Plurality: How Taiwan Managed to Unite Its People Through Tech - Audrey Tang & Glen Weyl. Ep. 553

Transcript

that one of the greatest forms of freedom is freedom from Politics the sunlow movement which is half a million people on the street and many more online um participated in a nonviolent demonstration occupying the parliament for 3 weeks uh to not protest but rather show uh what would it be like if you involve the entire Society in a deliberation the great thing about poce is that if there's like 500 people all very loud but Bel laboring the same point then it doesn't increase the area in polus it doesn't increase de statement's ranking because instead of sorting by controversy by default uh on PO is it's sorting by Bridge making by default if you have that sort of a

process and it search serves some things is bridging and some things is divisive both can be very beneficial because you don't take action on the divisive things but you form Community around the divisive things so people can come out of the process learning both that they have new friends that they can Ally with and that there's some things they can move for it on together welcome to epicenter the show which talks about the Technologies projects and people driving decentralization in the blockchain and Revolution I'm freder ANS and today I'm super honored to be here with Glenn W the founder of radical exchange and Audrey Tang the first digital minister of Taiwan who both wrote a book together it's called plurality and uh it's on the future of collaborative technology and democracy um it's ground it's a

groundbreaking book that explores how digital tools can enhance um Democratic processes and Foster a more inclusive and better Society before we talk with Glennon Audrey let me tell you about our sponsors this week this episode is brought to you by gsis gsis builds decentralized infrastructure for the ethereum ecosystem with a rich history dating back to 2015 and products like safe cow swap or nosis chain nosis combines needs driven development with deep technical expertise this year marks the launch of nosis pay the world's first decentralized payment network with

agnosis card you can spend self- custody crypto at any Visa accepting Merchant around the world if you're an individual looking to live more onchain or a business looking to wh label the stack visit nosis pay.com there are lots of ways you can join the nosis journey drop in the nosis Dow governance form become a nois validator with a single gno token and lowcost Hardware or deploy your product on the evm compatible and highly decentralized nosis chain get started today at nosis doio car one is one of the biggest node operators globally and help you stake your tokens on 45 plus networks like

ethereum Cosmos C last year and dydx more than 100,000 delegators stake with corus one including in institutions like bitgo and Ledger sticking with course one not only gets you the highest years but also the most robust security practices and infrastructure that are usually exclusive for institutions you can stti directly to cors one's public note from your wallet set up a white label note or use the recently launched product Opus to stake up to 8,000 e in a single transaction you can even offer high yield staking to your own customers using their API your assets always remain in your custody so you can have

complete Peace of Mind start sing today at chorus. one so welcome welcome to the podcast it's super nice to have you on good Lu time very happy to be on the podcast delighted to be here again it's been a few years it has been a few years so I think both of you are pretty welln so I I I will I will venture um a an introdu an introduction nevertheless so Audrey you were taiwan's first minister of digital Affairs um you are a globally recognized digital Pioneer you have been relentlessly promoting transparency and

um in Innovation and you have played a pivotal role in taiwan's digital democracy in initiatives you are also widely known for your contributions to free software and kind of uh for General advocacy in the digital space and Glenn you've been on before and back then kind of we talked about radical exchange the book you had just released you are a principal researchers at Microsoft research you are a rare breed in that you are both uh an economist and a social technologist um you have worked extensively on radical markets Innovative governance models and so on the recurrent theme is

um kind of this uh this uh uh conversations on how technology can transform societal structures and recently you you co-wrote a book with a bunch of other people called plurality that was a very long introduction by our standards so tell me about yourselves how did you two meet and what inspired you to actually write this book together uh well because of radical markets I met vitalic buer and we started working together he introduced to Audrey uh we started working remotely with each other on some of the applications of idas and radical markets in Taiwan and then had the chance to meet in person together in Berlin the same place we met you uh and

uh that was an experience that kind of changed my life uh Audrey um both who she is and what she was doing really changed my perspective on many things and my life started to sort of Orient towards that over the coming years and um eventually uh I managed to kind of free myself from the role that I had at that time which was fabulous and fascinating but constraining in my ability to speak publicly uh I was in the office of the CTO at Microsoft and when I did sort of the first thing I wanted to undertake was to write about in a more sort of analytic crisp way with Audrey um the things that she has uh done and that's

um was the Genesis on plurality book yeah and when I uh received the initial outline of the book uh I was uh just um kind of wrapping up my work in the counter pandemic uh which included um you know mosque provisioning and contact tracing vaccination all that uh and so the teams that I interacted with the most uh during the pandemic uh we decided to band them together so whether it's Universal Services um spectrum allocation uh platform economy um Department of cyber security data pipelines and so on uh we decided to just B them over into a single Ministry called the ministry of digital Affairs um and so it is kind of

a blank canvas uh and Taiwan never had such a dedicated digital Ministry with such all Inc compassive roles so in a sense a glance outline of the fundamental human rights uh that we should guarantee on the internet from a human right perspective of not individuals but rather uh Associated individuals in groups as well as this idea of plurality or collaboration across diversity became kind of the blueprint of the ministry of digital Affairs that was in 2022 wow there there's a lot of information here so we we we know you both knew viik uh so Glenn because you had re co-written your previous book with him um H how did you meet him

