 I have a very loud voice and this room is not so large, so I'm not going to use a microphone. Can you guys in the back hear me clearly? Okay, perfect. Great, so thank you for joining our governance breakout session. Thank you also to all the other presenters. We actually have a really exciting lineup today. You're going to hear from a number of people, a number of very hardworking, very intelligent people who have thought a great deal about governance from a huge variety of projects. I think we've had about 10 projects represented. That's quite exciting. Before we kick off, before we dive into governance, I thought it would be very helpful if we start with some basic introductory material. So I'll just take a couple of minutes now to introduce this concept of governance, share some basic terminology. Do we have, is it, Tak-san? Are you checking the time? Do I have the timer? Yeah. Okay, and you'll give me... Yeah, you have 19 minutes. Thank you very much. Maybe you can use the microphone because of the recording, right? Sorry? Are the recording the one? Yeah, yeah, yeah. Do I need to use the microphone? Thanks, thanks. Okay, this is for recording. Hello, hello. Okay, great morning. Cool, all right, let's dive in. Very briefly, by way of background, my name is Lane Reddy. I first and foremost am a blockchain core developer. I've been writing software my whole life, and so kind of the way I got into the blockchain space was through the technology and the code. It's still the main lens I use when I think about the blockchain. However, so I sort of have had this crazy journey over the past two years. I worked as an Ethereum core developer for about 18 months to two years, and more recently I've been working on another project called... What's his basement? It's early, I'm on my first coffee. But so what I kind of discovered as I was working on some of the core technology is that, you know, technology is not a solution by itself to anything, okay? Technology is a tool that we can use to solve problems, but fundamentally those problems are social problems, and I think it's very critical that we attack the social aspect of those problems first. And one of the problems that I think is sort of unsolved today in blockchain is governance, and that's why I care a lot about governance, and I've done a lot of thinking and reading and writing and speaking about it. So this is what we'll talk about in this quick talk, so I'll just share some goals for the talk. We'll talk about what governance is and what we're trying to govern, some terminology, a quick glance at the agenda, and if we have any time left we can talk some more about Ethereum governance as a case study. So my goals for this talk are pretty straightforward. First of all, everyone just warm up, get some caffeine, no more faux pas. I just want to get people on the same page about governance, right? Because governance is a very subjective, complex topic, and the reason I call this talk what we talk about when we talk about governance is because many people mean different things when we talk about governance. So just to get your juices flowing, we'll kind of talk about some of the different angles of governance and some of the different definitions of it, and we'll get some common terminology as well, because you'll hear a lot of these terms used in some of the talks you hear today. So let's start by answering or trying to answer this difficult question, what is governance? Does anyone here want to offer a definition? You're paying attention, you seem awake. What does governance mean to you? How we make decisions. How we make decisions, that's a good one. How we deal with other partners to get something to be done. So Max has a very good point, which is that it's very much about making decisions, and you also offered another piece, which is that we're not alone. It's how we do this with other people. John, what does governance mean to you? There are some good answers to collective decision making, the process, the execution of decision making. Yeah, process as well, cool. So I don't intend you to read all of this, but this is just to show you, so this is from an exercise that John and I and several other people in the room participated in recently around-table conversation about governance, where one of the things we tried to do in the beginning is we tried to define governance, and we came up with all these different answers, grouped into categories, so it's partly about decision making, it's partly about collective action problems, it's partly about power and authority, rules, change management. So it's a very, very complex topic. Again, we don't have time to go through all of these, you're free to review the slides later. But the key takeaway for me is that what governance really boils down to is exactly what these two folks over here brought up, which is we have to make decisions together as a community, and that's a very difficult thing to do, because we all have different needs, we have different values, we have different preferences, and so we're trying to sort of aggregate those preferences and make the best possible collective decisions we can and enact those decisions together. So another very, very important question to answer before we dive into governance is like, what are we trying to govern? And I'm going to use Ethereum as a case study here because it just helps to have concrete examples, and I'm most familiar with Ethereum governance. It's kind of all these things, right? So, again, just picking on Ethereum, there's the governance of Ethereum Foundation, right? And the resources it controls. So that's kind of a meat space idea, it's very, very specific. There's also like the community, which is also meat space in the sense that it's, you know, human beings in a building, in a room, you know, around the world, but it's much favor, it's very difficult to define the community. There's the governance of the core protocol, right? That's a separate question, and if there's a specification, like the yellow paper in case of Ethereum, there are independent client limitations, right? Ethereum has something on the order of seven or eight of these, and the ETH2 has others, and they each have their own governance processes. There is the network itself, right? So this is, you know, individual actors running nodes and software upgrading those nodes. There's the data inside, like the blocks, that's the blockchain, and so this is where we have questions about accountability and things like that. There's applications, I'm sure you could add more things to this list. For the purposes of today, we'll mostly be focusing on these two things, and obviously each speaker has their own take on this, each project has their own take, and you'll hear about other pieces, but kind of the most relevant and concrete pieces are the protocol, how do we govern the protocol, how do we make changes to the protocol, and the data, right? Because again, there's very interesting questions here around immutability and around property rights. So this is sort of the main takeaway, right? Is that governance is complex, it's multifaceted, it's subjective, this is my own opinion. There's no one, like when we talk about governance, we have this desire to kind of say there's good governance, right? Good governance is inclusive, good governance is democratic, good governance is representative, and bad governance is other things or none of those things. It's really not that simple, right? Governance is very situational, and I think that different initiatives need radically different forms of governance. If we in this room here, the sort of 30-ish people in this room were to get together and we had a task, we need to build something or construct something, maybe we don't need complex governance structures, maybe it's a small enough room, maybe we trust each other well enough, maybe we speak the same language, that maybe we decide to do something in a more efficient fashion, and we elect a leader and let that leader be a dictator for some short period of time. And that's maybe perfectly okay, again, in the small context of a small group of people in a small time period, but of course when we scale out to the level of a blockchain or even further to the level of a society or a continent or a nation-state, we need very different things in our governance. So I just want you all to reflect a little bit as you hear these talks today on, like, what is good governance, what does bad governance mean, what are the criteria, like, while there is no absolute good or bad, there are certain things that I think we can say or sort of properties of many or most good governance systems, and these are things like inclusivity and participatory, maybe permissionless participatory nature of these systems. Another very important point that I hope sets the stage for the talks today is this interesting question, right? So when we talk about governance in the context of blockchain, there's two very, very different angles on it. There is the idea of governance of blockchain and governance by blockchain. An incredibly incredible researcher named Primavera DiFilippi who has done a lot of really just mind-blowing the excellent work on governance in general and blockchain governance. I very strongly encourage you to Google her, read some of her papers and watch some of her talks. She spoke very eloquently about this topic. Governance of blockchain is how do we govern the blockchain itself, right? And that's kind of what I showed you a moment ago when I was talking about what we're governing. Governance by blockchain is a much bigger idea which is that once we figure out how to govern these blockchains, then we can use them to govern things in the real world, right? Maybe they offer an alternative to some of our existing institutions and governance structures, whether it's monetary, whether it's, you know, ways to build better democracies, a lot of the ideas that are being explored in the radical exchange community, things like that. But for the purposes of today, we're going to be focusing primarily on governance of blockchain, right? So, how do we govern these complex projects? And my personal goal was that, you know, hopefully we come back a year from now and we figured out how to govern blockchains and then we can begin using blockchains to govern the world using projects like Democracy Earth, which is not these projects. So, shout out to the blockchains. Yeah, that was executed for precision. All right, cool. So, let's just try to talk about some very basic terminology. There's a lot here and I have very limited time. I'm not going to go through everything, but just the core, most important concepts that I think you're going to hear about today. So, very, very, very briefly, incentivization is a big part of governance. You know, how do we incentivize people to vote, to participate, to have skin in the game? How, you know, in the context of blockchain, are we offering block rewards or validated rewards or something like that? Legitimacy and authority are just really core, essential questions in any system of governance. What is kind of the source of legitimacy, right? Why do we believe that in the context of, I mean, my home country in the United States, right, why do we vest power in the Supreme Court? Why do we vest power in the office of President or in the Congress? What is the source of that? Is it the Constitution? Is it a set of shared values? Is it something else? What gives those institutions authority and what does it mean for them to have authority and execute authority? Transparency and accountability, you know, they're fairly self-explanatory, but accountability is really, really essential in governance as well. So I would say a property of most good governance systems is that they are accountable and no individual, no governmental mechanism or body is above accountability. Everyone's accountable to someone else, right? And that's what the system of checks and balances means. I think that that's a very essential characteristic. So I mentioned checks and balances, right? Kind of different folks, different mechanisms, keeping each other in check, balancing power. Stakeholder, this is a very controversial, contentious kind of question. Who is a stakeholder? Is it just people who have a seat at the table? Is it people who are directly impacted by governance? Is it people who are indirectly impacted by governance? Again, in the context of Ethereum, Vlad Zamfir, whom I'm sure many of you are familiar with, who has a lot to say on this topic, has suggested that sort of all human beings, alive today and at any point in the future, are potentially stakeholders in Ethereum governance. To the extent that Ethereum becomes a network that governments are built on or that's used to govern things off the blockchain or that money is built on, it's a very expansive definition of stakeholder and so people have a much more narrow definition. This concept of on and off-chain governance, so very, very briefly on-chain governance, means when we make decisions through the protocol and they're implemented or enacted on-chain, this takes different forms like coin voting. Tezos is a very maximal version of this. We're going to hear more about Tezos today. Off-chain governance is the old-fashioned type of governance with meat space, you know, people having conversations online or in person. This is tightly closely related to a concept called tightly coupled or loosely coupled, which is also related to binding versus non-binding governance. Vitalik has a good article on a lot of this stuff, but again, just this question of like, when we make decisions, do they take effect immediately or is there some interpretation layer around those decisions? We have signaling and voting, right? So signaling is generally regarded as non-binding. It's just a signaling input into another governance mechanism versus voting, which should be binding. I think if you look at what's happening with Brexit today, it's a very interesting question. There was a vote, you know, and there was a single vote and, you know, is it the case that the government and the entire, all the citizens of the United Kingdom should be bound by that vote or should there be another vote, right? Formal informal governance. We have a lot of informal governance in networks like Ethereum and Bitcoin today, where decisions get made in some kind of complex ways that aren't super well documented and understood versus more formal systems where we have, you know, legislative bodies and election processes. You'll probably hear a lot today about different types of forks, hard forks, soft forks. They can be contentious or uncontentious. We have this concept of like a user-activated soft fork in Bitcoin. And finally, an idea I really, really like, which is wet code and dry code. This is an idea that comes from Nick Szabo, who is a legal scholar and a computer scientist. He's the person who invented the concept of a smart contract as well. And so he proposed this idea by wet code, is code that is interpreted by human brain, right? So that's like legal code in that respect of the word. Dry code, on the other hand, is code that's interpreted by machines. And so there's this very interesting balance here of different types of code and how they interact in the context of blockchains, which we can talk about for hours. It's super interesting. Shruti's going to come up and join us and share a bit on her work on system and governance. It's a super interesting and important topic. Introduce yourself as well. Hello, I'm Shruti. Can everybody hear me? I'm Shruti Alia. I'm a crypto economics researcher. And I bring to this space of blockchain all of the concepts that I've thought about in the field of complex systems. So I'm going to talk about governance as a means of coordinating multiple agents within a system. And there's this concept of guided self-organization, which is basically when you allow agents to evolve by themselves and to govern by themselves, but you guide it a little bit so that they end up in a desirable state in easier manner. There are several ways in which you can actually guide these agents. So the first way is through interrupts. And then the second way is through the design of the dynamic of the system itself. So by implementing some sort of systemic control. And then the third way is by constraining the structure of the entire governance system. So I'm going to kind of mathematically characterize this so that it will be easier for you to grasp. Okay, cool. Alright, so interrupts. So I'm going to start with Audrinos. Show of hands. Okay, cool. So then you might be familiar with interrupts. It's basically this piece of code that whenever something goes wrong within the system or if the Audrinos is running out of processing power, then that code will come in and it will modify a certain parameter. So basically interrupts is just when you want to change, for example, if you have this control variable X, then you change it to X prime so that you put the system in a more desirable state. So for example, if your system gets stuck in a local minima like this, if your system gets stuck in a local minima and say this is X, basically you go and intervene into the system to put this back here so that it's in a more desirable state and that the governance can just happen from there on so that the agents can just coordinate themselves from there on. So then