 So welcome everybody, thank you for coming this afternoon to the talk. Our topic today is the Dow of Bisq. Bisq is an open source project on top of Bitcoin. So it's a big change from the previous talk, no Ethereum. And Dow is one, how has this happened? Okay. Dow is one of the components of Bisq, which is right now in testing phase and is going to be launched in near future, so a few weeks I hope. A few words about myself. So I'm a, as I said, I'm originally from Israel. I live in Berlin in the last many years. I'm an mathematician, I have PhD in mathematics in the area of dynamical systems and differential equations. And in the last few years I was doing consulting about cryptography, decentralization, consensus protocols. And the other stuff that I was doing, not for profit, is being involved in open source projects. For example, Bisq. And the specific thing that I did in Bisq is being part of the team that designed the Dow of Bisq. Bisq. I'm curious, so if you can raise your hand, how many people here have heard of Bisq? Or know what Bisq is? That's a much more people who, when I gave a talk two months ago, it was less people. That's good, that's good, that's good. Bisq is a peer-to-peer marketplace for exchanging Bitcoins with other currencies. And when I say other currencies, I mean crypto currencies like Bitcoin Ethereum, Bitcoin Monero, but I also mean Fiat. So you can also use Bisq to exchange Bitcoin with US Dollars or Bitcoin with Euro. Now you may say that, okay, this is very boring. I mean, there are lots of places that you can exchange Bitcoin for other stuff. What's special about Bisq? And the thing which is special about Bisq is that this is a software. So from the moment that you downloaded the Bisq software, we are out of the pictures. There are no services. We don't give services, we don't communicate. This is also why I personally, Bisq is at the centralized project, so people have different views. I call it a marketplace and not an exchange because when you say an exchange, people think of very fast transaction. Some operator that does something, even in the centralized exchange, usually the operator does something, has a website, interface, something, we don't do anything. You download Bisq and you're on your own communicating in the peer-to-peer network with other people who download Bisq. So what does it give us? First, it gives us privacy. When you exchange something with somebody, the only person who know that you did it is the other person. So it's not a non-imity because if you exchange Bitcoin and Euro as somebody who knows your bank account, it is a privacy, which means that nobody besides the sides who are in the deal know about it. The other thing that you get here is censorship resistance. So we are not part of the trade. We cannot do anything. We cannot censor you. And the third thing that you get is if the software is well written and the protocol is well designed, you get security because your account cannot be hacked and exchange cannot decide to lock your money. The other part of the talk is going to be DAO. So I don't think that I need to ask how many people here have heard about DAO because this is a much more common, common concept in blockchain. DAO is a decentralized autonomous organization. And it has three parts. Organization is like, I think, people know what it is. It's a group of people that decide to work together towards some common goal. And in the normal world, traditional world, non-blockchain world, usually if you're a group of people who decide to do that and you want some kind of security, you have to register yourself as an official organization with some legal jurisdiction. And this solution is great. I mean, it works very well. It's not always fit for some kind of new internet communities because if you look at BISC, we have people from all over the world, from the States, from Europe, from Israel, from other parts of the world. Some people I don't know where they are. Some people I don't know their real identity. If we decide to register a legal, like a traditional organization, it is unclear so much in which jurisdiction we do it. And the moment that we also choose a jurisdiction, it will kind of influence us, give some people more control because they will be closer geographically to this jurisdiction. The other thing, sorry, so what we try to do in a doubt generally, in my point of view, is to take this legal framework and replace it with cryptography, decentralization, peer-to-peer network, and protocols. And then you may say, well, that's nice. I mean, to replace legal system with cryptography and protocols, but why do you need it? I mean, I just said that BISC is an open source non-profit project. We don't give a service. We make a software. We can just put it in the internet. We don't charge people to use the software. Why do we need it? And there are three reasons that we need a DAO. One of them is, though we don't have income, we still kind of have of internet assets. We have a domain name. We have social media account. We'd have an email. Somebody needs to manage that and we want the people to manage that not to steal it. And the DAO is one of the things we're supposed to ensure of it. The other thing is that while the organization, the software is free, the organization doesn't give any service, there are people in the community who give services like mediation to other people in the community. And we want somehow to give the people who, so we want somehow give reputation and the DAO ensures that. And the