Neighbourhoods Metagov
Metagovernance Seminar Archive | 2025-10-21 | Unknown
Speaker 1: K. I'm pleased to introduce, Michael and Emmeline to talk about, the neighborhoods project. Take it away.
Top Keywords
- holochain 0.011
- neighborhoods 0.010
- reputation 0.008
- basically 0.007
- state 0.007
- social 0.007
- unreviewed 0.006
- state machine 0.005
- sense making 0.005
- accepted 0.005
- making 0.005
- sense 0.005
Transcript
Speaker 1
0:00 – 0:00
K. I'm pleased to introduce, Michael and Emmeline to talk about, the neighborhoods project. Take it away.
Speaker 2
0:15 – 0:15
Great. So I will, start us off. So I'm Emeline and I'm part of the neighborhoods team and I usually talk or write about neighborhoods. And I've been doing the same sort of thing, like communications and ecosystem development, in some fashion in the DWeb space and in particular in the Holochain community since 2017 or so when I was kind of, radicalized, I guess you could say doing a dissertation in psychology, That was sort of like a social take on Internet addiction. So, Michael, do you wanna introduce yourself later when you do the demo or pipe up now?
Speaker 3
0:30 – 0:30
I could go now. Yeah. I'm Michael. I'm also in Neighbourhoods. I'm lead developer. And, yeah, I've been kind of in crypto e land since, like, 2017. I was more blockchain blockchain y, worked on Cardano, and then had a similar kind of, like, radicalization period where I was like, why are we making hyper capitalist individualist tech, and is there something else? And that kinda led me on this long, rabbit hole chase thing. And I found Holochain and then neighborhoods, from there. So, yeah, maybe that's enough.
Speaker 2
0:45 – 0:45
Okay. So I will share my screen.
Speaker 4
1:00 – 1:00
Let's see.
Speaker 2
1:15 – 1:15
Okay. Can everyone can everyone see? Yeah. Okay.
Speaker 1
1:30 – 1:30
Yes. Thank you.
Speaker 2
1:45 – 1:45
Excellent. So I'm really excited to be here today to present neighborhoods and kind of what, how, what we call social sense making relates to what y'all might call governance. And I use these like angular brackets as kind of like a more than and less than, but also like a back and forth sort of relationship. And I'll talk more about social sense making, but, I wanted to introduce these guiding questions. The first one is kind of relates to the project as a whole more. So thinking from the like Hirschman voice exit loyalty paradigm, one of our guiding questions is can voice be amplified to respond to the generic culture pushed by platform capitalism? And then the question that relates more to this particular conversation is how do features of Holochain and neighborhoods as a sort of Holochain ecosystem development project, impact how we go about governance? Okay. I'll be real brief here, but these are some of our kind of like assumptions and things that we've seen in this space. So again, with the voice exit loyalty. So just starting out with loyalty, what we've noticed and something that's really guiding us is that the platforms of web two, basically the way that they invoke loyalty is through lock in, and basically caging users via scale and also from holding data hostage or providing it in forms that aren't ultimately that useful. And whenever we talk about Web3 loyalty, I think vibes, but then of course also token ownership, whether that's vested, or or otherwise. And then when we think about exit in this space, one thing that has been like a provocative source of conversation for us on the neighborhoods team, is that there seems to be sort of like a technical idealization of low exit costs. And this idea that, you know, if exit costs are sufficiently low, then regrouping or forking and walking away, from a particular network becomes a lot easier. The assumption seems to be that this puts pressure on groups to treat its members fairly, like an ideal market scenario where there's that pressure. But when we thought about this more deeply, it seemed like, okay, maybe that's the case, but then also what that's produced at least in a lot of crypto communities is these kind of namespace battles, you know, you have Ethereum and then Ethereum Classic. And, you know, one group inevitably keeps hold of this namespace that is kind of what's garnered a lot of legitimacy,
Speaker 3
2:00 – 2:00
and
Speaker 2
2:15 – 2:15
the other is left wanting. And then the other part that we kind of joke about is that like, you know, low cost exit in real life, like that's kind of, that's just ghosting. Right? And in my opinion, ghosting has pretty high psychic costs, even if it's perceived as being low cost in other senses. Right. So, this sort of like notion of exit and low cost exit, I think kind of contributes to the idea that communities are disposable. So of course, that brings us to, to voice and I guess that's kind of what we're all focused on here today. So of course, you know, this is non comprehensive, but broad overview, you know, in, in Web3 and crypto world, voice tends to be kind of formal and expensive. You know, there's kind of the the very well known now familiar problem of of token voting, and delegation that I think is kind of like a necessary offshoot of token voting. And this is the norm, right? Like