you have the problem of or another way of guiding the self-organizing system using systemic control. So this is when you don't just change the parameter, you change the function that controls the parameter. So you change f of X to f of X prime or f prime X. So instead of changing or moving this point to here, you change the function itself. So if you had an initial function that looked like this, you can make the new function look like this or any kind of functional modification. This is usually manifested as hard forks on the blockchain. And then you can have entire digital constitutions. So this is when you want to change the entire structure in which the governance is operating under. So you can go from a democracy to oligarchy or anything of that sort. And this is when you pre-define the structure of the governance itself. So now you're changing the entire set of governance policies that is defined by f one X f two X all until f n X of all of the policies that determine or underlie the entire governance mechanism. So you change it to G prime. So this will become basically all of the different sets of functions in which it's operating under and then you basically transform it to things like this. So that is that although it creates a huge amount of change, you can see how limiting that is. So basically in the whole field of complex systems you have this, you have this kind of self-organization where you allow the agents to govern themselves so that you can observe emergent effects. However, if you go ahead and if you change the digital constitution itself you are guiding the system too much in such a way that the governance of that system or the behavior of that system will be too constrained or too limited in order to actually observe any emergent effects. So although that's high impact and highly constraining it's also. Systemic control is kind of in the medium level. It has some execution challenges. So the execution challenges with systemic control is that it requires a lot of coordination amongst agents. Do you arrive at a consensus for executing a hard fork? Whereas with interrupts it's very low impact but at the same time it will definitely create a change within the system. But at the same time it's also highly flexible in such a way that you will be able to observe emergent effects even if you do change the interrupts or if you change the control parameters. So those are the three ways in which you can mathematically think of how to guide self-organizing systems and I think that next we'll be talking about the different ways in which constitutions can or structures of governments can be designed within self-organizing systems. Thank you Shruti. You're going to do a talk I think later today so we can dive more into this with you later. Cool, so make sure not to miss that. There is an enormous amount that we can learn from researchers such as Shruti who are exploring these ideas of network theory and systems theory and complex systems. I've just tiptoed into this space myself and I encourage all of you to do the same check out this article. Is this your article Shruti? That's I think the Thai's article for the digital constitutions. Cool. Okay, so I'm going to end this talk but I wanted to just show you guys a brief snapshot of what the rest of the day looks like so up next we have a talk from Fede who's going to talk about the history of democracy and random selection which is really exciting. You'll hear about history of blockchain governance. John, I don't know if that's an okay title I think I may have thrown that up there but is that accurate? Yeah, my presentation approximates that. So there's a bunch of sort of lightning talks on the different protocols and then we'll kind of conclude with a few more talks from Shruti and Kate really getting myself. So my understanding is that this session officially ends at 12 but there's nothing here after this and the venue is open until 2pm so I'm happy to stick around and anyone else who can stick around we can go a little over time and maybe we can do a panel or something so feel free to go out and grab lunch come back if you're interested in sticking around. I'm out of time so I'm going to put this on hold for now let's continue with the other talks and if we have any more time later then I'm happy to share more on Ethereum governance with everybody. Okay? Thank you very much for listening. Hello, how are you? So good to see so many friends here and quite good that we kind of staff are being sorted I'm going to introduce myself, my name is Federico I'm CEO at Meros I'm a decentralized quartz I'm going to talk about this now I'm going to speak about this later but I'm going to speak about now is some of the ideas which Ethereum is based which is random selection so how when we think about blockchain governance we have to think that we are thinking about governance and humans had to solve governance problems for basically centuries and by linear right now and one of the key features of the early date in public was that they were based on random selection of people for different office. Is this okay? The internet is very slow, it's loading About my background I studied economics critical philosophy then I did a PhD in collective intelligence and this is like what brought me to this type of topics and so it's good The question to start is if rationality is always a good thing for making decisions we tend to think that rationality is a good thing for making decisions but there are some time where reasons can be very bad when there is racism or some groups that try to just move forward their own interests and what randomness makes is that we have a pool of options and the option that is going to be chosen is not going to be chosen because of reasons it's going to be chosen because of randomness so any of those are going to be selected and so the black break what it introduces is the break of what it's called irrationality a non-rational decision or in the selection process This means that a lottery decision is going to be impartial, it's going to be amoral and a lot of different things that may be important for some political systems especially when we fear that bad reasons may be used to make a decision so this is called a lottery principle you may want to consider lotteries when you fear that some bad reasons could interfere in a decision process, right? so we spend a lot of time blockchain governance debates about what is the best way to create a voting system like quadratic, like X voting system in the end it's a question of how we aggregate preferences so voting is a process that is based on reasons which may be good or bad and lotteries is a process that is based not on reasons, right? and this excludes both good and bad reasons so let's see now a bit of some examples of how this has been used in the history of political systems because you know we sometimes think that we are inventing stuff but we are just discovering things that have been working for a long time in different regimes so who knows this guy? so this guy is the guy who created what we know as sanctioned brick democracy there was this detector called Ipias he overthrew Ipias regime leading a popular rebellion and then he created what we know as the brick democracy and how people vote etc and you know what he did after doing these reforms? yeah of course, but anyone who would disappear so no one knows where he went but the thing is that he created the system and then left it for the citizens to manage so you know who did that, right? so this is a basic structure of the ancient brick system citizens can participate in the assembly or allowed, provided they are of course male and not slaves and the assembly is going to choose courts, council and magistrates and look how important was random selection in choosing people for party coffees too 99% of all the people who were in party coffees were chosen by lot and only 1% chosen by election who were in very high levels in the army generous because they were very specialized in order to vote so the council chose the assembly chose 500 citizens selected by lot and every year it rotated the council was charged like the executive power it was charged day to day running on the police and they had this really specific selection process in Antion it's a Greek ID token and citizens had one of these with their name on it and when there was the day of the lottery process they went to the lottery machine you know how it's called? I've been talking to Sandy for years about this machine and this is a machine that they use for random selection you want it with your token it's like a bronze plate on this side and then some guy from the political process put some white and black dice into the machine but you fix it on the side and they have like this logo here but it's like an icosahedron dice and if you have on your row white dice you were selected, black dice you won't call and this is where the count is so why was this used? why was this done this way? so the Athenian had like different families that weren't extravagant for power and they were very afraid of one of them concentrating too much power into the police so the best way to do this was okay instead of making elections which can lead to patronage let's just select them like randomly and they use this method and on top of that if there is some so randomness sometimes can be a favor of someone it changed every year so they always rotated so every potential game that could be done in the short term by some very lucky random selection was then destroyed the next year when it changed all over again so the two keys of the Greek democracy were random selection and rotation if you want to learn more about how this process works I recommend this is a classical book about the the classical books from the 1930s and if you want to learn more about how the whole political system worked in Greece this great book by Josiah Ober and a lot of the things that we are discussing in blockchain governance not these guys who knew already okay let's move forward some time to the renaissance and another important city where there was the highly used of reputation was Venice Venice was a trading empire and it was known as the Repulica Serenissima why is that? when the Republic was intended to give all the merchants of the middle age the confidence that rule of law was enacted there remember Shylock's merchant of Venice there is a trial scene and in the trial scene the judge says I know that this contract is not cool, a pound of flesh but a contract is a contract and if you don't comply with contract you are going to come to Venice to trade right? so the thing is that this image of Venice was from a very strict rule of law place and very densely replaced was built on top of like a gunpowder because the social structure were again of some families that were struggling for power and these families you know what they had private armies and they hated each other so they had to find a way for this guy to coexist in peace in Venice and preserve the rule of law that was very important for all of them to prosper for trade you know this painting? from this one? Canaletto and what is this building? we went to Venice Palazzo Ducati and who lived there caused the dogs of Venice the most famous dog the dog was the president of the Venetian Republic and the interesting thing about the dog is most of the political positions most of the office were selected by lot in Venice but the dog selection process was very specific and we did very well in the audience so this is how it is described that damages are like like a blueprint for how you select the dog and it started in the Consiglio Grande this is one of the 1,000 people who participated here in Palazzo Ducati and then you had a lot of selection of all the 1,000 people 13 members selected there although this 30 guys went to a different room and draw again by lot only 9 of them then went out voting to nominate 40 members out of all the 1,000 members and they did at least 7 votes a piece so these 40 members they went to a different room and then they selected by lot 12 members and these 12 members so you think it ends here these 12 members selected 35 members by voting and you know what these 25 guys did? they selected 9 members out of the 25 and these 9 members then nominated 45 members by voting and these 45 members drew 11 of them by lot and these 11 were called the Undici and the Undici nominated by voting 41 members and these 41 members you know what they did? they selected the dog sophisticated random selection process so they had in the Republic of Venice to make sure that no one could tamper with the process right? it was like a shuffling machine for people right you know and the problem they had in Venice was that you don't want them to run for office you want the office to select your money that you have it right? and this made politics another topic of career for people because you couldn't do nothing to become the dog that you could run you could not try to create your alliances and the random selection also gave all the families the chance to rotate in the dog so this time it's you, this time it's you and so it was rotated and this created peace among them so you want to learn more about how the process works and the mathematics behind it there's this very good research paper it's a doge election but who knows and these guys and the thesis is that the doge election process okay sorry you were taking a picture they called that security theater okay when you were merchant and you are thinking of Venice as a good place to do business when you see that they have this very sophisticated process then you can trust that the rule of law will stay right? that they call a security theater even if the process didn't work very well it still gave the impression that these guys knew what they were doing and what it is if you want to learn more about the renaissance and the sortition I'm going to do this because of time but Mattia Bellini whom we all know Prince in his discourses on living he was a Republican everything he wrote of the Prince he actually did as a politician he was a Republican and he believed strongly in people having the right to the government and he used the use of randomness in selection so this takes us to the United States the partition was important part of the discussions in the time of the early days of the country so you know they were throwing the image rule and now they have to figure out how they create what works basically the first time ever at large scale democracy and we have these guys the federalists were basically intellectuals politicians writing on okay how should the US work how should the political system work should have a president, a king a congress, two chambers, one chamber all these questions they wrote in the federalist papers and there is one guy that is not a very well known it is well known but not as much as the others and it's called Thomas Paine and this guy among other things he created some ideas based on universal income are based on some writing from him and a very society and he suggested a method of selecting the president in order to avoid a federal government being captured by the elites of the powerful colonies so this is what he suggested and the colonies would choose delegates to send to congress and then when congress started they would select by blood one colony out of the 13 and then those 13 members from that colony would choose one member who would be president for one year right so this is what he suggested because he was thinking of how the Greeks did it the rotation plus from the selection would tend to break consideration of power and this is the last time that the sortition was really discussed at a serious level it didn't pass of course and then it declined why did the sortition decline in political thought history because of two reasons one because Locke started to be seen as an application of responsibility so he decided to look more like gambling than some legitimate way to select people for office and also because the classical idea of the republic working for the common good fell and what we have is liberal democracies where you have people selecting some representative and this guy governing for a couple of years and then the people selecting again giving them more time if they did well if they didn't do well so this ended in what we know as partisan politics and that's exactly what the ancient wanted to avoid when they used the sortition so the ancient Greek wouldn't recognize our systems as a republic these days there is a movement that is bringing back the ideas of sortition if you want to learn more these are called the clarotarians you know Helen Landemore a French one but these guys are writing about this and how this works today is you have what they call mini-publics you select a number of people from citizens and then these guys they are selected and they are consulted in policy making they are actually doing one of these in France now and this is for making policy for climate change and it's just a place like last week and the thing is that all these ideas of sortition and how to create these political processes that are more transparent and less prone to pollution aren't very likely to pass in existing democracies now but the thing is that we have the chance since we don't have to reform all legacy regimes we have the chance of when we overthrow these monarchies