Sorry.

Cool. Thank you, Isaac. So my name's Tara. I'm here today with Sofia and Jamila, who are my colleagues. We're all three PhDs working on this project called Blockchain Gov. And maybe I can just kick off by giving a quick introduction to that and then Sofia and Jamea you can both just introduce yourself and Jamea is gonna kick us off with the, presentation. So Blockchain Gov is a five year research project, hosted by CNRS, which is like the national research, scientific research foundation of France and the European University Institute in Italy. It's founded by the European Commission, so the European Research Council and led by Primavera de Filpi who's one of the cofounders of Medigov and has been writing and publishing quite a lot when it comes to blockchains and governance and and DAOs in the past year. And the three of us are basically working as part of this project, which broadly explores topics related to blockchains and governance, specifically also focusing on a legal perspective as most of my colleagues in this project, apart from myself, have that type of background and and very deep knowledge when it comes to, codifying sort of governance processes and rules, especially keeping in mind this sort of legal background of, of doing so. A lot of the project is based on us doing ethnographic research, so that's like participating and observing, communities of practice in the blockchain community and how they're approaching governance. We're gonna share part of that experience just now. And then usually what we try and do is draw on these insights that we and the questions that we get to surface through the ethnographic research and try and theorize and model around that from the legal perspective. I guess just hearing from you, Isaac, and knowing in the medical community, there's also always a bunch of awesome like people with technical knowledge, and I think there's a lot to be said in the feedback round and discussion on kind of like where the intersection also of legal approaches to remedy some of the governance challenges that we identified and technical approaches can kind of like merge and bring a better picture. So, yeah, that's the project. I'm gonna pass the link to the website just now if you wanna check it out. But maybe about myself, as I said, I'm a PhD student. As part of that, I've also been contributing to medical quite actively for a good year now. My PhD research focuses mostly on the topic of exit to community and as I said I'm not a lawyer so I'm the ethnographer and as part of my research, I mostly hang out with communities who are currently in the process of transitioning to community ownership and governance and doing so in the format of a DAO, that's my specific interest. And yeah, with that, I would pass over to Sofia, maybe.

Can you hear me? Hi, everyone. It's a pleasure to be here. It's my first time joining one of MediGolf meetings. So, yes, I am a member of the BlockchainGolf and one of the four or five members that have contributed to this report. And I am a PhD candidate in legal theory and legal tech. My particular interest in my research revolves around blockchain based digital identity. So, obviously, this report has come very handy to deepen my knowledge into that area. So I can leave the floor now to my colleague Jamila.

