DAOs: The Good, the Bad and the Ugly

Transcript

Wonderful. Now, please help me welcome up to the stage Bill Laboon of Web 3 Foundation for a controversial topic perhaps. Uh, Dows, the good, the bad, and the ugly. I think we all have some opinions on this one. All right. Thank you. Oh, I'm sorry. Where's the clicker to? Okay. Yeah, no worries. Thanks. All right. Uh, hi everyone. Uh, actually, I don't even know if I need this. It's not a very big room. Uh, so my name is Bill Leon from Web 3 Foundation right down the train tracks here in Zoo, Switzerland. And I'm going to talk about Dows and in particular one specific DAO, the Polka

Dot DAO. Uh, the good, the bad, and the ugly. So, just a very quick uh, you know, overview for those people uh, who aren't familiar with Polka Dot. So, Polka Dot is essentially a set of blockchains. It is not a single chain uh that work together uh and we have a relay chain which secures all of them and is secured by a very large DAO where one dot is one vote in this DAO and just very quick overview of sort of your types of Dows that are out there. Uh I think there are really three basic types. Uh the first is you can think of just a telegram chat and a group of voters uh where people

vote on something and they ask some devs to do something, right? We have we see this often, right? It doesn't even I I would say this isn't even really a DAO, although a lot of places will call this a DAO. Just I've got a bunch of Discord accounts or Telegram accounts that are making decisions and then usually asking someone else to do something on chain. We can then have this essentially the same thing although in a smart contract where you have people that vote on something and then you can prove on chain that a decision was made but you're still asking you know please devs do something then you have uh how it works in polka dot and I would say this is really uh

how a dow should work is that when you are voting on something you are not actually voting on some text you are voting on a little program, right? You're voting to execute some code. So if that the DAO passes something then like here setting the validator count to a thousand or changing some other you know network parameter it's automatically executed. There is no please devs do something. It's really autonomous as part of uh the DAO. So uh in Polka Dot we have everything controlled by this. Everything uh up to update excuse me uh upgrading the software is controlled by the polka dot DAO modifying the parameters. We have a

very large treasury that's handled by the DAO. Uh I don't really want to get into arguing on whether or not it is the world's largest DAO, but we certainly are very significantly sized uh with about uh we've actually had over 2,000 decisions made by the DAO at this point. Uh and we've got over hund00 million in assets in the treasury. So all of that said, we have I don't want it to sound like I was just saying like, wow, Polka Dot is awesome. There's all this cool stuff. I do think it is a lot of cool stuff, but we have seen, you know, good, bad, and ugly things from that. And so I wanted to just go toward through some of the uh just some of the details of what

we've seen in in this uh DAO because we're really hitting all of the different angles, I think, of what Dows do. One is technical upgrades. You know, what's the validator count? What are particular parameters we should use? uh what you know what code should we run for uh the next runtime upgrade. We have treasury decisions which are often also done via Dows. You know we have some money uh this this pot is actually filled constantly by a percentage of staking rewards and we have to decide where to spend that uh that dot and other uh tokens. Uh we also have social decisions that are sometimes made things like all right what should uh

uh you know what what should our uh position be in regards to interacting with some other chain. Uh we have meta decisions about when you're setting up what we call a bounty essentially asking someone to do something for the DAO you know what should we expect from proposals. So we have a lot of the we've a lot of different uh types of decisions that have made been made with the Dow. And so the first thing that I would say when you are putting together your own DAO is that you should expect that in general people are going to be rational actors which does not seem like something that should be so surprising but a lot of times people think and just assume that people are going to be generally good or generally act like you are as a

developer of that DAO. And that really is not the case. People tend to do what is good for them. Uh and so what we see is anytime that there is a chance that you know someone will be able to take some funds and not do a lot of work for it then they will do that. If someone can take a lot of funds and do nothing in return because you don't have adequate checks and balances they will do that. And so I would say if you're setting up a Dow be very very careful anytime that you are sending anytime that you have funds especially are being sent out uh because people especially not not especially anonymous or

pseudonymous actors but we have seen actually real life people that have been KYCed uh that have essentially taken the money and run. Uh I I always find that quite uh quite interesting that people would do it for a relatively small amount of money. Uh but we have and ruin their reputations but according to them that is something rational. Uh similarly though you should don't expect everyone don't expect everyone to be rational. Uh a lot of people actually are trolls.

Uh and the larger your Dow gets the more you will see this. And so we actually have quite a few trolls. It is, you know, sort of a a troll-rich environment uh who will cause problems uh with anytime there's a a uh a vote on something. We see a lot of people do things essentially just to cause problems, right? Griefing attacks. And you have the the normal things you would you would expect with like a rational attack. So for instance uh very famously in in our ecosystem at least in order to make it look as though there was more support for something we had uh account a send their dot tokens to about a

hundred sub accounts and then at each of these accounts you know vote with a very small amount of dot and say oh wow and then they could point look how popular this is everyone is voting for this and then if you look on chain right then you can see this actually was just trying to drum up artificial support, uh, grassroots support, make it look like it was more popular. You know, that that's rational. Uh, you also see people actually wasting a lot of money just to kill bounties they don't like, wasting their own money, their own fees, uh, and then basically putting them in putting them in a dead space. It actually has no benefit to that user. It has some dis unbenefit, what's the word? top of the

drawback uh for people like me that are trying to make sure things work properly and so we have to issue other governance referenda to to fix it. Um but I I'm actually quite amazed at the amount of pain people will put themselves through if they can cause more pain to others. So maybe that's rational in some other other mechanism. Complexity is the enemy. This is something we have seen uh quite often uh in the polka dot dow. It's quite elaborate. Everything is handled via the Dow and this causes voter fatigue just in general. People having to look through all these kind of referenda.