as the US did in the 18th century we can't build the institutions that we need to build for the coming time of the internet to be better governed so this is the blockchain so we are building these institutions and we can choose whatever we want because there are no limits now right and no legacy to reform so I think that sortition can be a really good way to do it and how we should think about this well when do we want the reasons to be part of the decisions when we see that the rest are very high chance of bad reasons becoming part of the decision then let's use sortition so we sanitize those bad reasons and when we want people to express their preferences as we don't see the possibility of bad reasons being important then yes we can use traditional voting systems and quadratic or whatever to be discussed so you want to learn more about this idea there are like two viruses of random selection one of them is the leg of the draw by Peter Stone a professor of the University of Dublin this is great and the other one is Oliver Dublin the political and this is a history of how random selection was used so how is history of politics we can learn a lot from this so thank you very much yeah I learned a ton from that talk John I think you are actually using your laptop I sent you slides I'll download them you might take a minute thank you all for coming so early on Friday but last day at DEVCOM this week was great so my name is John Light I work on governance for the Ergon Network and I'll be giving a presentation about that a little bit later but my talk this morning is about a brief history of blockchain governance so as the title implies this is by no means going to be comprehensive although blockchains have only been around for 10 years 10 odd years there's been a lot of really interesting events as it relates to governance in different blockchain networks and so I'm going to cover just a brief selection of some events that have stood out to me over the years but again this is by no means comprehensive we could probably talk about this all day so thank you Lane did you need a picture? yeah the first slide that's a good one so the first one I'll cover was CB 2010 in August 2010 so this was a bug in bitcoin that caused about 184 billion bitcoin to be created basically someone just exploited this bug to create some transactions that created all of these bitcoin out of thin air and people noticed this understandably alarmed and a patch was quickly developed and circulated in the community Satoshi kind of gave the patch his blessing and it was very quickly adopted a soft fork for those who aren't familiar with the term is a fork that restricts the rule set of the consensus rules in a blockchain and in such a way it's backwards compatible so basically transactions that are created under the new rule set are compatible with old nodes that have not adopted the new rule set yet but transactions that violate the new rule set that are created by old nodes are rejected by the new nodes so basically this fork said that any transactions that create these 184 billion bitcoin or create these overflow transactions are invalid and everything else is valid and they basically kind of like rolled back the blockchain to before these 184 bitcoin were created included all of the valid transactions and then continued on from there and this was kind of done in again kind of like a very community driven fashion so this is the patch that Satoshi kind of published and put out and the blockchain pretty quickly overtook the bad blockchain so then there was the one megabyte block size limit soft fork in 2010 so this one was done kind of quietly Satoshi didn't really announce it he just kind of like published this made this commit in the codebase and then pushed it out people adopted it it was bundled as a soft fork again along with some unrelated changes and then it was scheduled to activate at a specific block height so once the block height was reached in the blockchain the soft fork activated and from that point forward no blocks could be created that were bigger than one megabyte so this was kind of done basically by Satoshi almost in secret and this is the exact code where it adds this limit so you can see you can see if the network height is greater than 79,400 blocks then and the block size is bigger than the max block size a constant you get returned an error but yeah this was done kind of surreptitiously Satoshi did this a couple of times too he just kind of like slipped in consensus rule changes so that it wouldn't be too controversial and they wouldn't have to you know kind of debate it so then there was a proposal to increase the block size limit because people were like this isn't going to scale so of course once the block size limit was activated people could then kind of talk about it in public and so Jeff Gardner comes along and he said he introduces a patch that would enable Bitcoin to miners to produce as a type of produce valid blocks larger than one megabyte and the community kind of pushes back and says hey people don't like download the or install the patch they're going to become incompatible with the network because it would be as hard for it at that point it wouldn't be backwards compatible the network would split if basically everybody did not install it at the same time Satoshi kind of also pushes back on the idea Satoshi says this is going to make you incompatible with the network don't run it but here's how it could be done however it was never done the way that Satoshi suggested it and so the one megabyte limit persisted for quite some time and we'll talk about how that's affected Bitcoin so here's Jeff Gardner's post saying this is how we could do it and then Satoshi is like don't use this patch if you install it you will make yourself incompatible with the network to your own detriment we can phase in the change later of course if we get closer to needing it so then there was the Bit30 soft port again a soft port whereas before Satoshi would activate it by block height this one was activated by kind of a flag day so it was based on the time stamp that a block was created rather than the exact block height Bit30 I think this was basically a change that prevented two transactions from having the same transaction ID with the exception of two specific blocks that they didn't want to have to like revert a transaction that was confirmed in the past but after Bit30 was activated from that point forward two transactions couldn't have the same transaction ID and so this was activated on a specific day and it says right here like this rule applies to all blocks whose time stamp is after March 15, 2012 at 00UTC and that was published by Peter Wheeler so then there was the Bit50 hard fork in 2013 you'll notice most of this is Bitcoin because like Bitcoin was really the only coin of significance at this point sorry to altcoins so then there was a Bit50 hard fork in 2013 March 2013 so this was kind of the first like big bug when Bitcoin was really worth a lot in value like this was around the time that Bitcoin shot up to like $150 $200 something like that so it was happening in a lot of like in the middle of a lot of like price drama Bit50 at there was a bug called Berkeley DB bug that caused a chain split it was the longest chain split in Bitcoin's history so far that accidental chain split and then Bit50 was kind of activated on the Bitcoin network using a flat day hard fork which created a basically backwards incompatible change to fix this Berkeley DB bug and like here's the code to do that once again you have this like rule where like if the block time is greater than one whatever this time is 11th March 2013 then you implement this rule and then after May 15th we'll actually relax the rule and so when they relaxed the rule that was a hard fork because anything created after that point was incompatible with software that was running before that point so it's practically a hard fork although some people might debate whether that was actually a hard fork so then there was a really interesting event in a coin called Faircoin it's the first rollback that I know of that actually kind of like was almost like a a kind of like a justice related rollback so basically there was an exchange called MinPow that was hacked and 30% of the Faircoin supply was stolen now Faircoin was a proof of state coin so if 30% of your proof of state coin is held by somebody with perhaps less than good intentions well they decided we're going to rollback the blockchain and reverse the theft transactions which is very controversial in the Faircoin community and has anyone ever heard of Faircoin we can see how successful that has made the blockchain but in any case a lot of people did think that it was a good idea because again 30% of your coins and a proof of state blockchain being held by an attacker is maybe not the best thing so I thought that was really interesting it was the earliest kind of example that I could find of a kind of reversal of a theft rather than a kind of fixing of above the way the earlier buffer overflow rollback was done in Bitcoin so then moving back to Bitcoin Bit101 was added to BitcoinXT so BitcoinXT the main Bitcoin client in the Bitcoin network today and back then it was called BitcoinQT or BitcoinCore this was the derivative of or I should say descendent of the client that Satoshi Satoshi himself wrote BitcoinXT was developed by a different developing team so a former BitcoinCore developer named Mike Hearn had this client called BitcoinXT that he was doing some experimental stuff with and then there was a bit or a Bitcoin improvement proposal that was implemented to increase the block size to 8 megabytes now BitcoinCore did not adopt this bit it would have been a hard fork and they were against a hard fork to increase the block size at that point but there was some movement within the Bitcoin community to increase the block size limit and so BitcoinXT implemented this and was basically trying to convince people hey you know BitcoinCore is it responding to the community's demands for an increase in block size for BitcoinXT instead and so this was kind of one of the first examples of a Bitcoin client going rogue and going against consensus in the Bitcoin network so later in 2015 miners and businesses were signaling support for bigger blocks so mining pool operators signed a letter saying we all support up to 8 megabytes or even 20 megabyte blocks in June and then prominent businesses kind of did something similar a bunch of popular Bitcoin wallets and payment processors and such also signed the letter saying you know we support Bit101 or 8 megabyte block size limit this is what the minor letter looks like you can see F2Pool and Pool4B and there's probably a BW and kind of a date stamp there they represented a pretty significant share of the hash power of the network and then also blockchain, BitPay and a bunch of other businesses signed this industry letter also supporting bigger blocks nothing really became of this because the Bitcoin network has started to change so in 2015 August 2015 the dash dial goes live so I think this was perhaps the first example that basically took a portion of the block reward so the new coins that were being generated and then gave it to a specific kind of entity and said do what you will with this and that specific entity was a network of what are called master nodes which are nodes in the dash network that have locked 1000 dash tokens in their wallet once they've done this then they get voting rights to vote on how this portion of the block reward that has been split off gets spent fast forward to July of 2016 Ethereum splits after the Dow incident so basically here someone figured out that they could like unilaterally unilaterally withdraw money from the Dow which was a smart contract on Ethereum that they did not deposit and they use this to try to drain most of the money out of the Dow the Ethereum network reacted by basically creating a hard fork that was trying to reverse this change or like reverse this event and the network ended up splitting into what we call Ethereum today which was a new Ethereum chain where the funds were returned to the original owners and then Ethereum Classic which was at the time the original chain where the person who originally tried to withdraw money from the Dow actually got the money and on the Ethereum Classic chain that history is still preserved today so then in 2017 the block size debate in Bitcoin was kind of coming to a head there were kind of three main competing proposals at various points in time throughout the year one was called SegWit, UASF or User Activated Software one was a BCH or Bitcoin Cash the other was a SegWit 2X and these were all backed by various factions of the community so the story here is that SegWit which was a change in Bitcoin that fixed transaction malleability and also happened to increase the block weight limit to 4 megabytes was originally intended to be activated by a minor activated soft fork so basically the idea was that minors would signal up to like I think it was 70% support and once it got to 70% support the change would activate while SegWit failed to gather enough support from minors so Bitcoin users decided to activate SegWit themselves basically they would run software that says at a certain block height or whatever we will start rejecting any blocks that don't conform to SegWit sorry minors SegWit 2X was a group of folks that proposed to compromise but we will increase the block size limit like double the block size limit and implement SegWit everybody gets what they want well that didn't work out when it was looking like the user activated soft fork was actually gaining momentum and was going to get activated some minors that felt threatened by this decided to just create their own hard fork to increase the block size limit that we now call Bitcoin Cash or BCH and now with a new chain that had a bigger block size limit and SegWit activated through a user activated soft fork the wind was completely taken out of SegWit 2X's sales and the SegWit 2X fork was eventually abandoned by the people who originally proposed it like two more slides so Decred then in June of 2017 activates its first upgrade using an on-chain governance mechanism so this was a way that basically what are called like Ticket Hold blockchain could vote on a change to the consensus rules and then just by voting the consensus rules would actually be changed automatically people didn't have to upgrade their clients which is recognized oh this is the way that Ticket Holders voted these are the changes I'm automatically going to kind of accept these upgrades so this is kind of the first example of successful like on-chain governance in action or like on-chain tightly coupled governance over the consensus rules specifically and then most recently fast forward to June 2018 EOS was launched they also have this idea of like on-chain governance over consensus rules they launched with like a constitution that was intended to bind how these rules would change and of course block producers who are like the equivalent of minors immediately violated the constitution they decided to unilaterally freeze some accounts that had been identified as potential scammers the violation of the EOS constitution which it actually intended for that kind of action to be taken by a separate arbitration body so they kind of like routed around the constitution to extra legally impose justice in this case once again a fast forwarding to April 2019 early this year EOS illegally scraps their constitution for a new user agreement the reason why it's characterized as illegal is because they adopted this new user agreement with only 1.74% turnout despite the fact that their constitution actually required them to have a 15% turnout to make this kind of change so so far block-chain governance is not going so well for EOS and this is the end of my presentation because we're in 2019 now and there hasn't been anything like incredibly new or novel that's happened since then so with that I will end the presentation thank you all which happened quite recently and I think we'll hear more from Tezos that was this year the reason why Tezos wasn't included is because it's while certainly a landmark event in itself it wasn't the first time that that had ever happened we have Alfonso up next speaking about Alfonso also speaking about Polkadot governance so this is the first of our lightning talks and from this point forward we're going to hear a series of talks to learn and do a little bit of a deep dive on much different protocols so we'll go ahead and get set up hi everybody so I have the ingredients to present a little bit about the governance in Polkadot where it's actually just a values so I'm a peer scientist I'm a mathematician I've been working things related to demising algorithms and also game theory and I'm also working so here's the overview the governance in Polkadot is based in stakeholder voting or current voting so it's not perfect but it's the best we have I would say some of the problems that people like to point out with coin voting is that they usually learn to not it's slow governance and it tends to be a protocol that's to give power to people with a lot of coins so I'm going to present ways in which we combat these problems so stakeholder voting we're going to use time locks which means that it matters not only how much how many