last thing which is very important is that, again, while it is free, it is also a software in a very charged area. And design choices of the software are important. I mean, if the developer makes the default, the developer has a lot of influence of how people use it. And we don't want one person or one small group to do the defaults. And the DAO is, again, one of the things that's supposed to decentralize this governance and decision-making process. So I would love to go straight into the DAO because that's my passion, but I cannot speak about the DAO of BISC without speaking about BISC, so part one, BISC. So I heard about BISC a long time ago, I think 2015, but I was not involved in any way and then the first time point that I got involved was a year and a half ago in this place. This is Parallelny Police. It's a hacker space in Czech Republic. This picture is cool and I have to say that the building in reality is almost as cool. So the picture does it justice. They call themselves Institute for Cryptoanarchism and once a year in October, they have a conference. So this conference is half technical, half political, half economical. I do not identify myself with any political point of view, but I go there because I get to see there's lots of really interesting stuff. So there's lots of libertarianism there and though I am not so much into that, I like to hear about it, I like to hear about all the alternative stuff. And a year and a half ago there was a talk by a co-founder of BISC. His name is Chris Beams. And I think the title of the talk was how to boost up a DAW. Back then I was doing kind of a side project of DAWs which was never down like most side projects, but I had lots of thoughts about it and I had lots of doubts. And I knew the problems of how difficult it is to make a DAW which actually works, scalable, functions and all this kind of stuff. So I came to the talk and next to me set another BISC developer and during the talk I got into what happens to me normally in technical talks, I get grumpy. Like he says something and I'm like, no, no way, let's not talk. And after the talk I stayed talking with this BISC developer and while in general I'm a very nice smiley person, when I get to kind of theoretical discussion I become intensive and sometimes it gets some, people need sometimes to get used to it. And I think that I was very intensive with this guy I was asking question but how do you do this? But this wouldn't work until at some point he just told me, hey man, you wanna maybe look at our design papers and see what you think. So I took the papers, I went on the red like for three, four days then I spoke with another co-founder of BISC, Manfred Karer and we changed some things, we updated a week pass by, we made a few more talks, a month passed, they asked me some questions, I get some more input and as time goes I found myself more and more involved in the design and development of BISC and after eight, 10 months I think that I was already saying that I'm a BISC contributor or a BISC core developer however you want to call it and I'm not saying telling this story just to show that I was in a cool place although I was, I'm telling it because I think that like in first time I meet people who are in an open source project and they know how you join an open source project but then I also meet all the people who want to join but they don't know exactly how you begin with it. So that's one way, you take some component which you like, you think about it, you speak, slowly you get into the community until eventually you are part of it. This is what BISC looks like. It's a screenshot from the software. No surprises here. I mean if you saw some kind of exchange or a place to buy and sell Bitcoin that's more or less how it looks like. There are asks or bids. The difference begins in behind what happens behind the scene because there is no server there is peer-to-peer network and also what happens when you actually do a trade because there is no optimization here. There is no operator that does automatic matching and makes sure that the money transfers, et cetera you have to do it yourself and what the BISC software tells you is it tells you how to do that. So let's take an example. That's me. That's a picture of me that was drawn when I was 20 years old. I already didn't have much hair but I still didn't have a hat. And I want to buy Bitcoin. So I broadcast via the BISC like some kind of announcement that I want to buy Bitcoin and then some international stranger comes back to me. Now I don't know what the international stranger looks like but I assume that he looks like this because I don't trust international strangers and neither should you. We discuss. I mean, he took my offer and we agree on some kind of exchange. I give him 100 euros. He gave me a 0.1 Bitcoin which is a fantastic rate for me. So the international stranger is actually very generous and then we go to the BISC software and the BISC software tells us actually how to do this transaction. So the first thing that I do is I put a tie because I don't do business without a tie. And the second thing that we do is we go to the BISC software and we open a multi-sig account in Bitcoin. So the software actually has like a Bitcoin wallet embedded in. We usually use this one. In the multi-sig account we put first thing of all, shit, the amount here is not updated. So now the trade is a bit better for him but we put the Bitcoin that he was supposed to sell me which is now a bit less, which is not good and security