Tools for the socio-economic renaissance

Transcript

[Music] all right good afternoon everybody um first off I just want to say thank you to the organizers and also thank you to the ethereum research Community it's a fantastic Community where I've uh learned a ton and I wouldn't be here without them and finally just want to thank these folks that I have listed on stage uh Christopher ghost D and John girl and other people who work on aoma uh while I will be talking about tools for socioeconomic coordination uh I will not be talking about anoma specifically so if you want to talk about that uh you can come find me after all right let's get

started so in this talk we'll have three parts uh in the first part we'll talk about socioeconomics and cybernetics in the second part we'll talk about some crypto problems and then in the final part we'll conjecture some tools uh that can help handle some of these problems potentially it's going to be rather speculative so uh bear with me all right tools so this last week I was in uh eth Berlin and when I got to eth Berlin they gave me a a carryon bag kind of like a toiletry bag that you would find in an American drugstore and I opened the bag and it had a black pair of slippers and a head and a neck scarf and I said to myself what why are they giving me this but then it sunk in these are tools for a digital Nomad and as it turns out I

had actually left my toiletry bag at home prior to the trip and later that evening when when I was in my hotel and I wanted something comfy I put those slippers on and when I went for a run I used the head scarf to tie my hair back similarly if you are building a house you might use tools like a hammer a saw a paintbrush nail gun and if you are an ethereum dap Builder you might use tools like solidity Foundry ethers or open Zeppelin so one major thing you can do with tools is that they give you the means to change the system they might seem like they have a very minor impact but that minor impact can expand similarly like a butterfly in

India that flaps its wings and eventually causes a hurricane so I'm going to read you this quote and doing some research for this talk I was reading this paper uh on art and culture in ideological systems and I found this to be uh quite relevant so in an ideological system agents can have heterogeneous goals beliefs and beh behaviors and interact with each other through communication and persuasion socialization and mobilization and these interactions create feedback loops that reinforce and modify the behavior and level of use of resources all right so this has to do with socioeconomic systems and what's interesting about socioeconomic systems

is that they involve policies that are usually set by some policy makers and they also impact people um and it's this interaction between the policies and the behaviors of people that form somewhat of a feedback loop and this feedback loop can be thought of as cause and effect whereby you have some abstract system you have some inputs that go into the feedback loop generat some outputs and then those outputs feed back into an input sometimes in the uh crypto Twitter discourse you might find the term virtuous flywheel to describe some defi token ponics or uh more academically you might find it in uh the writing of George Soros where

he used reflexivity to describe the boom and bus cycles of Foreign Exchange markets and to this all I can say is you know these are not the feedback loops we are looking for we're not talking about uh tokenomics we're not talking about Ponzi games instead we're talking about cybernetic systems and cybernetic systems are concerned with measurement and control of complex feedback system systems so in this framework I've proposed five parts to evaluate a cybernetic system you have control inputs you have outputs you have a target of the system so these would be like your target outputs you have an unknown function of the system Evolution one common example of this in ethereum

landscape would be like me and you have strategy to adjust the control inputs for the targeted outputs so how you would change the inputs to get the outputs of the system that you want and what's really useful about cybernetics is it can help us simulate some of these models and design adaptive control systems and also help us build more resilient systems systems that don't break and systems that get stronger perhaps even antifragile so given given these prefaces I'm going to make a conjecture and the conjecture is such that if you define a

Renaissance by notable shifts in art music literature culture religion everyday life you can probably conjecture that we're starting to see one of these Renaissance take place in our own lives in modern times and some of the talks today and and in this week kind of allude to this and some of the topics which are bringing this about could be like AG life extension space exploration synthetic biology many others not list here and in order to cope with some of these challenges or cultural shifts that we're facing we're going to need some tools perhaps an enhanced or a different set of tools to survive but before we can go about

solving the world's problems um we need to solve some of our own problems in the crypto industry um we have to kind of clean up our own house as it were and three problems which I think are quite relevant today that we see in the current market dynamics are number one Capital formation number two communication complexity and number three credible commitments or promises for organizations and we'll go into the three of these so Capital formation often times in this space Builders are faced with a particular dilemma and the Dilemma is usually to take one of two paths path one is perhaps more of the altruistic path you apply for a ton of Grants