coins you have but also how long you're willing to lock these coins because that shows that you are very much committed to the system we have several governance bodies that has some advantages the first one is that it increases speed because there are more channels to make it and the second is that there are more checks and balances because they can check each other and we try to give power to minorities and we obtain that in two ways while it's a proportional representation making sure that minorities are as well represented as we are and the second is that we delay I would say more about that later so who are these stakeholders I hope you are aware with Polkadot and I don't have time to go into it but there are validators they are the ones that run the consensus protocol there are nominators that are coin holders who support the election validators and they get rewarded to benefit the validators and there are parachutes parachutes also have to hold dots because they need act to secure a parachutes slot and slots are going to be given out at least some of them are going to be given out through just action and there are also of course the restores there are holders and they are all coin holders and I would like to point out that the first ones are locked and the second ones are liquid and we want to give a bit more voice to the ones that are locked because as I said people with locks they have more skin in the game and we can trust that they are going to care more about the future of the network ok, so decision making governance is going to be based in referendum they are going to be there is only one referendum once a month and the timeline there is a public proposal anybody is free to propose there is a public proposal so any coin holder can sponsor a proposal that they like and then once a month we are going to take the most popular proposal and raise it to referendum there is going to be voting here in one month and there is community and documentary that is the time between deciding something deciding a proposal and actually implementing and this is important because power to people who didn't agree with it to actually leave the system so that they know that something is going to take place in two weeks so they decide to sell their locks in their own like so this is some protection for this gives some protection to dot holders there is a voluntary time to open so everybody who votes and wins has to lock the coins in which they voted for at least two weeks but if they decide to lock for a longer period then they get a higher vote so I am running out of time I am going to be faster there is a this body called council it is going to be elected by stakeholders their function is to represent the interests of the stakeholders and something I want to point out is that the way we elect the council members is through something called proportional representation which basically means that if there is 10% of coin holders that have similar interests then they are guaranteed to be represented by at least 10% of the seats in the council or the same with 20% or the same with 5% and this is very different to some other kind of representation systems where only maybe the two biggest majority are represented so we really want to stress because also this gives checks and balances because the council has some power and we want to make sure that it represents everybody and if there is no an image in the council then they basically have no power and that's a good thing it's very hard for the council to have supermajority of the silence on the things if there is some nice co-graph that represents this proportional representation you can imagine that every color represents some minority that has some different point of view and each one of them is going to get one representative in the council the gentleman over there I'm not checking your time I'm sorry so I would just say what the council can't stop can't mitigate a controversial adventurous and malicious proposal we can't fast track important proposals but as he said most of the time it doesn't have a lot of power thank you so just brief instructions so a lot of people know obviously a lot of people are here so it's pretty long basically what it is but basically a very powerful process of applications has like, you know, it uses a group of steak not depots but it has multi-stage on-chain governance mechanism and one thing I want to point out at the beginning is that it's also an architectural thing to its governance it's actually a metaphor to call it can technically instantiate any blockchain based protocols theoretically but basically it tells us the underlying way that blocks can create it and form consensus basically through this notion of baking you have basically and that's actually a graph of bakers joining Tezos the first year obviously there's a big increase in the first few months and then sort of a study going off it has delegation so people can move their steak to and assign their steak to these various bakers around the world but it's not depots in the notion of the way of US or China or anything like that it's all obviously funded by the amount of inflation that we appreciate so far and so then the Tezos Bakery practice is actually almost 500 validators in about 35 countries at least that's what we know publicly not all validators state where they are what they're doing and so you have to sort of the participants in this governance system are basically across many different institutional folks joining really complex cloud-based setups there's folks just with ledger and treasurer you know in their entouragement and there's about 110 of these public delegates you know of these four, you know, 450 to 500 of the various bakers in town of them are public and vote on other people's behalf and then you have about the percentage of the network that's involved in this process about 71% of the digital supply I think it's about 80% of the activities so then what's the actual process you know very complex but it's actually much simpler than that it's basically comprises four periods and some of this is changing potentially in a week if we activate a new upgrade but basically it involves it starts with this proposal period each of these four periods basically runs for about 23 days I think basically I was like yeah and the proposal period is this period where basically you cannot put as many proposals as you'd like I think Adrian will go into some more detail on the next talk but basically when you have this four phase period it's a very conservative, unlike the sort of the polka dot model it's a very very conservative sort of a governance process by which you basically have to hit this quorum you know a very very high bar of consensus it started about 80% and some of the original Athens votes it was as high as I think 82% of the quorum I would reach out I had about some percentage of the stake it was about 86 or 87% of the stake actually voted or participated in the vote we basically have this proposal period you can have a yes-no maybe vote yes-no or pass vote really in the second period a testing period which you're not going to get into either a promotion vote period which is another yes-no or maybe or should I say confirming that a protocol in our proposal should make it to become the new protocol and so I decided it was about three months to complete so going back to you know what you mentioned earlier so what was Athens Athens was our first member proposal from the end of February to the end of May it was basically just voting on changes to constants so it was just changing the roll size which was just the unit of account for proof of stake in Tessos that I mentioned earlier and it was doubling you know basically a whole call of doubling the gasoline votes a little bit more than that but basically the participation I mentioned earlier 87% of the stake in the expiration vote about 84% of the stake in the promotion vote and it's so high is because bakers are you know mediating the votes for folks there's not is or why every single person holding a token to participate in this process it's mostly symbolic so there's just a little you know we have this notion of inflationary invoices won't go into that right now but it's mostly symbolic just about 100xdZ reward and it was really as I said just changing some constants and one only one team was really involved in the creation of the protocol amendment and but at the end of the day obviously pretty landmark about this is that it involved coordination et cetera 35 countries and about 200 bakers participated and so then the newer upgrades are the new upgrade that we're going through now it's you know what I mean just like changing constants to actually like the notion of constant upgrades actually hopefully every 3 months maybe it's every 4 months, every 5, 6 months you have some regular cadence of varying the chain based on feedback and new innovations that have been refined in you know engineering basically engineers that various core dev teams have been proposed to stakeholders also this is the vanity hash of the original proposal and now we obviously had an issue in this one so we're actually been voting on a slightly different one it's another topic for another time I won't go into all these details what's in there but basically there are a lot of challenges to this process so one of them is that it's very decentralized there's actually nearly 500 different entities that are all around the world participating in this and there's also sort of some minor areas of confinement that we've found we've been doing this in the wild so for example what happens if you find a bug halfway through the process we've been going through that recently and the most recent case we found it before the final confirmation vote but in other cases you could imagine this being having been a consensus threatening situation and so then also there's other questions of coordination so you have all these different forms in which people are discussing amendments and you have across 35 countries so a lot of language barriers cultural differences including pre-existing experience with voting, pre-existing experience with democratic norms all these different things and then also we've had adversarial proposals the city of Breaston in France and this is Baron Harconin from Dune so you sort of have this notion of people sort of trying to get taxed so it's potentially gainable but your proposal if you want to fork off you can just get a proposal not approved in the Tesla's process and then you can basically that gives you social energy to fork off for going and serving this kind of dynamic for a while and then other questions obviously about Tesla and other mechanisms that formally seek is what about the rest the answer here is really I think a little different than that the question of it's actually tangibly how do we ensure that we have flexibility against sort of hard on-chain roles so you look at things like Bitcoin where originally mining was actually widely distributed over time there became economies of scale it's not quite analogous but the point is you don't actually know who's going to come to if you have hard coded rules you don't have to say who's going to come to control the system long into the future so you need other mechanisms to ensure that you have flexibility and so a few of these that have been discussed are things sort of inspired by it's sort of now inspired by Daustack it is like you can do prediction games on proposals to rank proposals you can use this to mitigate governance by state potentially you can also do this if you want this to be this economic infrastructure that can upgrade itself you can basically set higher approval thresholds for serving parameters or other characteristics of proposals that you don't want to be able to change your flow potential and then the other question is basically our small stakeholders and smart contractors just curl in this system so if you have whales who's looking out for other folks whose voices you don't have as much insight into and so just to wrap up that's one of the things we've been working on for a while as well which is basically how do we get easy for many many stakeholders to understand what's going on in the process and also to express their preferences so we have initially focused on the governance explorer, the integrated forum but later the idea is to build a full suite of signaling tools as well as from the visual and prediction markets and then also things like Daustack for governance as well as for commas governance things like domain names and registries and other kinds of things like that and then lastly things like IDM to sort of more sensor resistance so yeah if you have a question I think it will be there after we said after the talks cool, awesome, thank you yes, so I have two slides because I'm mostly going to whiteboard this so I'm Adrian and I'm one of the penises in Criptium Labs so very briefly Criptium Labs is a protocol engineering team that also happens to run validators due to some historic occupants because a lot of insights into how these governance systems work because we instead of having to run every single approval state chain we only run the things that are technically interesting and we're very heavily involved in the technical governance of these systems so I'm mostly approaching this from engineering perspective of like my approach is the engineering perspective of how these protocols work over time since Jacob said that I should briefly talk about the technical specifics I'll tell this governance process I'll do that as well so this pen works and is for whiteboards cool, all right let's start with a brief overview of what blockchains are right, we have some networking stack normally I think my question may be off here then we have a slide here, then we have some consensus stack afterwards we have some economic state machine state machine, this is quite important because this is completely agnostic, so when you think about what nodes are networking networking, RPC calls everything that is local to your node it's totally up to you that way you have many third-client limitations that blockchains generally have many clients the only two things that they actually need to agree upon is consensus and the general protocol so for example in the example of cosmos this would be 10 million consensus and this would be the cosmos hub or guia depending on who succeeds and naming it correctly we'll see there's currently some contention about this there may be a new governance proposal about this as well but specifically the economic state machine so it says what is about transaction, how do you get to move funds around how does RPC stake work okay, cool, now that we know this and when we talk about governance in many cases we talk much more about this and this, so not many systems have modified their governance their consensus systems over time those are mostly stable but I think we should mostly experiment also with improving them over time regarding Tesla's governance this is quite interesting because it affects a lot of this part as well and so what happens is that there's a Tesla shell which you can imagine that it's a local thing that everyone can have their own version of and it as input takes some protocol where the protocol is a combination of consensus and economic protocol and this protocol is written in a kernel and what that happens is that the node will, the shell and oh, all of this makes a single node what happens is that the shell takes a bunch of kernel source files and then runs a restricted subset of the kernel compiler to turn the protocol into something that can be dynamically linked into the running shell and then executes it and then the Tesla shell receives network messages from other peers, for example new blocks new transactions and drives the single threaded state machine which is the protocol the cool thing here is that due to this you can build a binding engine governance system where someone or anyone in the world really can inject a bunch of kernel source files everyone then votes on the hash of those source files and if everyone in the network enough people whatever your acceptance criteria is and Jacob talked about this earlier agree on the source files the node at a predefined block again given in the upgrade will decide so the shell will return the protocol will return at the next block we should be using a new protocol and at that point the shell will use the sources that it currently has compile them on the fly and then dynamically link them into the running binary and that way the protocol is incorporated the protocol is doing effectively the same thing with WASM where people vote on a bunch of WASM code and then no dynamically switch to the new protocol but this is how you get binding engine governance because in order to opt out of this change people have to modify the shell and then they can say well the shell shouldn't upgrade to this new protocol but the thing is in most cases you will also have to slash a significant percentage of your token holders because what can happen is let's say you don't like version A like protocol A which is the new protocol and so you locally go and modify this and you now have your own little fork maybe your own 10% of network is running this but you didn't modify the balances on the new protocol A anyone can re-inject what the community wants to