{
  "metadata": {
    "transaction_key": null,
    "request_id": "metagov:tank-metagov-20230726",
    "sha256": null,
    "created": "2025-10-27T23:38:02.738063+00:00",
    "duration": null,
    "channels": 1,
    "models": [
      "metagov-manual"
    ],
    "model_info": {
      "metagov-manual": {
        "name": "metagov-manual",
        "version": "2025-10-01",
        "arch": "manual"
      }
    },
    "warnings": null,
    "summary_info": null
  },
  "results": {
    "channels": [],
    "utterances": [
      {
        "speaker": "Speaker 1",
        "start": 0.0,
        "end": 0.0,
        "transcript": "Hi, everybody. Welcome to another MediGov seminar. Today is 07/26/2023. And today, we are joined by Ashka Tank, who is going to be presenting on governance archaeology. It's a very wide ranging research project looking across time at different communities and the ways in which they've approached the question of governance. I'm gonna keep the introduction short because I think Ashka's presentation is going to be amazing, but this has been a long project and I think we're all very excited to see the results of the research, and I'm looking forward to a really stimulating and interesting conversation. And then I'll also say, if you haven't had a chance to introduce yourself yet, please just in the chat, name, pronouns if you like, where you're calling from, anything else you wanna add. And if you have questions along the way, we're gonna have twenty minutes or set up presentation from Ashka, and then we'll have a moderated discussion. So please jump in, with your questions, and I will be moderating that towards the end. And with that, I'll pass it over to Ashka."
      },
      {
        "speaker": "Speaker 2",
        "start": 15.0,
        "end": 15.0,
        "transcript": "Thank you so much, Sant. So, yes, it's been so much fun working on this integration, and I'll be talking about how we sort of try to diffuse to very disparate databases, government's archaeology and database. And, also concurrently, Riley has been working on sort of taking a more analytical quantitative approach to, sort of extracting findings from governance archaeology. So while it's not directly related to the integration, I think it would be an interesting complement to, Ali, if you'd like to speak about that after I'm done, that would be really amazing too. And, yes, that's it, really. And I'll just share the presentation."
      },
      {
        "speaker": "Speaker 1",
        "start": 30.0,
        "end": 30.0,
        "transcript": "Okay."
      },
      {
        "speaker": "Speaker 2",
        "start": 45.0,
        "end": 45.0,
        "transcript": "I hope you can see it."
      },
      {
        "speaker": "Speaker 1",
        "start": 60.0,
        "end": 60.0,
        "transcript": "Yep. We can see."
      },
      {
        "speaker": "Speaker 2",
        "start": 75.0,
        "end": 75.0,
        "transcript": "Great. Right. So what we started out with was this. And since the name is quite a mouthful, we just ended up calling it GA plus b. And so I initially started working on governance and archaeology last year over the summer. So I'm an undergrad at King's, and Frederica teaches there. So she was one of my lecturers, and I was very fascinated by taking this historical approach to governance. So I worked on this project, and I've always been very drawn to sort of more gov based oriented themes like blockchain technology, decentralized finance, and how that can be structured better. So this seemed like a very great fit. And initially, Josh and I sat down and took made a plan of how we could potentially structure this. But unfortunately, we saw that there were many roadblocks that came up with the initial approach. So we decided to completely overhaul the structure of the database more than halfway through the project. And now here we are with the final result. So this is essentially governance psychology for those of you who haven't interacted with it. So essentially, it has all of these historical communities, and these communities have sort of subgroups of institutions, and they're linked to mechanisms. So for instance, this community uses matriarchy to sort of aim at least one or more of the institutions, and that's why that's there. So when I initially started working on governance archaeology, I added more communities to the database. And based on a suggestion Federica got from my colleague, she actually encouraged me to build this column of heterogeneity, which I think was a really fundamental experience when it came to categorizing these very, like, broad, diverse, and fairly complex structures into neat little boxes. And I don't think I could have dealt with the tagging had I not done this in governance archeology. So that was incredibly helpful. And then the first conversation we had was about what we were trying to bring to the table from governance psychology. So it became clear very early on that we would look only at mechanisms and not this column of culture, values, and norms for two very basic reasons. Firstly, Frederica thought that, you know, it needed much more detail and even more work even though I think it's amazing the way it is at the moment. And also the boundaries of these are fairly foggy. So sort of categorizing them and integrating them seemed too premature, whereas mechanisms were more discreet. And as you can see, you know, numerous as well. So that seemed like the right way to go and essentially see what the payoff of integrating mechanisms was as an initial approach. And I think from the very beginning, I was very conscious of the trade off between the comprehensiveness and the complexity of the integration because we wanted to include as much of the depth of information as possible from governance archaeology, but not make chunks of information so large that it was impossible for a casual reader or browser to actually absorb and implement them. So the the initial sort of strategy was to map on the fields in governance psychology to the fields in GovBase. So as I'm sure you're familiar with the GovBase structure of you know, the first table is organizations, and those were an obvious match for communities because they act as, you know, platforms of firms in themselves even though they're prehistoric tribes and very far flung little communities. So, yes, that was fairly straightforward. And structures would be the mechanisms or the metamechanisms that were at play. So what I did was I picked up all the