compete in many hackathons you position yourself for like GitHub public goods funding or maybe even like an rpgfan the future um this is a very good path to take but it it does have some trade-offs and the second path is you go to investors and you promise them a token and some blockchain abstraction that's going to allow them to like stake their token or sell it in a liquid Market so let's talk about path one and we'll use the cybernetic framework that we introduced so an input could be you make an effort to apply for the grants you seek out these funding programs an output could be maybe you get funded

maybe you don't doesn't necessarily have to be binary it could be like you wanted $300,000 to scale out but you only got 50k something like this the target would be the funding objective that you want the unknown function of the system Evolution would be competition things like alignment availability of funding and market conditions and the strategy if you're not getting funding could be to you know position yourself um for these public goods funding by advertising on social media or making Forum posts just Word of Mouth going to conferences or sometimes like you just might Rage Quit because it's just too hard and so the path dependent outcome

of taking this first path is it's often not sustainable for projects that want to scale for Mass adoption you know so you can hack together something really cool on a weekend or you can string together grants to get a small team going for a while but if you want to reach millions of users often times you do need more capital and this leads to Too Much Delta between the output and the target so let's now talk about the second path in the second path you make efforts to raise money from investors similarly you might get more funding or less funding that you want and you're seeing that with some of these like seed rounds and series a at these massive valuations um the target might be the funding that you want the unknown function system Evolution might be like

competition narrative Wars memetics availability of funds market conditions right like if there's a global market crash it might affect your ability to raise funding and your strategy could be in order to get funding is build what the VCS want you to build or pretend uh until you get funded and promise a token and the path dependent outcome here it may War incentives to build something useful and the downside of this is users don't get what they want and Builders are flush with cash overnight and sometimes they like buy yachts and retire early and delegate to the community

right so let's talk about the second problem now communication complexity so all of us today we we all have uh laptops cell phones most of us and one of the first things that we do is that we open some type of messaging or or media application and if you happen to work in this industry and or around it there's like six seven eight ways to communicate with people and everybody has different preferences um and it's a bit much and so let's let's look at the cybernetic framework so an input would be use multiple apps to communicate some of the outputs here are like dopamine drains fatigue you become emotional um a Target is just you want efficient

communication with your colleagues and peers but it's just hard to get um and then an unknown function of the system Evolution could be like likes retweets random dialectics solicitations um gossip things like this and your strategy on how to cope with this is often you know check all these all these applications because you want to keep up with your friends and your colleagues your family members but it has a cost and so what this path to pendant outcome leads to is like a shelling point which is Twitter um and it it it just leads to more of the same and it leads to a lot of dopamine addiction and burnout um which isn't great so let's go to the third problem

credible promises for organizations so both internally and externally so internally in a lot of our organizations we have somewhat of a flat hierarchy and with a flat hierarchy often times it's very difficult to understand who's responsible for what and it's also difficult to hold folks accountable for their commitments and also for folks within the organization to understand the accountability of those commitments and externally it's very difficult to demonstrate competence to other organizations that you want to collaborate with investors or just even the real world so again let's look at our cybernetic model an input could be a flat organizational hierarchy an output

could be incomplete work delayed timelines on shipping things that should be shipped or even at worst case a toxic culture which leads to a lot of employee turnover a Target could be just you want to ship and iterate and satisfy workers and customers you want to have a great work environment um but an unknown function of the system Evolution could be Talent gaps you work on the wrong stuff employee turn as I mentioned or resource misallocation you wind up spending you know $100,000 for a podcast appearance something like this and a strategy could be you know you reorg every few months until something works which I mean it can work but it's not great and to the

outside world you just kind of say hey like it's all good we're going to ship at X8 right TM soon and the past dependent outcome is you overpromise or you under deliver until you ship or you run out of money and you break so it's also important to note that these problems aren't just isolated they actually kind of when you have many of them they start to interact and they form these larger feedback loops and a Japanese meteorologist talked about this phenomena in um in weather and to describe two Cyclones kind of merging together he called it the fujiwara effect and these two Cyclones actually become stronger over time

so we've outlined some of these problems um and I'd be lying to you if I said that I had solution to some of these and you know some of these takes might be a little bit naive but we're going to try to use protocol tools to design applications to solve some of these problems and one of those is capital formation so we're going to suggest public signal another one is communication complexity multi chat and for the last one a tool called uh promise graph so public signal public signal in simple uh one sentence is intent Centric Kickstarter with demand side aggregation and dominant Assurance contracts so what does that mean so in Kickstarter today if you want to back a project or if you're an entrepreneur we