Alright. Thanks. Could you switch to the next slide, please? Right. So to the question, what sort of dynamics led to the proof of humanity DAOs decision to fork and their crisis? As Jameel said, we kind of placed that in the context of Western liberal democracies and we identified six high level principles that we wanna relate it to. I'm gonna walk you through the first three, the first one being constituting a democracy. So we read around a little bit and usually when we think about what does it mean to constitute a democracy, it's predominantly discussed in the nation state sort of context and there people talk about a democracy has a demos, so that's kind of a group of people that have some sort of national identity like shared culture and language and whatnot. They get together and they they declare a constitution. So this constitution kind of sets out the the main principles and values and, and purpose of of whatever this, new democracy is gonna be about and and usually what they also have is some sort of institutionalized mechanisms to deal with conflict and disagreement across the community. Relating that back to how proof of humanity started and we have a little graphic here around all the stuff that was launched at the same time and the two different organisations that to me already mentioned involved in that. We found that none of these three things, so the demos, the constitution or the conflict resolution mechanisms were really in place at launch. So the first one being that, obviously, proof of humanity being a digital identity system or proof of personhood protocol, it was something that was building its membership or its its identity solution over time. So when the project launched, there was very little of the community that we see now already present. And instead we just had these two founding organizations, which is Cleras and the Democracy Earth Foundation. What's interesting to know here is that both organizations have, you know, have long standing sort of involvement in the blockchain community, but from rather different angles. So Democracy Earth has been concerned predominantly with ideas like liquid democracy and just in general implementing digital democracies and what does it mean to improve them. It's very much identifies as like an activist organization, it's deeply rooted in the Latin American context. And yeah, that's a little bit the background there, whereas CLERAS is a cooperative registered in France, which is also interested in digital democracy, but also very much involved with building blockchain based solutions for governance, particularly conflict resolution. I'm not sure if some of you may have come across the the Cleras Court system. So decentralized dispute arbitration, stuff like that, where technical values such as civil resistance and security are very, very important. So we have these two organization that's at the founding rather than the demos. And we had this project, which was kind of like to build the most civil resistant democratic proof of personhood protocol or whatever it is, which is in itself a really cool goal. But what we identified is that there was no demos, and there was also what we usually see in the constitution or what Eric Alston, also a member of Medigov discusses as in terms of what he calls an animating purpose. It was a little bit like the goal here was maybe to build, like, the first democratic DAO or the biggest decentralized identity solution, But it's it's not really you can ask the question, so what? Right? Like, what really is the animating purpose? What do we want to get to? That was not clear from the start. So there was no constitution animating purpose. There was no clearly defined demos. And there was also a very minimal governance design, minimal governance design being something that many in the blockchain community value as it abstracts away complexity, but in our report, we kind of argue that it wasn't just minimal, it was also incomplete, Incomplete in the sense that it's, for example, did not specify some sort of dispute resolution mechanisms or where do people go when we disagree, when we're not all in one page on what happens next within this project. So that wasn't present. Instead, all they had at launch was a big blog post announcing it, a snapshot page where people could vote on proposals and a and a kind of forum where you could submit them, but the all the rules around doing so and voting and whatnot remained rather murky. So our main finding here was that at the start of proof of humanity, there was mostly just these two partnering organizations who turned out to have very colliding expectations around what the animating purpose of this project should be and that the governance design at launch was not just minimal, it was actually incomplete. Can we go to the next slide, please? So another kind of key pillar of Western liberal democracies or that we found throughout our literature review was the idea of the demos and I've just touched around upon it. It's the, it's kind of the people who come together to form this democracy. In nation state literature, it's often assumed that these people have, as I mentioned, some sort of shared culture, background, language, stuff like that. And we try to translate that to what happened in the proof of humanity case. As I said, there weren't the people at start, but people started coming in incredibly fast. As Jamila said, it was like 16,000 registered members when we started the project. I think when I last checked, it was like 18,000 or something. So this community grew very, very rapidly, including like Vitalik registering and whatnot, stuff like that. And this community started coming from all sorts of part of parts of the world, which, you know, is a great thing when it comes to diversity, but which also made it difficult or which kind of like emphasize the conflicting expectations or unclear sort of way forward for the project. I think interesting to mention here is that there's especially a large Latin American community that got onboarded. So there's also a bit of a language barrier there because many of them speak Spanish, whereas a lot of the other stuff was primarily conducted in English, made it hard to communicate, and it kind of led to almost factions arriving arising. That's something we also speak about in the report. It's not quite clear what the factions are supposed to be, but the people that we spoke to and the way we observed it is that everybody perceives factions, even though it's not like Republicans, Democrats sort of any clear distinctions, but this kind of perception manifested itself unfortunately, not in like celebrating the diversity of this community, but also a little bit of mistrust. So what is the other side doing? And I don't like that and this must mistrust kind of got further emphasized, right, that's the other point, by the fact that even though this is an identity system and a proof of personal protocol where it should be quite straightforward to identify who the members are because they're in a registry, A lot of the governance discussions didn't just take place in token gated environments or through snapshot, of course, but they also happened on Telegram and, yeah, this this goes already into the third point and other forums where the mistrust was increased because people started saying like, hey, this person isn't even in the registry and they're just on this group for trolling. And then there were some like Telegram polls that were taken very seriously and but that couldn't be verified in terms of, like, who actually voted there. So the whole conception of, like, what the demos was in proof of humanity is quite murky, and it kind of led to growing mistrust. This growing mistrust, as I said, led to people not wanting to compromise with each other. We call this a convex deliberation mindset riffing off of one of Vitalik's blog posts. It's mostly the idea that you just, even though there might be aspects that we can find compromise on, people were not in the kind of mental space to want to compromise. It was just like, no, it's either or, and not we can find a shared solution. The kind of split in terms of then coming to some sort of shared agreement on the way forward was emphasized by the use of Telegram groups and the use of the kind of platforms that they chose, which, number one, always suffered from this notion that Nathan Schneider calls implicit feudalism, meaning that there was admins and owners of groups that could ban and moderate and censor certain content, which again, increase the sort of mistrust and the convex mindset. And also the aspect of here there's some like screenshots of the various telegram groups, and here there's like five of them, but overall there's like, I don't even know, there's many, many, many different telegram groups, which sort of like completely fragmented the public sphere or the space of deliberation within proof of humanity, thus making it even harder to to come to some sort of shared understanding and compromise a way forward. Over to Sofia.