Audrey oh no I did not co-write my previous book with viall I wrote it with Eric po but vitalic uh picked up on that book and a big part of its reception in the world became because of his interest in it and the things we built based on it together so it's it's not literally true that we wrote it together but he did end up writing the preface for the for the paperback edition so ah sorry it's kind of like quadratic voting and kind of harberger taxes and so on they they are just they are just in my brain they are very tightly connected to viik I'm sorry for the Mis misattribution yeah it's actually very funny because recently uh some several people have come up to me and said have you heard about this idea from vitalic buin called quadratic

voting and uh I was like I actually heard about that quite a while well before vitalic uh heard about it but I'm delighted that you heard about it now as well so well we kind of preempted this uh dynamic in this particular book by relinquishing the copyright so plurality Donnet is entirely in the public domain so it doesn't matter if you take a couple paragraphs and say it's your work yeah so attribute it to whoever you want to attribute exactly unless you attribute it uh to yourself and kind of uh you get done in for for plagiarism which uh is uh always a very real danger in Academia no the the the the whole point of of public

domain really is um that the ideas uh are the protagonists we are just the vehicles uh the vessels right inhabit those ideas before we go into the ideas that kind of form core of of your new book Audrey let's talk about you for a little while because I I think it's it's exceedingly rare that kind of people who kind of I would classify as kind of like hackers in some sense actually go into government right so I think this is this is something that um because working with governments often very bureaucratic and frustrating um H how did you end up in this very unique position of first first digital Minister well first of all

it is not that rare in Taiwan to have actual experts as ministers my successor the current digital minister is not just i e fellow but also a expert in privacy enhancing Technologies and zero knowledge and all sort of The Cutting Edge things so I think that the point I'm making is that um Taiwan had a long history uh of working to think of democracy as a social technology not as a like 200e tradition well we only had our first presidential election 96 so at that time already the wari web is part of the conversation now as for me personally um I've uh worked as a

reverse Mentor a young person younger than 35 um to advice cabinet ministers in 2014 and the reason why I was involved uh was that in 2014 the Administration was enjoying uh a 9% trust rate uh from the citizens because of its opaqueness handling a trade deal with the Beijing regime uh the cssa the crossray service trade agreements now and so the sunflow movement which is half a million people on the street and many more online um participated in a nonviolent demonstration occupying the parliament for 3 weeks uh to not protest but rather show what would it be like if you involve the entire Society in a deliberation about this tree deal with a

lot of facilitation a lot of live streaming and all the Cutting Edge Technologies at the time all so because of the demo was successful at the end of three weeks the head of the parliament essentially say okay if that's what people want would just ratify it so it was one of those very rare successful occupies and so the cabinet members then tapped us the young uh facilitators and so on to serve as reverse mentors and I work with them for two years always saying that I'm working with the government not for the government uh with the people not just for the people um before being I guess promoted uh into a minister myself two years after that that was in 2016 so I mean you travel the world a lot do you feel that um the Democratic especially now uh she it was harder for

her to travel before but now it's uh easier yeah I I've just been in like seven European cities and now in South America do do you feel that the democratic tradition or I mean it's not that much of a tradition because as you said it's not been around for all that long but do you think it marketly differs from Western Europe or the or North America well I would say that I was just in Finland uh and there's a lot of similarities I would say uh in the sense that um it is not about polarization it is not about Like Us Versus Them it is more about a matter that everybody thinks together uh and a high trust now

in Taiwan uh in the institutional capability um to think Beyond polarization to co-create so I would classify taiwan's democracy as uh embracing the conflict so whenever people see a incoming conflict instead of thinking that half of the population is against it and half the population is for it um it we're usually about creating a space where those new found conflicts and so on can be tapped into energy so one metaphor I usually use is that we have a lot of earthquakes because we're caught between the tectonic plates of uh uration on one part and Pacific Plate on the other but because we have very good resilience in our building coat each earthquake

actually push the tip of Taiwan a little bit upward um toward the sky so to speak so to make a more synthetic way of looking at things whereas I would say in many more fossilized democracies people feel that they have to wait for four years uh to make a change and you see a lot of this flip-flopping like One S side takes it and four years later the other s win and four years SI the other side wins and so on there's less of this Dynamic co-creation potentials yeah absolutely I mean I would say this is definitely definitely the case uh in Germany where I'm from and Glenn I I I'm sure you would agree it's most definitely uh the case in your

