On Chain Governance Panel with Jorge Izquierdo and Aron Fischer

cool all right we're gonna change gears slightly up next we have a sort of a panel conversation on the topic of governance using blockchains for governance its moderated by gabriel who is a lawyer from the web 3 foundation so we have some legitimate legal experience as part of the conversation which I appreciate and our two panelists are Aaron from Colony and Jorge from Aragon are you guys all ready to go all right come on up thanks ladies okay cool right welcome so yeah no real need for intros for you guys and certainly not for colony and Aragon I'm gonna ask questions that try and shone a light on some of the design processes and design thinking that you guys have have been going through how your particular governance solutions are going to come into being what the principles that have been foremost in in designing those and I want to start with maybe a framing and tell me if this is right or wrong so web 3 obviously working on polka dot with parity and there is a non chain governance element to that in that it's an interoperability protocol that in order to work has to have some governance right but the governance is in service of the protocol right to make sure it it exists and evolves you guys are in the business of designing governance frameworks for the use of any number of use cases and deployments in the world so how what design principles when you're sitting there with your blank sheet of paper have many months and years ago did you apply in that first instance well I Argan started like with a very specific use case like we were trying to put like your typical Silicon Valley type start up in the on chain and we actually built like very traditional governance models into the head like just like sir Sir holder type Odin and when we kept advancing we realized that we could actually have like a lot of different organizations that organized him very very different things and that it was like not the best idea to tied our entire system around a particular way of organizing because we really don't know what how organizing and governance and chain is going to go so then we went back to the to determine paper and we we decided on building this this new system called are gonna ways that it's built on top of the assumption that we really know what the best governance models are going to be but we are living it open so that we can be like very interesting governance models and talk without having to change the entire infrastructure yeah so at Colony one of the things we thought about very early is that even though we can use the blockchain to cook up all kinds of complicated procedures of voting's and what-have-you when it comes to like day-to-day decision-making in let's say a club or a company it's not all adversarial and you don't always have to vote on everything we wanted it to be fluid and permissive and easy to get things done like the the procedure shouldn't get in the way and you know occupy too much of your time so we try to find the right balance where it's really easy to get things done to work together on projects and sort of use the on chain governance well the smart contracts that can enforce a procedure only as needed so only when there's dispute and so to try to write find the right balance between that so they both have a sort of a thin desire as a thin profile as possible to achieve the ends that you're trying to produce for people without necessarily being too dogmatic about how they achieve those ends is that yeah it should be very flexible to have accommodate many different groups many different well you should be able to work as you want and it should never stand that the governance framework chef should never stand in the way of consensus like if everyone agrees to do something then it should be possible to do that so it really is only so these smart contracts and one of these frameworks they should be able to help you move along you know sometimes you need some structure just to get a lot of people to move in a certain direction but it should never be a hindrance so there should always be a way to you know change whatever you've done before and go off in a different direction if there's wide consensus that you want to do that so yeah a smart contract can force procedure but that's sort of the only thing they can do everything else has to come from from the peep using the procedure so we still need people to communicate to talk to each other to have video chats whatever an on chain boat is not a decision-making procedure just a procedure that records a decision that has been made in you know in conversations and outside of the framework okay so in terms of procedure so the the law which is a background from which I come has lots of them it likes them to fetishizes them in some ways and there are tests for whether a decision was made well in certainly in England and Wales the jurisdiction I come from whereby if you haven't followed certain procedures the decision is reviewable right and to some extent if you are therefore making decisions which are not reviewable so in other words if you followed the non bias rule if you followed the rule to be way only good evidence and kick out about evidence in the process of decision making that's the formula for making good decisions other ten commandments of decision making that on chain governance has something particular to say about or do you borrow from these kind of existing scholastic or practical examples when you're building yeah I think something that pure on chain governance if you have an organization full of pseudonymous entities something that you means from the from the traditional world is that because of everything is happening this my contracts and if you follow the procedure everything can happen but there can be some sort of decisions for example if you if you have a traditional company there cannot be a majority boat to kick out so minority and remove their research from them and this is something that were missing because like if if the majority in an organization can do basically everything you need to put some processes in place so that these minorities are protected that in the real world you you have protected you're going to have people do these sort of things so for these they are like basically to two different solutions that you can go with one is allowing people to exit so if the if the minority is like feeling oppressed by D by the majority if for a decision is made they can exit and take their proportional part of the of the assets of the organization with them that that protects