There can be 20 30 I think the maximum we have seen was 76 referenda at a time. Uh most of which have like you know lots of different uh you know pages and pages of documentation and comments and what exactly they're looking for. And so uh this has caused a lot of problems in terms of turnout. people don't want to go through this or when they do especially if they're not being paid for it or if they're a relatively small holder of DOT then you know again going back to rationality dealing with all this extra complexity is not worth it so make things as you know as simple as possible uh this is something we try to do as we've been uh continuing with open open gov our uh you know our governance uh mechanism is we've seen things like

uh having to renew bounties every 90 days uh this this is something you extra that needs to be done. That's just you one more thing that every bounty then has to remember. Uh so any little bit of complexity you add remember that this is going to be happen for every single referendum and for every single person who votes on it. So make it as easy as you can. We like to think that we all live in a decentralized world but we're still in the process of getting there. And uh what we see is that you know with this this complexity and with this decentralization often you do still need coordination. You still do need some people who will go and talk to various actors in the ecosystem. Uh different

voters uh like people whose job it is essentially that doesn't have to be an official position but whose role is to make you know uh hammer out compromises get team A to talk to team B realize they should talk to team C etc. So one thing I think we didn't do was provide enough coordination at the beginning and let the community sort of handle things themselves and in some ways they did a very good job but we eventually hired a governance coordinator and I think that that's actually made a huge difference in in monitoring our Dows making sure that people are uh constructively criticizing and constructively working on things and coming to the right right

uh decisions. you need to think of the long term. A lot of people, especially in our industry, uh it's a very it can be a very fast-paced and uh industry and thus you have very short-term views of things. And when you're but when you're doing something for a DAO, especially when it's something very large like keeping the Polka Dot network afloat that you need to think of not just all right, what does this mean for the next five months, but rather what does this mean for the next five or 10 years? And so one of the things we do is that the longer you're willing to lock up your tokens, the more powerful your vote is.

Uh we also, you know, try to encourage talking to people that have already held their DOT for a long time, you know, and coordinating with them. Um but something that uh we have seen actually is actors often with a lot of DOT who have very very very short-term uh uh uh goals. And this is something that actually be very disastrous to a DAO because something that is very useful in the next month uh may actually cause a lot of problems uh in the long term. So I heard about this actually in the old days of BBS's uh so I don't know if anyone here is is as old as I am was on BBS's but basically forums uh the 1101 rule right so for essentially you know

for like everyone who's willing to be like an admin right there's 10 people maybe that will like someone that's like really into governance there may be like 10 people that vote you know a hundred people who actually read uh what's happening in the voting and a thousand people who just hear about it vaguely. You know, they one sentence that they read on Twitter from someone who may or may not like you. So, something to keep in mind, right, as you are discussing things, and this goes sort of back to keeping it simple, is realize not everybody is going to be as deep in the woods, deep in the in the the rabbit hole as you at this one or 10 level.

There going to be a lot of people that take whatever you've done for the last two months and have to sum it up in one sentence, and that's what they're going to hear about you. And so, you know, a lot of people that are in Dows are not the ones that are talking in in the chat every day. They're not making, you know, government governance referenda every day. They're just sort of, you know, vaguely involved. And even more so, a lot of people hear about what you're doing, but in a very condensed format. And so, trying to make sure that you do things that make sense, not only with a, you know, 20 minute discussion, but in a one-s sentence discussion, uh, can often be difficult.

be sure to shard. So, you know, Polka Dot is a a heterogeneously sharded network, but I mean this sharding in terms of of personal responsibilities. One of the things that Polka Dot lets you do is have a referendum that is voted on by all DOT holders. But it turns out that this is a very inefficient way to do things. Uh, you know, I know nothing about marketing. I'm I'm a technical guy. So, you know, asking me questions about marketing is probably not the right thing to do. If we have everybody vote on everything, they burn out very quickly. But if you have like small groups, right, if we have a we have for instance a marketing team, an events team, uh you we have a technical fellowship, etc. that focus on their uh particular niches than niches,

however you pronounce it, uh then it it actually gets it things much done much more effectively. realize decentralization and effectiveness are often opposed. Trying to do things, getting coordination among so many members of a DAO, not having you one dictator who can say, "All right, this is what we should do." Often this is a very effective way in the long way long run to get things done. But whenever there is an emergency, whenever something needs to be fixed quickly, uh oftentimes it's it's quite difficult. You don't have one person who's accountable. So trying to put together this accountability actually is a way of uh solving this a little bit. But uh you do need to realize decentralization

often has you know these drawbacks of getting things done you know very quickly when there's an emergency. Democracy is messy. People's opinions uh and and emotions often run high. Uh and so dealing with this you know trying to keep things calm. Uh I don't know that any country has solved it. Switzerland comes close, but as you may tell from my accent, I'm an American and there's a lot of messiness in in American politics. So, I don't know that we have also uh solved this. I think it's sort of human nature. Uh but I do think, you know, trying to be a good role model, a good example of keeping things calm is is very useful and necessary.

This sort of goes with something I said before. It's an adversarial environment. You have trolls. You have people attacking you for personal gain or just for the lols. So keep that in mind as you're designing your DAO. And even if you follow all of my guidelines, please expect chaos. There's going to be things that I forgot to mention that you're going to forget that everyone's going to forget. It's a very complicated thing that we're doing here, you know, making dials and trying to run things, but I do think it is the future. All right, so thank you very much. Again, my name is Bill Leoon from Mud 3 Foundation and really appreciate having you