flip it and you want your project funded you go to the marketplace you say Hey you know I am writing a book or I'm making a movie or I'm holding a conference um I'd like some funding for this and it's up to backers individually to fund your campaign and if you don't meet your campaign's dollar targeted threshold you don't get funding so sometimes what this leads to is entrepreneurs asking for too little and taking on too much for that little amount or they ask for too much and they don't get funded and often times uh Studies have shown there's a strong correlation between the entrepreneur and their ability to get funding based off of their social media now imagine if you could take this problem and flip it on its head and allow communities to come together and

say hey we would like you know some cryptographic libraries produced or we would like to fund a research Collective working on life extension research or we would like to build a Crypton native Hardware wallet or we would like to build a Crypton native phone or snark proving asex things like this the idea of this application is it allows users and communities to aggregate demand and express those as intense to qualified entrepreneurs and Builders of these things and so you can meet together in this Marketplace place and one cool idea here is that you can use Alex tabo's research of dominant insurance contracts now dominant insurance contracts are these things better known as refund

bonuses and so they did some uh testing whereby if you as uh a backer donate to an entrepreneur you're guaranteed to get some risk-free return back over what you put in so if you put in $10 you're guaranteed to get $1 back now the entrepreneur takes on a little bit more risk in this model so you'd probably want to have like some type of quadratic funding or public goods funding to you know pre-load that um but the studies actually showed that you went from about 40% funding rate to about a 66% funding rate uh in the laboratory setting so very exciting and we talked about this there'll be some different types of

intents that you could have you could have an identity based intent um you know you'll donate one eth to any project falic donates one eth to information based incentive based or even threshold based and this could be kind of like a mockup you know vitalic logs in there's a funding feed of different accounts that he's following um you can see some like pending contributions you could see some recommendations this could be like an app that you have on your phone or it could just be like a simple desktop uh browser application as well so let's talk about multiat briefly so multiat embodies this idea of separating in protocol from operators and so what I mean by that is if you

think about a lot of the traditional messaging applications we use in web 2 there are gated modes that are put up and it's very difficult for you as a user to take your data from one application to another and it's also very difficult for you as a user to interoperate with one application to another and uh folks like the folks at Matrix have been trying to solve this interoperability problem for a very long time um and kudos to them they're doing a fantastic job but it's still a really hard problem and the one of the trade-offs that signal provides even though they provide incredible privacy guarantees for users is they're centralized um and you are relying on signal to act honestly um which also isn't really great and so with multi-chat you'd be able to build a

messaging chat like application over the enoma peer-to-peer Network and you'd be able to build different interfaces for the different types of communication you want different organizations could customize their interface it could be more slack like Discord like telegram like and you be able to interoperate with anybody else that's running aoma node on the network using a messaging app and one of the cool things is you'd be able to specify precisely what information you want to reveal to whom uh and this is a concept we call information flow control there's a deep literature in uh programming language Theory um that touches on this and also security research uh in particular we have a blog post um on our blog if you want to find out more information about this one and we should have a prototype

out fairly soon and in the the article I do a comparison table um and just kind of talk about some of the trade-offs between a theoretical multi chat signal um and Matrix the last one I'll talk about uh is promise graph so promise graph is a language and a structured accounting Logic for making and managing promises in and across distributed organizations so you could think of it like a dependency graph so on the end you could say like you want to ship ZK evm or like you want to ship your project and then you as the founder of your project or you as the founding core team could decide okay what are the things that are needed one could be a specification one could be a prototype uh one could be

like marketing and Business Development so then these become like subtasks on the graph and then Within These subtasks you need to Define what are the specific tasks that need to get defined and so on and so forth so you could think about this kind of like a patrici Merkel tree and what's interesting about this is once you get to this subtask level you start to set up commitments between what's called a customer and a performer so a customer is somebody that needs something done so they might need a cryptographic library built and the performer is somebody that's capable of building this now the customer and the performer are going to negotiate and the customer could say to the performer I'd like you to build this library in 3 weeks and the performer could say no way it's going to take me 3 months to do

this and I don't even think it's the right task so you can have a negotiation process around on this to get some clarity and what's great about this is that other people in the organization can see this commitment so if they want to collaborate with the performer on something they know what the performer's workload is they also know what the performer is working on and it can reduce conflict within the organization because you can now globally see what people's priorities are and you can have input and so it kind of looks something like this uh where at the top you kind of have this unplanned customer starts planning and then you have the negotiation process that I talked about after the negotiation process happens you have this progress where the performer is