deposits. I put security deposits, he put security deposits. The security deposits are in Bitcoin. This already means that to do use BISC and to do the whole trading protocol I need to have some Bitcoin to make the security deposit and that's true. We needed for security and I'm not sure if there is any way around it. So if you have zero Bitcoin and you wanna use BISC you should get something at least to pay the security deposit first. Then once we did the multi-sig account so there is money locked there. We go to the software and the software tells me to give 100 euros to the bank. Great. The bank, when I tell the bank to give the 100 euros to the international. So now he has my 100 euros. I don't have the Bitcoin yet because it's still locked in the security deposit. So in the next stage he tells the software that he got the money. The software tells him, well, you guys go and you unlock the security deposit, you unlock the multi-sig account. I get my Bitcoin that I bought plus the security deposit. He gets the security deposit and we are all happy. This process is slightly simplified, of course, because there are security risks here but I don't want to give the full trading protocol. I should have it. We have about 1,001 security measures in the protocol and I hope it's well thought. We are, of course, happy for any security problem that it has. We have, for example, beside the security deposit we also have some ideation protocols and arbitration protocols and we have, if you wanna sell Bitcoin for euros in the first month, you think you can only make offers up to 100 or 400 euros in the second month a bit more and third month a bit more. So you cannot just begin and do huge deals with it. Bisk, so the work on Bisk began in beginning 2014 right after the Mt. Gox accident. It might be a connection, I'm not sure. And it took about two years of hard development until the first release was in 2016 and, of course, when your first release as you find out that nobody use it. Then the numbers grew up a bit with time. In December 2017 there was a jump because the Bitcoin blockchain community went crazy. So all the exchanges were happy. Met the bidon afterwards and in the towards the end of the year it started to grow again a lot from the Monero community who started to exchange Bitcoin and Monero. In January there was the launch of Greencoin which Bisk was one of the places where people actually traded at. So we were very happy. And I should say that we have smiley, nice community right now and I think it's a great time to join the community because it is big enough such that if you go to the Slack, if you go to the forums, mailing lists, stuff are happening. It's not boring, but it's small enough such that everybody knows everybody at least by nickname and you are not overswept with information. So it's easy to follow. Bisk is developed, it's written in Java. So Java developers, please come to our Slack and join the process. We need basically also non-Java developers. We need everything of everybody. So people who write documentations, who do social media, who do videos, everybody are welcome. We are a group of smiley faces. So please join us. So there is a smiley Bisk developer in the crowd right now. Now we can go to the Dow of Bisk, chapter two. And I already talked a bit why we need a Dow, but in general when people ask me why do you need a Dow for an open source project, I begin with a story that you know. You are at home, you write a hobby project. You make a tiny demo and you show it to your friends and your friend says, oh, that's quite a cool idea. Let's join. And now you're three people. So you work together, you make an MVP, you release version 0.011 or whatever version it goes. And suddenly more people with it joins you. And then a few users and you become more people and more people. And then this happens in almost every big enough open source community that doesn't have any structure. And this does not happen because engineers are strange or social people. I think it happens because we are human. So there was a book that I read earlier this year. It's by Yuval Noah Harari. He's like, he's an Israeli. He's I think kind of the poster child of Israeli history world. He, like, there is a course in the Jerusalem University of introduction to the history of the world and this course was traditionally given by a dozen professors until he came and he was just teaching the whole course from beginning to end. And in this very interesting book, there is a point about communities. And he speaks like, you know, how big community people can make just with communication, direct communication. And he gives there a number. I think it's 50 or 80 or 100. It also depends on personalities of people, et cetera. But there is a limit. That when you reach to this limit, you need some kind of social structures, some institutions, some stuff that people knows a way to make a decision, a way to people to know that they belong to the same community or they don't really never spoke to each other. And that's where you need a structure, a DAO. So not a DAO, an organizational structure. If we are a big enough community of developers, we can have a traditional organization. That's option one. And I already spoke about it. It's a fantastic option. It's worked for many cases. I think that in the case of biscuit was actually caused problems because we are all over the place. And also we don't have a bank account. An organization needs a bank account in some country. Who of us will do that when I have