them against like a majority and then you have the the other aspect of it that is okay you can have like you can add human subjectivity into the mix that a smart contract will never be able to tell you if the payment that you're about to do is like given a share to all the all the people that both voted YES or it's actually a quote unquote legit investment for the for the organization so for that we feel very strongly that there's the need for like a human subjectivity or I call or like a decentralized Court so that if there's a minority that that is feeling oppressed and you cannot go to like a traditional court with a decentralized organization this they can rely on this entity to protect them from the from the oppressive minority said that at that point that these concepts of you know what has weather what has gone on is fair and whether the right considerations be given that's where these come in is it yeah I mean this is like to evaluate some sort of like rules that you cannot encode in in a smart contract because for the smart contract it's just sending sending money someplace or another but it's like you cannot evaluate what it's what is going on in this Marc contract so do you need like some sort of human evaluation of like some written rules in a paper or a PDF that everyone agreed to and said okay we are this is not allowed even if the this my contract theoretically allows it we're not going to do that and if this happens someone can raise their voice and have a human like a set of humans evaluate that and have a process for stopping that from happening yeah I mean I jump in on that you said like you can check that law was passed according to the correct procedure and if it wasn't you can raise procedural points now that's not really the problem with smart contracts because the one thing they can ensure is that procedure gets executed as written regardless of content but if you have agreements you know so if more lofty goals that in a majority whenever you know do some I mean you don't want to have a dictatorship for the majority or so suppose you have some code of and some kind of a decision gets made that's according to procedure looks correct but is against some rules that you sort of thought the community I'd agreed to then it's not a procedural question so we can ask is there an arbitration or can there be a certain appeal and so if you have if your smart contracts are coded to have an appeals clause that somebody can trigger and then the decision goes into arbitration then that procedure will be followed so to go to your point yes there is a space for this kind of reviewing decisions as they're made but not for procedural questions but for content and in colony we do have a lot of these kind of ideas of you can appeal to a group to arbitrate on a decision and you know unlike in a company where you can appeal to a higher boss you can just appeal to a bigger and bigger group of people so the appeals process is social but yeah in general we can imagine that any decision making governance procedure in on etherium could have these exit clauses which if triggered take some decision over into an appeals process and you could have arbitration as a service you can plug into other governance systems but yeah it's never for procedural questions because that's the one thing we can we can ensure that procedure is followed including the arbitration procedure okay so then thinking about how I mean we we didn't start this conversation with a question of what governance is and boring interminable issue because so many different ways to take a cross-section of this of what we're actually discussing discussing but let's say we've agreed that you know there is there are there is on chain governance and there are going to be different approaches to it and so those paradigms do you have in mind a way for applying value judgments to you know comparing these different paradigms and if so what would those judgments be based on so you know you could say that are good for a governance system would be the way that it it maintains integrity of the original community or you could say this is a particularly good governance system because it makes nuanced decisions in away other the things that strike you is better or worse in in in the design of these things I mean I think the two most important aspects are like one is the process like are the people involved think the process is legitimate and the second one like are the decisions that are being made by the governance process like good for the organization or the initial charter that that organization wanted to make so I think that measuring indies in these two variables are the most important aspects that you can do by the in terms of evaluating whether a governance process is good or bad I think you can only you can only test it in practice like you can write you can write a lot of papers and theorize how are people going to behave if people are going to make good decisions but the truth is like you can only experiment with it see how how people react see when there's a disagreement when there's a conflict how how it gets resolved and whether the decisions that the organization is making through the governance process are actually good decisions and the organization flourishes I mean I'd say there's two things you should look at once these these were launched and people are using them as one are these tools being used to actually get stuff done like are they can people use them to you know run a company run a colony and that connection has great projects and produces work and the other thing I'd really look for is people who lose contentious decisions do they stay within the system so does the mint as a system retain the JIT a'mma see even to those people who have lost so you know I think there might be systems where if I lose I'd say you know this whole thing was rigged and unfair and I don't want to be part of this anymore or you might have the feeling you know it was fair I lost but you know we'll move on so those are the points we really want to look hone in on look for in in their projects once they're launched because that that sort of way the signal would be first be that something's amiss if people don't feel that you know it's it's okay to lose but the process itself is still legitimate notwithstanding the fact that you lost your your way was not voted you respect the process sufficiently well to say I'll stick with next I don't know propose different proposals and and write it that way yeah when you don't go shouting for hard Forks every time you lose right right legitimacy is sort of the