J3 AG 02 Vlad Zamfir Governance ♥

hey everyone welcome welcome to my talk about watching governance but mostly it's really just to talk about governance in general the point of this talk is kind of to give you guys a basic language for talking about governance trying to like basically understand what governance is and and and I and I'm not gonna be normative I'm not gonna say anything about what I think watching governance should be like I'm not gonna I'm gonna try to be like like dial the controversy dial to zero and the like obvious dial 200 so I can I goal is to try to say things that are only kind of very uncontroversial in an effort to kind of give it's your language that has helped me reason about governance and I think will help you reason about governance regardless of what your governance agenda might be so here's kind of the outline so we're gonna talk about you just see everything everything great we're talking about like some basic facts about blockchain governance and then we'll just go in for most of the talk speaking about various concepts in governance there's so somehow like I'm gonna start with talking about blockchain as if it's a decision-making process and then I'm gonna talk about their like participants and stakeholders then I'm gonna talk about governance as if it's a coordination problem then we talk about the role of knowledge and coordination then we talked about like legitimacy formalization and institutionalization of governance then we talked about like kind of when things go wrong the legitimacy disenfranchisement apathy revolution forking and then I'm gonna talk about you know what I really think governance is and I didn't want to spoil it so you know this is some some suspense here so blockchain governance what is blocked in governance well it's about governing the blockchain mmm what is the blockchain well it's a bunch of software running on a distributed network mmm at least alright this is something that we kind of know that every single block chain has in common no matter what from now on till the end of time block chains will always have software running on distributed network so today they're like nodes running on the internet running software from some software repositories nodes can be kind of consensus forming they might be responsible for our mining blocks or forming consensus around blocks they might be whose they might be service providers they might they might kind of you know be like a exchange or a block Explorer or like an application provider it could be basically anyone anyone who has a bunch of users who use kind of a service that is specifically using the blockchain and they run the full mode like that synchronizes with the blockchain then like um you know they're gonna be note operators infrastructure providers like in furo is probably the most important note in the etherium space just because of the sheer that massive web 3 queries that they serve there's people who run knows just for infrastructure just like relay nodes and then there's people who run knows just for like this kind of oh I want my use of the blockchain to be trustless sorry I hear my I hear my music going off I just stop it so you know and this is kind of like the true cypherpunk users right though they'll run the full note and check all the blocks validity to make sure that they all sent authenticate the state and so each of these types of users will have different incentives different reasons for running full notes and they own the notes and so they're gonna have their own processes for governing them namely for deciding what software to actually run on the nose so different nodes have different senses and differ in their and then they do different roles but somehow they are the people who run the software on the network there are the ones who kind of are running the blockchain but kind of hot running code and there's a software repositories which space leased or like the source code and so these are governed by separate kinds of processes there because the decisions there are changes to software repositories they might be governed by like Oh an individual company they might have like a particular you know leader or like core dev who may like to who has the right access and determines what its merged they might there might be you know I know of a source community with open source offered at all their practices or you can even have like you know something a lot more a lot more kind of official and mediated by like a Sanders body so there's lots of ways the software repositories could be could be governing and nodes kind of provide an important check to balance against the people who make changes to software repositories if you are node you kind of have a choice about well software to run and so somehow like the developers don't have free rein because they don't control all of the all the nodes and the node operators generally aren't all developers this cookbook there's like a separation of interest and responsibilities and also like I think that nodes quite clearly provide a check against the power of developers because of their like choice to upgrade an arm but this doesn't tell us much at all actually this is just stuff that's kind of always fundamentally true and but like all watch chains the bachchans will have different norms and different different ways to govern their repositories in their nodes the relationship between the nodes and their repositories will be different between block chains and somehow we don't know yet like I haven't said anything I haven't really have enough language at this point I'd say to talk about like what is the kind of objective function like what do we what are we aiming for in determining like how do we govern our nodes how do we govern our software repositories so this is kind of you know the end of the first section where I say like oh look the blockchain is something that can be governed and there's different resources that need to be covered at least to write this at least knows and there's at least suffered repositories in practice there's a lot more but everybody know how these two in common so now I'm going to talk about governance in general and this is going to be like the the the kind of I think interesting portion of the talk so I'm gonna start off by considering governance as a decision-making process it's basically a process by which somehow decisions are made on that'll affect something we'll call it the governed so the changes to software repository changes to what know what software is running on a node each represent decisions that affects some resource be it like the repository or that node and I call the resource that governed and like the governance process like makes decisions on it so this is like a nice start it's kind of similar what people normally think of as governance in terms of like a governance model which will output like decisions based on usually votes things that are upstream of the governance process I'm sorry things that can affect the governance process I refer to is upstream things that are affected by the decisions made by the governance process are called downstream and basically if you think this as a causal upstream and downstream if you can constantly have an effect on the govern on the decisions made by a governance process than your upstream if you're constantly affected by it then your downstream and you can be both upstream and downstream so actors who are upstream are called participants actors who are downstream are called stakeholders so so far we kind of talked about processes we haven't really talked about actors the idea somehow is that I go you know might not be that easy to formally give a definition that distinguishes actors from processes but actors we kind of ascribe with a much higher degree of ability to set goals and change goals and have intentions and have a kind of agency that we don't usually ascribe to processes so there I'm going to talk about upstream processes or am i talk about participants which are upstream actors and actors may play a part in different processes but somehow the actors are assumed to have agency maybe intentions and incentives whereas processes are assumed to not have intentions or incentives but just be kind of chugging along doing like the execution