despite some pretty cool exceptions. And then there's also the fact that we all take notice of, which is that you know, the kind of juicy meaty deliberations sort of stuff tends to happen off chain, and usually uses centralized platforms like Discord, maybe even Google Docs. So this is kind of just our broad overview of the space and kind of yeah. We'll we'll help you get a sense of where we enter in. So we've been thinking about voice primarily through this idea of social sense making. And I'll talk a lot more about social sense making in the next few minutes. But some of our presumptions basically is that, you know, if a social space and neighborhoods as I'll talk about is like the socials and group where project built on Holochain. And so some of the assumptions we're operating off of are that, you know, if there's a really clear purpose to social spaces online, if there are sort of constitutional but iterable thresholds for state changes, iterable thresholds for state changes, if there are customizable metrics and tags and other sort of social primitives, And if a group is community or mission driven rather than ad driven, it's possible that participation in social spaces online can become sufficiently efficacious, basically meaningful enough to participants, like in the sense that you can actually see the impact that your input has on forming a space. That groups may require, less formal governance, and maybe even less intensive moderation when moderation comes into play. So there's a lot there and I'm going to sort of unpack that from the, from the neighborhoods perspective. Okay. So like I said, neighborhoods is a, it's a framework, for doing groupware, with social inputs and kind of just general, like most social use cases online. This is like the framework to do that using distributed Hello chain architecture. So just to give, it seems like some of y'all are from the space and probably know this pretty well. But just to show a little bit about, Holochain's affordances, which neighborhoods basically takes advantage of before we get into neighborhood specifics, that kind of impact on, on some of the, you know, on some of the problems, with governance in Web2 and Web3. So the first is that Holochain has a native architecture that is agent centric. And so it essentially has this sort of strong native capacity to integrate with, you know, with IDs, with public key infrastructure and with kind of like larger build outs of, of reputation systems. It also has a programmable redundancy, meaning that there's no need to hold the global state of the network as a full node, which makes it cheap enough to do that juicy meaty deliberative stuff I was talking about, or data and social scientific techniques, on chain. So this is really a lot of what makes it possible to do like whole distributed social networks or, you know, projects using groupware in a distributed fashion. And then Holochain is also comprised of distributed hash tables, which is where that redundancy is shared publicly, essentially. And so through these distributed hash tables, there's sufficient validation of records that are countersigned between agents. And so what this affords is basically that agents can leave with their public data without removing it from the network necessarily. And there's a lot more nuance to that, but basically for me that and the agent centricity bit gesture toward the importance of portability, social data portability, which is one of the sort of key values of the neighborhoods project. So back to how neighborhoods sort of utilizes this stuff, We use this mantra, generic tools, specific culture to sort of explain, to sort of explain our vision. And so what this means is that what we're trying to create is basically like a marketplace of super lightweight, like tiny, basically functional, modular Holochain apps. You can think of like calendars or chat or collaborative docs. We'll see an example from Michael today in the demo. And we think that those should be, in order to form neighborhoods, to form online social groups, that those should be kind of decoupled from the start from culture design elements and metrics. So these include things like entry and exit rules, engagement incentives, upvotes and whole series of ratings and tags, things that influence visibility, and that influence the location of objects and agents access to different membranes, that sort of stuff. So, so basically what we're trying to do as neighborhoods is to make it as easy as possible for people to, you know, for, for what we call community activators to sort of pick and choose from these different elements, and to, to be able to, to bundle them basically, and iterate on them in, in creating their neighborhood. So then just to go a little bit deeper into this specific culture bit, if you're creating a neighborhood, our ideal is for you to be able to customize like reactions and tags and other inputs that we call sense making inputs, including reputation metrics. Basically just to iterate on this question of like, you know, what do we wanna optimize for, in this space? And so what we're calling social sense making, which is basically what we believe contributes to the possibility of greater cultural