native us as well except I'm also a German citizen so both are native to me both have similar is no idea yes fantastic um yeah but how do you create this atmosphere where people kind of start creating together rather than kind of entering this tribal thing where kind of it becomes about kind of like identi identity and kind of how you identify and kind of who who is wrong and who is wrong and who's right on a very primal level yeah I think uh a lot of the recent polarization can be traced uh into this uh like button and the share button and the retweet button nowadays

called the repost button uh because uh a lot of those buttons do is to surface to the front only the viral ideas that puts people into a very quick impulsive action of clicking you know like or uh retweet and so on and so in Taiwan in 2014 uh we observed as part of the sunflower movement that the social media is somewhat antisocial in a sense that it manufactures counterpower very easily but it doesn't build Bridges easily so we decided to build uh or adapt our own pro-social versions of social media so uh at the sunflow movement we tapped into a lot of co-creative tools um at

the time hackpad lumo as so on many of them are not uh even the r any more HEC pad become part of uh Dropbox as but one thing stay uh which is the ability to create the toolkit of bridg making pro-social social media so one example is this system called polis which we deploy in 2015 um the idea is that you can be Pro Uber you can be Pro taxi you can be Pro this or that but on this social media you can post your comments for other people to like or unlike but there's no reply button and there's no retweet button so there's no way to dunk on each other um the only way to get viral so to speak or visible is if you propose a statement that was also liked

by the people where I like you so the more that you unify the more Bridge you build the more visibility your ideas get and people see uh very clearly how the different islands are and how they're brought together by ideas for example that you know raising prices is fine during a price surge but under cutting existing isn't fine or that insurance is important as so so very Nuance very eclectic ideas and we honor the top 10 ideas by reading them out to the multi-stakeholder meeting and have the minister deciding the law uh based on those crowdsourced agenda and so giving voice to those Bridge makers has always been the core idea collaboration across

difference and we're very happy now that the community notes of x.com previously Twitter or nowaday YouTube and so on are all picking up on this ideas yeah super cool I I kind of I want to go into the polus thing in just a bit I just wanted to ask there was um a short story on the internet a couple of years ago maybe like five or 10 years ago um it was called sort by controversial I'll put it in the show notes I'll send it to you guys later so basically um it's it's um this idea that um it's a short story so basically it's it's it kind of comes across as kind of like a comment on social media or Reddit or something where someone describes that they worked at a social media

company and they had made a startling Discovery um namely that there is such a thing as scissors statements so basically statements that you can craft such that they kind of divide the PO population into more or less um equalized um chunks um and half of it will agree with it vehemently and the other half will kind of um defined the notion of that statement utterly abhorent so kind of like there's no middle ground at all um and kind of they they they kind of they mention examples of kind of like where this was dropped and kind of it's it's um it's kind of like it's like this conspiracy leaning

thing where where um kind of governments use them to kind of cover up things that are really going things that's really going on and so on and to a certain extent I actually think the the author actually picked up on something that actually really does happen so it feels like there's a lot of manufactured conflict out there oh yeah definitely yes there's actually very clear research on this you don't need to uh you know use it hypothetical Molly Roberts Gary King and Jen pan have a wonderful article showing that the most effective and in many cases in their sample of research at least primary uh approach of the Chinese government is not censorship

instead is the manufacturing of scissor statements in context when Collective organizing might otherwise occur so that uh the group that might organize is distracted and divided against itself by the scissor statement it's what we call polarization attack or mass destraction or whatever yes what you guys kind of advocate for is kind of like this um this culture of kind of like bringing everyone together um at the same table and kind of just surfacing the differences to kind of make sure that kind of a compromise can be reached that kind of that honors everyone that everyone's equally unhappy or happy with

um how how do you but how do you actually get to that stage how do you actually get people to the same table I wouldn't say that it's uh trying to get people equally unhappy you're equally happy I would say that it's coming up with something we can all live with uh so the difference here is that it's not trying to reach a conclusion or a decision but just some lower hanging fruits that let people see that after all we think similarly as our neighbors on some very basic things right so for example um we uh sent um 200,000 SMS uh in my previous job as digital Minister uh to random numbers in Taiwan um so

anyone who with a phone has a chance to get a SMS number from 111 which is a recognized government only number uh and of course contain with it a survey uh asking what do you think about information Integrity online what do you think about the def fakes and cheap fakes and fraud and whatever right so the idea here is not for us to come up with a preset um compromise positions and ask people to pick their position uh the idea really is to ask people how do you feel about the current situation and there's far more potential for overlap for feelings instead of solutions and then after that uh we chose uh through random stratify sampling 450 people that

are statistically representative of the Taiwanese population uh and randomly assigned 10 uh people into the same room so for 45 rooms of 10 people each and they just deliberated just have a conversation um and there's no facilitator the room is the facilitator so it nudges people who don't speak uh he allows Interruption budges for 5 seconds it knows the common points but also knows the points of differences make a whole transcript and all that but the end result is very very nuanced it is not getting people to a place where they all agree 100% but most people actually agreed for example that if you uh are a large advertisement platform