of the rules so you know you will see one like right now why we kind of do this right because we want to consider governance as a coordination game and games are normally understood to be between actors right between kind of players that have some sense of wanting to make a particular choice and having agency about reasoning with about like other players and making decisions you know in like this kind of way that like we assume or agency than we normally would for a process even though somehow it's like quite difficult to define the difference between process and agent so in the governance game then it covered a governance process a lot of the time it can be understood as a coordination game a coordination game is a game where there's different equilibriums but you can't necessarily get them just by choosing the thing you intend because the equilibrium is a function of other players behavior and so we have we have multiple equilibriums that we can only get to if we choose strategies in a way that corresponds to other people's so for example if we have a voting system that requires a threshold on a proposal before passing there needs to be correspondence between the different votes namely that they're all voting on the same proposal for any proposal to be passed and so in order to avoid a outcome where nothing happens you have to all kind of coordinate your strategies so any kind of game where the coordination of strategies is required for getting intended outcomes is a kind of coordination game the tricky thing about coordination games is oh you don't know at least like the traditional ways to interpret them is like you don't really know what the other people are gonna play right are they gonna go to the pub or the coffee house or are you gonna go shopping or you know see the boxing match and sometimes there's a tension and conflicts of interest between what people what what people liver and people want what outcome people want and sometimes people just want to decide on something they don't care as long as the other side of the same thing so basically the structure of the process that we talked about as a decision-making process is the structure of the coordination game it determines how like the strategies the strategy profiles the behaviors of the participants actually lead to different outcomes and so the governance process like the decision making process is the structure of the coordination game when we consider governance to be a coordination problem so the structure of the coordination games we just looked at is like a two by two matrix which just like says like oh look there's like four outcomes and we can do and we just like label them by the strategy profile but in like a more complicated game like a voting system you know we're not going to actually implement all possible strategy profiles and say what's like tails are instead we have like strategies lead to possible outcomes and then people have like different payouts for different outcomes and so the structure of the governance decision-making process structures the coordination game that the participants are involved in when they play where they kind of try to make these decisions let me stop for questions I wonder if is if anyone feels like I've said anything controversial yet so first governance is a coordination governance is making process oh look there's upstream and downstream processes and agents and the ones on top work are the ones upstream are called participants ones downstream are called stakeholders participants participate in a in a coordination game that is structured by the decision-making process right so I hope that that's you know uncontroversial so the next the next kind of interesting thing to note is that coordination games are kind of hard because we don't really know whatever what other people are doing and the but we all need to choose the same same or we need to kind of coordinate in order to produce outcomes and so actually the coordination games are always structured by the information that the participants have about each other they might have a whole bunch of different types of information that help them coordinate around outcomes a focal point is something like for example oh we've met there in the past or oh this is like a more central location and people will meet around focal points just kind of in lack of any other information and and these are often used to solve coordination problems when there isn't a focal point it can be much harder and focal points can be quite useless expectations are kind of like oh look I've seen people behave this way in the past and so I kind of build up this anticipation that they're gonna be there and and still like oh we've oh we always went to this bar and so show up that bar again to try to solve the coordination problem a lot of the traditional coronation problems are kind of described by people trying to meet up at the same location and this common knowledge common knowledge is kind of a much more interesting thing common knowledge is a distributed state of knowledge that allows people to behave as something as a fact as if something's a fact without like verifying with each other and knowledge in each other states of knowledge so it's like not only do I think something and you think it but we need to know that we think and we just know that we each know that we think it and in a way that we can each act like it's a fact so when something is common knowledge then it basically is something that it's almost as if like it's part of the rules of a game and like a traditional like normal form game we kind of like everyone everywhere everyone sees the rules of game and everyone knows all of them so can they all they all take it as a given so common common knowledge is basically like a distributed state of knowledge where we don't have uncertainty about each other states of knowledge it's kind of we all know that we know and it's super useful it turns out in the formation of governance norms so governance norms get rid of forcible expectation of a kind of behavior for participants basically it's a norm is something where like if you don't follow it you can be no necessarily penalized or reprimanded too much but encouraged or corrected or you know just someone can point out to you that you didn't you didn't fall in line with the norm without fear of their own reprimand and so norms are safely reinforcing and the way that this often happens is with a common knowledge that you know we all know this is wrong and therefore when I tell you that it's wrong I don't we're not gonna have a conversation about oh is it really wrong because we all know that we all know that it's wrong and so it's like easier to cut the call to call it out so norms are kind of a kind of common knowledge that we do in practice structure governance the structure governance as coordination quite a lot because people will coordinate around the norms like for example in Bitcoin we have a strong norm against contentious hard forks and that norm really really heavily structures the coordination that people do around the coin Bitcoin core even though that's not actually part of the process for merging software except for I was like this kind of norm so legitimacy is kind of really really important concept in in governance it refers to this kind of common knowledge that we're gonna use the decision-making process as opposed to replacing or abandon it it's kind of a distributed state of knowledge that we agree to use process a decision-making process it's this is really a norm to use governance process where our governors process kind of means this decision it's like process for decision-making legitimacy is kind of really important because it prevents people from having to debate how what to replace the process with for having to coordinate around a new process memory establish new norms the legitimacy also applies to norms and not just the processes and and it's it's it's it's probably the most important thing for the stability of a governance process because if the participants in the stakeholders decide that they're not going to use it anymore then that's that I mean that's the first kind of step towards having it be abandoned and replaced so formalization is kind of well firstly okay any questions