autonomy, and community self determination using distributed socials, are a bunch of configurable social primitives. And these are already really familiar. Like we're not trying to reinvent wheels here. So things like resources, agents, reactions, tags, assessments, social linking, like different thresholds that give access to contexts within a neighborhood and of course mathematical and statistical operations, on social data. And this impacts things of course, like feed ordering, it impacts visibility, and that's configurable from an individual and also from a collective like whole neighborhoods level, which hopefully will become clear. So again, social data portability is like a huge element in all of this and using Holochain, it's pretty easy to, you know, to create a scenario where meaningful individual activity, is stored on that person's device. And this makes it more kind of easier and more reliable to share into other neighborhoods. And that's something where Holochain's native validation rules really come into play. And UI for this is, is on our roadmap, but it's kind of far out, but we're looking at things like open badges, for standards for these. And again, that's in service to building bridges across neighborhoods, which is a really high priority for us. And the way that we hope to put this all into action, is again, by offering these like sense making and reputation libraries, and also kind of a framework for plugging these into the DNAs, like the basic code of these little HAP modules, little Holochain applications that I've been talking about that only include these generic functionalities. And yeah, just going a little bit further about how we wanna accomplish that. Like we envision kind of a setup wizard, if you will, for choosing these modules and also for choosing these sense making reputation data types. And then, you know, once a community, once a neighborhood is set up, then that setup wizard would basically become, would change context a little bit and become like a dashboard for, for maintaining the community and iterating it, iterating on it. So basically like our idea is to use these social primitives as like the basis for governance. So I'll just finish up here basically. But, but yeah, but governance in the neighborhoods context then, like when we're focused on social sense making really becomes kind of just this question of, you know, what do we want to do with our collective data? Right. Because it's not natively the case that it all exists in some place to be operated on. And also like, what do we want to change about daily life in our neighborhood? So Michael will show just how easy it is to sort of like, iterate on the way that computations are done and the way that rules are set. And so as everyone here probably knows, there are plenty of great tools out there, and research done on different ways to designate authority, you know, everything from democratic decision making to mounting more, you know, moderation, benevolent dictatorships. But basically in neighborhoods, these tools overlap with marketplace modules and we would want these to be usable in these community maintenance dashboards. So let's see if if the demo will kind of bring some of this to life a little bit. I'll I'll turn it over to to Michael and he'll introduce it. Yeah.
Speaker 3
2:30 – 2:30
Cool. Thanks. I will share screen or try to. Sorry. I have to grant permission. Okay. I have to quit and reopen. Sorry. I'll be right back. Okay. Let's see about it now. Okay. Can everyone see two windows that look the same? Nice. Yep. Okay. Yeah. Yeah. That's a relief. Okay. So this is the NIH paper widget, and this is one of our early widgets. Our widget is a Holochain app, which also supports Neighbourhood's modular reputation architecture, which is still being fleshed out. So Widget is kind of a nascent concept, but it's sort of a subset of Holochain apps, which are neighborhoods compliant. And so this is, like, a early rough it's it's somewhat rough as you'll see, and the UI is not as beautiful as, you know, it eventually will be. But this the idea here is a sort of social tool for reading and annotating papers. And we had this idea when we're thinking about Medigov and what Medigov people might be into reading papers is our best guess. And so, yeah, so, basically, I will just do some oh, and as as additional context, like, the window on the left is connected to one Holochain conductor, which is Holochain's, like, node, I guess you could say. And the window on the right is a separate conductor, and the two conductors are both running on my machine and talking to each other. So they'll you'll see that updates on one side appear on the other, and vice versa. So I will do some initialization stuff. Please ignore. And then upload a paper, and it will appear. And so say okay. I'm reading this paper about microgrids, and they made a typo. They said cow. They should have said chicken. So now I've created this annotation, and we'll see it's a little clutter y UI, but the main thing to look at is the state unreviewed. And I can submit an action, which moves us into the accepted state. And, basically, this is a this a very simple two state state machine where there's unreviewed