right yeah or they're they're part of how how many shares they

have of of this jackpot so so so basically like out of these 1,000 people like 5 500 people might be owning shares for Hillary Clinton right and then uh when they when they bought the share so they might have sent let's say 50 Cent to the jackpot and B and purchase the share when the result was not out or the price might also have evolved over time but then once the jackpot once the event is out that set of smart contracts takes takes input from an oracle like um Oracle I and then uh it it gives away all of the winnings to the people who correctly predicted that particular event right right or yeah all all the

people who hold shares of the winning outcome right so shares will turn into into money and do you have any any particular views on what kind of oracles you will you will use for uh for gnosis at the beginning right so uh so what we what we are doing um we will create or we already create an oracle market so um different Oracle providers will will sign up and um and offer their servers and we already have agreements with um with reality keys uh we spoke with orac there's smart contracts um they they also provide

Oracle service but also we are talking to W from alpha they are interested in providing their data um for ethereum uh and also companies like Bloomberg so I uh totally expect that in a few years um or maybe even less uh Bloomberg will provide their uh data feed uh with a signature and that's all you need so you not necessarily need um you don't need those oracles to know about this prediction market so so they they do not have to interact directly with a prediction Market they even do not have to interact with with ethereum itself

all they need to do is uh somewhere publish um the outcomes with a signature um that that everyone knows that belongs to them and as soon as they do that uh then um basically this signed this signed piece of data that says the price of the US dollar compared to Euro or whatever was this and this at at that point in time and there's a signature of of of Bloomberg Beyond it then you can take this piece of data and put it into the ethereum blockchain anyone can do it and the contract can evaluate it step by step and say yes this is the right signature um and those are the data

points and then it can um execute whatever uh whatever is is built around uh around it I totally agree I thought about this quite a lot and mentioned to various people and had these discussion I think Bloomberg is in such an interesting position there uh to provide exactly this kind of service now I'm curious here how would a company monetize that because if you have that kind of dissociation from like publishing the data signing it and and the actual markets like there doesn't seem to be a clear way to directly monetize it yeah um that's a good question um and I don't have um I don't have the

perfect answer yet I mean the obvious one is is to to charge per per um per data point um but but it would not allow um this model would not allow to charge a fraction of of of the market so it would only um it would not POS be possible to distinguish between um between a data point that settles something worth $10 or a data point that settles some something worth hundred million so that might be a challenge for for a business model personally I think uh I suspect uh

I suspect the the uh the fundamental feature that can drive uh monetization in this market is uh reliability so uh so let's say let's say I I I build a prediction market and that prediction Market needs the input of some event in the future now uh I ideally want there to be a strong guarantee that the data is going to come from a certain Source right and uh if if I just choose to choose to take uh data from Bloomberg without paying them anything then they have no uh need to guarantee reliability to

me and it's it's this need for my reliability that there's certainty that I get the data from Bloomberg and it will happen that might drive me as the market Creator to pay a bit of uh of of of whatever profit I make to Bloomberg now so that's one thing and the second thing is um even in this Oracle Market I have actually talked to a few employees at Bloomberg whether they they'd be interested to do something like this and it turns out that Bloomberg is uh is a company that is completely focused on selling these Bloomberg terminals so these are for people who are not aware for for Traders uh you can buy a terminal which is a piece of hardware

and then on this piece of Hardware you can receive all sorts of financial information and uh Bloomberg doesn't offer currently any any choice if you want to just receive a subset of information so for example you buy a Bloomberg terminal you put like $2,000 per month into it and then you receive all of the information over all of the markets but if I just want to receive say data about London print crude market then there's no way today I can get just Bloomberg data will learn print code I would have to spend the Bloomberg terminal and they have kind of stuck to this business model for over over 10 years even though many different markets have demanded different solutions and so

it it seems interesting that in this market wolf from alpha actually seems a a much more interesting player which which is more suited to to this to this kind of application where we need data about very specific things only at very specific times this seems to be something that uh the they are already doing rather than Bloomberg which is in a completely different field it's good to have competition and there will be competition and so um we'll see do you think do you think reliability drives monetization here um yes um so reliability is of course extremely important however um we also have thought about things um to to um to kind

of make a system work even if the uh even if the Oracle is is not 100% reliable so the um the default setup we want to use in noses um is that you use centralized Oracle like uh blomberg or like um uh W from alpha or in the first days reality Keys um but there will be a short uh kind challenge period so so the the Oracle will will publish this data uh data point um and then there will be a short challenge period And if everyone agrees with this outcome uh it is fine

um and it will just be executed money will be paid out and then it's kind of irreversible um but if during this challenge period someone puts up um some uh security deposit they can trigger a decentralized Oracle uh um and they could if if this challenge is successful um they could um revert um the outcome so even even if even if the if the Oracle is not 100% safe or trusted um then you uh could still use it in the model however you still have a point if the if the um if the uh um Oracle is