mechanisms from governance archaeology and put them into the structures table in GovBase. And then I used all of the columns of GovBase to try and sort of plot those mechanisms into the language of GovBase in a sense. So we have the ontology, which is obviously government psychology, the subclass, superclass component, and property columns. Those were the ones I would wanted to focus on. So I was trying to establish links between the mechanisms of government psychology and all of the other structures in GovBase and simultaneously try and create internal links between the mechanisms in GovBase. So if you see here, benevolent dictators, the GA, mechanism and autocracies in GovBase. So I would try and figure out whether this is a component or a property or subclass or superclass. But from the get go, this approach presented a lot of challenges, and I don't know inefficiencies as well. So for some, I would see things like an agent or actor and roles or positions. And these are from different ontologies, but they're very similar sort of tags. And the end result was that I was ending up with, like, 10 subclass, superclass tags, and none of them individually contributed an exceptional amount of value to the understanding of the mechanism. And similarly with this, like, you have values, ideologies, and incentives, and then you have institutional values. So I felt like I was paraphrasing the same concept over and over, which just slowed down the process quite a bit. And I thought that was fine. That was a necessary evil of integration, but there were also more substantial issues. So for instance, there was some subclasses that were very difficult to sort of pigeonhole. So if you look at fines, I thought that was an obvious case of pain damages or compensation. But in a discussion with Federica, I came to realize that I was viewing fines as the sort of restorative remedy. Like, if you give a fine, then, you know, what you did is forgiven. It's alright. But it was retribution in practice. It was more of a punishment. It seems like a, you know, very granular distinction, but it completely changed with subclass I was putting it under. So if I it was if it was restorative, then I would put it under compensation. But if it was retributive, then it would be punishment or damages. And those kind of decisions were not really mine to make, I thought. I would pour over all of the historical communities, try and glean some context from that. But I've always felt like I was, you know, at the end of the day, even with a lot of information, just making very flimsy judgment calls. And also more tricky technical questions arose. For instance, is matrilineality a component of matriarchy? So these columns of components and properties were especially restrictive in a sense because even matriarchy could potentially exist without matrilineality. And even if we looked at all the communities that were in governance archaeology, we could not tell if tomorrow some new finding of the historical community appeared, which showed us that, in fact, inheritance and large relations were not a necessary aspect of representation in the matriarchy. So I was really straddling the boundaries between what I was seeing, what could potentially happen, how the database could evolve, and it was just getting very messy, very reductive. And at the end of the day, I felt like there was not much detail being captured. So I we all finally came to the geolization. And in a very long winded email to Federica, I bought something like, I think that I'm trying to fit a round peg in a square hole. And that was that very simple realization sort of sparked this new structure. And that was really fundamental to, you know, how we are now approaching the integration. So now what I did was that instead of putting the mechanisms into GovBase, I retained the original structure of governance psychology and instead added a very crucial and really interesting tags from GovBase, which are these decision process types. So you see the mechanisms here, all of them, and then there's a clear description. And these tags of types, actually Federica came up with this, and this is all originally from governance archaeology. And I think she presented this framework to me after I'd finished this decision process type. And this was the point when I realized that, okay, this integration actually has worked because where the decision process types were not very accurately fitting the bill, the tags were were an appropriate lens for understanding the mechanism. And where the tags fell short, the decision process types would add a lot of context and weight to the way the mechanisms work. So I thought that this was the integration that we were looking for where both systems of thinking were adding value to one another. And, yes, so essentially, all of these mechanisms are linked to communities where they have been implemented. And there's also the institutions table, and a lot of crucial institutions have specific mechanisms linked to them. So a community could use, I don't know, election, assemblies, checks and balances. And if you go over to, say, like, a meeting group or a seminar, that would be an assembly in itself, and it would not have any other mechanism tags because it's just an assembly. So that's the way this generally works. And I think this example would make it a little clearer on how we're trying to make this more usable. So, essentially, we start with a mechanism. For instance, let's say, a dual sex political system. And if you click on this, you see a description and you scroll down and you see all of the communities which have used this mechanism. And so you click on a specific community, like, alright. How did the Igbo people use this specific mechanism and what other mechanisms do they use? So you see all of these. And then if you click on a specific community I mean, sorry, specific institution within this community, you go up there. And, again, you see all of the other categories it's linked to. And the best part about this is that you don't have to leave the mechanisms table in order to do this. You're just clicking on different sort of, like, pages within the the sort of panel that opens up, and that's why I think it should be straightforward to use. And also for this example, specifically, what we're looking at is in a dual sex political system. I mean, I'm not sure if you've seen the sort of very long brief I put up. I sent a sentence now in the description for this meeting. But that essentially talked