upgrade to and then they do exactly the same vote and by default all nodes again will upgrade to the new protocol so you effectively either have to keep half forking to prevent new protocols from being injected or you slash 80% of token balances of the people that actually want the new change this is why all of this is cool and there are many different ways to implement this Tesla uses a restricted subset of your kernel compiler Ocaro will use Wasm as long as the important thing it has to be deterministic and properly sandboxed because otherwise if this isn't properly sandboxed you're building a massive botnet where anyone can upload code to and they will then automatically execute everyone's machine and if we mistake this it's specifically scary because all of these machines hold hotkeys that are using consensus signing and can incur significant slashing losses if they ever compromise so yeah we have to make really sure that we have properly sandboxed a kernel compiler and properly sandboxed Wasm interpreter for all of these protocols cool without this is a quick intro to Tesla's hard to look technical perspective one minute all right then really quickly we call it this Cosmos Governments in one minute Cosmos Governments are hot new whiteboard series never should do that so there's a 40% quorum you can take two minutes 40% quorum and strict majority so 50% 33% video power as a quick explanation so Cosmos the way it works it's a signaling process where anyone in the world can upload a text proposal they have to pay 512 atoms as a deposit this prevents spam but once they've uploaded it anyone can all the validators can vote on it importantly delegators can overwrite their validators choice it means that if you want to delegate your voting power to a validator you can but if you're validator then there's something batched and insane you can overwrite your statistic in order for a quorum to be valid it needs to have 40% and of those 40% at least 50% at least half need to out yes it's not half out yes it needs to be rejected and importantly if something is rejected you do not get your deposit back it's $50 if that happens and also due to the way BFT works if more than a third ever vote no vote no with veto and everything it won't get accepted either because in either case if one third of the network doesn't like something they can always halt the chain so this is why there's a no with veto fresh out here as well right then the cool thing about having signaling proposals is that you can have many at the same time so doing these kind of governance upgrade binding onto governance upgrades where you vote in code you can't have many at the same time because it is incredibly difficult to figure out how you merge two distinct code bases into each other to form a coherent protocol that still maintains properties like the desired properties of either so with signaling though you can have many running at the same time the deposit phase runs for two weeks and afterwards the voting phase runs for another two weeks the downside is that it's very unclear what should happen after a vote has been successful quite because we don't actually know the specifics so due to this the way you upgrade the code base of the Cosmos Hub is that you first have a bunch of votes on we want these new features and then hopefully someone builds those new features and afterwards you have a secondary vote where you vote on the specific hash of the code that you want to run and if you're following Cosmos governance at all this was the very I don't want to say funny I spent like six hours sitting in our data center anyway there was a vote for the Cosmos Hub 3 which turns out wasn't properly tested the migration procedure wasn't properly tested and due to that the new Cosmos Hub 3 didn't actually start and so we had to roll back to Cosmos Hub 2 this cost down to two hours and massive months confusion I was very sad because I said let's figure out the data center cage for six hours while trying to figure out what to do there so in my opinion honestly some of the best governance models will combine both so you will have some sort of signal proposal because it allows everyone in the community to have a voice to just change us to get ideas of high level goals that you want to achieve like do we want to upgrade to wasn't do we want to have a new consensus mechanism is how do we want to reward people and once you have sort of broad consent or a broad agreement over where you want to go then someone builds it and once they've build it you vote on this very specific implementation so that everyone agrees and everyone upgrades to it automatically yes okay cool I think I covered most of the points and I'm only thinking about five minutes over time so thank you very much I'll be around later if you have any questions I can talk to you I can talk to you about either lockup governance systems we happen to spend a lot of time on all of them so if you have questions please send me up thank you very much and so for those of you who are unfamiliar this here is Stakey in the Decorating Ecosystem Stakey represents a ticket which is a float on the network and it's kind of like our community affiliated in MassMod now before I begin the talk I just want to extend a special thank you to the Ethereum community for extending an invite to us Decret doesn't have any explicit link to Ethereum so something that I do find particularly admirable about the Ethereum community is not only its positivity but also its willingness to welcome ideas that exist outside of its immediate ecosystem as I know that the cryptocurrency space can often be quite tribalistic at times so for the talk today I'm going to focus Decret's model has quite a bit of complexity to it so I'm going to focus less on how it works and more so why we implemented it that way ok so for a bit of historical context beforehand for the developers in the room you might be familiar with some of our developers previous work so prior to Decret many of our developers spent over 15 years hacking away on the OpenBSD project they then moved into very early bitcoin development releasing BTCSuite BTCD which at the time was the first alternate full node implementation of bitcoin then they moved on to release cross chain atomic swaps so that was released and made open source by Dave who is one of our lead developers and tends to be one of our more prominent open source contributions so the reason why I bring this up is that for Decret governance wasn't really an afterthought it was something that was carefully planned because the team had spent quite a bit of time prior dealing with open source communities and having to deal with the various conflicting stakeholder groups within the bitcoin ecosystem and so for that reason it's lost my whole chain of thought but we'll just go to the beginning so Decret was originally conceptualized around 2014 so that was roughly around when the team started to transition away from their bitcoin development then we later underwent about roughly two years of development we later released it quarter one of 2016 then in quarter two of 2017 we had our first on chain voting process so John alluded to that in his earlier talk really the first time we actually got to demonstrate the core of Decret's governance process which is our ability to implement user activated hard forks now to my knowledge this was actually the first time a user activated hard fork had ever been observed in the cryptocurrency ecosystem and then since then we've started to focus a little bit more on off chain governance which is specifically relates to the coordination of the various different contributors and the projects resources so whilst we were fairly early in recognizing governance as a problem I want to make it clear that we're not so arrogant to assume that we're some kind of governance panacea I think that just like blockchain itself there are many correct ways to implement governance and depending on the particular problem that you're looking to solve the effectiveness of a particular governance solution is going to depend very heavily on that so in the context of Decret we focus specifically on on chain coinholder led governance specifically as it relates to the consensus change process and so I'll just briefly give you an overview of how that works basically we utilize a hybrid model that blends both proof of state with proof of work so probably the easiest way to visualize this is in the form of something like an assembly line so what happens is whenever a miner produces or is producing a log they have to select a number of tickets or votes at a random of the stakeholders into their block now in order for that block to be considered valid in order for them to actually successfully post it and receive the block reward a majority of the votes they include within that block need to be in favor of whatever particular block type of their signal and those votes are selected at random so the miner can't go and just select every two of the votes exclusively in favor of whatever is their signal now so the whole process is on-chain and binding so once in a approval threshold of at least 75% of the votes within a particular voting window have been accumulated then whatever change gets automatically implemented so an example of this might be pro or anti segregated witness okay so the thing about coinholder led governance is as Layne alluded to it doesn't necessarily take into account all the possible types of stakeholders that might exist that's definitely very true but what it does do is that it does cost a vastly wider net to proof of work miners alone and so that was really the point behind Decret is that we didn't want to eliminate miners because we felt that they play an incredibly important role and we're still huge advocates for proof of work as a consensus model we just felt that there need to be a greater level of democratization in the governance process so in the context of coinholder led governance of course the distribution of the currency itself is incredibly important and that's why earlier I stressed that for us governance wasn't an afterthought it was something that was very carefully planned and the hybrid model and the distribution of the block reward was carefully planned such that we would have as wide and as fair as possible of a initial coin distribution and it's also for this reason that I kind of urge the Ethereum community not to give up on proof of work just yet because it's an incredibly effective tool for implementing fair and wide coin distributions so certainly if you're considering a coinholder led governance model it's not something that I would recommend moving away from just yet but something else that's oftentimes overlooked with proof of work is that it extends a lifeline for those who might be politically persecuted so I suspect that for many of us we're probably pretty fortunate to have the liberty to pursue an interesting group of currencies but for a lot of people around the world they don't quite have that luxury and for many of them the idea of having to submit your personal identity in order to participate in an ICO or to access an exchange platform can oftentimes be putting there either their liberty or even their lives at risk and so something that oftentimes gets overlooked about proof of work mining is it actually provides a discreet mechanism for individuals to obtain cryptocurrency using hardware as a proxy and so in the context of a coinholder led governance system where you really want to have a highly democratized access to the currency for the governance process again proof of work is an excellent mechanism to use so just a few closing thoughts before we end for those of you that are considering implementing coinholder led governance systems a few suggestions that I would have is implementing a coinholder led governance system becomes exponentially more difficult after you've released your protocol so if you can try to really really carefully plan out your initial coinage distribution and your governance process before you launch the second one is that to make sure that you consider your projects ongoing resourcing so I think a lot of projects that I see particularly who ones that are raising outrageous amounts of money in the form of an ICO is that I think whilst that does provide some instant gratification something that a lot of them probably are considering is that those resources can actually go quite quickly and then you can find yourself in the precarious situation of having financial pressures being applied from external special interest groups which can get in the way of your governance process so for us we had 10% of that of our block award goes into a decentralized treasury fund and that's where the off-chain component of our governance comes into play we released a governance platform we call Politeia and so Politeia is the Greek word for a system or ancient Greek term for a system of government and Politeia is just an interface where stakeholders can vote on where the various different funds get allocated to and funds our ongoing operations and contribution efforts so if you want more where can you go to find out a little bit more information definitely the documentation is your best resource we don't actually have a white paper so this is the closest thing to it the next best resource would be our forum and our reddit and then of course our bridge com so the bridge communications channel is kind of cool we actually have our slack discord and matrix chats all bridge together so message on one will show up on all the others and this is a really cool place to hang out because this is where you see all of our governance activities being and operations being coordinated in real time it's very transparent it's very cool and finally go and state it just something that I really would have urged before we end is don't move fast and break things we should sort of transition away from that Silicon Valley mentality shift quality code and consider its implications because implementing major consensus changes in this space is not easy it's not it's not quick and no matter how good we make our governance systems I don't think it's ever going to be quick so really really carefully consider code before you ship it and don't compromise don't compromise to fit gimmicky outrageous transaction throughput and silly features don't forget what we're here for it's censorship resistance and decentralization and thank you if anyone has any questions or wants to know more about specifically how Decret Decret's governance model works I'd be more than happy to take questions afterwards and comment on it anyways thank you very much that was super interesting you guys are some really exciting stuff so thanks for sharing and I really like your closing thought that we can't rush governance we need to have patience and be very thoughtful and move forward kind of incrementally I couldn't agree worse wrongly with that alright up next we have the one and only Litecoin again John and I talking to us about Aragon governance I'm sure probably everyone in this room is already familiar with Aragon probably needs no introduction but they're a project that has been really blazing a trail and just doing some of the most interesting and exciting work in governance and maintaining space with DAOs and many other things I'm going to pull off your slides but go ahead and feel free to kick off great thanks a lot so for folks that were at Web3 Summit earlier this year or perhaps saw the video that came out the presentation is going to be kind of a repeat of the recap of Aragon network folks that have happened so far I'm going to kind of briefly cover how Aragon network governance works and then just kind of recap how it has been working so far for a more complete review of how Aragon network governance works I did a presentation about this at our annual conference Aragon earlier this year and you can find that video on YouTube but so yeah let's jump into it Aragon network votes recap so some background Aragon network is the DAO that oversees the Aragon project so this is a decentralized organization governed by holders of the Aragon network token Aragon network is funded by proceeds of the sale of that token back in 2017 there was a sale and that's how ANT was initially distributed so but before Aragon network token holders could govern the network we needed the tools to govern the network and so what we've been working on building is tools for creating and governing decentralized organizations on Ethereum and we have one of those tools live today at app.aragon.org so once we had the tools live on main nets that Aragon network token holders could use to participate in governance we decided to formalize our governance process and so first we wrote kind of the foundational document of Aragon network governance called the Aragon manifesto this was ratified by ANT holders in June of 2018 and then later that year we wrote up how the governance process could I'm serious I'm enjoying your work but thank you for clarifying Serious and so later that year we formalized how the governance process would actually work under AGP-1 which is the Aragon governance proposal process which was ratified in November of 2017 so this is roughly how the proposal process itself works there are six stages now there are the first three stages are just basically information gathering and proposal drafting