and accepted, and you can move from one to the other and then vice versa. So this is annotations on papers can move through this simple state machine. Basically yeah. They can they can go in, unreviewed, get accepted, and then get demoted. And this is, you know, maybe interesting. Maybe it's part of your group's culture of annotating things. You you only really want a simple notion of, like, is it new? Has or has somebody looked at it and said it's good? And what I guess yeah. Okay. I will now show a more complex state machine, which has four states And so this box here is don't look too hard at it, but this is this language that I built, and it's very verbose because I've not added the language features to make it nonverbose. But this is basically, there's an interpreter for this language, and because we have an interpreter inside of the HAP code, basically, we can update the computations in the state machine functionality on the fly instead of, for example, having to redeploy the app and the whole network of apps, which is much harder in a distributed context because you'd have to get everyone to update kind of in sync. So here we can basically, this text that I just pasted in and submitted is program text. And as soon as it propagates around the network, everyone has moved to a new, like, rules of the game, instead of having to, yeah, update update software. So right. And I'm also oh, sorry. There's I have to fix some weird JavaScript. I got the wrong quotation marks in my thing, and it'll mess the demo up. So sorry. One second. Okay. So we now have four states, and I am mapping what they are. So we now have unreviewed, rejected, tentatively accepted, and fully accepted. It these are just kinda arbitrary choices I made. The point is that you can change oh, and I can show over on the right that things are indeed gossiping around. Okay. So now we've updated the state machine and now you're in the one state. This means rejected. We can go back to unreviewed. We can go over to tentatively accepted. We can go oh, wait. We can go to fully accepted. We can go back down to unreviewed. So hopefully, that makes some sense. I'm moving kind of fast and but the point is to show that the rules of the game on annotations can be updated live without having to redeploy the code because we're sort of yeah, we're kind of cheating by having an interpreter which can dynamically interpret new rules. So and these are kind of simple examples, like, with just the state here is just an integer, basically. It's yeah. Either zero or one or two or three. We can also do like, this is basically a full programming language in the like, it's turn complete. It's just very limited right now. So we could but since it's turn complete, we can do richer stuff. So we could we could have a richer state like this. And again, it will be painful to try to read. But basically, we're operating over now a two full of integers. I haven't added, like, more complex types, but that's coming. So we we start out in the state zero zero. The first integer is a tag and the second is account. And so and let me update the last. Yeah. So we can now have counts of things. Whereas before, we didn't have we didn't have counts. We only had, like, a single integer state. So now looking down here, the state of this new annotation is unreviewed and count is b n zero. That just means it's a integer in the language. And so for unreviewed, it doesn't the notion of count doesn't exist really or it it's not meaningful. But if we submit a one, we'll move into the voting state. And now the count tells us how many up or down votes we have. And one is upvote, two is a downvote, just for demo purposes. So I can upvote this and downvote this. And if it goes below the threshold of five, which we will do right now, now we're in the rejected state. So just showing an account got reset. We could also make a new annotation and upvote it off off the top to get into accepted. Should've made it lower threshold. Okay. So accepted. Yeah. We got to accepted. And so so this is just showing a, like, slightly richer state where you can have a count and you can do things conditional on the count exceeding some bound or whatever. Basically, just kinda standard integer inequality comparison stuff. So, yeah, that's I guess, it this is kind of a very specific example of or the this whole demo shows a very specific notion of, like, metrics or computations that a group could hold, like, for for their group context. And it is a bit of a, like, stretch or it's a ways off from the higher level notions of reputation that Emile was talking about. But, like, these very specific low level detailed notions, and even be like, beyond this, what I'm showing here, we could imagine relatively easily, I think, building, like, how many sorry. Your standing in the paper widget reading group could be based on how many annotations you have that got accepted minus how many got rejected and then some other yeah. Adding in other relevant details. And all of those could be kind of say, like, you're you're standing in the group could similarly be a dynamically updatable computation, which is held by the group and can be updated live similar to how the annotation state