to trust them? It's a bit of an issue. The other thing that you can do is just use an online platform, a service. So GitHub as an organization option, which we use, and it's good. And you can add to it more feature. And eventually you can really manage your project with voting or whatever the online platform offers to you. And it's okay. And I think at the first time it's okay to say that my experience with online platforms is that it's good for a short while. So when I join a platform or some service, usually I am happy for a month or two, a month, a half a year, a year. Every service that I ever had at some point I got annoyed by. They did something which I understood that I don't have control. They changed the terms of usage. They locked my account until suddenly I have to pay and it was impossible to migrate. So I don't like platforms, but I use platforms because in the internet you cannot really do things alone nowadays. I cannot run my server at my home. But as a community, I think that you can be alone. So if you are many people, you can do stuff alone. So P2P network can be a bit like a server. And everyone does a few things and you can really manage your internet stuff alone. So that's for me the DAO, the alone solution or at least an effort. There are many ways to look at the DAO. So DAO is very popular right now in the blockchain world and you go to different people, you got different definitions and even I kind of always change here. How I look at it, it's all point of views. Not one definition. I'm gonna skip it because of time and I'm going to go to what? So we have a DAO, what does the DAO does? For this I have screenshots, which definitely could have been better for resolution, but I hope you see something. What you have here is people making compensation requests. The BiskDAO has a token and you use the token for governance. So for voting on the season, you need some kind of token. You get the token basically by contributing. So the more you contribute, the more you have, the more tokens you have and the more tokens you have, the more you can vote. And this is compensation request. For example, this is me, you don't see, but here I am making compensation request for 100,000 Bisks and here after the proposal stage is over in the DAO, people go to the voting stage and downvote my request because they don't want to give me 100,000 Bisks because definitely I tell more jokes than contributing. What else we do is we have kind of bundled rules. So this is all from the test net. The data here is like fake. It's not live, but we have bundled rules. So like Twitter operator, mediators, community managers, if you want to be one, you make a bond in tokens and then if you somehow misuse your trust, the community vote to take your bonds. So it's kind of supposed to make us a possibility to find people from the community who misuse the trust of stealing the Twitter account. We also can bond reputation. So if the internet stranger, for example, would give a big bond for his reputation, then I would feel a bit more secure to make transaction with him because I know if he cheats me, I can go to the community and prove that he cheated me and then they can take his bond for the reputation. So that's another thing that the Bistler is supposed to do. And the last thing which I didn't take a screenshot, I think, is a protocol parameter, which is super important because we have a trading protocol and we want to update that. Right now, big updates are kind of a hard fork. People really have to reinstall the software and if you just take parameters from the Dove and it's all about voting and nobody really makes a long decision, the community make a decision and things are being updated without a software update, just the Dove is being updated. Nowadays, when people speak about Dove, so there is this tweet from Alexander Sander, I think from two weeks ago, he's a film developer and this was a fantastic tweet. It's a series of tweets and the first one says that 2019 might just be the year of the new Dove on Ethereum. I would like to highlight a few on-chain governance experiments that I'm excited about and then it goes on for a few Dove stuff on Ethereum and each one of them in a really nice few lines, he kind of sums up what they do. We are not on Ethereum and our governance, I'm not sure if it's called on-chain but my point is that nowadays, when it comes to Dove, people are really curious about governance. So I have to speak about governance. It's another kind of part two and the goal of Bisk is to have a meritocracy. So people who can't attribute it to the community should make the decision. This is our definition of meritocracy and the question is how you do that, how you make decisions, the answer is voting. How you decide who votes, well, you do it by tokens and who gets the tokens, you get token by contributing. I'm not going to speak much about tokens because I saw so many very nice, great projects that once they entered the token, this was the only thing that they spoke about and I don't like when it happens. So the last thing about the tokens is that you can also get tokens so people can exchange tokens and the question is how do we do the exchange of the tokens and what we do is kind of outsourcing. So we do collared coins on top of Bitcoin. I think that's it for about tokens. Tokens are necessary for us to know who can vote and it's necessary for people to want to contribute and get compensated and maybe able to sell the tokens so they can get compensated in Bitcoin. I don't