acknowledgement that you will continue to use this decision-making procedure right so on the shouting point then you know we it's probably a fuzzy barrier between what is on chain what is it I mean it's actually very hard distinction but you know when you talk about a governance paradigm generally there's there very clearly smart contract governed parts of it and then there's the shouting bits and the fuzzy bits around the side what is the role for you know people in the meat space and how far can you isolate these - how far is it desirable to isolate these these - from each other I'd say one of the things that we don't really talk about enough in this governance space is how to actually talk to each other because we like talking about procedure and the stuff that you can encode in smart contracts but sort of the role of a plenum of a parliament of a physical space where you talk to each other or even the online space where we where we talk to each other to come to decisions we know that reddit is terrible we know that Twitter is terrible but we don't really know what's good so if we're working in a small team and a lot of us are in small distributed teams we probably have you know a daily stand-up channel or a weekly video chat and once or twice a year we meet in person because it's so important to talk to each other because that's when you you know learn about in these conferences you talk to people you learn about other opinions that you would never have learned online even though there's plenty of material out there it's just too much noise and so to complement any kind of procedure that we could encode in smart contracts there is this whole other space of talking to each other and really debating issues so we can come to agreement you know come making a decision rather than just recording it and that's a really important aspect and it's kind of I understand why it's not so present in the discussion because we're talking about how to implement the tools and the smart contract but sort of always assume there's this other half of oh well we already know I mean we already know how to have video chat so we don't need to rewrite that we already know how to do calls and how to coordinate a team but the larger and larger the project gets and the largely a governance problem the bigger that issue is because the more distorting a few votes vote if voices can be on Reddit and Twitter and the more important is to have like large plenary discussions even structured ones yeah so I mean that makes conferences like these really important even though you know they don't leave a trace on the blockchain for like a governance procedure they the Parliament has two functions it you talk about things and then you execute the procedure to vote on them and like we need to we need to encompass both and so a ip0 for example is a good example of something Aragon was behind the organ or one of the co-organizers of that I think and that was that was very very good to talk about like in particular one very contentious issue on only theory I mean I know and it's it's I think as Aaron was saying it's very clear that the online platforms that were using for like communicating and it look it's it's not an accurate representation of what people what people are thinking and they're definitely not a place where ideas are like evolving and people are trying to talk to each other and like try to come to consensus on some ideas everyone's just like talking they're talking to book like being very very very aggressive so yeah I totally agree that we definitely need a place for for actually having these dialogues about like governance or decisions and where people can try to like convince others but like you know in a very nice way you know like I'm in this stream during this stream and we're just like shouting at each other but that doesn't go that doesn't go anywhere and this is something that I think we we could do a way better about it and almost everyone is focusing on the decision-making part like a tea that should be should be the end but not like okay let's talk about the issue and hear everyone's opinions now yeah and maybe you did mention the role of neat space and people and so I'd like to touch on a different direction that a lot of the time of my writing about decentralized decision making it's always this pseudonymous entities online without real identities coming to some kind of decision and of course it's important that we talk about civil attacks and all these kind of technical issues and why but in the real world there are a lot of people who we know and who we trust and a lot of people with real world reputation that's not some number in some smart contract and when it comes to sort of the public debates the role of people like that are really important and in especially if you maybe want when we talk about arbitration then the identities of the people doing the arbitration are important and there might definitely be a role for having panels of known people that you can call to I mean if you want but sort of this idea that everything has to be pseudonymous and we were because we can't guarantee everyone's identity that means we're not allowed to work with identity at all I think that's you know a flawed dichotomy so no I don't really know where I'm going with this but I feel very strongly about like designing this system so that they work in like completely if only pseudonymous entities are interacting and you don't have to rely on knowing something someone or knowing that if they do something bad you can like sue them in a traditional court that that that is very important that we we preserve that and we don't just like the hardest case yes you've got that you can yeah but it's important that we don't design our system saying okay we can always fall back into into like human like reputation because they're like a lot of very interesting use cases for decentralized organizations in which like the people interacting in those organizations like they cannot reveal their identity or they would be in big trouble if they do so or maybe they don't even have like quote-unquote real identity that they cannot kyc with whoever is the centralized kyc provider so the mindset is almost we deny ourselves access to these meatspace options and that helps the developed drive the development is that well that's with from the security analysis you always have to assume the worst case you don't know who controls which account and who which accounts are all controlled by the same entity and make sure that they can't team up and so from yeah when you're designing your protocols