then stage four this is slightly out of date stage four is a community review period that all proposals have to go through after the community review period the Aragon association which is the non-profit that kind of oversees the project right now review the proposals and then any proposals that get past the review go on to a final ballot and get voted on by ANT holders in the network vote and any proposals that are approved get executed and any that get rejected go back to square one so that's a high level overview of how the process works today so far we've had three votes for the Aragon governance proposal process the fourth is coming up at the end of this month actually 33 proposals have been submitted on time so like before the deadlines 26 have actually made it on to the final ballot and then 21 proposals finally were actually approved by ANT holders and so far about 8.5 million die have been distributed through this proposal process and about 3.5 million ANT participation rate on average across all the votes is about 4.75% of all ANT tokens in existence participating in the vote and our highest turnout rate was 7.84% the top three most contentious proposals which I define as like the closest ratio between yes and no votes was a funding proposal for a team called ATARC so this was a team that was seeking a relatively large grant to join Aragon 1 which is the team that I worked for as a full time team working on Aragon project software their proposal was relatively close in yes and no votes with 54% yes and 46% no Agp16 which was a procedural vote about modifying the governance process itself also relatively close it ended up getting rejected and then AGP37 which was a retainer for an ongoing Aragon network security partner this would be like auditors that are auditing Aragon software at an ongoing basis and then maybe like a honorable mention would be the proposals to reject or prevent the Aragon project from building on any blockchain network that wasn't Ethereum that one also had I believe it was like 69% no and 31% yes so some lessons that we've learned so far it's it's good to have a direct communication channel with voters to cut through the noise so we set up a mailing list called the Aragon Network Vote Alerts mailing list so that voters can sign up to get like very low volume very high signal alerts about when important events are happening in the governance process in particular when to actually vote because the Aragon project is doing a lot of stuff if you're following all of our channels you're getting a lot of noise and you might not see important announcements about the vote another thing is to put authoritative information about a vote in one place and then repeatedly tell people where that one place is located so they can find it again it's just good to have an authoritative source of truth that people know they can always go to this spot and find the most authoritative information about a vote rather than looking on all these scattered channels and maybe finding out-of-date information another thing is to keep the process as simple as possible to lower the barrier to participation and reduce the likelihood of mistakes I'm sure we could spend time designing some elaborate like group Goldberg governance system that looks sophisticated but doesn't actually accomplish a lot people just have such a hard time navigating it that only either extremely dedicated or manipulated people would actually get through the process and finally ask participants for feedback and use it to make the process better after all of those that we've done so far we've sent out a survey asking people several questions about their experience whether they voted or not if you want to hear from them and as well as suggestions about what it would improve about either the governance process itself or the user experience of interacting with the governance process we've gotten a lot of great feedback from those surveys and have actually incorporated that feedback into our product development roadmap as well as our governance roadmap so this has been really helpful for us and then some open questions that we have that I would love to discuss if folks want to chat after this would be like how do we increase participation rate while maintaining or improving decision quality one thing we could do is just pay voters to show up that will probably get more people to show up but will that actually result in better decisions how do we protect against various attacks like vote selling bribing or collusion bad voting how do we run meaningful experiments while protecting the integrity of the Aragon network as the last speaker mentioned we don't want to move fast and break things we don't want to break our governance process but we don't want to stand still we want to continue iterating so how do we run meaningful experiments without breaking the governance process and then finally how do we continually improve vote communications and user experience the feedback answers have been helpful for getting some direct feedback from the voters but how can we make sure that we're continually doing this so that we're always improving so that's what I have today about the Aragon Network Governance Process like I said we have a vote later this month so if you want to see it in action just kind of follow our channels jump into our forum and you can see how the governance process is working in real time this month thank you again John you have a very valid point which is that a lot of the governance ideas you're hearing up here today are sort of still on the design phases but I think what's really exciting about Aragon as well as D-Credit and some of the other folks of course is that you guys have been live and doing this for over a year now so jump in and participate and observe governance and action it's been really exciting to see actually I'll be a co-founder of NIR protocol yeah NIR is another really exciting project you guys are very close to launching I mean I think right you guys have a very different vision for you know future block change and charting and all that fun stuff but also governance so yeah take it away about NIR so I'll give a quick spill but that's kind of not the main part of the talk so NIR is kind of the way we started we wanted to build applications on blockchain and when we started looking at kind of current space of block chains yes there's like scalability issues and there's a lot of people working on that actually one of the main concern that we had was that it's pretty much like really hard to build applications on current system and coming from that too building applications with like fireplace parts with like different clouds there's so much tooling there's so much libraries and make a system kind of developed here as soon as you approach this it becomes extremely hard and it takes weeks or months to actually report and then on top of this you can actually target real users in a pretty small community so figuring out how to get people to use this is important now the thing is like obviously you cannot figure this out from the beginning, from the start like this is an evolutionary process and to do that you need to process how to evolve your protocol how to evolve your whole ecosystem to have to governance now like from our perspective defining governance so we haven't like when asked about governance but the question is how do you define governance like what is like hey our governance is blocked like from our perspective it's actually about defining the mission like what are we trying to achieve and then the culture of how to make decisions towards the adult instead of like oh it's unchained governance we're going to be voting and stuff it's like hey we want every single person in the world have a wallet we want every single developer to be building applications that is a mission and then the culture is how do you actually get there what is the what are the rules you follow so on the other side you still need tools like we are doing pretty much software space and we all love building tools and the thing is like tools can be very different we have Swiss donations which is just a tool how to organize people how to actually make decisions and be compliant and actually follow you know still regulations of physical world there's often tools like Twitter, VATCon it's great actually to communicate and make some decisions and there's like old chain voting and different new proposals kind of content like that but those are old tools to actually facilitate the process and I mean over time incentivize it right if like let's say it's loading right when there's some process of rewarding time so kind of one thing that's usually overlooked is that different decisions actually warrant different methods right like if you think of it the best decision process kind of is not going right it's actually consensus when everybody around the table agrees on the same thing they kind of come into this different view and then through a conversation and through a kind of figure out what is the difference in views they should figure out what is the best option to kind of align everybody together and obviously this doesn't scale right you cannot put you know hundreds of thousands of people in one room and actually make a decision even if it's 100 people that's already pretty unskillable and it actually works way better in person because like we have you know millions of years figuring out how to communicate to each other in person like chats and Twitter etc I think it's true we'll just say trawling over time now the other option is this kind of meritocracy where you elect somebody who presumably is fast in doing this function right and this is for example like for technical decisions right you can elect you know Vitalik who knows a lot you know how they're thinking about this a lot and can make these decisions and this is great and like this is made sense for a lot of tasks but in kind of more global consensus right this can be complete to a stalemate where like one person cannot have like kind of kind of capture all possible roles in this ecosystem and then the third option right is voting is kind of what we used to which is great to get decision to make to go to decision from but it's currently divided right like it's IDPS right now like there's no way to kind of have a nuanced conversation figure out maybe there's a better proposal and also simple very often the proposal itself is written by a single person or entity and kind of currently has some biases of that entity themselves right given that it usually has some kind of vested interest so the idea is to kind of combine that right and figure out how to kind of follow some principles and get to you know the mission and the goal we have right so like just to be clear there's no solution like later you got to talk a little bit about the tools we want to use for this but the principles we want to follow is to make sure that there's clear goals like clear mission and clear kind of OKRs measurable metrics that you can target as a community right so everybody clear like hey we want to get to more users we want to get to more developers and be able to measure it that's important and coming to developers right now actually being able to see an ecosystem to me in a case that's a little bit unclear you want to make decisions transparent right so right now it's actually really hard to make technical transparent technical decisions because it's a really nuanced conversation it's way better done it was in the room right it was like more technical discussing like making it actually transparent recording this putting it in public showing how the decisions are made not just like hey this is obviously safe for community safe for brands you want to make the strategic decisions right kind of by bringing people who are representing different roles and like roles we'll talk about this again later but in person right and get to a consensus like for strategic decisions you want this consensus where like everybody agrees that this is the best option right and they bring whatever the concerns they have up front and you can invest them and for tactical decisions right like let's say hey you know like literally like we need to fix some by right this is not something that like you need like huge community support right this is something that you can have kind of elected people who are held accountable but at the same time have can have really quick decision making process and to kind of finish up is that like all the tools that you're building like Aragon has done a great job right like those deep red but they all kind of experiments right we don't really know what works and what doesn't and what we need to do is we need to test them first we want to give them kind of a little bit of state and see how these people will organize and how like what decisions they'll make is it moving towards the goal we set up and then if it does that we can actually have like a clear pass how the state goes in the system so you can imagine like let's say a doubt where you put in some amount of money to say hey you can manage this now as before making decisions you want towards the progress of the goals you can set up and then over time as this doubt makes more and more decisions to improve the metrics you actually can fund it more so this is all like to be done programmable and kind of this is the idea and to just finish up like one of the things we've been doing is actually trying to bring some of the knowledge in this space while being transparent is through a whiteboard series where we pretty much interviewed a lot of protocols and a lot of projects I think it's certainly right now in a whiteboard series and we want to continue doing this as well as we're publishing more content around different spaces like government etc so check it out and if you have some ideas what other content will be interesting I'm just going to let you know Federico is up again to tell us more about Clara's governance Hello, how are you guys again? It's curious how many projects I use like metaphors about the great rights it's not something clear about what you're telling us Yeah, we're we are looking for answers somewhere Good, so the difficult part starts with the problem of incomplete contracts in blockchain, you know Len talked about white, white and dry code and there are some things that the computers cannot read some of the contracts are incomplete you don't need humans to do anything because everything is pre-coded but other contracts and many contracts in particular you need some human input for making them enforcement this becomes an important thing if you think that 3-5% of transactions are online in that in a dispute you would need some type of arbitration and that's don't have a personal assistance for this type of problems so let's imagine Alice and Bob and they have a contract for a website she sent the payment in a scroll account and they both agree that in case there is a dispute she won't go to Clara's Clara selects a jury of experts and they are going to see the evidence and the contract and they vote on who is right and then this says to the scroll contract you pay Alice because you were right understood? for now? so how do you select jurors? that's the question and then we tell you a bit about how it used to work of course in Greece they had this concept of popular trials every citizen would go to be a juror in a trial but of course there was a very specific selection mechanism so you wouldn't go to the trials with your friends to help them so there was a random selection of jurors so first you work to the court and they select you to be a juror or not and then after you selected they decided again by blood to see in which court there is a download selection do you know what they used? what machine they used for the selection? are you paying attention? Clara do not tell you very good so this woman wants to be a jury she's a designer so she needs to violate these tokens and deposit this token what is this token? do you remember this one? selection I think Clara has many courts and she deposits this token into a court of websites disputes and so a lot of people are going to deposit the token which is the the standard of how it works so the token stays into the court and many people stake the token and the token gives you the chance to be drawn randomly and some of them are going to be selected randomly and these guys are going to be the jurors in the dispute and the token stays long until a decision is done so how do you incentivize people to do a scope? how do you make sure that you know how many people go into a grant randomly A, B, A, B just to collect money and for this we use case theory who is this guy? it's a blockchain conference it's Thomas Shelling so Thomas Shelling he's a jurist of the 50s and 60s and he developed this concept of focal point but it's about how people coordinate when there is uncertainty and when they have to make a decision independently so each of you choose a number and if you choose a set of the majority you get 20 dollars which one do you choose? Thomas so that's the sharing point any number would be valid to choose but we expect others to choose 1000 so we choose that one and so this generates a coordinating mechanism into that number so if different jurors vote independently on the same dispute knowing that they have to vote at the majority then this creates an