that reaches more than 10 10% of Taiwanese people and you have advertisement but you do not do kyc like digital signature or whatever so that you open up the scammers uh to Cal people using def fakes then if somebody get C for $1 million you probably should be liable for that $1 million so what this technology does is that it gets into this very compressed but not lossy compression very nuanced positions that can go straight into a law it's actually already in the parliament now and to your question why people participate well because it's effective it is much more effective than waiting for four years and also we pay people some I don't know hundred us for their um time to participate in this process I you

know the thing I'd also say is uh I I think that s scissor statements aren't good or bad it all depends on context in fact it's great to expose people to the scissor statements in the context of understanding this is something that would divide the population here's of people who would like it here's who not there's also other things that would you know uh not divide the population you know the way I think about this is uh we're very grateful to the you know Chinese Communists for the you know party they did great research uh just like medical research you can do the research and you can find the vaccine or you can spread the disease right and so we just believe in spreading the vaccine which is that you you give a little bit

of it at low volume to a large population rather than injecting a very large amount at a very high volume to a small population right right you make it a social object so people can't talk around the fact that there is this division in our society but you do not amplify that division itself it is similar to mRNA vaccine just as Glen said yeah or or with you know my my children it's actually great for my children to have intense emotions and to reflect on those emotions and understand how those emotions affect them it's not so great for them just to completely experience without any reflection those emotions that get lost in them right yeah I I I totally hear that if

you kind of look at these Taiwanese examples of these deliberative um processes I I kind of I hear you say that it's a representative sample of of society kind of when we go to social media where kind of a lot of kind of like this um these Echo Chambers ex exist I feel it's very much not a um a faithful representation of society it's kind of just very few people who are very loud um do you think there's a way of kind of like tuning that down as a society and kind of giving more um giving more weight to the to the quieter people yeah I think so I mean part of the design of poish which by the way

isn't just used in Taiwan I was in Finland and Finland have this uh what do Finland think conversation that has almost a million votes on Pol bigger than any Taiwanese Pol single yeah it's bigger than any Taiwanese polish examples and the most divisive Point by the way in Finland uh a year ago uh was this statement that says nowadays you can't say the words like jokes or whatever as you could anymore without offending someone end of quote uh as so which uh really portrays you know uh the new generation maybe in the past 5 years or so on uh the awareness of how words can offend or trigger people so that's uh their divisive points anyway but by point being uh the great thing about

polus is that if there's like 500 people all very loud but be laboring the same point then it doesn't increase the area in polus it doesn't increase deer statements ranking because instead of sorting by controversy by default uh onul is it's sorting by Bridge making by default so if 500 people make essentially the same point over and over again they're really in this very small area uh and there's no bridge to be made and therefore um it's not visible at all uh on the scoreboards the scoreboard only highlights really the ones uh that Bridges across large divides which is difficult to come by so the idea of a bridging bonus or just changing the

Sorting algorithm by default gives the voice to the quiet people and if you're uh the quiet people that doesn't think in a mainstream way it creates a ey of people that is quite distant from the majority clusters but it also means if people manage to bridge your view with theirs then the bridging bonus is actually larger and you see this dynamic in Pol in community notes and nowadays in YouTube noes as well and by the way I don't think there's anything wrong and polus does allow it for those uh divisive things to exist it's just that just by screaming louder they shouldn't be the only thing that gets seen it shouldn't emply

by getting more uh louder and screaming it yeah so we've talked about polus quite a bit maybe let's kind of dig into what that is so kind of like in my understanding polus is um this clustering algorithm that kind of takes things into n-dimensional space where kind of you have nend statements and everyone says yes no to the statements and then then it kind of clusters you in that it clusters people in that n-dimensional space so kind of it shows which opinions kind of clustered together and um where um where different clusters kind of actually meet is that more or less that's more or less it okay Des description yes yes uh and it's coupled with a visualization like a group selfie

uh that's dynamically updated so that people can see what new uh divisions are being formed uh the uh mathematical term is uh a dimensional reduction so that you can always see what are the like two most div Ive points that currently separates people aparts yeah the way I one way I think about it is there there was a regime in the United States that's very famous and often hearkened back to called the fairness Doctrine and the Hutchins commission and the idea of these was very simple it was that what media was required to do was surface and report as news things that people generally would agree on and for anything that was not like that they had

to show the different sides of the debate right so is just a way of doing that so like the facts that people agree on the come on facts um the divides so take one take two uh and also not confuse the facts with opinions so what do you feel so your own feelings about it so it's these four different segments this this is a very simple thing to do but it used to be very expensive to do this and require a lot of gatekeeping and top down and whatever and po has just do a way to like cheaply do that in a social media environment rather than have the social media environment be curated by some other set of principles that applied before the Hutchin commissions were which was called yellow journalism where you take the thing that is most divisive and you and and