is that also allege it Amit yes I'll repeat your question great so actually we're gonna talk about the imposition of governess trustees in a second but re so the question is like oh what if this governance process that like everyone agrees to use but it was imposed on a particular community any answer as well if they do have the common knowledge that they're gonna use it rather than replace it then it has the it has legitimacy legitimacy doesn't mean that it's good it doesn't mean that we like it it means that we all know that we agree to use it it like in an ideal world where people only agree to use things that are good for them which is far from the world we exist in then I would say that like yes legitimate legitimacy would entail consent but I don't think that as a matter of fact or as like you know a matter of like Oh getting utility out of my definitions that legitimacy contains the approval of the like the the kind of approval in the sense of a you know an enthusiastic consent of like participants or the stakeholders my favorite example of this is authoritarian regimes and things that are kind of blatantly unfair although still people have a norm to use them and I I think like it actually is kind of common that a process will be established it'll be you get legitimacy because people think that it's good for them and it will retain legitimacy just because of that inertia well the reality becomes apparent over time that actually it's not good for them as perhaps the system degrades or gets captured things often are legitimate for much longer than they're actually fair or agreeable so I saw the cool so the comment is that I go okay and this means that like there's if there's a there's a there's a it's possible for like a norm to like maintain it just continue being a norm while losing legitimacy and I think that's true but the important thing is that like the legitimacy that's being lost is because of the mute of the common knowledge that's being built about that norms inadequacy it's not because of people individually without talking to each other come to the conclusion that it's bad it's a they actually have to form common knowledge for it to lose legitimacy if everyone is just independently thinking it sucks but but all believes that everyone else thinks it's great then I will have legitimacy and and and and actually this kind of mismatch between like our expectations of other people's opinions and actually people's opinions ends up having a noticeable effect on governance decisions and like you know even like day-to-day decisions a lot of the time so I don't think it's should be that controversial actually that legitimacy is a more about a common knowledge or a collective state of knowledge I supposed to just agreement right so agreement is we all believe the same thing common knowledge is we all know that we believe the same thing right it might be that we all hate you know the legislature in some particular country however we might all believe that we all are going to go along with it and because we haven't had the kinds of conversations and things that would make would make us but like believe that has become like that is actually less legitimate because we're in the sense of that we all that we agree that we're going to replace or been in it because remember kind of I definition legitimacy here is that like it's common knowledge that working on use rather than a replace are been in it John so I was wondering like if legitimacy it doesn't your definition doesn't capture it being true to the intent of the group rather this would seem like a result of legitimacy oh yeah so somehow like true the intent of the group is something that like would be nice and often I think legitimacy is established through things like a compelling argument for why this is good for the group however like I like I kind of mentioned like oh I think that can Genesee can land last way being way for way longer than like the actual like viability of the arguments for well for why we should use this I think that like it's easy to conflate the factors that lead to legitimacy arising with a legitimate legitimacy itself so for example you know fairness might be something that leads to legitimacy and people might think that like Oh legitimacy requires fairness but I but I like prefer that way that these definitions set up now where somehow I can have a unfair system that is legitimate and a fair system that is legitimate and I'm talking about building legitimacy oh they rope in all the kinds of things are thinking about in terms of making sure that it's kind of in line with what people want so that they'll agree to it and like you know be able to form common knowledge that they want but I think those things like are the means by which legitimacy is achieved so I think like oh I'm not sure if we really disagreeing about like the underlying reality just like what do we call the Gintama see is it like though like you know is it like the charisma that led people to believe or is it like actual common knowledge that people have that they'll continue using it so you know are you satisfied No okay I'm not gonna repeat that sorry oh so you know when you say that about regimes where its authoritarian it is regimes that are authoritarian like dictatorial or something they are legitimate but oppressive yeah exactly so the thing is that it is not also that it is legitimate because there is common knowledge that's always the fact but there's something added to that which is it is also that what you said that that preference optimization which is when I know that like you of us know that you are seeing something wrong but I'm waiting for somebody else to first say it out loud so in May in cases of common knowledge what happens is because I know that otherwise known others know that I know that we are not supposed to talk about it so even if I so someone has to go out of the way and first say it out to make it to create a ripple effect so sometimes it becomes a problem of coordination cost because it becomes too expensive to get into another coordination focal point where you can have the new state of legitimacy so that's why it becomes economically more viable to accept a regime which is authoritarian so wait a second it sounds like you said two things one of them is oh look switching between switching from like an illegitimate system to illegitimate system can be extremely high coordination cost and very difficult for a number of reasons and the other thing that it sounds like you said was something like there can be additional norms against reducing the legitimacy of a governance process and that's a fun thing of coordination cost because it's reflexive right it's not sure I can you fun thing about coordination cost about common knowledge because it's not just I know that you know and you know that I know but I know that you know that I know that I know that you know and and in the way is true because and in a way we all know there is reflexive knowledge so like I mean you know we agree about that yeah exactly so that's the point the point is that it is the the thing that makes it legitimate is also the fact that it becomes a coordination problem to change the legitimation because of the expanding small group it's easy to so the comment there is basically like oh look one of the things that gives legitimacy to the process is how much of a royal pain in the butt it is to change it to another one which i think is completely true right I mean one of the reasons why we agree to you something is because like oh we've been using it for a while and it seems to work and why it's screw with something that works and that's more or why screw something when it can take so much work just switch something else and see in that case is increased in in regimes because if you create bureaucratic costs really high then people would not be able to coordinate together to change the regime easy legitimacy of the regime that's why in some cases bureaucracy is actually like knowingly the dictator would allow and actually motivate people to bureaucracy more to the point ironically when the bureaucracy so much that dictator himself is trapped in his own bureaucracy that also happens I mean I know yeah it sounds like the RPC can increase coordination costs