machine rules can be updated live. And I haven't yet built something to showcase that, but it's a, like, similar in principle. So yeah. I think that's that's kind of the yeah. That's, like, the the meat of the demo showcasing that dynamic ability. And then I guess one other note maybe that here, all the conductors, all the nodes participating are bound by the same, like, rules. Basically, the whatever this state machine computation is on annotations and whatever the starting value is and what the states are labeled, that is there's consensus on that. Everyone has the same rules on annotations. But, like, the process of updating that is totally anarchic or anyone can update at any time, and the yeah. Anyone can update at any time. So this is not how you'd wanna do some actually mature group process because it's a bit too chaotic, I think. But we could imagine needing like, in order to suggest a new state machine computation, you need a certain threshold of credibility, and maybe something, like, hinting towards Stack Overflow if people are familiar with their very ornate system of, I forget what they call it, like, rep or something. You need a certain amount of reputation points to be able to, like, downvote things or cause trouble, and they kinda meter out the ability to do things, based on members demonstrating that they deserve this, like, extra power. So we can imagine and a thing that gets me excited about this is the ability to update the rules as you're going and in a cheaper way than synchronized distributed software updates. It would be moving more at the speed of data and people sending around different computations and agreeing on them and or forking off or, yeah, moving more swiftly and kind of iterating the social constitution. Yeah. So maybe I can end there if you wanna take questions.
Speaker 5
2:45 – 2:45
Yeah. Do you need to keep sharing screen or would it make sense to
Speaker 3
3:00 – 3:00
Yeah. I can stop. Yeah. Super. Cool.
Speaker 5
3:15 – 3:15
And I can try to make eye contact with people who must have their cameras off. And we want to put foster comments in the chat. And I'd like to ask in particular if there's something you've seen that was either particularly pleasing or impressive to you, make a note of that in chat. And I'm gonna scan over this and see that John had a question or a comment maybe. And, Greg, you were replying, you know, in in the chat log about the nature of the the the NH credits. Is that something you wanna address out loud for the group, or does that not seem important to you?
Speaker 4
3:30 – 3:30
If Emeline or Michael would like to talk about that, that'd certainly be cool. I don't know if that fits with the flow of the conversation.
Speaker 5
3:45 – 3:45
Yeah. It could be it could be a bit of a sideline, although it's certainly interesting. And, Emily, you'd suggested that this is policy making without the formality that usually comes with that. Can you elaborate, please? I'm not sure I followed you.
Speaker 2
4:00 – 4:00
Yeah. Sure. And this is this is something that, you know, we've been kind of trying to frame well because there's something that feels like, yeah, there's something that feels really important there. Like, there's kind of, I mean, my, I guess my deep seated assumption is that like, there is this realm of human activity that's like capital G governance, And everyone kind of like gets on their best behavior and kind of thinks from this place that is not like the place of that is that is a different place from the place of the like the daily group activity and the goings on kind of like within the context of whatever the group was set up to do. And so one way I guess that we're thinking about kind of limiting that sensibility is by trying to use this idea of sense making to kind of bake governance in or, like, what we would recognize as governance into daily activity. Right? Like the only reason, I mean, ultimately, like there's a lot of, conviction polling that happens pretty much all the time on social media. Right? But we don't call it that because like, you know, what you decide to like, or what you retweet or whatever, like has a limited and a particular influence on the way that your feed is ordered on what suggestions you receive. Right. Because there are other, namely like paid basis for, you know, for, for how that stuff is sort of ordered in the last instance. And so kind of what we're trying to develop is this idea that,
Speaker 4
4:15 – 4:15
that
Speaker 2
4:30 – 4:30
we want like governance to be, to be gaseous and we want regular social activities of opining, of stating preferences, of giving feedback, to be like efficacious enough and to actually change the nature of the activity enough and the visibility enough to, yeah, to give people a sense that like, oh, okay, like my actions are meaningful. We don't need to do this. Like, we joke about like the angry march to city hall. You know what I mean? Like there's a lot of general sense of, of powerlessness, I think in, in, in socials these days, because, because indeed, you know, you're stating your preferences all the time, but they're not being taken into account meaningfully. So
Speaker 5
4:45 – 4:45
So Or Yeah. Please. Good to hear.