want to focus on tokens because for me personally it's not the main thing that I'm in the project or I'm into DAW. Sorry. So the third part is how does the DAW work? There is implementation and the most important thing about this DAW is that the second order DAW. So it's supposed to be more scalable. The question is what do I mean by a second order DAW because it is not a well-defined term and one way to look about organizations for the purpose of this talk is a series of records, series of decisions and you can like order the records. So this set of records was make first and this second and this and this and this. And in general you have records plus some organizational rules. That's a way that for the purpose of this talk I look at an organization. Now if you took all of this thing and you put it on a blockchain then you have our own chain DAW which is great but not scalable plus we use Bitcoin. We cannot put those things on Bitcoin. What we can do is put hashes on Bitcoin. Many, many, many, many, many, many, many hashes. Every proposal, every hash of every proposal is put on Bitcoin. You can use like all pretend to put hashes. The hashes of the votings are put on Bitcoins. This means that it's a lot, a lot, a lot, a lot, a lot of small transactions. So is it scalable? Kind of, kind of. It's more scalable. It's scalable enough. I mean it's not going to be a method for Google because Bitcoin has a finite amount of transaction that it can do. So for a huge organization it won't work. If you now have 10 fast organizations using this method it probably also won't work. It works for BISC, for this stage and this side, this size. So it's scalable enough for us. It's scalable enough for probably more 100 organizations like BISC, not scalable forever, but it's a first step. Then our BISC block, so this series of decision is carried in first phases. Every phase is described in a number of Bitcoin blocks. So we have some kind of amount of Bitcoin blocks where people can make proposals. Then an amount of Bitcoin blocks when people have blind voting, blind voting such that nobody can manipulate the results by looking at what other people do. And then there is the last stage of reveal votes. And the result of all of this is a BISC block. The BISC block, like the whole BISC block cycle takes about a month, more or less, depends on the BISC, on the Bitcoin exact, on the block creation times. And it results in a new BISC DAO block where the information of the block is in the P2P network. And the hashes of all the information is to verify that in the Bitcoin. There is still a question of how do you do it because how do you manage the votings? How do you know exactly which data is on the block? What happens, especially what is very dangerous if somebody puts a hash on Bitcoin but does not publish the data on the BISC P2P network? And for this we do two assumptions in the design. One is that we can get the Bitcoin data. So everything which is on Bitcoin, we assume people can read and it's available. The other thing we assume that the stuff on the BISC P2P network is available for people if a big amount of the participants of BISC of the P2P network has it. So when we design the protocol, we have to make sure that if a big amount, either with a big amount of people have the data and then we know that it's available for the rest of the community or if people don't have it then it's not so important. Like let's go, sorry, I'm still in this phase, let's look what happens. Like if I want to now make a proposal, compensation request or something, I hash it, I put the hash of the proposal on Bitcoin. So in the end of the phase people go to Bitcoin and know exactly what they are supposed to vote on because there are like some amount of hashes. They may not know what is the proposal because it may be that I put a hash on Bitcoin but I never really broadcast the text of the proposal on the BISC P2P network. In this case, the default vote is no. So if people don't have my proposal, they will outvote it and then it doesn't matter that other people will not have it in the future because it's not going to be part of the organization. Then when this phase, proposal phase is over, I go to the blind vote, I take all the proposals that I know from the Bitcoin and I take the text that I got from the BISC P2P and I vote and I encrypt my vote. I put the hash of the encrypt of the vote on Bitcoin. So then I know how many people voted. There is a risk here that I will not publish the encrypted vote in BISC P2P. So maybe people will think that I voted but actually they will not have the encrypted vote and they will never be able to decrypt it and know what I wanted to do. Which is why in the last stage, we do two things. One, we put a decryption key on Bitcoin. So we do it in one transaction. Put a decryption key on Bitcoin so people can decrypt our encrypted vote. That's great. Second thing, we take all the encrypted votes that we got. We combine them together in one block. We give it, we hash it and then we also put this hash on the Bitcoin. And then people can go and see which hash is the most and this will be the hash of the block that most people in the BISC P2P network has which means the data should be available. It's again a slight simplification naturally of the full protocol and if you wanna read more about the BISC DAO there are two places to do that. One of them is the BISC network slash DAO. You get all the information you want for people who are users, technical, not technical, links to technical