that's the use case you want to secure yourself against but that doesn't mean that most the day-to-day use will be different in I mean if we're running a colony yes it might be a huge colony but it's divided into departments of the partners or domains and there could be workgroups and when you're working with like you know a small workgroup every day you know exactly who you're working with you talk to these guys every day you have conversations you don't need you know so in that small context it's it's no longer that adversarial autonomous context it will be when you go to larger and larger groups but you don't want to make you don't want to add complexity that makes it harder for people to work together when they know each other and don't need you know every possible safeguard against civil attacks when there's five people that know each other and so to fluidly allow you know the the transition between them is very important so like that so that's been one of our guiding principles make it easy to just get stuff done right but be secure in those adversarial cases when you need it but it's only as needed otherwise you get no I come from academia and if you've ever been to a university student Parliament you know how voting and procedure can be the most horrendous experiences the torturous they're slow and they never get anything done and we definitely don't want that and so we find like the governance smart contracts they should help shoehorn a bunch of people into a certain direction to get things moving forward but should never you know you don't want a vote a thon every week cool I want to go back something called hair that you said before when we were discussing value judgments about different paradigms and one of the things I think you said was how good it is for the organization or it was to that effect and I want to talk about the problem of self-destruction and ask is it is it a problem right so to compare it with a well-known governance paradigm the US Constitution no matter how sorely it is being tested right now its bearing up pretty well because it has you know sufficient checks balances it's it's thought ahead about the problem of someone coming from within through the system using the procedures and governance and then doing malicious or nefarious activity which undermines that process potentially to the point of non-existence right so you've got in the in the real world if you like you've got that example what do what doesn't change governance have to say or does it even need to say anything about the self-destruct problem that for example you know people could vote according to various you know the proposals are made they're voted on and you know before you know it the organization the Dow has changed out of all recognition leaving the original community just wondering where it went I think having the like if if that was like if everyone involved understood that that could be an outcome and the organization and sub deciding to solve distract or transform into into a totally different thing I think that that's totally fine and we should we should definitely allow that I I think like a very very important property of any governance system is that it has the ability to upgrade upgrade itself to something to something else that potentially potentially is better but yeah if we if you try to stick to okay this is the governance model that we had at the beginning and we're gonna stick it with it for D as long as the organization lives I think that that could become more of a threat than the threat of okay we can evolve how the organization is working and if that ends have been in a very dark place then then that's that's definitely unfortunate but you can always have the D safeguards that we were talking me before about protecting about very bad outcomes or abuses within the system yeah there's at least two kinds of decision right there's the kind of decision that we're using our rules to come to decisions on whatever we do every day and then there's the kind of decisions to change the rules by which we come to everyday decisions I can you know passing a law and passing a constitutional amendment to go by your example and both of those have to have certain thresholds built in like it should be much harder to change the rules and to come to a decision on something within the rules and to find the right balance it's not always easy so if it's too easy to change the rules of decision making then it's it you can easily get it just should be easier for somebody to gain an undue majority and just take over an organization if it's too hard you run the risk that that the procedure loses legitimacy because you've got an overwhelming majority that wants to go one way and the rules just say you can't because then they will just mean you called revolution on one side and hard fork on the other but it's essentially the same thing when you've been the decision-making procedure no longer allows for what you have a very wide consensus to do then it's a decision-making procedure that loses out and not you know the 90% majority so yeah those two bout those two levels definitely exists in Coney and arrogance well and it should be much harder to change the rules than to use the rules but both them have to be adaptable so Thresh so the threshold criteria for changing the rules for example would be you know particular types of majority or what are the actual concrete ideas around what are these training wheels or these safeguards actually look look like I don't know if that's these are questions that you've tackled here well the exact thresholds are not set in stone yet because that'll be a lot of empirical data that we need to determine we have a procedure for rolling out this kind of stuff where we imagine that at first any any changes to the rules are controlled by us you know when we're programming colony we're not gonna give it all to a dhow and see how it works there'll be a special colony that that runs the network and at first it'll just come up with it can come up with changes to the rules but we'll have to explicitly allow them then in the next stage the colony can come up with changes to the rules and we still retain a veto to make sure that enough you know the nothing's breaking too much and then so slowly we're ceding control to a decentralized mechanism because because we don't know if these thresholds are set right we don't know if it's a 2/3 majority or a 90% majority that you need because this kind of stuff like there's no not enough empirical data out there for how you know large groups of of crypto nerds interact right and we'll