Speaker 4
5:00 – 5:00
Well, I
Speaker 6
5:15 – 5:15
was gonna say, or they are just being hijacked and used for a kind of policy making that you don't have. Like, I I would argue that it is actually already the case that we have algorithmic policy making. It's just that the constituents of most of those systems have no meaningful say in that algorithmic policy making. So our day to day activities affect these data feeds. These data feeds play into algorithms that decide what we see and what we're allowed to do, etcetera. But you can't even go into city hall and storm it because what you're actually doing is complaining on Twitter about Twitter's algorithms, and Twitter doesn't care. And then Elon Musk makes up some shit.
Speaker 2
5:30 – 5:30
Right. Right. Exactly. Yeah. Yeah. That's exactly it. And that's why we're kind of, like I I view us as being in, like, a lucky position. You know? I mean, we're doing, like, the social primitives ontology, but there's nothing really to reinvent because all of the, you know, like all of the preference formation, like all of the forms of tags and reactions and assessments, like they're already there, in the web that we know. It's just a matter of having it meaningfully be accounted for.
Speaker 5
5:45 – 5:45
So this idea of lightweight, maybe governance is the wrong word, a lightweight way of achieving a social consensus about, you know, do people say excuse me or not? Or do we just, like, stand up and leave? Or do we greet people as they come in the door? Or do we just nod silently? There's all sorts of little routines and rhythms about human life. And then you can get up into more meaningful things like, you know and and who what state of of your lawn shall you permit in this neighborhood? And at what point does your neighbor come out and mow your lawn for you because you've let it go too far? And there's no, you know, there's no HOA rule about that. But for heaven's sake, Thomas, look at your yard, dude. You're making us all look bad. And so if we can't live that way on the blockchain, we're gonna live that way else somewhere else, and the blockchain will be the poorer for it is is the sense I'm getting from you.
Speaker 2
6:00 – 6:00
Yeah. Yeah. Very much so. And it's interesting too because, like, we're not I don't think I mean, certainly the neighborhoods team is not necessarily like we don't spend a lot of time and not enough time, I think, talking about how important some of these aesthetic features are. Like, for example, like if you have a I don't know if you have like a, say you're a community of surfers, you know, like having your, having like a little wave emoji, like be your like, like these sorts of things that create like a real sense of belonging and sort of like have the, I don't know, have the, the texture of like, well, this is, you know, this is really how we do things around here. This is really our community. You know, like little things like that can, can be quite sufficient for, for loyalty often provided, you know, everything that we just named about preferences being meaningfully taken into account is is also happening.
Speaker 5
6:15 – 6:15
And and clearly your focus is on enabling folks to do what they choose to do rather than you having a plan for what they ought to be doing. I'm wondering whether you've read or are familiar with the work of Dan Coyle. He released a book not long ago called The Culture Code, which talks about signals of belongingness and the role they play in building a a culture or a society. I I suspect that it'll intersect interestingly with the work you're doing. I I don't know it well enough to comment on it further.
Speaker 2
6:30 – 6:30
Okay. Yeah. If you don't mind dropping the author Yep.
Speaker 5
6:45 – 6:45
In the chat, I'll check it out. Absolutely.
Speaker 2
7:00 – 7:00
Yeah. It's quite exciting to be kind of on the heels of stepping into this big culture design stuff. Like we're really trying to nail down these primitives such that we can kind of, get deeper into populating said marketplace with, with fun stuff, basically.
Speaker 5
7:15 – 7:15
And I'm gonna ask for my speaking a second, I wanted to acknowledge you for giving a back end demo, which these are always notoriously unsexy. Front end demos are sexy. Back end demos are never. And I I was a back end guy my entire technology career, so I I feel your pain. Michael, comment.