more trusted then you could have this challenge period shorter and that would uh free free Capital earlier so you have little bit less Capital costs so of course reliable Oracle um and a yeah basically a reputation as a Oracle will be something you can monetize let's take a short break to talk about Jax jackx is a cryptocurrency wallet created by the people at the central now there are two cryp currencies that matter at the moment one is Bitcoin and one is ether but using them can be tricky what wallet you use how do you secure them where did I leave my umbrella it's all a big mess and that's where Jax comes in Jax is a

unified wallet it works across all your devices it works for the Android phone Apple iPhone it works for your desktop computer and they have browser extensions for Chrome and Firefox and it works for both currencies at the same time it works for Bitcoin and it works for ether one of the things that makes Jack as delightful as walking through the fifth a mall of Paris on a Sunday morning and getting a whiff of fresh pastries is how they leverage HD wallets so they use a 12-word single backup seed for all three currencies and make it super easy to sync your wallets across all your devices so if you're using the Chrome extension or the desktop app you just going whip out your phone scan the QR code and boom your wallets are synced

and plus uh the people at Jax take your security very seriously it's open source so anybody can look at the code and plus they never hold any customer funds all the keys are stored locally uh on the client side so go to ja.io that's jx. download the Jack's wallet right now and understand what it's like to use a Next Generation wallet we'd like to thank Jacks for their support of epicenter Martin there's also been uh plans about agnosis Dow agnosis Fund in a crowd sale can you tell us about those yeah of course so um so to build this um to build this uh decentralized prediction Market uh

platform our um yeah our path forward will be to also do uh crowd sale and of course part of the reason is um to raise to raise money um to do this but also uh it it's also about uh um creating the the right incentives to have this one platform that Aggregates all the liquidity to build such a platform you want to have as much uh stakeholders and shareholders um as possible you want to align the incentives of everyone um uh who who who is in the space uh to agree let's coordinate on this platform

and let's uh achieve by this the max the maximum possible um uh efficiency um so yeah that's why we will um that's why we will announce or are announcing now that we are preparing a crowd sale so we will We aim currently for the weeks around Devcon um the next uh Devcon 2 Shanghai the big big ethereum um uh conference and we we want to uh fund two two entities so first um the nosis

Dao um that will that will um build the platform um that will run the platform that will some degree govern the platform um and the second entity will be the nosis fund um that um that will fund projects and we call it skins um that that can or will be built on top of the platform so you can build uh kind of endless things you could you could do with a prediction market so you can you can do insurance you can do uh Sports bding you can do

forecasting of um of the weather you can do forecasting the next Supreme Court decisions so so there are endless possibilities you can do uh with with a prediction market and and each of them although they are they are fundamentally using the same the same mechanisms they will have very different user bases they will need very different um um interfaces super important to to to to create interfaces that people understand and and are easy usable um and they maybe even need to um they need to be in different jurisdictions to to um to

comply with with different uh laws and regulations um so those um so that that is our second entity um we want to crowdfund uh this nosis fund that will uh invest in those uh Ventures so let's say let's say I'm uh I'm um I participate in the crowd sale and I say s some ether and get some tokens in theosis Dow right now um does in the Gris now do I expect a dividend of some sort in the future is it the case that the Dow will make profits of some sort and give me a dividend or is it like Bitcoin which is which has no dividends so it it it will have some uh

some utility so the tokens you you will get um in in the nosis now uh will have three uh three properties so they will first be used to um to govern the platform um they will be second secondly they they will give you the right um to create uh markets um on top uh of of the platform and they will give you the right to create markets um without paying a fee on those markets and that's that that's super important I will I will go in into this

uh a little bit deeper but yes they will also um give you uh um will give you a fraction of the shares of the fees that that are earned um on the platform so so may maybe let let me describe a little bit um this fee model and and why we think that that is um that that that is super important and it plays back to this this idea how can you create a platform where everyone um Can can agree on um and and as much as people could could agree on such a platform so you could have um a model where

um where you would charge specific fee 1% half a percent whatever um and distribute it to to the Token holders however uh if you are today and you let's say you you you raise two million to to to build a sport Bing thing um the question is would you really build on top of a platform where you would forever uh need to pay um even if it's a low percentage of of fees so it's it's to some degree questionable and and often uh whether whether you would build on platform where you you would have to pay pay fees

um uh for forever so and creates an incentive to split no into four right right right so so often the question that comes is uh why cannot someone just copy uh copy the the platform and remove the fee from it and and and then um run run it without the fee and and and that is why we want to um offer an option to um to to buy in into the platform um um is a long-term stakeholder and shareholder and then use it without uh paying a fee and and our claim uh to make it simple is