basically we the US moved from one regime to the other and and pol is just a way to move social media in the same way but basically for that to work you actually have to agree on the facts first right no no because they get surfaced by the algorithm as there's no prior definition of facts facts are the things facts are the things that people all agree on agree that other people agree on exactly okay and basically if if if there's too little com Common Ground so kind of I mean because kind of like I think this is something that kind of we have all noticed over the last couple of years that there there are there's now a very significant number of people who kind of deny obviously true f

facts well I I mean it depends on how you see the conversation uh I remember at the beginning of the pandemic uh there were a lot of different takes different um scientific opinions even about the efficacy of musks but one of the communication statements that we push out in the early 2020 uh was a very cute dog as Sheba enu putting the Paw on her face saying wear a mask to protect your own mouth against your dirty and washed hand um so I mean for that people generally agree on because it doesn't talk about virus or aerosol or whatever it's mostly musque as a you know handwashing reminder um and so I think it's called overlapping consensus you

can almost always find some lhing fruits like that that people maybe laugh at it maybe it's humorous but at the end of the day nobody really seriously dispute that and and by the way you know let's take about as divisive an issue as it gets the Israel Palestine situation Community notes contrary to what many people would think has actually been enormously valuable in that situation of course there are many notes that get posted that are very opinionative that don't make it out but I Community notes has played a huge role in debunking a huge range of things because there are a range of just verifiable things uh using what people in this audience might call the blockchain but other people would

just call time stamps you know uh that that just can show that certain things are not true and there there's a wide range of things where community to play the key role in yes so I think it's the thickness right in Polish you cannot unsee the bridges in community notes you cannot unsee the top notes that is posted along with the Tweet you you literally can't take it down um and so when people become in that context aware that there are some things that are bridg making then they cannot think anymore that we live in a Posta world I actually love the approach of community notes um so and the way that I'm I'm not sure whether everyone's familiar with it but the way that it

kind of works under the hood is that kind of if someone posts a community note then uh then kind of like you can opt into kind of seeing it first and kind of up voting or down voting it but basically if uh people who have very different um social media patterns actually upvoted it gets a lot more visibility rather than kind of if like uh 10,000 people who do the exact same things and like the exact same things up voted then kind of that that gets seen much much less than kind of if it's a diverse crowd is that fair right it's the same breaching bonus it's actually inspired by poish yeah super super nice yeah you you guys have many many examples of kind of these

uh um these collaborative Technologies and kind of um H how to um how to um use them to Aid um this notion of digital pluralism but I kind of I also want to talk about kind of how you see kind of like web 3 playing into this so and for that kind of I actually want to start super simp simply kind of how would you yourself kind of De Define web 3 what does it mean to you oh um so as you know i8n is just a shorthand for internationalization so to me web 3 all overc case with no space web 3 is just a

shorthand for desc calized uh so whenever I see decentralized and I don't want to type that many letters I just write web three how do you think we can use um web 3 Technologies to kind of Leverage four outcomes you would like to see in kind of the the democracy and collaborative Arena well let me let me give you an example of a tool but conceived in a very different way so like the default imagination for many people of what a web through technology is is like Bitcoin like a like a open public fi fully financialized blockchain but here's something that uses two web3 technologies to build something very

different imagine that we have a group of people and they all uh are exchanging messages or maybe even payments or something like this um and all of the messages are on a shared Ledger but it's protected by a designated verifier proof that is uh the only people who can have confidence that the message was sent correctly are people who are part of that community so that is a distributed Ledger and a zero knowledge technology put together to create something that's actually almost completely opposite to the notion of a financialized public blockchain because that group of people is the only people who can know what's on it and they it

can't be permissioned by a Global Financial value it's permissioned by some social certificate of that community so I think you know tools can be very useful but they can be used in a variety of different ways yeah uh and uh in my previous job uh when we faced unprecedented uh denial of service attack in 2022 um on our uh websites um we uh went to this technology called ipfs interplanetary file system uh and we use it just as a super CDN uh where people can just pin our website at ipns uh cola. ov. tww to keep us afloat uh so that when people even in very low

connection um capabilities want a copy of the official websites and Communications and so on um they can actually go ahead and do it without relying on any centralized CDN systems now of course we still use cloud layer and many other CDN Solutions but having the ability for people to participate and through participating verifying that we have not tempered um the official website that is actually very very useful uh because that means the code um is all on JavaScript is all on the front end we have nothing to hide when we publish um my you know transcripts of lobbyist visit or interviews or whatever it lives on your computer too and you can verify that I have not tempered with

it uh because tempering with it will change the content addressable hash code again uh I don't earn file coin this is not a cryptocurrency application but wins trust by trusting the people okay so I kind of what what I hear is transparency how do you guys feel about co-ownership because that's another thing that kind of like web 3 technology enables at its schol right so kind of like when you're on the internet today um most of the value you acrw you actually acrw to the same five companies do you think kind of we can turn this thing around um by leveraging um decentralized technology well I I think cl's example is is actually co-ownership uh the ownership is around the Integrity