Speaker 6
7:30 – 7:30
Oh, yeah. So, actually, this kinda connects to that. I really appreciate the sort of decomposition of the architecture in
Speaker 4
7:45 – 7:45
such a way that it can be extremely expressive on the lower on
Speaker 6
8:00 – 8:00
the back end, so to speak, and yet still lower on the back end, so to speak, and yet still provide people with, like, sort of concrete assemblages that are useful and maybe, like, easy to deploy or sort of preparameterized. Then they can, you know, iterate on them themselves to sort of, you know, adjust the policy making to meet the needs of the particular neighborhood. But, like, this is a really difficult challenge because it's not just culture code. It's like, you know, culture language. You have to create the primitives or distill the primitives out of the observations of how people choose to organize themselves, but then you need to simultaneously solve for the a degree of expressiveness of the language and the sort of ease of use, so the the UX associated with, you know, deploying and maintaining an instance of of a of a community or of a neighborhood. And I just wanna really acknowledge how difficult that is, and so I appreciate the the actually relatively clear segmentation between the requirements from the language design perspective or the the sort of ontology, you know, system building layer versus the efforts to build specific products or or enable specific assemblages that are useful to specific communities, but under their control.
Speaker 3
8:15 – 8:15
Thanks. Yeah. Definitely, it feels like quite a challenge as you pointed out.
Speaker 5
8:30 – 8:30
And Seth had said something in chat. I'm not sure if we've already covered this in the verbal dialogue about permissionless or anarchic state changes in certain environments. Is that Centri in a place where you can elaborate on that? I know John, for instance, can't because he's in a public place and can't talk. But oh, you've unmuted. Seth, please.
Speaker 4
8:45 – 8:45
Yeah. Sure. I can speak to that. Yeah. So I guess kind of what I'm thinking about here is I like that you've built this interpreter that allows the state chain to change dynamically and in real time, like using kind of like the gossip network like you mentioned. And that kind of has like a potential playfulness or lightweight quality to it, a kind of gaseousness, maybe. And I'm curious, like if there's any thoughts that go into sort of like what a kind of like crowd play or like community play could be that isn't trying to kind of set up a type of permissions administration of that dynamic state change through something like reputation scores. Like, the the kind of I think I'm thinking of something like, like Twitch Plays Pokemon where, like Mhmm. It's a version of Pokemon played on Twitch where the the players type in the commands for the buttons and it like it just takes all the information in immediately and starts running and it's it's kind of really an Arctic. But there are certain places in the game where it's not possible to play that way, and so the creator of the system set up a scale and there's a threshold where you can. Araceli Emi If enough people in the chat right democracy, then it'll tip over from anarchy into democracy and then like the way the system works is it all like there's a time buffer and the amount of like if, like like a. Gets the most number of inputs during that time period that's what actually gets executed in the game, and so it creates a way of kind of like harnessing that can Arctic energy of that crowd play. But like without having to like rely on sort of like reputational systems, we start to kind of take on a little bit of like the heaviness of the formalism of what we were talking about with governance with a capital g. It kind of removes some of the gassiestness. So I'm just curious to hear maybe some reflections on on, like, what are other places where weight is being introduced in the system? Are there other mechanisms for, like, community play that are that could take use of this dynamic state change? Maybe if it's even just, like, in a sandbox environment.
Speaker 3
9:00 – 9:00
Thanks. Yeah. Yeah. The Twitch Plays Pokemon example is really good. Thanks. Yeah. One thing that one of our other team members was thinking of recently was, like, maybe kind of similar idea, but for producing, say, like, image or video montage by like, say you're in a neighborhood where your focus is, like, videography or, I don't know, photography or something, And you want to produce as a kind of group produced artifact, like, some slideshow or video montage, and a set of rules can be constructed, hypothetically, that would allow, kinda similar to this Twitch Plays Pokemon, like, I push my button and it somehow feeds into what happens emergently at a group level. Like, maybe there's a separate, I don't know, quasi Instagram like widget thing, and people who get a certain threshold of likes or whatever it is can be featured more prominently in the aggregated slideshow thing. And so there's there's some way to yeah. And here, there's lots of question marks in my head about how we specifically implement it, but I think, yeah, we're kind of trying to think in that direction where yeah. Like, I how to say personally, I'm very interested in stuff like this where it's the group is making a thing together, and then the reputation is, like, just kinda part of instead of being this standalone, like, I have a 100 Twitter followers, and it's kind of meaningless. The my weight in, like, how this montage gets produced is kind of a more contextualized meaningful thing. And if that's dictated by how many upvotes I got over here somewhere else, then it all kinda ties together into a process, I guess. So, yeah, hopefully, that was some answer.