Thank you very much. So another element that we looked into is the, quote, unquote, the law. Obviously, again, to recap, we're looking at western liberal democracies as the closest actual example of a democratic system. So, obviously, we'd know western liberal democracies have body of laws that comprise primary rules and secondary rules. Rules about, let's say, substance or how we have to tailor our behavior and rules about how to amend or repeal or pass other rules. So we looked into the proof of humanity, Dawa, to see how the legislative, again, quote, unquote framework and the legislation prevented or not the polarization of the community. And what we found is that the legislative framework and legislation were both incomplete and vague. Aspects that we highlighted in the beginning that were missing were not passed or discussed as time passed by. So the DAO never really had an agreement on what the animating purpose or the shared values were. Right? Was it supposed to be innovation? Was it supposed to be inclusion? Was it supposed to be both? And what did the community understand by each one of those terms? Nothing similar was passed and proposed. Similarly, there wasn't really an an explicit agreement on who the, quote, unquote, citizens of the DAO were. Were the people registering the registry? Were other enthusiasts joining the telegram groups or both? And, likewise, there wasn't an agreement on what the, say, rights and obligations of the members were. So was voting a right but also an obligation? And what about the right to information, which is something that came up in the community a lot in terms of transparency and lack thereof of the off chain practices? What about, the right to renounce my citizenship or erase myself from the registry? What happens with that? So none of these none of these things were actually passed by community members that could have prevented a lot of the heated discussions that transpired later on. Another issue that was lacking was a clear or a clearer, let's say, principle of separation of powers between the different government bodies and officials. And finally, a process to amend rules, let's call them that were constitutional or rules that were deemed more fundamental. That was also not addressed as time passed by. And and and we believe that that could have prevented the community from eventually, polarizing to the extent that it fought. Another problem was that the legislative framework and by that would mean there was a proposal called HIP five, which was sort of the rule to make rules, was pretty vague. So the DAO said for a proposal to pass from phase one to phase two, it had to gain sufficient momentum. That is what the the role was saying. And because momentum is not something very clear that we can measure, it was easily sort of weaponized with one side of another to their own convenience. So if we think about the community more and more coming closer to a football match between, like, two football teams, well, they literally weaponize these these ambiguity in the rules that existed to their own benefit. We can and that is sorry. Before we change the slide, that is what we usually refer to as law firm. Usually using a a legislative framework to sort of attack and oppose infection. So we can now change the slide. Another element is or issue is the government. So, again, going back to our historical example, we know that we have in western liberal democracies three branches of power, an executive, a legislative, and a judicial. And we wanted to see, okay. So is that is that transpiring into the law? Do we have certain bodies and officials that perform the same functions or not? And how could have they been how could have they been sort of, like, planned differently sort of to create an environment that was less polarized and would have been inviting to it, like, a deeper, let's say, deliberation and culture. And what we found is that the word government bodies and officials, but they were quite informal and partisan. This is super interesting because our intuition in the first finding was that because the governance design was quite incomplete, what would happen is they kinda like the underlying power structures would inevitably emerge, and it actually that that is actually what happened as time passed by. So you had a a mission board as one of the government bodies, which basically contained members of the two partner organizations, which was Clarus and DEF slash UEI. And, also, the people that were working, for instance, the developers or staff members that were working into kind of, like, transposing the proposals on chain or onto the smart contract or managing communication channels that were also affiliated with one or another faction. So DDAO in itself, the the interesting thing is that it behaved like a state a little bit. So you had, you know, all the members were sort of the legislative. They had the power to legislate. And the mission board had the power to interpret the law