doing the work and then you get to this review process so the customer ultimately does review and approve this but the performer can say to the customer at this point in time like hey you know I didn't really feel great about this I don't I don't think I got everything I wanted out of this or I don't think this is the right task I think we actually should do X instead of Y and in this process process you can do one of two things you can drop the task Al together instantiate a new task or send it back through the feedback loop and have new timing and um this is actually something we've been dog fooding at heliac right now heliac is the company uh employing people to to build anoma and namata and we currently have this built uh with Git

and we use it with GitHub um we have task actions like creating planning starting completing confirming rejecting and different types of task representation and we found this to be really helpful um for internal communication between research and Engineering teams and since we've started this about 8 months ago um we've iterated on three separate versions of our anoma specifications and we've iterated on multiple versions of an internal prototype leading up to a decentralized test net sometime this summer um so very exciting and and and we hope to share this with other organizations um we take some inspiration from the informal systems workflow communication and uh galloy uh as well so what are some takeaways from

this talk we've covered a lot of grounds it was quite abstract maybe a little bit too ambitious um but few things to remember if you don't remember anything else uh number one cybernetics can help us reason about socioeconomic systems and problems there's a lot of feedback loops that um are evident every moment of the day and if we just kind of examine those and the interactions um and we think about the control system the input and the outputs we can kind of tweak our strategy to get our outputs more aligned to the end goals it's also fun to solve real world problems and I have a tremendous amount of respect uh and I'm humbled uh by folks that you know come on the stage and and talk about things that they're doing for North Korea Liberation um you know I I I can't pretend to solve those

problems and I think a lot of builders in this space can't quite pretend to solve those problems yet but we can solve a lot of our own problems so I would challenge anybody listening uh building in the space to just think about that what are some tools that we can build to to help people better coordinate uh within our own ecosystem and then we can start taking those tools to our communities in the outside world um and we talked about Capital formation communication complexity and uh credible promises and new tools to create new feedback loops so I'll share with you a brief anecdote um you know before my time is up um this past week I was in Berlin and I I had the privilege of getting to talk with Audrey tang and uh Glenn whale about their new book uh

on pluralism and you know one of my learnings in the conversation was that communities are always kind of bundling and unbundling changing over time based off of how they feel um based off of different factors that are important to them and very much so with pluralism we want this idea of kind of a minimal set of shared values that allow us to interoperate with each other and it's with these minimal shared values that we can then bridge and Bridge our community and share some common language and I think this is like um a really Noble thing to work for especially for ethereum um vitalic just wrote an article about l2s kind of talking about

this and while there might be very specific differences in some communities about technology or about different ways to proceed with building or Community formation um just remember that we're all in this together and we're just trying to make the world a better place starting with our own communities um if you have any complaints feedback or questions you can direct them at me uh my Twitter handle a priori ZX uh we have a lot of writing and research on these uh resources for anoma and uh thank you for your time very much appreciate it thank you a priori thank you very much uh any questions from the field there we have okay we'll get the

mic to question coming up hi thank you so much for the great presentation uh you talked about new feedback loops and I'm quite interested about examples if you have anything in like social AI things that can benefit from uh you know provenance and anonymity decentralization and all of that yeah this is a really good question uh quite broad um I would say that I'm not really that good a thinker in terms of um artificial elligence I'm still um very much a novice uh there but what I would say

is thinking about ways to give more power or control back to individuals and to communities users I think maybe like a good answer here would be along the lines of um diac um you know like vitalica suggest it basically giving tools to protect uh the sheep from the Wolves rather than than like giving the Sheep weapons um and I think you know if we're going to continue to evolve uh as a society and kind of escape this like top down fascist control um we need a lot of preference entropy and we need a lot of Tolerance and I think giving people more tools to express their preferences whether it be you know with

llms or whether it be with like some human brain interface um these are things that I think can be helpful in kind of altering the current feedback loop as they exist because right they're like this new um evolution of the system um that comes in and then it it Alters the strategy about how people interact so um I'd say yeah like uh I think falic has a lot of really good ideas in the dacc post that he wrote recently thank you any more questions out here NOP seems for good well thank you very much stay with us short break and upcoming I think thank you aori if very interesting topic upcoming so stay with

us thank you [Music]