uh oversimplification is the claim we want to build a platform that creates value and does not extract value what what do you mean what do we mean by this so the idea is you you you have your money you want to you want to build something and and you might decide well instead of forking it and and having the costs of of of forking it and then maintaining my own fork and having the um disadvantage of of not sharing the liquidity pool not sharing the same platform I just buy into the platform so I buy so and so much tokens of the platform and that gives me the right to create let's say 100 markets a day and

don't have and I don't have to pay a fee on those um 100 markets so what does this mean so that means that there will be as long as people will want to want to use this platform there will be demand for those tokens um because you need to hold those tokens and you need to hold them for for maybe you need to even lock them down to to um to uh to create those markets for free so the tokens by only by this by this demand will will have value uh even if no fees even if no one would uh do the other model so so the other model is if you don't uh want to invest uh capital and

you don't want to have make an upfront investment you still can create markets and then pay your 1% or 2% fee on those markets but even if everyone would use the tokens and no Revenue would be generated whatsoever the tokens would still have a value because you just need them you just need to hold them um there is is by demand for for them um so if more and more people want to use platform the tokens will also gain in value and that that's our claim we want to create value uh instead of um yeah extracting value yeah I mean so what you're saying is uh you can imagine this as like a a country club or gentleman's club or

whatever right like so there's a there's like a club of people a group of people that can create markets and do trades for free but in order to enter that uh enter that group you need to have tokens above a certain amount I I assume right right okay and so so this is the argument that um the utility of the agnosis token is it's like an entry ticket like it is it is actually reminds me of the utility of ether itself that uh you hold ether because you want to pay uh pay gas fees it's like an entry fee in order to do things um in on the ethereum platform and very similarly um gnosis tokens will be an

entry fee to for this club in order that you can trade on it for free yeah Al although if we would uh if we would stick to the ether example then it would be like let's say you would need to hold those uh ether and you would need to lock them down uh and that would give you the right let's say it would be like if you hold one ether um for one day that gives you the right the right to to do one transaction uh for free so so that would be kind of the an analogy so so then there would be still demand for for eer and it would be still uh have have value although it would not be spent or there would there

would be no transaction fees and I guess that that would be also for ether an interesting model okay so do you have any interesting governance prop for the doo itself right so so uh obviously as a prediction Market um we uh we are super excited of of the model to use a prediction Market itself um as the governance model and I mean I guess that that has been this concept of futari or I never know exactly how to pronounce it um this has been well disc discussed for for quite a while and I guess we want to be the first entity that actually um uses this uses it is is using it so so um so

the model we we we we we believe that this model of Turkey um will not and cannot be used for um for small day-to-day decisions and I would argue even voting is is is is not is uh is is is not suitable for so it's in my opinion not a suitable model to have 10,000 token holders and and and to let them vote on on every small decision and with futaki it's even more problematic because you can only in my opinion uh make good decisions with futaki if there are long-term uh bigger

long-term implications of a decision um so the way we will um set our Dao up is that we have a core team and this core team can make um by default proposals and by default those proposals uh will be will be accepted um however it's always possible to challenge or to kind of veto um such a proposal and start this um this futari mechanism and this futari mechanis Mech ISM would work uh let's say there's a proposal bigger proposal to spend so and so much

over the next years or over the next month on on those projects and someone disagrees with that um then this futaki mechanism would would basically create two versions of um of nosis tokens it would create uh nosis tokens under the condition um that's kind of a that's kind of how a prediction Market works it would create two tokens one under the condition that um the decision is implemented and one that it is not implemented and then there will be a trading period for a week or two um and then and then automatically that decision will be made uh that would

maximize or increase the value of nose tokens so if nose tokens are more have higher value under the condition that it will be implemented then it will be implemented otherwise it will not be implemented and eventually um this this mechanism uh can also be used to exchange the core team so uh we we will um suggest that that we will will will be the for first um or we will be the core team that that can by default uh make um decisions and make um um those proposals however with this futaki mechanism it will be possible um to

challenge or to to exchange um the core team with the same mechanism so if no tokens would have a higher value under the condition that there would be a different core team then that can be done yeah that's that's very exciting and I look forward to to actually seeing the the futar key mechanisms in action and see how they they play out I'm sure it's going to take quite a lot of time and playing around and figuring out especially user interface and user experience to get this right but I think once once uh once we get it right it will be super powerful um so you you mentioned a core team and that kind of brings up an an important issue here right so as as you're now doing a token sale what's

going to be the relationship and the kind of business model for for you guys uh the founders and the company uh and and how does that relate to the token holders right so um also there we we we came up um with the with the Innovation um so our idea is that those who buy the tokens um Can voluntarily decide um in the moment they they buy a token so so they will spend one ether get 100 tokens and then they have the option they can decide on their own to create another to uh another more tokens up to 100 um so so ranging from zero to 100 that would