about how, you know, modern communities could take up a mechanism that hasn't been seen and say smart contracts with the blockchain world, but actually worked very well hundreds of years ago and potentially apply. So I think there's also more interpretive way of looking at mechanisms. For instance, in dual sex political systems, it comes down to two groups with different areas of focus, if not opposing interests, and how they sort of manage their interests. For instance, in the case of the Igbo people, the females always wanted to conserve grain and food, and the men always wanted to trade them and get weapons and, you know, conquer more territory. So this sort of conflict between economy and expenditure was the reason dual sex political systems were needed within that community. And I think those sort of different views would obviously exist in a more modern context, and so it could be very interesting to see how something like this could be applied. So if I look at this and I get a clear idea how dual sex political systems could work, and I wanna see that. Alright. So if you look at the decision process guide, this is a group decision making process. So it does not involve consensus where it's not a consensus process in the sense that in a group, if even a small subset of people disagree but a majority agree,"
      },
      {
        "speaker": "Speaker 1",
        "start": 90.0,
        "end": 90.0,
        "transcript": "the"
      },
      {
        "speaker": "Speaker 2",
        "start": 105.0,
        "end": 105.0,
        "transcript": "the decision goes ahead anyway. And that's very interesting because that means that sometimes the interest of men, if they're a larger number, could outweigh that of women. So that might intrigue someone, and they might see how that sort of balance between the powers of the two groups is"
      },
      {
        "speaker": "Speaker 1",
        "start": 120.0,
        "end": 120.0,
        "transcript": "maintained using other structures. So if"
      },
      {
        "speaker": "Speaker 2",
        "start": 135.0,
        "end": 135.0,
        "transcript": "you go deep into the community, you see that So if you go deep into the community, you see that consensus is actually a mechanism that's used within this institution, which is an elite assembly. So the wisest of men and women are elected to this elite assembly, and then complete consensus is needed in the kind of decisions that they make. So that's how it comes down to the interests of both groups being equally represented. And so you have a balance between making decisions, but also including the voice of all potential groups. And by looking at the historical approaches to that, I think a lot of tasks can be added to the governance mechanisms we see today. And so I think the main things I'd love to hear from you are, whether you think people will actually go for the mechanisms and how deep they'll go. Because the way it's structured right now is that the mechanisms table is it. It's I think that's where people will begin and end, whereas the community and institution tables are there because they need to be interlinked, but they're very sparse in terms of content. So I would love to hear if you think people would actually go and look at individual communities just out of sheer historical interest and whether you think I should include, like, more columns, more details there as well. And, also, I would like to briefly go over the tags I picked from GovBase. I think from the failed first attempt, this was the best thing that came out of it because I had a very clear numerical idea of which tags from GovBase were being used most. So it was not me making arbitrary calls on the relevance of each tag, but really picking out tags on based on data. So the first one is consensus processes, and then there's delegation, voting group decision making protocols, and regulations and rules and norms and conventions were an interesting call, I think, because they could in the future, if we decide to extend this to culture values norms, then this could also be a very relevant decision process stack. And big user informal decision making processes are something I think would be very useful as well because there are things like alliances and handshakes, which, you know, you can't sign a contract or a legally binding document to sort of cement that. And understanding how historical communities achieved that without rule of law or many technicalities or central enforcement mechanisms is very insightful because they did it extremely well. And so I think that would be a very as interesting aspect to gain insight from. And then there's unilateral decision making process, which would be something like an autocratic chief or a, you know, tribal leader who's who has divine authority and is just elected on the basis of that. So I think that would be marginally less relevant, but it's also important to code for that. So people could actually use this to filter out those kinds of mechanisms maybe and just, you know, focus on, say, consensus or group decision making. So, yes, that's it for me, Jilly, and I would now love to hear from you. Thank you."
      },
      {
        "speaker": "Speaker 1",
        "start": 150.0,
        "end": 150.0,
        "transcript": "Great. Thank you so much, Ashka. That was a really amazing presentation."
      },
      {
        "speaker": "Speaker 2",
        "start": 165.0,
        "end": 165.0,
        "transcript": "Thanks."
      },
      {
        "speaker": "Speaker 1",
        "start": 180.0,
        "end": 180.0,
        "transcript": "Great. Yeah. So much to think about. We already have some questions and reflections coming in. I'd like to start with Drea. I think it's a kind of pretty straightforward question, but I think maybe it'll help to kinda get the conversation going. And then, maybe if, Seth, that it appears that your two inputs can maybe be stacked together, we can turn to you next."
      },
      {
        "speaker": "Speaker 2",
        "start": 195.0,
        "end": 195.0,
        "transcript": "So one second. If if Andreas Andreas question is the one on the sortation mechanisms?"
      },
      {
        "speaker": "Speaker 1",
        "start": 210.0,
        "end": 210.0,
        "transcript": "Yeah. Yeah. Yeah. Sorry. She just said I was gonna invite, Drea to speak to it."
      },
      {
        "speaker": "Speaker 2",
        "start": 225.0,
        "end": 225.0,
        "transcript": "Okay. Great. Yes."