of the context so in the group of five only these five people own the contextual integrity and for anyone outside of that group uh anyone has deniability right uh if you get a screenshot or whatever it it really doesn't work that way right so I think ownership is not meant for me as something that is necessarily about a financial value of things that you can sell or things like that is more like the collaborative governance that people would not exceed the expectations of each other so that we co-own this context but we not necessarily assign monetary value to it yeah I think um there are many different ways to

conceptualize co-ownership and I think that often in some web3 communities the only way to think about that is financialized and like there there are some limited use cases where Financial ways of thinking about things are are the right way to approach them but uh I would argue that that's actually a very simplistic way there's you know it's kind of choosing one out of a million and then putting all the focus on that one uh which really is kind of a edge case that you might want to use in a very limited set of contexts um so I think that that may be the mistake is that to take a good idea like co-ownership and then assume that it

means you know monetary very financialized spongeable blah blah blah I would subscribe to that and I think kind of to me the monetary element is just as much about not not um feeding the man in a way so kind of not actually by by kind of being on Twitter which is the the only social media platform I'm actually on uh but to to not kind of acrew all that value that I create to um Elon Musk I'd much rather kind of have it go back to the to the entirety of the Twitter usership rather than and I and I think kind of the this

um hyper finan financialization where kind of you kind of you know the price of everything and kind of there's like a token behind everything and so on I think yeah that's obviously that's a misstep I think it's kind of just because people love to speculate right I mean people love number go up but um the the fact that you can actually create some that is legitimately co-owned by tens of thousands or hundreds of thousands or millions of people I think this is fundamentally new right well I mean I I'm not sure I would say it's uh all new but it's made it could can be made more efficient so um there's a old literature and sociology that is all about

different symbolic media so uh money is one voting is another these are two ways that we take sort of inputs from people and then allocate something based on them but there are many others there's like the way that if you've worked at a particular restaurant that you might be able to work at another restaurant later in another country and or there's the way in which if you've written a paper and it got certain kinds of citations that that might enable you to get a job at University which might let you write other papers like all of these are different uh maybe imperfectly or in incompletely articulated mathematical functions that take various kinds of inputs and turn them to outputs you know

there's a game civilization sex or civilization in general where there's many different media there's like faith and there's money and there's like scientific whatever and like I think that's a very nice like simplistic model of the notion that like yeah there can be many currencies but they're not all like exchangeable in the same ways they can be used to do different types of things and I think that's existed forever you know that that's the way that most things work it's just a very easy thing to talk about money and because it's easy to talk about it gets talked about more but it's not actually like that runs most things in the world and so I think the misstep that we've taken is taken the simplest easiest to describe thing not the most useful not the most relevant not the most helpful thing and only try to describe that one

thing which other things do you think are helpful to look to look at in this context well I I mean Community notes is a really good example because it's a open- source decentralized thing right V like set up his own node a very fire node a validation node of community nodes um so it can conceptually exist in a separate layer uh that is not tied to x.com as seen uh recently by the YouTube uh adopting something like that right so I think um a a community node layer uh on the entire social network it is something that we can't imagine already now uh in a decentralized way um and because I already said that I just

abbreviate decentralized to web 3 so I would say that it's a web3 layer than around um social media and that I think uh addresses one of the over concentrated um Nexus of power exactly as you said uh that my posts my accounts as on x.com um should not be simply for somebody's taking if I somehow want to exchange Community NS across the fediverse or across many other interoperable product I mean just just to give you a very simple example you can imagine that there's a person like three people all of whom have made the same amount of money one is like a social entrepreneur Who's involved in lots of big world

causes a second person is someone who works at a very standard company and spends most of their time with their family and a third person is a local industrialist who's done great things for a town but is not widely known outside of that all these people might have made the same amount of money but the types of social roles that we would want to place this person into are radically different right um like the first person might be a good candidate for like a un leadership role or to be on television you know in a global way the the second person might you know reasonably have a good degree of material Comfort but it's not someone you'd want to really be in a social leadership role and the third person is someone who might be a very good mayor

for a local community but probably not in either of the other two roles right and so clearly and and that's like that doesn't require any moving away from our like that's just true in our society but the problem is our money systems don't capture that like you you need to read a biography of the person and run advertisements that's a very inefficient way to gather that information so on the one hand we have this very efficient thing called money that captures very little it really just doesn't tell us about most of what we care about and then we have these very inefficient things that are capturing more with social richness so like why rather than just like doing the very easy and I think kind of stupid task of making money like you know go more and more places and faster and faster why don't