um be issued in addition and would be sent um to the core team and the idea or the reasoning behind it is um that maybe tokens are uh more valuable if there is um is a core team or um I mean it's it's like a startup so so so uh you can if you want to if you want to invest in a startup uh of course you you on the one hand you want to have as much Equity as possible um but it's also it's also a good idea or some some people say they would only um put money into something where there is a core team

that has at least 10 or 30 or 40 or 50% um in own interest okay that is that is super fascinating so essentially what you're saying is people put in money right I I get my 100 tokens and I say okay I'm going to give 10 additional or 30 additional Etc to the team so essentially it's sort of a choose your own delution model uh but also of course what this brings up is that you have a big free rider problem right so economically probably the rational thing to do for me would be to not give any tokens to the so it's not it's it's not the free writer problem because you will get if you put in one ether you will get

100 tokens anyway anyways so you don't have to use your own tokens uh to give a fraction of your own tokens um yeah to the core team you would just generate another um X token so if you decide that ratio is fair then to some degree everyone would uh would um would would a solution yeah yeah you're totally right yeah no got that wrong no that's that that's very true and that that's extremely interesting model does that mean those additional tokens would go to the the founding team sort of as like equity in the thing and then they would get paid from The Ether you know their

salaries and things like that I I guess it's important to note that that those that that we we try to go or we will go this through that we see those tokens as as something as a software that is necessary to use uh the platform or with so um but but yes it will go um it will go to the core team so the idea is that and and we will have um kind of long-term uh vesting so only people who at least spend three years uh working on it will in the core team will eventually get um those tokens but yes in addition the money The Ether that is collected

will be will be used to pay regular salaries now a last topic that we should probably address because many people will have that question there is a prediction Market or a a project for prediction Market at least on ethereum already that has gotten quite a lot of attention because they did a successful crowd uh funding campaign as well which is augur can you just very briefly address like what's the difference between gnos is and or sure sure so um I would say there are maybe three three differences uh first of course they they similar thing it's also working on on um on ethereum and

they also have the goal to build a decentralized prediction Market um however so our um our Oracle model is very different so so they they have this um this model where every um every augur token holder would vote on every uh on every outcome and we think um that is simply not efficient so so for for fet for example uh currently there are 2,000 events created per day uh and the claim of augur is every month uh every token holder would vote on on um on the outcome of events you can

simply calculate 30 so so in 30 days um only F and and fet is not is not um that big yet it would be 60,000 events where you would vote on so so that that that would not work in in in my opinion or in our opinion or at least it would be very inefficient um so again our model is to use um the is uh as a default use centralized oracles um like again Bloomberg W from alpha um and only have as backup mechanism um the

decentralized uh the decentralized Oracle so that that is point point number one uh Point number two is that we uh Focus we focus on uh being uh a platform where it should be super easy to uh to build your um applications uh on on top so again there's so much you can do with a prediction Market some insurance flight Insurance markets some Bounty markets for finding uh uh finding bucks in a contract some um credit peer-to-peer lending credit default swap market and and so on and so

on and uh we want to provides a tool set to make it very easy uh for for developers and eventually even make it so easy that you don't even need a developer uh to build um such such markets so our claim is eventually it should be as easy as today setting up a WordPress block so if you're interested in a topic with a few clicks you can set up your WordPress block on a specific topic and similar to that it should be as easy to set up your a prediction Market on a specific um topic with nosis okay that that was point two and point three is I guess this um this other platform model where it hopefully

will uh be easier to coordinate on to to agree um that those rules uh that that that you would like to make midterm or long-term commitment to those rules because you know if I am share token older to that degree then I can create unlimited markets without losing value to the platforms and still giving the platform value that that's beautiful part of it in my opinion okay Martin um thanks so much for coming on that was extremely interesting talking about uh nosis prediction markets fairly and the exciting future that's ahead of us now of course we will link to uh the website

and we look forward very much to the the news and the developments that will come up with the the crowdfunding campaign and I'm sure a lot of people that might not be as familiar with gnosis at the moment will become a lot more acquainted as the project kind of comes out and and and does a a crowdfunding campaign that hopefully has a happier ending than the last Dow so thanks so much for coming on Martin right thank you thank you and with that that that's it so thanks so much for listening we will be back next week so EP and Bitcoin is part of the ltp network you can find this show and lots of others on let's talk bitcoin.com uh we put out new episodes every Monday you can get it of course with all your

favorite podcast apps or watch the videos on youtube.com/ epicenter Bitcoin and and that's it so we look forward to seeing you back next week [Music] no [Music]