yes remember yeah I just take the microphone from a new camp so in the case of an authoritarian regime in which anyone that actually even state just communicate somehow that it's against that regime will be immediately killed will you consider this to be a legitimate system just because no one is actually expressing themselves as long as there's common knowledge that people agree to use rather than replace it then like yes I consider legitimate however the existence of that common knowledge is actually may not be that easy to detect right so like you people will act like they all agree to go along with it for their own safety even though they all kind of know that they want to replace it so if they know they want to replace it then I think legitimacy has definitely been undermined but if they kind of are you know oppressed but not but but like they don't have this shared intention to replace the process then I think that like it's not then I don't think it's illegitimate however I should also note that like this is all defined with respect to the particular group that has the common knowledge and question right so to us looking externally we could be like oh that's alleged because it doesn't meet our standards right and and I think that that is like a completely you know completely valid instinct to say look the you know like this and but the instinct they're basically essentially is look we intend for this thing to be either abandoned or replaced or like you know changed so much that it's you know yes so the group that we're talking about is maybe different right so like you talked about the college within-subjects common knowledge within like the selected common are between you know from like the international community is very different kind of story No so so so the question is like Oh am i you I mean sustainability by legitimacy and and I don't I really don't think so right because I think III think that like you can have an illegitimate government government in its process be sustainable just because people you know have this intention that they're gonna abandon or replace it doesn't mean that they're gonna be able to for a number of reasons I think sustainability is weaker I think I think legitimate or it's I think I think I think you have like you know sustainability without legitimacy and legitimacy without sustainability you know I don't think so I don't think it is at all and I think that actually this the definition of legitimacy lines extremely well with a political science definition of legitimacy that you can just like firing on Wikipedia where there's not sustainability but they call it agreement to use the teacher to like like the subject to or to use a particular government well yeah of course I think common knowledge is a better definition because the agreement is kind of I think more like a these kind of heuristic so somehow like I actually don't think this is that controversial I think the reason why we feel so sometimes people feel like oh this can't quite be right is because they refuse to acknowledge the legitimacy of a system that contradicts their principles right they they have like they have like this kind of they they don't agree that people should be subjects they don't agree that people should be treated badly and therefore they think that like any system that where these things are okay or is something that should be a ban in a replaced and so I think that and and they and they and they probably have common knowledge like themselves as like people who say care about human dignity and and I think that like that is totally normal right I think actually in the modern world like legitimacy a lot of the time comes from you know consent and like effectiveness at representation however I I don't want to put it in my definitions because I think that like it's kind of I mean it's a much more complicated it gets much more complicated right so far this definition is very very basic if I start to add stuff like people being treated nicely stuff like fairness this these things blow up and it becomes a much much less come much more cumbersome language for talking about governance so like I like I really think that like legitimacy in this sense of like the common knowledge that we're going to use a process is a very useful concept you know even if like I'm not bringing with it the things like fairness that you're used to being associated with legitimacy I think only because we live in like such a sophisticated time where like the legal tradition has developed to a certain point and we're like you know our kind of like idea of justice has developed to a certain point our deal so I think there the fact that legitimacy requires all these nice things today is a is an artifact of our like level of like you know like or a level of sophistication of the culture around governance sorry you need to have um so this so the need the need for fairness is okay yeah sure but like like when I say fairness I meant like really like procedural fairness and stuff like like I'm much more like question about like oh the law being applied equally to all people and you know fairness is kind of a like very difficult to define relative to like this kind of stuff also but so again what I'm saying is like look well the degeneracy often comes with a bunch of stuff that you recognize as making things legitimate but like the fundamental thing of legitimacy is is the common laws that we agreed not to replace our ban in the process that's that's I don't even I'm not quite work with this out with this definition of legitimacy he's asking if it's a legitimate definition of legitimacy but it's this definition is not a governance process so all right do you have any more any more objections I'm happy to hear no not for me one news not from you [Music] okay so the like according to Wikipedia allegedly the like the definition involves not just agreement but also consent and I would also like I still kind of believe that that's all that sounds like incredibly modern right that's an incredibly modern notion of at least modern in the sense of like Oh more than like it's not like ten thousand years old it's more like it's like less than ten thousand years old so some somehow like a bow or what I think think like people needed to coordinate and have legitimacy around shared decision-making processes for much much longer than then we valued liked and needed consent for that for that kind of agreement and and I think but I think one of the main reasons why I'm like why resist so strongly this idea of making adding this definition is that the notion of consent is I think like a you know like defined for a much more narrow scope of phenomena than the notion of common knowledge and I'm trying to be kind of as abstract as possible with my definitions and so if you if you if you require you don't like it but the more you require that your agents are able to express the less kinds of coordination systems we can talk about legitimacy for but you know I mean I mean at the end of the day we can just make up new terms for different things and so if this ends up being a problem for everyone we can we can we can call we can call we can we can call this like something else but I think that like it is still a very helpful definition so can we move on what do you think all right so apathy is kind of when participants decide to like stop participating usually because of the belief that or like some kind of belief that like their their participation isn't helping or isn't going to lead to the outcomes that they want disenfranchisement is when like participants are removed from their role in governance so disenfranchisement you know can happen without an apathy apathy can happen without disenfranchisement legitimacy is on the other hand Oh oh it did oh I'm sorry so well let me finish this one another back formalization so illegitimate see is kind of the opposite of legitimacy like a strong opposite not just not legitimacy but it's common knowledge that we agreed to or that we will I mean tend to abandon or replace a process it's kind of like oh we know that we should replace it we know that like this sucks and we and we and we and we have like a distributive state of knowledge that the process is out of date or some for some reason used to