      },
      {
        "speaker": "Speaker 3",
        "start": 240.0,
        "end": 240.0,
        "transcript": "Awesome. I tried to get most of it in there. So I'm curious about so sortition was used in Athens. It was used in in, like, for the Vatican and things like that, and it's a new mechanism. I'm curious. Is that, like, a subpart of a larger mechanism? Like, where in your mechanism structure would it show up?"
      },
      {
        "speaker": "Speaker 2",
        "start": 255.0,
        "end": 255.0,
        "transcript": "So it is still separate mechanism. And, yeah, it's not so it's not a culture cultural value or norm, obviously. It is a mechanism. Like, checks and balances is a mechanism. So some of them are just bigger than others in a sense, I think. And that's where Federica's classification really helps because she has something called principles of rule. So things like rule of law and I know power fluidity dynamics, and those go in sort of overarching scenes, which are also mechanisms in themselves. So I'd have to check where the sortation label actually goes. I think it goes within voting, though. So yes. Though she she's an Athenian expert, so I think she'd be a great person to have this conversation with."
      },
      {
        "speaker": "Speaker 3",
        "start": 270.0,
        "end": 270.0,
        "transcript": "So it's more in the in the gov based side is is where that shows up. Because this is I can be orthogonal to or at least independent of other things. You can have sortition with a consensus. Yeah. Discussion or a sortition with voting. You can have sort yeah."
      },
      {
        "speaker": "Speaker 2",
        "start": 285.0,
        "end": 285.0,
        "transcript": "Mhmm. Exactly. So there I think there are a lot of this Riley could also potentially speak about this. But in my understanding, sortition could coexist with, you know, many different kinds of voting mechanisms. So, again, another thing I love about governance archeology is that it's so customizable. You could look at the co occurrence of certain mechanisms and figure out how they sort of blend in together and how they're sort of keyed into separate institutions. Or for instance, consensus could be linked to another institution, and sortation could be linked to another one, which does not potentially use consensus."
      },
      {
        "speaker": "Speaker 3",
        "start": 300.0,
        "end": 300.0,
        "transcript": "Got it. Nice structure. Thank you."
      },
      {
        "speaker": "Speaker 2",
        "start": 315.0,
        "end": 315.0,
        "transcript": "K. No problem."
      },
      {
        "speaker": "Speaker 4",
        "start": 330.0,
        "end": 330.0,
        "transcript": "Thank you. It's really fun to watch this come together. My my questions are kinda granular. I think they're fast. It's like a a bigger question. Bigger question. Gut based, I've I've never been super satisfied by its kinda coverage. You know, it kinda clumps, like, big things and little things. And and and any, like, real world kinda complex, you know, institution you're close enough and just, like, all those tags are gonna fit every institution. And so where do you draw the line and, you know, of course, every formal institution has informal stuff going on. Everyone, you know, might have a little committee that does have, like, a single decision maker. Right? You you've got informal or formal levels of authority. So, yeah, kinda where do you draw the line and, like, what what's where where does the judgment call stop? Or is there kinda were you able to find some way of slicing, some kind of protocol, to to make it to create something that feels complete that doesn't include stuff that shouldn't be there and to, yeah, have have some signals instead of everything tagging everything."
      },
      {
        "speaker": "Speaker 2",
        "start": 345.0,
        "end": 345.0,
        "transcript": "Yes. That's I think that was the main thing that I struggle with certainly because some categories like values, ideologies, incentives, and norms, I think, that's five words and that's one tag. So I completely get your point, and that was obviously what I was grappling with for the first stage. But the way we sort of zoomed in on that was through a combination of governance archeology tags, which are, I think, incredibly granular. So they go into access, decision making, and enforcement as three broad buckets. And within that, there are subcategories. So that's very detailed and precise. And the decision process type tags is it's I think it's the essence of GOV. So they are very broad, but not so broad as to completely eliminate all sorts of context. So, yes, ambiguous and informal processes could include a whole range of stuff, but I think it captures more of the, like, fuzzier mechanisms like rule of the old or, you know, alliances and handshakes. So those are mechanisms which are purely based on, like, softer values, whereas hard ones, which are encoded, would be in some of the other categories. So, obviously, I didn't want to zoom in that and make more value judgments there. I I always have this, like, intellectual humility complex when I'm looking at governance psychology because it's so vast. So I didn't want to make that decision, but I just put the fuzzier bits there and the more discreet bits in another section."
      },
      {
        "speaker": "Speaker 5",
        "start": 360.0,
        "end": 360.0,
        "transcript": "Thank you. Can I have one really quick thing to add to to Asha's comment?"
      },
      {
        "speaker": "Speaker 2",
        "start": 375.0,
        "end": 375.0,
        "transcript": "Yes, please."