when there was any doubts, and the clearest humanity courts had the power to adjudicate in cases about the registry. And then you had again the devs and certain volunteers, which were performing administrative tasks. However, the problem was that, like I said, they were crystallizing the existing power structures, and the DAO never really hired neutral, let's say, yeah, neutral, quote, unquote, DOWL workers. It was intended to to hire them, but it never it never happened in the end. And it also didn't have any bodies or procedures to deal with things such as internal disputes. Internal disputes that were core to the governance of the DAO itself and not necessarily the registry. So we think that these could have prevented the community for from from, again, polarizing to to the degree that it did. So we can look into the next slide. And I think this is the last one. We finally looked into governance system and governance sites or surfaces. Governance sites and surfaces is actually a terminology. Surfaces that we borrowed from Saragam and Kelsey, colleagues that that we we we share many spaces of deliberations with, and this came very handy to the observations that we made. So what we found is, number one, the the governance sites or surfaces were inadequate. What do we mean by sites or or surfaces? Every every instance of governance happens somewhere. There is some sort of place in which it it takes place. Right? So western liberal democracies can be a polling station or a legislative assembly building. So in the case of the proof of humanity down, it heavily relied on off chain or non blockchain platforms. And by that, we mean they had a website, which was a a forum for deliberation, and then they used snapshot for voting, And they used to you're also heavily relied on Telegram groups for the liberation as well and polling nonbinding polls. Sorry. And Discord, a little bit not much so. But all that to say, it was very, very heavily relying on off chain and non blockchain platforms. And what was the problem? Many of these platforms were quite inaccessible or hard to navigate for many members. Right? The website wasn't that clear to understand. And Snapshot, for example, not many knew that it you didn't have to pay gas fees to vote on Snapshot. What is the problem when we design UX that is not accessible? Usually, the neutral members don't want to participate. Right? Because, like you said, it's already such an effort to try to understand it. So what you attract is kind of like a bias sample of the most heated and sort of, like, politically committed people, which ends up polarizing the community even more. And then other surfaces, we're designing ways that were kind of, like, not conductive to their purpose. So if we use Telegram for mass debate, it it is it is a bit of a mess to put it in in plain language. Right? It is not as handy as if we use Discord or Slack, for example. And, also, we can easily create a million avatars and try to influence the public opinion. And the DAO heavily relied on Telegram groups. Secondly, for instance, they use Telegram groups for nonbinding polls. And, again, you cannot ensure one person, one vote in this in this platform, which is a problem, particularly because the DAO was a democratic one. And then a snapshot well, a snapshot is not really anonymous because you can see the public address that is voting. So if you just go to the registry and you do a quick a quick search, you can actually see how each person voted. So if you already have a community that is highly polarized, well, I think that these tools were not the best ones to attract visions that were perhaps more neutral. And, also importantly, the proof of humanity that didn't dog food its own technological or, let's say, yeah, technological innovations. So this was the first registry that was human resistant, yet it didn't have any, like my colleague said, any sort of token gated space for people to deliver that were registered in the registry alone. There was no way to ensure that only the people that have submitted the biometric data were able to participate in in in debates. And, also, there was no sort of system decentralized dispute resolution systems to show any of the disputes that came up as stamp passed by. That's a a technology developed by Claros. So we found that we found that interesting that they didn't think about it, and they didn't implement their own technology for their own governance. Finally, then the governance system. So the beauty of of this case study is that it is the first large scale liquid democracy system or experiments. So a liquid democracy and I know there has been there have been many talks, I think, about liquid democracy at Medigolf. It is sort of a hybrid between representative and direct democracy. So, basically, you can either vote directly or delegate your vote to a person. What is the problem with liquid democracy? There are two inherent dangers. First one is the so called tyranny of the majority.