be either replaced or abandoned and legitimacy can cause certain types of problems for governance because like participants may not agree to participate in an illegitimate process things downstream from the governance process might not decide it didn't might decide not to accept the like and what kind of relay the decisions so illegitimate see is kind of the you know what happens when things lose legitimacy too and like and then continue become like losing legitimacy come to the point where people actually want to like replace them any questions issues yeah no no so the disenfranchisement can happen without common knowledge so so so formalization is kind of the process by which we like create or or like the formalization is like a description or a model or like some kind of easy to communicate encoding of a governance of a decision-making process so formal people love their love formalizing formalized governance models because they're easy to communicate and share right well the lots of governance processes that arise on more of an ad-hoc basis end up being less easy to communicate and share than the ones that are formalized and the institutionalization of a process is kind of when you when you take a formal model or like a Mont and you and you and you like by kind of deliberate choice whether you know collective or more declarative by like someone who has like a relevant like Authority will or like the relevant like you know mmm ability to create a legitimate process or like something that people actually want to you will actually agree to use certain when you when you when you when you create a governance process out of model out of a formalization or you have like some you you know regardless of how formal it is you have a governance model and you go and try to bring it about in the world I'm calling this like the institutionalization of the model it's [Music] people love for some reason to come up with formal models and institutionalize them the the formal formality makes it easy to share about which can make it easier to build to build legitimacy because people will have like some ability to understand why something how something works and and and it'll be easier perhaps to be for it to be fair because people will know how to interface with it because there's like a description that everyone goes with however the institutionalization of a formal model and creating just a like a focal point around the model and so like people will both when they're doing when they're running through the process try to comply with the model and we'll expect that when they interrupted the process that it is as the model describes so the institutionalization of a formal model creates some stickiness just around the fact that everyone uses this kind of formalization to understand the governance process so when you have like a bunch of decrees about like how things are gonna are getting made that usually corresponds to the attempt to institutionalize a formal governance model sorry what definition asking about you asking about the literacy definition again ill is generously definition okay so so sure so you're asking about like self amending governance processes right you know it just has to do with the with the kind of the way the thing is in terms of intention i like-like-like intend to change it or not know whether it will or not or how exactly i think is like not really the main issue although although like you know we can get into pedantic conversations about like whether changing a process by which by like a means that is allowed in the process corresponds to the like illegitimate see if the piece of that process that gets rectified or the whole process was legitimate because we didn't replace it and I think that like I think I think that we can we can split those hairs or not and it doesn't it's not really I don't think it's gonna be this like no I'm but I still think we can talk about like the the legitimacy of the parts of the process that we want to change oh yeah of course and I mean so then don't worry we're gonna get to that in the later slide I mean and this is I mean partly one of the main the main thing about legitimacy is that it leads to behaviors associated with the attempt to abandon or replace a governance process there's revolution revolution basically represents the idea of the idea that we're gonna replace a decision-making process with another one that makes decisions on the same resource for King on the other hand is the idea that I go instead of replacing that process we're gonna create a copy of the resource that it is governing and create a new process that will govern that copy or alternatively a split splitting also corresponds to forking so imagine we were governing a bank account and then we have like some falling-out and then half the funds go to this process and half the funds go to that one I kind of also consider that forking so forking basically corresponds to either way it's like a clean splitting or copying of the thing that's being governed and the establishment of a new governance process on the forked resource or the kind of copied resource as opposed to revolution which is about replacing a governance process and having the new one make decisions on the same resources that the original one have was responsible for governing forking is fundamentally easier than revolution because revolution kind of requires that you like unseen to process and replace it with another one kind of in place whereas whereas forking with forking the initial process can continue going on and and and you know perhaps this legitimate to some subsection of the community and you know with the the other section which believes is illegitimate and fork off and form their own independent thing and it's kind of much more peaceful than than a revolution to the end of the talk and it's oh I'm about to drop like I or I think is I mean so this is like my current best definition of governance and it kind of relates to some of the questions the news is just asking so I think governance really is the process by which we maintain and establish the legitimacy of our decision-making processes so it's not actually the decision-making processes themselves it's not the actual model that we use for like counting the votes or whatever that like leads the trigger being called it's the process by which we like establish and maintain the legitimacy of those decision-making processes it's the way that we come to agree on a process the way that we deal with when that process becomes less legitimate how do we make like the way we make like a little the process by which we maintain the legitimacy of our systems which may involve things like amendments things like making changes to them but while still kind of keeping them in place and like not well rather than you know the the replacement and abandon an option where we're kind of instead of maintaining the legitimacy we kind of either failed to maintain the legitimacy or decide to abandon a process because it's become a legitimate and it's not maintainable and there's a reason why we maintain is establish the legitimacy of governance processes I mean firstly like if you if you don't establish legitimacy of a process when you when you try to when it when it then or if somehow let me rephrase that if governance process doesn't have legitimacy then people won't really believe that other people will participate in it and people won't really believe that other people will respect the decisions that it makes downstream if a governance process has established legitimacy then they will kind of have that expectation that people will use it and so participants are much less light much more likely to participate and stakeholders and people and downstream processes are much much more likely to accept the decisions made but and and and so like the maintaining of the legitimacy stops you from having to do all of this work of replacing the process with something else maintaining the legitimacy of governance involves stuff like for example you know the consent of the participants and the stakeholders at least in a system where we do where we do have like you know some degree of Independence and like dignity and kind of like you know the right to be kind of to like have our intentions matter