we try to take some of these things that are totally inefficient and try to find some way to capture them better so they can scale a bit more so so that that would kind of uh take the form of kind of like um some sort of reputation score social graph or how how would you kind of envisage that yeah I mean like in the book we have uh quite a lot of description of a range of different algorithms like this but you know like one simple example uh of this kind of thing is um the recovery of a key so you might want to recover your key and you might want to give some Guardians that let you recover your key

uh but you might want these Guardians to be diverse relative to your social graph which is to say you might want to choose five guardians who the only point of social connection is through you uh because then it makes it almost impossible for them to collude with you like almost any way that they would try to collude with each other would have to go through like either contacting you or like someone who's a very close friend of yours so that's a very secure way um that is based on a reputation but not based on a reputation that's like this person is good and this person is bad which is just as stupid as money it reduces everything to one dimension instead it's based on an understanding of the social context of these people and the notion that when it comes to being your form of security it should be

something that connects only through you rather than through other routes yeah that's that's super interesting and I personally have thought about recovery a lot because uh I started a project called nosis safe it's kind of an ethereum wallet a multis wallet and um we we once did a survey among users kind of asking them kind of how they would choose Guardians to kind of for key recovery and there the thing that we heard most often was um that in principle they would kind of love to take people like from their real lives kind of their sisters or brothers or fathers best friends and so on um but

they don't want to do that because their balance is visible so kind of like anyone who kind of who who would be who would be a guardian to their account could could immediately see how much funds they actually have on that address which I thought was um so kind of we thought about kind of like how would you choose them so they wouldn't collude with each other and so on and this wasn't people's concern at all like almost no one actually brought that up so uh I and I think kind of I mean obviously those are kind of like things we will uh we will move past kind of once we have uh better privacy on Shan and so on but in principle it kind of I think it kind of shows how important it is to kind of like to to to speak with

people to see what what it actually is that that bothers them so um yeah very interesting I'm not sure how into kind of web 3 governance uh you guys are but it's terrible so kind of like despite the fact that kind of we often say that um leveraging decentralization for better governance is kind of where we want to go and kind of how we want to leverage this technology actually web 3 governance is terrible so kind of there's this continuous uphill battle of low voter turn out and a voter fatigue and kind of attention as the scarse resource and so on and I think in

some some way it actually makes sense because um kind of like in a way democracies are also just very convoluted algorithms um and yeah they could probably do with um uh a little refactoring but in principle a lot of parts are there for a reason um and kind of actually making people vote on almost everything is not a very scalable way of kind of like running any sort of operation do do you think it kind of it it will require a more orchestrated push to kind of get some of these things we already know work in kind of like traditional democracies um into this decentralized world or do you think kind

of we should continue doing what we're doing now which is mostly kind of like very kind of like evolutionary in terms of its you know algorithms just starting from scratch and making it marginally um more complex here and more complex there and kind of adding to it rather than kind of replicating things that are already there I mean Hana arent famously said that one of the greatest forms of freedom is freedom from politics um because most people don't you know you know want to spend too many evenings working on socialism and so the question is how can you make it very easy for people to express the things that really matter to them and not

impose on them to do everything else that they don't care about for example uh you might think I'm a very politically sophisticated person I'm involved in these Global issues I have no idea about local politics where I live uh and I really wish I could uh give that over to people in my Community who are knowledgeable but with some things that I feel distinctively so we need to design ways that make it possible for people to do that that make it possible for people to opt into participation without the converse of that which is that of course then the most interested people can get too much of a voice and that's a lot of the problem of of Designing these things is

taking those issues seriously and designing systems expecting that that is part of the design rather than thinking that we just want some sort of idealized simplified model uh that and the thing that performs best in that case yeah and I also think it's far easier uh to look at one statement from your fellow Citizen and say I feel this way or I don't feel this way um which is the Pol way uh or to be in the room of a video chat of nine fellow citizens uh and just listen what they're feeling and share your own feeling in both of these modalities um it you get to know view

points no new people and there is no kind of decision making burden on you like you're the final Arbiter or the final referendum Caster or things like that is much more exploratory and very interestingly we have discover earlier in the process uh of a decision making the more you ask people in the explor moratory way Discovery in agenda setting way the more interest they have uh in learning because they get something out of it they they get Love Actually uh Community love that is to say they get to know know more people and the more you're finally at a point where you have to count the votes um you have to make the final decision referendum and so on