Speaker 4
9:15 – 9:15
Yeah. That's that's interesting. I I also have more thoughts on reputation. There was a question that I asked her or a point I made earlier about, like, exit and loyalty and kind of like web two dynamics of like people trying to kind of like push back a little bit against like this kind of archival fever of web two or like this like impulse towards branding. And like go like once you get, like, a a 100 followers, for example, they'll just, like, close off their account, make an alt, and, like, the kind of the real people who, like, know will, like, funnel over to that. And they'll just keep exiting multiple times on this Web two platform to kind of avoid capture. And so, like, there are there are, like, cultural strands of, like, kind of, like, constantly, like, reemerging organic emergence that resists this kind of, like, formalization into reputation. And I'm sort of also curious, like, how yeah. This is, like, a a broader question, I think, that I'm kind of trying to work through of, like, what is like this impulse towards reputation like and like what are the different ways that it's like stack like at the individual level or at the Community level or like some in between. So yeah. That that's not really a question. It's just more, like, a thought.
Speaker 3
9:30 – 9:30
Yeah. Thanks. I think it's I don't know. There's lots of complexity to investigate, and I think we're kind of all in this. I don't know. Trying to make sense of what could be possible or desirable.
Speaker 2
9:45 – 9:45
Yeah. Yeah. All stuff that we're mulling over and sent like, it really for me, it really changes everything to have like a more time sensitive perspective on reputation, you know, such that like a community or, you know, like a group of agents that are spanned across many communities have sort of like reputation, Jubilee type commitments, like you were talking about, or even to just have reputation exist within particular game spaces, you know, such that like when you're accruing reputation, it's for a particular activity, it doesn't attach to your, and web two Relic. Right? Like it doesn't attach to your your big identity. And but then also, you know, I think where it runs into the question of like, well, some of these things you actually do want to manage on like an archival sort of basis is just being able to find your pack across different neighborhoods or across different communities, which is like such not a given in the distributed world.
Speaker 5
10:00 – 10:00
So two minutes from the end of our time. I wanna make sure that we get to Michael's question. I wanna acknowledge the very active conversation that John is conducting in chat, and I'm hoping that there's gonna be, attention given to resolving some of the intriguing and unresolved questions he's he's posed. Michael, final question, please.
Speaker 6
10:15 – 10:15
Yeah. Actually, it's actually a a a props pointer to Ellie Rennie's work who's one of the Medigov core members. She's also conducting within and around Medigov some ethnographic work related to various reputation systems and contribution graph metrics into some of the stuff that you're evoking. And the reason I am bringing it up is because, you know, we all can talk theory, and it's actually super fun to reason about the theory of this stuff. But there's already been some preliminary findings about the differences between the way people imagine or design reputation systems to work versus the way that the people who operate within systems those reputation systems are part of the community's policy making end up being experienced. And so I would make a strong call out to maybe you know, you can connect through either directly to her or through me, but there's some ongoing research that I think is actually a little bit closing the loop oriented. Like, how do people experience these things, not just how we intend them to work.
Speaker 5
10:30 – 10:30
If there's a link you can put in chat to try Yeah.
Speaker 6
10:45 – 10:45
I'll go grab one of the papers. There's there's, like, one submitted, one under review, and then Or
Speaker 5
11:00 – 11:00
even just the person to to who's name to
Speaker 6
11:15 – 11:15
look for. Name. That'll be easier, I think, given it's ongoing.
Speaker 5
11:30 – 11:30
Love it. Love it. Super. We are at the top of the hour. I think it'd be appropriate to express some thanks and love to our presenters and to the folks who convened us today. Big thanks, folks. Awesome, awesome work. You're gonna unmute and reply. Yes, please. Three, two, one.
Speaker 2
11:45 – 11:45
Thank you so much for having us. I love that you guys do that.
Speaker 3
12:00 – 12:00
Yeah. Yeah. Thanks.
Speaker 5
12:15 – 12:15
K. I've moderated enough. Someone else can close us.