out Oh, okay. Okay. I'll go very quickly. So two problems. Tyranny of the majority and monopolization of voting power. So long story short, they did nothing to prevent this dangerous, and they actually materialized. Meaning, there was a very big group, the so called Argentinian group, that could pass any proposal because they had a bigger bigger numbers. And the leaders of each faction monopolized a great degree of the voting power in each instance of voting. So they didn't do anything to really prevent the dangers of legal democracy. Now final, I think. Final slide. Yes. So what does this mean? Basically, then our conclusion is what we found out is that the problem of the brief humanity DAO is that it lacked many features and procedures that are present in traditional democratic governance that could have prevented the degree of polarization in the community. And in the future, we're also working with both sides of the frog to help them constitute. So, yes, I hope that this hasn't been too long for everyone. Thank you for for listening.

Well, it seems to me first, I'm gonna start with an observation. It seems to me that the purpose should have been in the first place to develop exactly the governance structures that they were gonna need, or any other organization like them were gonna need to move forward. So the fact that they didn't have that purpose in the first place didn't bode well. The fact that they didn't see that this splinter into all these different platforms was it was gonna be a UX nightmare. It seems to me that designing something that is usable should be a main democratic objective. So, I mean, it seemed hopeless from the start looking back on it is what I'm saying. Do you not agree? Also, I wanna know more about the onboarding process. I forgot. I don't I how did they actually establish that someone was human? You know, did you just have to have a phone number? Could you go a little bit into that?

Do you want me to to go very quickly through the onboarding process as an actual community member? So, basically, it combine it's sort of like a social graph that requires submission of biometric data. So what you do is you take a photo of yourself, and you you took a photo of yourself, a video, you download MetaMask, and you send this biometric data. And when you send it, you sign with your public address of an Ethereum public address. And then other community members have to check your biometric data, the video and the picture, to see that you're not a a bot, number one. And number two, that you haven't registered before. And if they deem that you're not a bot and you haven't registered before, then you get a vouch from an another community member. And after three days, you are you are basically registered. And if not, they challenge your profile, and then the challenge goes to the Cleras humanity courts, which then deal with the case.

But you only need one member to verify you?

You only need exactly. You need one doubt, meaning from someone who is supposed to know you. Right? And then it is if anyone can verify you, what you need is is that no one challenges you in a period of time. Three days. And then you're registered.

Yeah. Let's do that. Yeah.

This is Let's hear one more question and then answer all of this. Yeah.

Yeah. I think this tags nicely onto Brooks. I was curious just in terms of any, like, methodological reflections you have of, like, course directions you had to change or things you would have done differently. And I'm thinking about this too as someone who's, like, I'm still mentally mapping out the idea of how to do this kind of research because it's so far and like, anything I've done before.

Okay.

Yep. So as Jamila said, like, both Jamila and Sofia were already kind of, like, in this community and new people and were aware of things going on. So I think that was, like, a big consideration for us or, like, that really also informed some of the methodology, like ethnography participatory ethnography being not to, you know, test a model on somebody and, like, come up with this grand theory, but really spend time, observe, participate, and experience what is going on here. So I think having both Sofia and Jamilia as part of the team was was tremendously helpful. On the other hand, this being a conflict, obviously, they were also experiencing, I guess, it's fair to say, like, some divergent sort of dynamics going on within the community. We try to supplement these experiences and observations by joining Telegram groups, following Twitter, joining a few talks, reading the articles, and all of that by really mapping a bunch of things. I think the most elaborate one in the report is the timeline. So being very, very analytical sort of in the what happened when, who did what, like, how did that relate to each other, which sort of proposals were referenced, what was the outcome, like this sort of thing. So I think this getting the the the temporal overview of the situation was very helpful. Another thing that we did that's not in the report is trying to start some sort of, like, institutional mapping on, like, what are all these different governing bodies and, like, things that are active within the proof of humanity ecosystem because it really becomes quite big. And I think, like, maybe reflecting from a methodological standpoint. Oh, and then we conducted interviews. We also had, like, group discussions with them. We started to share feedback with the community. Right? Like, so this report all went back to them. We had some sessions in both English and Spanish, with each side and trying to get feedback and and learn more about our findings. So it was it was really trying to be participatory. But reflecting back, I think what's really, really hard in observing and studying Gauss is kind of defining the fields. Like, what is it? What is it not? When is it there? Where is it at what time? Because these people are fluid and people are engaging at different times. And so so that's really, really hard. And I think making that more clear also within the report and our subsequent writing about this case, is gonna be important. Another thing that we did is got a lot of help from our MIT folks, specifically Kelsey Nabin and Eli Rennie from there, which was tremendously helpful. So I think, engaging more and pushing this engagement with other empirical researches in the Dao space is is very, very useful in trying to figure out things. Something else that we could have done that I'm thinking about now more is a a technique called situational mappings where you don't just map entities and people, but also concepts and trying to understand the relational dynamics between them. So this is not something that we did that but might have been helpful in taking this a step further. Yep.