documentation, long specification and then there is my analysis which I wrote independently and I put on my GitHub. It's quite a high level analysis but not so long, I think five or six pages of the BISC protocol. We're supposed to go to explain how it works and convince you it's safe. And I think that this would be it. So thank you for coming. Yeah, that's it. Hi, thanks for the presentation. Just wondering how do you handle Bitcoin forks and also can the voting be disrupted by miners for example by discarding certain transactions? So you say how do we handle Bitcoin forks and the second part is? Can the miners affect the voting by for example blocking certain transactions? Yeah, it's a good question. So I think there are Bitcoin forks that were like in the first stages of the BISC. It was early stages, it was quite central. I mean centralized, there were a few developers and we just made our decision. For the future I hope that when the DAO comes to action there will be kind of a voting inside of the DAO because there is also general option therefore general votings to make a decision which of the fork we want to use. And then the question about censorship. Yes, we cannot make Bitcoin miners to accept the transactions. However, this is a general Bitcoin problem and I'm not sure it's a problem because you do have right now enough miners to accept the kind of transactions that we do. So even if some don't want it, the transaction, the voting period is quite long. So hopefully some miner will agree to mine it. Hi, thanks for the presentation. Is that DAO going to deal with the trading fees of the system? So in which way are you going to deal with the trading fees? I mean BISC has also trading fees, right? Yes, yes, yes. So if DAO is going to deal with that or it will be just as right now? No, so right now the trading fees are for compensating. It's go for the arbitrators, not for the organization because organization doesn't exist and doesn't have any revenues. There are two things about it. First thing of all, there will be a parameter of the DAO I think which is trading thing so you can vote on it and decide to change it. And the second thing is that we are changing the arbitration mechanism to a mediation mechanism which should lower the trading fees and would also give it an option to pay it in BISC tokens to the mediators. So it should change. However, exactly how it will change, I think that the answer would be clear in about two, three months when this thing is really launch and we see what the community wants. So the idea is that it's supposed to give you a tool to make community decisions. We have here a developer one second. We have here a developers who write right now the new trade protocol. So how is it going to be? Can you wait for the mic? You hear me? I do think actually for your question that for instance transaction fees, things like that, those are the kind of parameters that the DAO will adjust. So yeah, that's part of the purpose of it. Any more questions? Thanks. I have the question. Is there any mechanism to prevent someone from voting multiple times or buying tokens to vote with? So you can buy tokens to vote with? Yeah. The idea in general, I think like most voting blockchain mechanism is that 50% of the coin holders are good people. If more than 51% people who hold the token wants to make bad votes, they can do that. There is one thing maybe to say about the whole vision of a DAO, it's not a blockchain. I mean, so I present a series of blocks, but that's not the idea. The idea is to represent an agreement in community. And if somebody will really buy most of the tokens and try to manipulate votings, the community can decide what they fork. So that's one thing. And the first question was, if somebody can vote multiple times, so I think the protocol allows you to vote within the period only once with each token. Yeah. So you cannot vote multiple times. Yeah, but if you create multiple accounts, then you could probably. Yeah, the votings are weighted by the amount of tokens that you have. So if you create multiple accounts, I mean, it's a linear sum, it won't matter eventually. But a good question, thank you. Thanks. Any more questions? All right, let's thank K.L. again. I mean, I don't come from cryptography, but every Israeli mathematician has to do cryptography at some point of his life. I use community. And I changed to cryptography, and now I do a little bit of consulting like enough for me for income, and I spent most of my time here. I'm not going to leave all of this. But yeah, I will try. Thank you, the first time I saw this. You didn't play that game for you. How long have you been living in Darlington? 9 years. I did that yesterday. Okay. My master didn't do it. I think highly of him. So long. Okay. Ready? Paris, France. All right. Okay. This is your bottle. Thank you so much. Thank you very much. Thank you. Thank you. All right. Have fun. If you need to put the microphone, I'm afraid. Is there something to talk about? Oh, A.M. A.M., I think you left us. Yeah, I didn't click. Awesome. Thank you. Sorry. Should I test? Can you guys hear me in the back? No. That's good. All right. I'm going to mute it. Now, before 15, you have iPhone hour slot, which means 20 minutes 40, 25 questions, and then 5 to switch to the last quicker. If you want, I can use 5 minute sign. Yeah, okay. All right. All right. Next talk, we'll start in a couple of minutes. It's by Sean Tabrizzi. It's about substrate and open source framework for building blockchains. All right.