DAPPCON 2018: Futarchy Curated Registries - Chris Whinfrey (LevelK)

Transcript

it's Winfrey from Louisville K and today I'll be talking about few turkey or more specifically few turkey curated registries so first we're gonna go over what is few turkey a few of the different drawbacks or possible limitations with few turkey which we're trying to experiment with and then our own experiment which is a few turkey curated registry which we'll get into and then we'll talk a little bit about how you can get involved and help us test few turkey and figure out how this new governance system can help us improve the governance systems that we have today so what is few turkey so few turkey is pronounced like the word

future with the suffix R key that's using words like anarchy and monarchy and few turkey is a proposed system of governance that uses prediction markets to determine which decisions will have the most positive effect so let's imagine we're creating a system of governance and say we're creating a system of governance for what's for lunch at DEFCON so we're trying to get you know which lunch option is going to be the best for the most people so the first thing we would do is we'd probably put it to a vote democracy is usually a very fair system

of governance but it doesn't always result in the best decision so let's take a look at why that might be the case so we have people here from all over the world today and we might not all know what the best choices are for lunch in Berlin so we have Tito's tacos la ratatouille Rick's Cafe krusty-krab and Pizza Planet so if I were to choose an option I I don't know what the good option is I might just choose randomly I might just go with my gut feeling is or I might just not participate at all so we have this problem where we have uninformed voters and that might lead to a decision that's not the best for the

most people and then I need to feel that it's worth my time to actually participate in this vote I'm getting a return on an investment with my time so we have another possibility which is low participation which also might result in a choice that's not the best for the most people so let's take a look at how few tarkie could be used to make a decision and obviously this is a very trivial decision but this is a new system of governance and this could be scaled up to any type of decision even governance decisions at the national level so the first thing we need to do with few tarkie is come up with what is

our success metric how do we know that we made a good decision so for our lunch decision we want to maximize overall lunch satisfaction so hypothetically if we were to survey all the attendees after dark on and get a rating on how satisfied they were with their lunch we could come up with an overall average lunch satisfaction that will tell us how good of a choice we made and now that we have our success metric we can start a prediction market for each of our options on on how on what the overall lunch satisfaction will be if we were to choose one of these options

and Sophie Turkey so we can see that we just launched a prediction market for each of our auctions and at the end of our prediction market we it's showing the pizza planet has the highest overall predicted lunch satisfaction so that's the option that we would choose in reality the the function that decides which option we choose is called a decision function and decision functions are usually more complicated than just showing which one is the highest at the end because that's very susceptible to manipulation but that's beyond the scope of this talk if you want to talk more about it come find me after but few turkey has its own drawbacks so it's

really hard to come up with a good success metric because your decision needs to have a large impact on your success metric so if we were say to have a success metric of overall conference satisfaction for our lunch decision then our lunch decision may not have a strong enough effect on our success metric to make a good decision and with normal market noise we might end up with a fairly random decision that's not the best for the most people and then if we scale it to talking about making a decision at the national governance level it can be very difficult to come up with one success metric that represents what everybody cares about most and maybe GDP and maybe health and maybe happiness but it's really hard to

come up with an overall metric for wellness and then we also have the potential for market manipulation so we need to make sure that our feet are key market is both liquid enough and has enough informed participants who are able to predict what how the decision will affect the success metric in order to be resistant from market manipulation so this is why it's so important to experiment with these new forms of governance we need to figure out where the drawbacks are and where few tarkie will improve our systems systems our governance and where it might not be the best choice

so for our experiment we're building a few tarkie curated registry which is basically a token curated registry that uses few turkey as its former governance if you've never heard of token curated registries I'm gonna get into it a little bit but there's a ton of material later today on token curated registries is a pattern we're all really excited about at level K because we think it has applications all over the place we we see token curated registries using are being used to curate a list of trusted ad publishers tokens that meet a certain criteria of transparency they're being tied into more complex systems like civils decentralized journalism platform

and really quick I want to give a shout out to Mike Calvin Eze and Emily Williams for because they put a huge amount of effort into making our few tarkie curated registry a reality I would definitely go find them if you're interested in the project and talk to them so really quick what is a token curated registry so basically it's a list in no particular order of entries and these entries could be anything they could be tokens they could be ad publishers and at any point someone can challenge an item on the list and that starts a new governance decision so in a typical token curated registry we would

start a vote and token holders would vote on whether this listing should be removed from the registry or be kept on the registry and the result of the vote decides whether or not that listing is actually remains on the registry or is removed so our future curated registry simply replaces that voting governance decision with a future key decision so when a listing is challenged we start a new free turkey market and for our future key curated registry our success metric is token price so token price is a good success metric because it's often what people care about

it's a good metric on the overall health of a decentralized system and so and then it's also very easy to resolve the market because we can just get a price Oracle from some sort of exchange and resolve it and we don't need to do any kind of survey or anything like that like we did with our lunch decision so when an entry is challenged we would start a new feat our key decision for is the price of the the Registry's token going to be higher if this list listing remains on the list or is it going to be higher if it's removed from the list and whichever decision shows the higher predicted token price is what actually happens to the list so we think few tark

be curated registries are a great way to experiment with few tarkie because they're very flexible once we get this project complete we can apply it to all kinds of different applications and we can test different list sizes which will help us determine things like how much our decision needs to affect our success metric so with a small list size our decision is going to have a huge effect on our success metric token price because adding a listing to a 5 item list is a much bigger effect than a listing to a thousand item list so we can kind of get a gauge on where few targeting might start to break down so

if you want to get involved in testing how few Charpy can help improve our current systems of governance there's a few different ways we can apply it and the least invasive the easiest to do is signaling so signaling can be applied to any dat that's out there today and in even governance decisions that are being made in traditional forms of governance you don't need any kind of permission to do this you simply start you come up with a success metric and start a few targeting market for that decision and then voters can choose to use that few tarkie market as a signal for which way they should vote so if they care about GDP for their country they can

use a few turkey market this to decide which way to vote in a certain decision and then if your Dow implements some sort of delegated democracy scheme you can actually allow voters to delegate to a few tarkie market instead of an individual so they could both delegate to someone else or they could decide to vote with the few tarkie decision so it's kind of like an opt-in future key and then there's full-on few turkey so if you want to get involved I encourage you to follow our progress we have two repos for this project at level K dev slash TCR and level K dev slash F CR -

project and if you have any questions or you just want to chat about future ki feel free to reach out at contact at level K dot IO thanks [Applause]