incentive in each of them to vote on the truth about the dispute good? these are some interfaces about how that works and then jurors vote with the majority they get the token back and arbitration fee and then jurors vote differently they lose the token that was deposited so this creates basically a situation where users could try to abuse the system like most irrandomly they would lose the token they would lose money on average as they participate in more cases who made the honest work of analyzing the evidence and voting correctly they would make money on average so the vision is to create this justice system that can solve all the disputes of all the doubts that are being created and Kleros has a cool side an arbitrable side arbitrable side is where you have dubs that have disputes and they send those disputes to Kleros and Kleros sends back the decision of those disputes this can be used for escrow transactions this is an interface we built you can create an escrow contract that is going to be sold by Kleros you can use this now or use it for paying our contractor for example then token list this is people can send tokens to a list and other people can challenge them if they don't compare with rules and Kleros in case there is a dispute and here I am pushing a kind of application of this to create a humanity list how do you make sure that people are humans who are not on the internet so this is a project if you are interested use Kleros for making this list and it's used tonight at what time it's short we have the meeting for the humanity six area of Kleros house many persons are going to be there to collaborate in building this and Vitalik this is creating something that Vitalik suggested into the his token is in New York I use of Kleros to create a list of humans on the internet if you want to come to this place if you want to collaborate tonight at 6.30 at Kleros house I can show you and finally Kleros can be used for all of these there is a dispute about what is the value two parties and Kleros can solve this problem so this is how the Kleros system looks like now and this is how we think it will look in the future when all of these other tabs connect into Kleros so basically is to solve the problems of the dispute of all of the labs and this is the standard way to pitch Kleros but there is more to this than the resolution and this is what I want to tell you now and I told you before how the great democracy evolved but I didn't tell you all the story because remember Kleros telling us reforms that he created the assembly and not that but the democracy didn't work after he did that nobody didn't work and the majority the assembly was voting like laws and it was crashing the rights of the minority and this ended in a regime that Plato called the Theatropracy imagine being governed by twitter by the crypto twitter every day like a decision that's how it worked in Athens right after the reforms like lifetimes like a very hard dictatorship of the 30 tyrants that killed a lot of people and then a restoration and you know how Athens became actually a real democracy but actually a war because they made a small change in their institutions they created somebody called the Nomo Tethi and this was a judiciary review for all of the assembly in Greece you have the country proposing listen to the assembly and vote in the assembly and then after the rule the law was passed it was posted in the middle of the agora where every citizen could see and could read it and they thought that this didn't comply with the constitution they could challenge it and when they challenged it so you have a jury of citizens that conducted a trial on the new law some of them being prosecutors other defendants and then a jury and a decision so this decision could end up with the law being rejected and it was all powered by the people this is why very famous if you control the courts you control the state because you control what laws can be passed in your community so a very important use case I imagine in Kleros in Dao governance is creating this judiciary review of all the proposals that get voted in these different laws but if someone comes and says let's raise the bitcoin limit or something like should X Dao invest or give money to some project so this can be passed by the community vote let's say in classes so those who voted yes so they have their token that stays locked into their vote and then the community can see the law that was passed and if they don't think that they complied with the constitution they can challenge that law and this we send it to trial of course to Kleros right and Kleros will create a jury of randomly selected users and these guys are going to decide if this proposal is compatible or not with the Dao constitution and the trial arguing about things and this jury is going to vote and if you voted for the proposal in the proposal vote and then this is overturned by the jury to lose your deposit and this is distributed to those who challenged and to the jurors who participated in the trial so this is how I imagine to be used in the middle of the Dao governance John was asking before how do we prevent but those who passed and how about the vote to be done and vote buying so if we have this layer of revision based on this rational mechanism then this is a solution for that so basically imagine that besides all we know about Kleros about centralized course or websites and that kind of stuff I think it's going to play a very important role in what is the governance of Daos in general as some kind of centralized supreme court as a service that you can improve and then you get the decisions made right and the good thing about this is a supreme court that is going to control by the users right and this is controlled by people who are in the Dao and I think it's going to be really important to have solid governance for the blockchain but who are you all about Zcash governance do you have no slides I guess did you send them to me okay sorry I will get them pulled up you are the executive director of the Zcash Foundation so we're really excited to have Josh here with us today why don't you kick off introduce yourself do you email the slides or did you Twitter DM thanks so much for having me and thanks to all the other presenters for their great talks so as Lee mentioned I'm the executive director of the Zcash Foundation as honestly I think this is probably going to be the most boring lightning talk of everyone's because it is our governance process is decidedly the most off chain it is the most sort of meat space consensus and actually it's also one that is much more legally enforced than others I'll get into a little bit of that so first off toast to governance the cause of and solution to all vice-president much like alcohol a little bit is great but too much and hopefully so let's talk a little bit more about that I think generally speaking in cryptocurrency systems you have this you have a number of set of stakeholders and it's obviously can be a little bit broader than this but based in my view in Zcash this is the way that I tend to divide up different stakeholders that may have different conflicting interests at times may have common aligned interests and what we're kind of relying on is that each group relies on trust in some fashion but they also have some kind of ability to ameliorate any kind of disagreement that you might have as John had mentioned before Bitcoin there is the UASF there's a clear demonstration of users overriding what appeared to be minor and corporate origin so what be in the Zcash at least on the Zcash foundation side and on the electric coin company side which is the for profit entity that really has been responsible for the vast majority of development in Zcash today we focus more on the development side of this equation and that's what we're going to focus on today so we have this sort of aspirational model in the community of this two of two multi-sig governments and the idea behind that is that you have these two separate but aligned organizations on the development side one of the electric coin company as mentioned they're the ones that really started everything on the Zcash side and they are answerable to shareholders we have the Zcash foundation that was set up a year later several years after company but a year after Zcash the protocol had launched and we're a non-profit that answers to our board and mission and we both are dedicated primarily to certain Zcash so in 2019 here's kind of a list of where I see the roles and responsibilities of the electric coin company on one side and the Zcash foundation on the other and where there's an intersection in between and where this really comes down to brass tacks in terms of how the protocol is governed is how the electric coin company decides to build and maintain the node software that users choose to run how we at the Zcash foundation choose to build our independent node software that we are in the process of developing now and then how we decide to interact with the zip process which is similar to the VIP process and how we come together to decide future development on the roadmap for the protocol so what I and this is not indicative of everyone in the Zcash community my personal view is that users running nodes are the ultimate expression of who has power in the system it's really up to users to decide what kind of rules they want to follow and a given protocol and I think that was most clearly demonstrated in the Bitcoin ASF kind of the situation in 2017 so in essence the most like real impact that we at the foundation have in terms of governance in the Zcash protocol is in our ability to build and maintain a consensus compatible node implementation for the Zcash and what does that mean for the zip process tomorrow time goes by fast so right now the zip process today for all intents and purposes we do have an editor on the zip process that's represented by the foundation but ultimately whatever the electric coin company decides to integrate into CBRD is pretty much what winds up defining the protocol there's a wrinkle to this that I'll get into about trademarks but at least on the pure development side that is what matters and for from the beginning until now a lot of that process has been kind of opaque and it hasn't needed to be more open because it's really been more about like technocratic decisions being made about things like privacy for instance so where it's going to go in the future is there's going to be this mutually kind of selective process about what ECC decides to add to Zcash and what we decide to add to zebra and then there's a mutual upgrade to sort of network so this is how you know we keep that band together there's that power the actual power of users to run nodes inside is the that plausible threat of an intentional chain split but there's this other legal enforcement mechanism with the trademark which is basically the right now the trademark is fully owned by the ECC but there is this there's a current agreement that's being negotiated in order to give both the ECC and the Zcash Foundation power in deciding what represents Zcash from a legal perspective so this is a little snippet we actually wound up delaying what I will be talking about in a second a community sentiment collection because there was disagreement with the ECC about moving forward on the trademark and this is like one of the most important things that we could have done as a foundation in order to legitimize the process by which we go through future upgrades and right now the reason this is such a hot button issue is we're in the process of discussing a future fund for the basically an extension of the Founders Award and who would stand to benefit from that sort of award and we at the foundation want to make sure that that decision is being made with a degree of legitimacy and care so just going into like how we'll decide what we're going to put into zebra there's non-conversial ones and then everything ultimately escalates to the Board of the Foundation but the Board can use valuable feedback and this is that community sentiment process that we were talking about earlier that I mentioned very briefly with the idea that there are many ways that we can collect kind of both on-chain and off-chain sentiment that can then be used to inform the Board's decision but like make no mistake that ultimately like whatever gets decided about what the foundation decides to put in its software is going to be decided by the Board and I suspect that it won't go against what we view to be community consensus but that's ultimately where that power lies users can still have the power to fork away and choose to and choose to operate and potentially engage in sort of a UASM style maneuver but ultimately there's this trademark issue that makes it harder to do that and thus really I think kind of compels both the ECC and the Zcash Foundation to try and reach some kind of consensus about where we're going in the future okay actually I made it in time wow I was a little worried there but anyway thank you everybody that was me kind of trying to blitz through that but there's there's a lot more that's going on on the Zcash side I wholeheartedly recommend that if anyone's interested check out some of the blog posts that come out from the Electric Point Company from the Zcash Foundation check out forms.zcashcommunity.com if you want to see a lot of the debate about the future dev fund proposals which we're currently in a standstill until this trademark agreement gets renegotiated which will happen very very soon so okay thank you very much cool so I'm going to talk about governance of games so let's uh so so anyway I'm sure give me up and I do research on complexity and I also build the cosmos in theory by direction bridge okay cool so what is governance? so governance can be defined as either decision making or coordination of multiple agents so that's what we discussed earlier today I'm sure you're familiar with decision making can be defined mathematically by just talking about a function which maps to a set of outcomes why why being the desirable maximum and coordination can be defined as a multi-agent system where multiple agents have to coordinate in order to achieve a certain goal cool so um so let's say we are playing a very generic prison salama game um and the game is defined as so so um the action space is basically co-opered in the fact and the A and B are playing the prison salama game um so the Nash equilibrium which is the equilibrium in through which um the either player doesn't have any incentive to move from uh is a defect defect so it's at zero zero so they're um top um like I mean the bottom right corner and it's defined by that by that uh function right there cool so how can this game be modified so this is the concept of game warping where uh the Nash equilibrium of a game can be shifted uh and this game can be modified if a third player has information about the game that these two guys are playing um is able to make an offer to change the utilities of the initial game and this is especially possible with smart contracts because smart contracts are publicly visible, strictly enforceable and also uh they are frictionless to implement means that you wouldn't have to expend any money with with um legal contracts and putting lawyer fees um and they're also strictly enforceable means that any threat that you make on this um particular work will be credible so let's say player C comes to be and says hey B I'm going to give you 1.5 utility points if you cooperate with player A while player A defects now even though this is not the most rational strategy for uh for B unless um C doesn't exist or if C did hadn't made the offer now by proposing this utility he makes it rational cool so so then we have this composite game where um since the C is able to offer 1.5 utility points for um corporate defect um C will be able to modify the game that A and B are playing does making the new national equilibrium to arrive at um A defecting while B cooperates so now it has essentially changed to modify the game in such a way that it does is not the traditional since it's not by anymore there's just one way in which games can be worked games can be worked even through pre-agreements made between two players that are just interacting with each other without the player in next listing and there are many other ways as well in which games can be worked which I wouldn't go into details right now uh but the point is that any smart contract that can be computed will be computed at this is kind of the double edgedness or cheery completeness of smart contracts because anything that can be computed will be computed so we arrive at the standard of do we consider good power to be money so um I've sort of loosely defined um how game warping configurations can be extremely vast so the number of offers that a potential player C or a game warper can make um is defined by the number of actions that exist within the game times the wealth of the maximum the maximum wealth of the warper times the combination of the players that the warper chooses to work and this will be divided by the computational power of the warper because that's a key limitation um that is going to take the number of game warping offers that one can make um so okay so I'm going to this is a game um I'm going to ask the room to pick the number between zero and 100 and the person that