      },
      {
        "speaker": "Speaker 5",
        "start": 390.0,
        "end": 390.0,
        "transcript": "I'll just say that yeah. So, is definitely there are definitely giant holes and giant allegiance or abstractions that are been put into place. Kind of the idea and I think Ashka referenced this in a different part of her presentation. But part of the idea of this integration is to allow us to do a very, in some sense, intellectually unhumble thing or just, like, as a first step to build a map, the kinds of things that gov base tries to understand, I e, vis a vis these online communities, and do kind of, like, a nearest neighbor mapping to, hey. This is what a traditional this is the closest thing, in governance archaeology, like, you know, like, what we think was, like, a historical historical community. And that's essentially going through I guess the way I with Oshka's integration and a little bit of creative querying, so we need to do some basically, take some equivalent sets, but that should be doable. I think there are other strategies as well, that we can leverage on top of this integration, but that's gonna take a little bit more work. But I think in the first pass, we we should be able to do, like, a minimum first pass on that kind of query, using this integration, which as far as I'm concerned, it's like, ah, mission accomplished."
      },
      {
        "speaker": "Speaker 2",
        "start": 405.0,
        "end": 405.0,
        "transcript": "Definitely. I agree with that. That would be, like, the cherry on the cake, the seal on the letter, for sure."
      },
      {
        "speaker": "Speaker 1",
        "start": 420.0,
        "end": 420.0,
        "transcript": "Great. Let's continue with the current stack. Val and then Francisco and then Nina."
      },
      {
        "speaker": "Speaker 6",
        "start": 435.0,
        "end": 435.0,
        "transcript": "Cool. Thank you. Thanks so much for the presentation. It's awesome. I'm curious if there's any, like and I've I haven't seen the governance technology database, so maybe I need to look deeper. But if there and also maybe this gets into more qualitative governance question, but, like, if there if you're starting to notice any patterns or links around, like, different mechanisms, like, achieving different results and sort of starting to, like, see yeah. Is that included in this research? Is it in a column in the database or not? Or, like, is it something that's still kind of in a more, like, qualitative part of the the research, and it's not reflected in the database."
      },
      {
        "speaker": "Speaker 2",
        "start": 450.0,
        "end": 450.0,
        "transcript": "I think Riley could answer this and back it up with a lot of data for sure because she's been working on analyzing precisely these kind of questions. But one thing I noticed is and this is probably just going through stuff, nothing very solid. But communities which tend to have a better balance of fuzzier, more ambiguous mechanisms and value based structures and also very clearly laid out rules. Those seem to work best and, you know, last the longest if you look at that criteria in terms of their time span. So, yes, I'm I'm very excited to hear about this as well and how we're sort of gleaning insights on this procedure."
      },
      {
        "speaker": "Speaker 1",
        "start": 465.0,
        "end": 465.0,
        "transcript": "Great. Thank you. Since we, you know, referred to Riley, if Riley is interested, perhaps it's up to the current stack. We can hear a little bit from Riley about some of the work that they're doing. Okay. Francisco."
      },
      {
        "speaker": "Speaker 7",
        "start": 480.0,
        "end": 480.0,
        "transcript": "Hey. Thank you. Ashka, that was an amazing presentation. I was looking into the into this project, and it was very useful to have someone explaining how this was built. And I have one question and one curiosity. And the question is related with, Paul's question, and I want to know that I know that in this current phase, you're lapping. But I want to know if somewhere you have some insights, know that if you have written something. I I'm not asking for an academic paper, but maybe you have some documents or somewhere where I can see your insights or what you have been thinking about this process. The curiosity I have is about scaling because, I'm an anthropologist, and I have, when I was studying, I I I read a lot about Indian populations and the Amazonas. And I I remember that it got my curiosity to see different ways of organizations to ways to solve disputes. But then you think about, like, well, these are a group of a 100 people, 50 people. But how do we implement this in a full scale society, like fully connected Internet with with fake news and all that? So my question is, do you have any insights about the scaling of these different governance strategies?"
      },
      {
        "speaker": "Speaker 2",
        "start": 495.0,
        "end": 495.0,
        "transcript": "Thank you, Francisco. That sounds really interesting, the kind of work you're doing. And I think I'll also link this answer to a question Andrea asked in the chat about wage quitting and sort of exile because that relates really well to scale. So one of the major reasons for failure of the community or, like, disintegration is exactly this. It just grows too large to be managed and for things like, you know, ostracism and, like, reputational risk to be useful, and they're very key parts of keeping the community together. So these mechanisms just don't work when there are 10,000 people. That's why for some communities, they've built this in. They could elect a person, a chieftain, and use that to split away from the main set. And they they're sort of interlinked separate communities, and they coexist amicably, but they're independent units. And in terms of academic research, I think the the sort of most insightful place to look at would be nested enterprises potentially and Ostrom's work on that because that chat tackles precisely this kind of problem. And I mean, I hardly have any answers. I'm equally fascinated by broad questions like these."
      },
      {
        "speaker": "Speaker 1",
        "start": 510.0,
        "end": 510.0,
        "transcript": "Great. Let's turn to Nina."