or our consent matter so so you know I think and I think that actually like if you understand that governance is this process as opposed to a design of the process of like just designing the decision-making processes and I think that you understand actually why governance is a much more political process a much more social process than actually just coming up with decision-making rules right if you can come up with a decision-making role by solving some math problems and you might have a really elegant rule but if you can't get people to agree to use it I don't think you have you know I mean I don't think you really accomplish the goal of governance right somehow governance isn't just about just coming up with decision making processes it's about how do we collectively agree on what decision-making processes you're going to use and how do we maintain the legitimacy of these of these dorms and processes over time so that to me like you know it's obviously we started off kind of governance as a decision-making process and then we talked about governance as a coordination game where the participants coordinate to produce different outcomes and now we're saying oh look actually the the real game here is the participants and the stakeholders establishing and maintaining the legitimacy of these governance processes that they later coordinate around so I see some questions but the I just have like one more slide so I'm just gonna go through so you know kind of let's act like it right you know it to act like governance as opposed for establishing maintaining legitimacy means stuff like for example don't try to unilaterally declare what the - model will be right even if you think you come up with the best model ever if you just announce this is gonna be the government's model like probably that's not the kind of public debate that will create legitimacy if you like there's the behavior that you engage in when you're designing a decision-making process is very different than the behavior occasion when you're trying to establish the legitimacy of some process and like were you and you don't and somehow the question about which one you pick in this framing is much less consequential than in the original framing we were just talking about oh it's just a decision-making process because a lot of the time for example more complicated decision making process will be harder to establish legitimacy around and that is something that is really a function of like people's ability to agree to stuff more than the pro that are the properties of decision making processes so thanks a lot really appreciate it and I'll do Q&A for the rest of the time please he's using mic for the question for the stream okay so wait wait for the microphone thanks Randy if I had been asked just before this toke to give a quick definition of governance you know I would have said it's just mechanical set of rules over the you know ruling over the decision-making processes for coordinating organizations or distributed systems so if you go back to two slides before your definition of governance you know which I find very interesting I would have said you know oh it's just a great definition of politics and you know but after thinking it over I think you know my sense is that it's an enriched version of definition of governance and it's a restricted version of a definition of politics you know I find it very interesting but you know just a bit like when you were describing legitimacy I felt that you know you and I agree with I can understand that approach you sort of cut off the ethical side of it and I just said you know legitimate if I understood correctly you know legitimacy is really for me the the fact that you know it's common knowledge that we expect the current rules to you know to continue you know for me this definition is good governance is sort of enriched in you know so you're going a bit the other way around but anyway one question I had also was with respect to you know metrics of legitimacy or you know or the converse of legitimacy because what I find interesting is just an intuitive but in the way that you know the governance rules and evolve over a complex distributed system whether it's a nation a community of so forth is that I think you know the the metrics of legitimacy matter like I've never seen you know any model where you sort of you take into account the fact that you know the way we behave like you know an organization you know like if you think your boss is going to be replaced if you think with a hundred percent probability that is going to be replaced you don't behave the same as if you you know you think that it's gonna be around for like two years or you know like forever so you know introducing some metric of whether it's through probability or you know of legitimacy in the in the system have you have you done that or have you seen anything you know around that ID well firstly like oh great comments thank you I think I personally never thought too much about measuring legitimacy I have made a cruder poll once or asked people if the EP process was legitimate and it turns out that like I think it was like Oh like like 70/30 yes or something like that but I think I think like measuring legitimacy and providing measures actually can either improve or decree the legitimacy of the system because they create a kind of common knowledge about people's like state at least a please share the statistics around or like publish them in a public place and so I think it is interesting although not as simple like not necessarily well advised if you're trying to establish legitimacy of a process right if you if you just create a new process like you probably don't want to ask people with determent right away because if they all say no then they might just believe that it's not gonna be legitimate ever because they once saw each other say now so is I guess I I think they're like it's interesting but also and also actually may play into the process itself okay so thank you very much I think that your talk bring a lot of questioning and I wonder if those questioning will not be less I would say long if you just restrict the title of your of your talk on governance on community of people who wants to be part of a community because of course when you are talking about legit legitimacy of authority on regimes and if you map that to political worlds I mean you have no choice if you're born in the country of course you have to accept if you want to go out of the country you have to accept what's going on so isn't it more for the governance of community like the crypto community I mean III would hope not I mean so somehow like I I I hope that these and I kind of expect and think that these definitions will apply super broadly including cases where the governess decisions that don't represent things that people voluntarily consent to I think that like these things like apply also in the area of political science however I imagine in political science a lot of things there's probably there's probably like the me like and like emphasis and considerations are probably more applied whereas I'm trying to be as abstract as possible and I think the it's partly just like my own style but I always try to be as distract as possible but like you know that also makes me feel like oh if I can't cover every piece of governance then I'm not like oh I haven't really done a great extra option but however like Oh tactically like oh I think you might be right because like I don't want to have to say that Arthur authoritarian governance can be can be Kimmi legitimate just just because hypothetically like you could you could have a common state of knowledge that we will go along with the authoritarian and so you know I guess it's a struggle and I'm gonna I'm gonna keep and keep having it yeah I think it's great definition of governance how do you uh I think it's critical to handle the focal points when it comes to taking decisions on a governance model so do you have any some sort of best practices to handle those focal points yeah sure so actually my my point of view on focal points and I'm gonna be a little bit normative even though I said it wouldn't I guess I was just for the duration of the talk I think like focal points and defaults are actually really