picks the number closest to half of the average wins alright go ahead pick alright are you guys done yes okay cool uh how many people picked above 50 show of hands okay how many people picked about 25 okay level one uh you guys didn't make a random choice if you picked if you picked about 50 it might be a random choice uh if you picked um about 25 it's a level one choice how many people picked um about half of 25 okay level two choice uh how many people picked zero okay cool you guys probably were the concept or you're really computationally smart um okay so level level k games is when um you are able to make inferences about your peers um about what your peers are um choosing so yeah yeah exactly um so so the rest of level k rate um brushality one of the one of the um key limitations for that is that humans are usually studied or empirically it has been verified that humans are usually level three and um computers can be as level high as um the compute power that they possess so also when we're dealing with uh a lot of these um blockchain or smart contract systems um I I have tended to work under the assumption that information is complete because everything is visible on a public blockchain our information is actually incomplete because a lot of beliefs can be private even when beliefs are broadcasted they can still be private because you don't know the true belief so you can't assign a high degree of certainty to it and um also reality is a fabrication of the collective perception of all of the agents within the system so there are things such as um the value um is also a perception it's something that we have assigned the definition of a thing is also a perception um it is also defined by the agents and um so essentially we are not able to meet um very definitive inferences about those things unless we do do so by empirical study in which case it would be a probability distribution cool so um and then there's also this big challenge of um um formalizing Bami Rashmati so so why do I call this human Swiss and machine Swiss humans it's basically human designers that are designing machine systems that are supposed to coordinate other humans um and the big challenge with that or mapping um through this is formalizing Bami Rashmati and the way we can do that is um two ways uh one you can use um Gibbs free energy in which case you look at it as a constraint optimization or you can measure a misfit using a KL divergence which is basically the delta between um the expected outcome versus the true outcome of the game um and the constraint optimization is basically when the constraint is a compute power that of the agent that's making the calculation and then we also have the complexity of value this is basically an argument stating that you cannot represent value um that a human has or what a human value is mathematically and thus it becomes hard to implement that in a simulation cool so um so we're essentially uh when we're talking about these kinds of systems they do act as BL3 networks meaning um there are networks where 80% of the nodes hold 20% of the connections and 20% of the nodes hold 80% of the connections um and there are ways in which these networks can be resilient but first we have to define resilience so resilience we can measure it as um as um either um the ability to not be fragmented or the ability to tolerate false or the ability to survive attacks so we are not going to consider random networks because usually um blockchains are more similar to scale free networks um so within scale free networks the attack survivability of blockchains is pretty pretty bad so that's the that's the one thing that we should consider when we're thinking about resilience of a blockchain so basically attack survivability is the fact that since there are such few nodes that possess most number of connections if you attack those nodes then the entire system essentially fails however with regards to network fragmentation and false tolerance because of this property of a few nodes having a high degree of connections the the network would essentially not really change if you um remove a couple of nodes from the system or if a few random nodes get influenced or attacked cool so um this is Laosu and um he uh in 500 BC he um proposed um self-organization and self-governance so how do how do we self-organize so we um aren't very good at self-organization as we have observed in current systems um so but then there are ways in which we can self-organize um and and we can also guide it so I have to give credit to my friend Virgil Griffith for um for introducing me to the concept of self-organization which he first implemented in terms of um studying some information theoretic systems um but essentially self-organization um can be guided in three ways so either you can um talk about or um change the control parameter of the system in which case the x will be made into another x which is better so earlier I explained it actually graphically I think that might um help a little bit better so essentially if you have a system where it has been stuck in a local minima and it's just not even moving from here you just take this and put it up here so so that the system goes back to uh dynamics that are that are more um or to a state that is more desired or you can have or you can actually change the function itself so this is changing x or you can change the function that governs x so instead of having it like this you can transform it to something like this or any other function so essentially you would change the dynamics which govern any specific control parameter uh and you can change this for any number of functions or you can change the entire structure of the self-governing system itself in which case you change the set of all the functions that um that constitute the government the governance model so all of the policies that are defined within the system can be changed in that case it is extremely limiting uh in the sense that it will it will um not allow for you to see emergent effects because you're constantly modifying the way in which the govern governance model is um whereas with the regards to just interrupts you can you can be able you can still observe quite a lot of governance effects um or emergent effects um despite there um being a little bit of intervention within the system so um those are the ways in which you can guide the self-organizing system to better um to better states or to um having a better governance model cool remember this guide um so um so think about um guiding the process in which um systems can be governed instead of allowing it to completely um self-organize by itself thank you I'm a software engineer at GORT and today I'm going to be talking about the other half of governance that Lane talked about uh whether a blockchain can provide governance and in this case whether a blockchain can provide rule of law so what is rule of law um this is a very nebulous topic so uh there's uh this great paper called Microfoundations of the World Walk inspired by Jillian Hadfield and very guest and in it uh they basically say that the rule of law is a classification institution so what they need by this is that it takes the facts of the case it's uh it's a function that takes the facts of the case and then it outputs a decision so in this case it would be is this lawful or unlawful is this constitutional or unconstitutional and they say that a good classification institution it has these six characteristics uh so it makes the decision-making logic publicly available so if you're trying to decide whether or not to do something uh you need to know whether you know what you're going to do is lawful or unlawful right uh it results in continuity, it's stable, it's personal um it's able to produce new rules um by soliciting information for users and it gives predictable results for novel inputs okay so they say they say importantly that it's a mistake to presuppose that government is an essential feature of law and legal order so um this is from Google DeepMind and you might notice that these folks are kind of acting a little strangely um but the reason why they are acting this way is because the DeepMind project didn't choose to uh instruct them on how to do things it just goes to give them a goal of moving forward without falling and so you can see that you know we got really really different behavior because because we focused on the goal and not how to do it and we can actually apply the same thing to law and this is an actual approach it's called the functional approaching comparative law and what it says is that law is comparable if it performs the same function so if it solves the same social problems I'm just going to give one example of this so this is the law speaker um and this was around the year 1000 or so in um Scandinavian countries like uh Iceland and what the law speaker did was um uh he as you might imagine spoke the law so this was how they carried the law um down through generations was that they had a law speaker it wasn't a problem so this is very different than what we're familiar with but it still counted as law okay so um given that definition let's talk about the social problems to solve so there's this great paper by Anthony Kerman who's a uh Yale Law Officer in contract law he has um this paper called contract law in the state of nature and in it he says that there's two problems and this goes back to Hobbes he says that one of the problems is uh vulnerability of possession so you know people can take your stuff right and uh the other problem is transaction security so this means that if you try to make promises with other people they may break their promises so how can we solve this so uh let's get some definitions out of the way so I'm going to be defining a blockchain very loosely and very abstractly as a state machine so it takes old state and new information it has a transition function and you get new states I'm also going to be defining uh on-chain versus off-chain assets so an on-chain asset um by that I just mean that a blockchain is the sole source of truth for ownership or access and by off-chain I mean that something else determines ownership or access so this is really about social consensus and source of truth okay so let's take a simple on-chain case for cryptocurrency so let's say that Alice wants to give Bob 10 tokens right so the old state Alice has 10 Bob has 0 it goes through a transition function now Bob has 10 right very simple let's uh look at an invalid transaction so let's say that Bob uh tries to take 10 tokens from Alice uh without her consent so as we all know this would not work right so uh because Alice is not signing the transaction it's not going to be uh validated by the block producers, validators, bankers, whatever whatever you call it in your system so what's really cool about this is that uh it turns out that cryptocurrencies uh because they do this they're actually enforcing property rights on the internet in code and without traditional law so uh medium on rate has this great paper called ownership and uh N8 talks about I think there's like six or eight different uh rights that are in the bundle of ownership and it turns out that cryptocurrencies uh have three of those rights so there's the right to use it there's the right to exclude others and there's the power of transfer so at agorac we decided to uh see what would happen if we tried to express more sophisticated property rights so we chose to uh implement a pixel demo based on red inside our place this is a picture of red inside our place and uh in this demo uh we decided to allow users to buy and sell pixels as well as color of pixels and can create whatever they wanted to create but what we discovered was that um this was actually implementing another one of honoree's uh uh rights in the bundle of ownership and this was the right to um to alienate the use right, maintain ownership but allows someone else to use it and it turns out that this is kind of the um the owner-in-tenant-sub-tenant relationship so this is very cool so we talked about vulnerability possession we talked about property rights what about transactional security in contract so um the simplest uh uh example of transactional insecurity is what I call a swap and there's this great example from Raiders of the Lost Ark it's not actually this swap it's what happens right after it um so I had a video but I wanted to try to play it for time reasons but uh it's all inactive for you guys but Indiana Jones is trying to get over this pit right and uh he has the idol his assistant is on the other side and has the whip that would help him get across and so um he says give me the whip and this guy says throw me the idol so Indiana Jones throws the idol over and as we might expect this guy is like Adios signal right so this is the problem of transactional insecurity so how can we solve it so we can solve this if we use a smart contract so um what's really special here is that a smart contract can essentially own the assets um and allow us to uh to transfer them so if Alice has x and Bob has y we can send it into the smart contract then the smart contract can actually use the dispersal um safely so this is really really cool so that means we've solved the social problem of transactional insecurity without using traditional legal contracts and so you might be saying well you know that was just a simple swap what about um what about promises that unfold over time what about more complex things and um and you would be right that's a good question so let's open up the black box of contract enforcement and so there's two steps there's a classification step and then there's applying the grant form and this comes from a great talk by uh Bob Professor Oliver Goodnuff called ambiguity and legal specification the process is that uh so there's the event classification step right that says has x happened or not and then there's applying the correct rule so if x happens then do y it turns out that humans are much better at classification uh than computers are so like for example um in uh in tenor plot it's really important to know that the apartment is habitable so if you were doing this with a machine you would have to have sensors for like carbon monoxide for you know you can for mold you can imagine all the things that you would have to uh track but you could actually just have a human being a human inspector go and say oh this looks right so that would be a lot easier uh so human beings are great at classification but they're terrible at legal application so as we all know there's a lot of corruption misunderstandings all sorts of things right so it turns out that we can have the best of both worlds and this is where things like Clara's coming so uh we can have a human classifier say yes this is habitable and that can be part of the input to the smart contract we can also have human dispute resolution system such that uh if something is triggered it gets inked out of the human dispute resolution and then comes back into the smart contract so we talked about learnability of possession we talked about transactional insecurity and how uh property rights of contractors solutions to this but do we actually have rule of law so going back to the rules of the good classification institution uh so number one is the decision making logic publicly available this is obviously true for a blockchain right we have to be able to replicate it so it has to be publicly available uh number two does it resolve ambiguity um so things have to be deterministic we can't really allow for ambiguity so I would say that this is yes uh number three is it stable so for anything on a blockchain to be uh used to solve the social problems it has to be long lasting so I think this still remains to be seen right uh number four is predictable results for novel inputs um so this is kind of uh yes and no um so on a blockchain we can um uh we can we can say if x and we can say if not x and that's the entire universe so the possibilities right but uh at the same time there's probably going to be things that we don't expect um so I think this one is a maybe we're working on it uh number five uh so a smart contract running on a blockchain it is impersonal right um so on a blockchain we don't even know who someone is we don't even know if it's a machine or a person so it's very impersonal um and number six it can produce uh new rules by soliciting information from users I think this one um is also one that uh the blockchain ecosystem is weakened um we can uh produce new rules by transitioning to new blockchains or new or upgrading to new smart contracts but that's very difficult and still remains to be seen how how good we are at doing that so in conclusion we took a functional approach uh we talked about the rule of law as a classification institution and we talked about the social problems to solve right so vulnerability possession and transactional insecurity and it turns out that uh for on-chain assets we can enforce sophisticated property rights and we can solve simple transactional insecurity we're still working on as I'm sure all of you know uh censorship resistant articles often and we're still working on solutions for off-chain assets but the really cool thing is that a blockchain as a smart contract platform can provide at least parts of the rule of law so thank you so much um and we don't have any time for questions I don't think so thank you