      },
      {
        "speaker": "Speaker 8",
        "start": 525.0,
        "end": 525.0,
        "transcript": "Yeah. Hi, Ashka. Thank you for the awesome presentations. Very illuminating and so much that I didn't know before. So I guess one question that I have is when I go through GovBase, what can I use as sort of basically an evaluation strategy? Because I want to think about implementing some sort of format that I think maybe is good for my use case. So, like, yeah, basically, like, different ways to evaluate things. What do you what would you and your experience recommend for something like that? Or is that something GovBase is working up towards some, you know, evaluation strategies, you know, something for optimization and evaluation?"
      },
      {
        "speaker": "Speaker 2",
        "start": 540.0,
        "end": 540.0,
        "transcript": "I think Josh could answer the bit about working, you know, what which direction GovBase is going in. But, from, you know, my struggles with it, what I've seen is that the filtering mechanism can be very useful. So what I did a lot of the times is just filter the subclasses the superclasses and look at the biggest category and then see what falls under it and then pick the light right level of, like, zooming in and zooming out. Essentially, I I I mean, I probably need to think of an example to explain that better. But, yes, for instance, if there's, like, governance structures, those could be dictatorships or democratic or oligarchic, but maybe that's not what you need. You need some specific sort of lens on how exactly that works and mechanisms within that and structures within that that makes sense. So I think it would really depend on the exact kind of use cases you're using, and I would love to hear more about that. And if if you, you know, would need any more input on this, I'd be very happy to give him."
      },
      {
        "speaker": "Speaker 8",
        "start": 555.0,
        "end": 555.0,
        "transcript": "Yeah. That sounds awesome. I'll be in touch. Thanks. Great. Thank you."
      },
      {
        "speaker": "Speaker 1",
        "start": 570.0,
        "end": 570.0,
        "transcript": "Yes. Thank you. Let's turn to Riley. I think, you know, some of Riley's work around the kind of analysis of this database is really salient to the conversations that we're having. And in the interest of keeping this conversation lively, perhaps five minutes, and then let's keep the conversation moving."
      },
      {
        "speaker": "Speaker 9",
        "start": 585.0,
        "end": 585.0,
        "transcript": "Cool. Thank you. So, yeah, I can give a little introduction about the dataset and, like, an overview of the analysis and then talk about, like, next steps and open questions. Also, just to introduce myself, if people haven't met me, I'm Riley. Pronouns are they, them. I used to work as a data scientist at Penn Medicine and then as a machine learning researcher at Google. And I've been basically came on this project to do, like, the data analysis piece. So when I came on, this dataset has already existed. I know Federica and, like, her team, has been, like, kind of working on compiling this. So, yeah, I think I can share the Airtube. Like, I think I can share my screen. So to give some background, like, the dataset's not published currently, but that is, like, part of the plan is that we want to eventually have this be a public dataset that anyone can look at, can use, can, like, download and and do their own analysis and research on. And, also, a big piece to that is that we want it to be able to be, like, contributable. So, basically, we wanna make it so that, people from anywhere are able to, like, add to it, basically. So let me just show you, like, little glimpse of what it looks like. And let's see. Do people see an Airtable? No. Not yet. What do what do you see?"
      },
      {
        "speaker": "Speaker 1",
        "start": 600.0,
        "end": 600.0,
        "transcript": "That you're sharing screen."
      },
      {
        "speaker": "Speaker 9",
        "start": 615.0,
        "end": 615.0,
        "transcript": "Let me try this"
      },
      {
        "speaker": "Speaker 1",
        "start": 630.0,
        "end": 630.0,
        "transcript": "again. Still nothing, but maybe it just takes a moment. Okay."
      },
      {
        "speaker": "Speaker 9",
        "start": 645.0,
        "end": 645.0,
        "transcript": "What do you see now?"
      },
      {
        "speaker": "Speaker 2",
        "start": 660.0,
        "end": 660.0,
        "transcript": "He's got Perhaps"
      },
      {
        "speaker": "Speaker 1",
        "start": 675.0,
        "end": 675.0,
        "transcript": "does Ashka have the ability to screen share the same thing and perhaps kind of"
      },
      {
        "speaker": "Speaker 9",
        "start": 690.0,
        "end": 690.0,
        "transcript": "Oh, yeah. Let me let me I know how to use Zoom."
      },
      {
        "speaker": "Speaker 1",
        "start": 705.0,
        "end": 705.0,
        "transcript": "It's okay. Every time Zoom is incredibly adversarial, I would say, to its users."
      },
      {
        "speaker": "Speaker 2",
        "start": 720.0,
        "end": 720.0,
        "transcript": "And I think it it has this sort of bandaccurate against Airtable. I don't know why."
      },
      {
        "speaker": "Speaker 9",
        "start": 735.0,
        "end": 735.0,
        "transcript": "Okay. Do you see a long Could"
      },
      {
        "speaker": "Speaker 1",
        "start": 750.0,
        "end": 750.0,
        "transcript": "you see? Very long. It's okay. Cool. Could you zoom in a little bit perhaps? It's kind of difficult to see."