U the less uh agency one have on it because at the end of the day um half people will feel they have lost or something like that right so I think uh the ear ier the engagement is the lower the threshold it is and the more like psychological support uh we provide in the context of just learning about each other of learning circle of sorts the more likely that people will see governance as something that they participate for fun or for um communal interest but instead of just for the decision itself and that can happen for both the divisive and the bridging things because if you have that sort of a process and it search Services some things as bridging and some things as divisive both can be very beneficial

because you don't take action on the divisive things but you form Community around the divisive things so people can come out of the process learning both that they have new friends that they can Ally with and that there's some things they can move forward on together in principle um and I'm 100% on board I mean these methods um that you describe they've been around for 10 15 years or so um why aren't they more widely adopted um by nation states and other decision-making bodies at this point well because facilitation and summarization uh are very very costly they require special training taiwan's uh used to be youth ministry nowadays

the youth Administration have a whole program that's been running for more than 10 years to train the facilitators and the summarizers and even with the best trained ones I've worked with some of them even if you hit say 150 people there's a limit to how much they can facilitate and summarize um but nowadays uh the cost is rapidly decreasing as I mentioned there are now technologies that can uh not just turn speech to text but actually take the non-verbal cues um in language models multimodal models and that does a summarization that doesn't feel like forc that compresses but without losing nuanced and that does the facilitation I would argue better than

many of the best facilitators that I have met so just a combination of natural language processing Technologies and the recent uh language models multimodel process made the cost essentially zero it's like uh before open whisper and after open whisper when it comes to speech recognition uh and so I think this is really the year where we're seeing a lot more uh interest by investing into the digital participation infrast structure as infrastructure without the prohibiting cost of human facilitation as summarization I I also think you know it's always much easier and requires much less imagination to use some new tool to just uh you know very rapidly reduce the cost or automate

things that are actually happening but it doesn't accomplish very much in fact you don't even see very much productivity gains from that kind of thing whereas the much harder work is to actually get people to change social practice along with the technology um that because that requires cultural imagination it requires actually involving people in the process of the change and I think that's where you're going to see much broader productivity gains much bigger transformation not by the way that I AI is usually talked about but actually the way that's hard to talk about AI but it requires people to think about it a little bit so it it it moves slowly and it repairs things rather than moving fast and breaking things uh but it can actually bring a lot deeper change a lot more I would say even economic as well

as more importantly social growth so there are the fast changes that are happening because of the low cost and there are deeper scaling deeply changes that are happening also can you describe the role of a human facilitator so what do they do and kind of how I think kind of becoming more of a facilitator for each well to each and every one of us I mean often people go into confrontative mode where really they don't have to do you think these skills that kind of you teach those people in the facilitator academies do they transfer uh to kind of like General Life yeah they do I mean there are different schools right but they all revolve around this idea of

like nonviolent communication focused conversation methods to ensure that people speak to word uh something we can all live with instead of winning at a moment and controlling 100% uh of things right so it is um a lot about uh the everyday um practice of listening of listening actively uh listening with care listening with taking one side over the other but listening to both sides or all the sides and so on so yes practicing facilitation even though in the age of AI makes you a better person uh in everyday life as well it's a general skill uh that transfers to life but what I was saying was that it is

also something that this um current generation of language model also does very well if you ask any uh language model that is pre-trained in the past year or so to facilitate a conversation like Audrey Tong it actually does a very good job fastic I will definitely try that so what can what can listeners do to contribute to a more pluralistic digital future well uh you know we in the book we lay out a whole range of things from like the easiest simplest short-term you know retweet Etc to the longest you know most in investment but you know we would certainly encourage everyone to read the book to check out the trailer of the film about Audrey's

life that's available online uh also just came out today a wonderful interview with her with Christian manour that I'd encourage people to uh to look at and then um uh there are so many technologies that we need to build I know many you know uh listeners are in that community uh the book describes uh an order of magnitude more things than already exist but many of them within near reach so we would really welcome uh entrepreneurship we would welcome people to contribute to the book as they build those things so that there's documentation and ability to share and learn and scale uh those practices uh we would welcome people

doing research on some of the infinite number of open questions that that we describe in the book uh we would welcome people financially supporting the communities we would welcome people telling their friends about uh the ideas we welcome all kinds of cultural creation that both instantiate the idea of bridge building and St sheet some of the technology associated with it and that tell the story of how that's all happening uh so those are a few examples from the book but there's many more yeah I just uh read um a novel by Mena Scott Al called human power which is like a fictionalized version of whatever we just talk about the plurality ideas and it's Set uh in a very immersive uh world

uh almost like um you know Earth sea or something like that uh and which I find will reach a audience maybe teenage readers and so on that is quite apart from this podcast readers but this kind of cross-pollination between cultural products is exactly why we chose to public domain M our work because we never can tell whether somebody remix that into a hip-hop song or a rap or something like that cool so where does the prity community gather is there some sort of forum or Discord or telegram groups or how how do you all talk with each other well for the book there's there's a Discord there's there's a Discord group

for the book there's also a lot of people are involved in radical exchange groups around the world um there's plurality Institute which is more of an academic community so there's many different uh communities but for the book the plurality Discord is a great one uh and uh where there's also a GitHub where people actually work on creating the book yeah you can find all this um plurality done it perfect thank you both for coming on it's been it's been a true honor um I wish you both the best for the rest of your book tour I know you've been on the road for a very long time now um so uh yeah thank you so much for thank you live L and prosper [Laughter]