bad and we should avoid them and get rid of them I think like the default to nothing the default to go with an upgrade will mean that and operate note operators and developers will have you know less of a requirement for action and less of the necessity to participate in the governance process so like them and which means that they have become they could basically they risk becoming solely disenfranchised by norms so imagine if we have a focal point of always installing the latest release of the software implementation and that focal point becomes like strong and everyone just expects everyone to do it and it becomes like a norm and so then you you even though you might disagree with it upgrade know that everyone else would just go with it and so you don't even bother and then everyone else could be in the same situation where they think that no one else everyone else would go with it even no one likes it but if you didn't have a focal point of following and if you all were at every time there was a patch you know you had to like manually say that you wanted it and there wasn't a focal point then there would be kind of a forced more arduous discussion which I think provides a nice check and balance so like to me actually I think like I think defaults and governance are incredibly dangerous and I would hope that we can remove them but that is a controversial opinion hey bud so sorry over here now thank you so you've made a choice in kind of in this description to use you know pre-existing terminology to set out maybe some more abstract versions of those I think we saw today that that kind of runs into a problem where you know people already have an idea of like what I think legitimacy is and vlad's telling me that it's actually something more abstract than what I think like ought to be the case and so I wonder about that choice about like do you think that it's the best way to try and help this community understand governance and get their heads around it to use existing terminology or to create kind of different terms that more established clarity that this is a different thing than what you might know already like we run into this problem over and over again I'm like the is ought or like normative versus positive statements of like you're just trying to say what is the case that we can observe and other people are taking that to me and you're saying like what should be the case and what ought to be morally the case yeah I mean you know you're not the only one with that concern obviously and I and I share it but I don't think actually I would lose very much if I just added the idea of like consent and like you know into into the definition of legitimacy I don't think that I would lose much in fact like I think it would just be just fine for both people's use cases in terms of blockchain space however it like adds terms to the definition and like I think this notion of legitimacy is quite useful and I think it's like it has like a real practical value just in itself and I think like you know I I think they're like the term Maps like you know kind of close enough onto existing terminology I don't I don't feel too guilty about it but you know if this continues to be an issue I might come up with a different word or I might like add more cumbersome stuff like intention to all my models oh one question ball in your definition governance is the process of establishing and maintaining the legitimacy of a decision-making process it doesn't imply that governance is not decision-making it doesn't know it doesn't imply that it's somehow like the gentleman there in the middle made a really good point this kind of represents an enriching of the governance model somehow oh like this is itself still somehow this is making process right it's just it's just a little more meta because instead of deciding how we're gonna make decisions so it's don't think Colin govern it's a decision-making process is necessarily a mistake however I think that like it's it's a mistake to lose focus of the fact that the decision-making processes that we come up with have to actually be established as legitimate for them to be adopted yeah so when when there's a legitimacy or when there's an expectation legitimacy is also when this legitimacy in the talk of a common knowledge that it would not be abandoned or changed so there's also the other side which is when you accept something to be legitimate you also there is also common knowledge what would happen if you do not consider to be legitimate there may be yeah so it which is almost always I mean there's the start of the definitions right I mean it's not part of the definition might be because kind of hints at it without saying it because for example in regime it's very easy like you know if you don't consider legitimate you will be able to violence of course in democratic process is this not because if you know that if you don't consider legitimate you are a law-abiding citizens you have to like for example I don't agree with the environmental policies of Trump administration but so so that's my legislation with legitimacy are more like not like polar opposites but more like a spectrum and when there's a common knowledge around it the other thing also always is there for people to be less disenfranchised or motivated to accept it you know what I mean oh my god I can't I can't I can't follow your sentence as a news one more question over here she's like mumbling I got I got it so forth like for the first half and then I was like so my point is that Bend is a common knowledge about legitimacy you understood you don't understand me I don't understand my accent is there is there common knowledge about legitimacy knowing there's a common knowledge that this is the norm oh you're asking like oh like do we have common knowledge that we have common knowledge yeah of course Norman knowledge of what would happen if you consider it now so there might not necessarily be common knowledge of the consequences of of announcing that a governance process is illegitimate there was one more question over here somewhere maybe oh wow okay I had a question I guess do you think that the price of Bitcoin or etherium is an accurate reflection no particularly no not at all because I don't think that like the prices I mean I mean basically I would say like okay the the price of Bitcoin relative to the price of Bitcoin cash is the is a measure of the legitimacy of Bitcoin cores governance process over a Bitcoin caches governance process okay but like if you talk about the like absolute valuations I mean I think I don't think that these valuations are indicative of anything other than people's insanity and level of hype you know I may be biased so there was a slide which mentioned that there are two ways of handling one was foking and one was resolution revolution revolución sorry so foking was the easier option to achieve but I guess that breaks the network effect and the real power of blocking is in the network so should focus actually be an option yeah so actually I am it well absolutely is an auction as a matter of fact the but the reality also is that because the network effects the the forking and the fragmenting won't go too much it won't go too far and there's gonna be basically well there might be a lot of but there's gonna be a kind of oligopolistic or like highly concentrated market where like some of them have a lot of weight my favorite analogy for this is like it looking at forks in the history of religion and they where they really have done it all like there's lots of forks like contentious non-contentious hard fork soft forks and and and and they've and there's been like just like years and years of like forking and the kid you have network effects and there are switching costs and and and they're kind of you know communities that like governed shared resources kind of analogous and their forks really because you take like you know literally like oh it's the Bible also for this woman for that one and for this one but like their interpretation or some key parts of their doctrine it'll be different and what you find in my history religion is like oh look and we still have religions with network effects and I think that like we should expect the same thing with watching I I'm supposed to have