      },
      {
        "speaker": "Speaker 9",
        "start": 765.0,
        "end": 765.0,
        "transcript": "Make it bigger? Like like, the this works? Yes. Yeah. So there's a couple different tabs. I guess the main ones are community institutions. So, yeah, similar I mean, there's gonna be some overlap to what Ashko was saying. So there's, like, the name of a community, a list of institutions that are a part of that community, some mechanisms that, that community may be using, some, like, culture values, norms. So these are kinda, like, the main pieces of info, and then we have, notes that are, like, manually compiled, time span, geography, size."
      },
      {
        "speaker": "Speaker 1",
        "start": 780.0,
        "end": 780.0,
        "transcript": "And then if I can briefly interrupt, there's a being at a good point. Would you like us to pause the recording, or is this okay to have in the recording as well?"
      },
      {
        "speaker": "Speaker 9",
        "start": 795.0,
        "end": 795.0,
        "transcript": "Yeah. I think it's okay as long as I'm not sharing, like, the whole dataset. I think just to give you a glimpse is okay."
      },
      {
        "speaker": "Speaker 1",
        "start": 810.0,
        "end": 810.0,
        "transcript": "Yeah. If you if you ever need us to just pause it, let me know."
      },
      {
        "speaker": "Speaker 9",
        "start": 825.0,
        "end": 825.0,
        "transcript": "Okay. Yeah. And I think the main thing I would add about this is, like, the I the the end goal is that this is a public resource. And on top of that, we want it to make we want it such that people can contribute to it. So also, sir, for some context, like, this dataset is manually compiled by, like, Frederica and, like, her team through her institution, of, like, I'm guessing, like like, research assistance. And so at least my take on it is that, you know, if we want this to represent, like, a you know, if if ideas like governance archaeology, we wanna cover, like, a a global geographic span. We also wanna cover, like, a lot of different cultures. And so that inherently has to come from, like, people contributing and people contributing from, like, different geographies and cultures and not just, like, from one institution in London. So and and that's something we've, like, talked about too. It's like, we wanna take a decolonial lens on this dataset. We inherently need, like, more, more distributed contributions. Cool. So this is just give you a sense of what this just a little glimpse for context. And so does this do you see a notebook, or is it still the Airtable? Notebook. Okay. So I think just really briefly, just gonna share a little bit of analysis I'm doing, which is a big part of it is just, like, getting this dataset such that it can be, like, analyzed quantitatively. So a lot of that is, like, cleaning up the data. And this is kinda just, like, data cleaning, data prep. So just getting stuff in, like, numerical format even. So one example would be, like, you know, like, turning the list of mechanisms into, like, a bunch of ones and zeros. And so that down the line, what I'm hoping to do is to do some, like, similarity clustering based on, like, what mechanisms are shared, what values are shared, what, yeah, what institutions. And then let's see. Yeah. So in terms of some analysis, like, the low hanging fruit is just, like, descriptive analysis of the dataset. So, you know, putting together some histograms of, like, what mechanisms do we have. Another one that was oh, yeah. The size of these communities. Another one that is useful, I think, is geography. And so at least for me, it was recognizing gaps where, like, you know, we have a lot of communities from Europe, not as many from South America or Oceania. So I think recognizing that this is not, like, a comprehensive dataset. Like, it's very much, like, we still like, we can't say, like, this represents all communities geographically and through history of, like, all these different cultures. Like, we have to acknowledge, like, the context of how we came across this data, how it was compiled, and, like, what might be missing from it. And then I think in terms of next steps and, like, some open questions that I have, Next steps, I'm hoping to focus on some collective governance institution specific analysis. So that's kind of, like, a lot of the data prep that I've been doing is just getting this stuff ready for that. And also doing some, like, clustering and similarities and, like, comparative analysis between a lot of these different institutions and mechanisms. So yeah. So that's kinda like a general overview. Some open questions that I have, I would be really curious to hear, like, what do people want to know about this dataset? So, like, I have questions and, like, the people I work with have questions. I love to hear from other people, like, you know, what are specific things that you think would be really interesting to get out of this. People have any thoughts on distributed data collection. So right now, we have, like, this Airtable. I was thinking of doing just speaking loosely, I was thinking of doing, like, you know, keep the Airtable, and we have, like, a form that hooks up to it. And then we do a little bit of, like, moderation or verification so that when someone submits that form, we can, like, quality assurance it. If people have other ideas, I'd love to hear them. Cool. So I'm I'm gonna stop sharing my screen, and then if people have comments, questions, etcetera."
      },
      {
        "speaker": "Speaker 1",
        "start": 840.0,
        "end": 840.0,
        "transcript": "Yes. Thank you so much for sharing that. Oh, sorry. Please."
      },
      {
        "speaker": "Speaker 2",
        "start": 855.0,
        "end": 855.0,
        "transcript": "Oh, no. But, yes, Riley, that looks really good. I think I was familiar with, like, the low hanging fruit of descriptive analysis, but this is new to me, getting a decentralized input. And this is based entirely on a personal anecdote, though. So a friend of mine who's the history undergrad was very interested when I was working on government psychology, and she suggested a lot of resources to look at as well. So I think another way of collecting data because I don't know about obscure texts lying in the British Library. Right? But she does. And so another way of collecting it would be to ask people to recommend books and, you know, works and have another RA do it because a lot of times, they just want to give you the information but don't want to actually code and put in the work. So somebody who's very interested in doing that like me, you know, an archaeology geek would enjoy that."