see it'll be exciting to see once we're launched just to collect all this empirical data and you know we we all came up with some thresholds for decision-making sixty percent here two thirds there eighty percent but those are you know guesswork I mean they're not arbiter they're not random choices but there are some in some said arbitrary some sense arbitrary and yeah we're we looking forward to seeing how they work and adapting as we move forward like this is version one we're rolling out it's not by any means you know the final never finished but in terms of sort of reducing the footprint is that is that does that describe arrogance process similarly or do you do to pretty different name yeah as for the individual organizations themselves like of course they can set their own thresholds for like how to how different changes in their organization might look like there's also detailing said it's it's hard to know it's impossible to know one chain whether like an upgrade to a governance model is upgrading to the same system but like without that bug that was there like some like it was like a purely technical change or was it like a change in the rules of the game and as for the big Aragon Network or Dow we're thinking it's a similar a similar way slightly burst in which like are typically named we will have the the foundation proposed changes to the to the governance of the network and then we will have veto power by the token holders signaling like hey we really don't we don't want this and then slowly transition to a fully fully decentralized model without the training wheels very interesting so let's let I'd like to move to a discrete area of governments models which is how proposals are formed and you know what sort of goes into that sausage if you like you know is it you know because the problems with plebiscites with referenda you know we know our manifold and they're usually center around the framing of the question is there some learning from colony in Aragon on on how these things might go down better or be made to function more effectively I mean like 20 proposal there's like two parts to it one is like what it's actually going to happen and you can like check that like okay if these proposal is passed then we're automatically going to execute the following action that has the default win side effect but then of course you have the like explaining of like what it's actually going going to happen why is this proposal being being made well there's very different types of proposals so I don't you know what so any any time you want to do an action and call me you're essentially proposing that you're going to do it but unless anyone objects it's just going to happen so there there there's there's there's times when the proposals just me doing something say hey I'm as agreed I'm gonna do this work and you create that and that's fine there's other instances where we really want to harness sort of a wisdom of the crowds one of the things which we've been talking about recently is budgeting like how do you collectively determine which parts of a call need how much budget and that's hard so it's not going to be that somebody just comes up with the numbers puts up a proposal and we all vote on it because the that you know maybe you'll have 20 different nobody you know it'll be very hard for people to judge which one is best so there's other games we can play to sort of tease out what the collective feeling is for which departments are how important to a colony and which ones need more funding so that's like the creating of the proposal is separate from the deciding on the proposal but you were asking my question was more and it might be my I might have conceived a bit wrong but how do you stop through an otherwise legitimate process a badly phrased or half thought through thing being put to a vote which may then have negative consequences you know other ways to specify the forms of the questions that have to be asked or I have half in mind and laying those all about this you know the Edith's process for submitting an EIP it's only if it comes in a certain form that you know it really gets the traction that's then brought into the the process is there some equivalent here that needs similar safeguarding or not so much well then that's a policy you could have and you should have that but the policing of that would have to be social at any time somebody puts a proposal that doesn't match your rules somebody just Flags it and as long as you know everyone agrees to automatically delete and reject the ones that don't pass the format and that's socially enforced because it's not something you're gonna write contracts for to parse corner you know so there is role rules there there's a role for that and there's a role for coming to you know high level agreements sort of in human language about what kind of things you want to allow or not allow but then it come it's it falls to the collective to police those rules now within colony definitely you will you can disincentivize going against those rules because every time you get you get your stuff deleted and voted down by everyone else you will lose reputation you will lose your voting weight so you can't just sort of maliciously do this over and over again to annoy people but there's no automated way to reject these kind of proposals it has to be you know the community the good thing as I was touching upon before he said when a proposal is being made like you can be sure like if these passes what are the side effects exactly in the same way that you can make sure that they the procedure was followed like point by point do you know exactly what the side effects are so like the human understandable proposal it's less somewhat less important because like you can know okay this is this is what's going to happen and there's there's there's no yeah cool well I think that almost brings us to today every time we've got time for a question or no one a question hey it's a pretty broad question but I'm just curious on both of your thoughts about how you foresee this playing out inside of the sort of regular regulatory environment sort of how do you see these entities fitting into just regular societal structures or legal structures sorry like how does it play out with the courts how does it fit into you know the German legal system I think the beautiful part of it is that it doesn't doesn't need to fit with the with the existing legal system it's something that if people are interested in using and they're it has real advantages in organizing this way people are just going to do it if you need like some sort