thanks that was really interesting so according to this definition of governance it seems that like a valid path to government governance is to suppress the knowledge of disagreement in the process thus by preventing common knowledge in the like lack of faith in the process from developing is that correct people do it all the time yeah absolutely it's not only like expected but visible right I mean somehow somehow like people are a lot of the time you know engaged in this game of you know who finds out when and they kind of try to like architect like kind of rule things out in a way that like everyone has kind of agreement at every step and by the time the people who would disagree would find out there's already such a large group agreeing that they're kind of forced to go over so I think I think not only is it like to be expected I think it's to be expected I don't I don't know but I also don't think it's to be encouraged all right I think is actually kind of manipulative and strategic as opposed to open and transparent and I think that if you can achieve consensus and using only public debate then that would be actually horrible I mean I think Eve more legitimacy than if you did this kind of process of like escalating agreement between like a kind of set that's like kind of private until it's big enough to win so thanks for the talk I really enjoy discussing abstract ideas and definitions but as we can see it brings lots of discussion and my question here is that it's good to discuss what we're going to discuss like the definitions of what we're going to do in the future but then where's this balance of going into understanding what we're doing but also actually designing the governing process and where is this with where's this time to stop and say okay this is what we need to do I mean that's a great question I it's hard for me to say exactly like how we're gonna be able to agree on governance processes it's probably not gonna be easy I'm trying to apply this definition so my miners govern their mining hardware and everybody did I Network that I see they if I support their governance I see that they make a decision as well then in troubles to apply this definition to separate separate government Porter from making yes somehow there's no definition here about the actual effectiveness of the governance process or like it's actually a my ability to make good decisions Oh or like its alignment with our philosophy or it's like somehow oh this is too vague to to tell us about like you know like like like like how we'll be sure that the that they make good decisions although often I find governance processes also don't can't really be that convincing it's more really the culture and the norms that support them that end up guaranteeing that good decisions are made have you thought about incentivizing the whole governance process for I mean it can be any process eventually but the process of incentivizing it well I mean I I have thought about incentivizing stuff a lot and it's very hard especially when you don't know precisely what behavior you want from your parties and in this case like just looking it's gonna be hard for the like just abstractly to tell you know what are the kinds of decisions that we want so I would say that like No or research there is like early stage or it's like or it's like just something that I think that I'm not too optimistic about I would say that like incent incentivizing something requires like a lot of domain-specific knowledge for to have like Eddy really strong guarantees that the stuff you're incentivizing is the behavior you want how do I handle bad acts I mean you need to be able to identify what bad acts are right and so if you don't you know if you don't know what kind of behavior you expect your behavior is like the right behavior it's really hard to handle bad acts especially you know I mean I as a protocol level in on the level of like social norms it's much easier because we don't need to formalize everything in sometimes we can just like notice something's wrong and I feel it in our gut and identify why without having I've said previously okay like here is our objective function this is what we want people to do when we peel off anything except that isn't that so there's Adam I think it's much easier for us to get these kind of guarantees and have reprimands extra protocol through like social means then actually to have like a formalized set of rules which we apply mechanically which will do this it's just way harder represents I would say like a like a relatively advanced state of knowledge about like what you want people to do or how to recognize that people are doing the thing you want to do which more specifically is the relevant thing what about influencing the vote of the consensus of other people that it's quite general I mean it's probably a bad behavior but how do you detect if someone's influenced if someone else yeah it's hard what if you detect well I mean if you detect it with a protocol or do you mean was like as humans I mean both in both cases if you somehow are able to detect influence one person maybe even that isn't it so it's not necessarily clear that like influence of one person or another is necessarily behavior we want to discourage right this probably is in some cases and not in others depending on do you like maybe the intention may be the like like nature of the influence like it's one thing to buy a vote and another to have someone be inspired by your opinions and follow you right so it's really not easy it's really not easy to incentivize these things all right one more from Peru bear what do you think she might as well end on a note about how bad my definitions are I think I just wanted to ask a mate a question because if the governance is the process of establishing maintain the legitimacy of decision-making processes then how do you actually establish the legitimacy of governance in that level I mean that's a great question I mean I think the answer I think somehow like actually if you if I do allow the idea of like intention then I then I think I could provide somewhat of an answer which is like oh if we come up with something that serves all of our intentions then and we can come and we can you know come to common knowledge on that fact then I think it could be we could all be happy with it just on an individual basis if it goes through and therefore even though we don't have a collective process for agreeing on the legitimacy because we are because of our because basically our incentives allow it we can independently make the choices that can bring about this this legitimacy will that if there is an illegitimate process by which I'm establishing and maintaining the legitimacy of a decision-making process do you actually consider that the outcome of this governance process even if illegitimate itself will actually lead to a decision-making process that can be considered to be legit well I mean I mean is if it is I mean I mean trivially yes in the sense of the definitions allow it so like it's really possible for a new legitimate process to create a legitimate one there's no there's no contradiction there although it seems weird the the question of what it takes to like so imagine you oh well you can always keep going meta on this question right and so like I'm some well at some point you're gonna bottom out to somewhere where you don't have a process and there I think is like where he kind of is interesting I'm like okay well if we do have a legitimate process for establishing and maintaining legitimacy of our proceeds then like oh well we maybe shouldn't maybe that's bad and maybe we should replace that process to something legitimate right and I think that like actually like you know like the whole like consent thing is super compelling because actually people's individual like a bottom as a way to bottom it out because people will be able to individually make decisions to go with it even without a process that they agreed to for how to make those decisions because it individually makes sense for them and so they don't have to like you know agree to an output of something that they don't they didn't it's some process they don't they don't believe microphone yeah I know we have to yeah well thanks forever thanks everyone