      },
      {
        "speaker": "Speaker 1",
        "start": 870.0,
        "end": 870.0,
        "transcript": "I'll also add, I mean, perhaps you had a chance to meet, Dushini, at the web camp, but they are doing work around, I think it's called lab in the wild. I shared a link to it, in the Zoom chat. And it's very much trying to figure out a decolonial approach to, data collaboration and data contribution. So very happy to connect to both of you with with Najini. I think he could be a really good collaborator for that initiative. There is a question here from Drea. I'm gonna turn to that and then see where we go with the conversation after that."
      },
      {
        "speaker": "Speaker 3",
        "start": 885.0,
        "end": 885.0,
        "transcript": "I think, actually, Riley just spoke to it at the end. It's like, okay. Yeah. We'd love to know, like, what questions we have, what questions do you have. And and having doing that having that public would be fun. I love seeing the data of, like, the the sizes and the I I I love the histograms. Thank you."
      },
      {
        "speaker": "Speaker 1",
        "start": 900.0,
        "end": 900.0,
        "transcript": "Great. So there aren't any current questions that is referencing here. There was a question previously from Steve about reputation systems. Steve, do you do you, you know, see that as still relevant? I just thought, was there anything either of your works about reputation systems that you found interesting, that you found surprising? I can't hear you though."
      },
      {
        "speaker": "Speaker 2",
        "start": 915.0,
        "end": 915.0,
        "transcript": "Sorry. Yes. One thing I came across was that ostracism was very, very strong when handing handling sort of more high value goods. So communities of merchants, which handle precious metals of resources for them, obviously, reputation was important, and that functioned, you know, independently of any legal mechanism. So it just seemed very shocking to me that without any other safeguards solely on the basis of doing trade again, it could work really well."
      },
      {
        "speaker": "Speaker 1",
        "start": 930.0,
        "end": 930.0,
        "transcript": "May I? Yes. Yes, sir."
      },
      {
        "speaker": "Speaker 4",
        "start": 945.0,
        "end": 945.0,
        "transcript": "I'm I'm curious about sources. So a lot of these sound like, you know, from the from the forties on, anthropologists have been really active creating, you know, a trap and really kind of comprehensive structured databases of cultural structure that includes governance variables. But it looks like the the sources that, y'all are, drawing on are a little bit more I don't know. And I see Stasovision there, for example. It it it seems more, like, convenience, based sampling. And then so I'm curious, like, what what's the basis for picking what you're using to populate the cases, and maybe what's the shortcoming of these sort of more traditional, you know, more or less complete existing sources in anthropology, religious studies, and so on that really have giant structured databases for things that are adjacent to this."
      },
      {
        "speaker": "Speaker 9",
        "start": 960.0,
        "end": 960.0,
        "transcript": "So I can speak I can't speak too much about, like, the like like, in terms of, like, the data that's already there, but I can definitely speak to, like, what I see as kinda, like, the the downfalls to it and, like, the gaps and what I would like to see, next from it is yeah. I mean, a lot of historical like, at least for me, like, I would really like to take a decolonial lens to this dataset And just, like, kinda, like, inherently relying on, like, more mainstream text for anthropology is kind of, like, recreating the, like, oh, this community is, like, exotic. And so a big part of, like, opening this dataset up for contribution is something we've been talking about is ideally if we could have some, like, grant funding to partner with, like, collaborators in different geographies and have backgrounds in different cultures and have them be able to also contribute to this dataset and, like, source text so that, at least for me, it's just, like, inherent to having, like, a distributed data collection process where, like, it's not just, like, one or two people from, like, this one institution in London. It's, like, having many contributors that we're able to, like, compensate from, like, a lot of different geographies, especially given that there are, at least from my perspective, like gaps in the dataset, like, underrepresented, like, geographies that very much have collective governance institutions and are just not being represented."
      },
      {
        "speaker": "Speaker 1",
        "start": 975.0,
        "end": 975.0,
        "transcript": "Great. I think if we hazard another question, we would go over time. So maybe let's take this this these last two minutes to thank our speakers today, Ashka and Riley. Really wonderful presentations. And, I invite everyone to, unmute and give a round of applause for our speakers. Three, two, one. Thank you. Thank you. Thank you. Yes. Very, very interesting. Such a I think Bea wrote it so nicely. It feels like such a gift. So thank you all so much for attending. I also wanna invite folks who are on this call, if you're not already, to join our Slack. I'm gonna post the link there, and I'm gonna make a thread where we can continue having a conversation about this. And if you have questions that emerge over the next couple of weeks or anything that you'd like to speak about or people who you think that we should reach out to for collaborators, please do find that thread in the seminar discussion channel and contribute. And with that, I'm gonna end the recording and close the meeting. Thank you all for coming."
      }
    ],
    "summary": null
  }
}