Balancing Efficiency with Decentralisation: Tradeoffs in DAOs

Transcript

[Music] um so my name is sinz uh and I'll be talking about balancing decentralization and efficiency in Dows now I will address the title because it's a bit misleading and it's on purpose but first let's talk a little bit about me and about l2b so you can get an better understanding um I'm a Governor's representative at eltu be which means I represent L2 beat uh in different Dows that we participate participate in as delegates I have a background in business administration and and finance been in crypto since 2018 uh and in Dows since 2020 but more full-time like over

the past two years or so now l2b is a public goods open source company we provide analytics for different l2s and we seek to be like an independent Watchdog over ethereum uh l2s we try to provide partial assessment uh and our thing is mostly risk but we do a whole bunch of things you can learn more about uh about LTB lb.com now I'm going to address the elephant in the room um in the proverbial like room of the title and the title implies that if you have decentralization you can't have efficiency or if that something that's decentralized cannot also be efficient which is not what I'm trying to argue here what I'm trying to argue

is that there can be some things that are more dent centralized and there can be some things are less decentralized for the sake of efficiency it's not a zero sum game it's not that you can have either or it's just a matter of figuring out how much of which you want to have in which occasion so this is kind of like how you can think about it and how not to think about it uh and by about it I mean the title of the presentation and the presentation as a whole now if we think about um balancing it as trying to figure out which things in a dow should be more more efficient and more uh centralized in terms of decision making versus which things can be more decentralized and open and not

necessarily rely on someone to make a decision but rather look at the whole Dow for that decision to be made versus things um or rather the opposite side of the spectrum what which I don't want you to think about it uh is that all those things are kind of like in vain and if you have something that is decentralized it cannot be efficient and it doesn't work and we'll never make it work that's not kind of like what I'm trying to argue here there is room for decentralization there is room for efficiency and it's just an act of finding the right balance all right and kind of like more illustrative purpose like there is a zero sum Spectrum where you have decentralization and efficiency it's about finding the balance in kind of like the golden

pocket of if you're in Game Theory 3 comma 3 um this is this meme that you probably have heard uh over Twitter but it's about find in the win-win situation where we can have both decentralization up to the extent that makes sense and efficiency up to the extent that makes sense and have something work now I'll take a step back and ask the question of what it is a dow now most of you I guess will already know what a DA is uh or have some notion of the idea of what a DA is but often times I run into people I talk with people and they get it somewhat wrong they thing a DA is this entity in the middle that you know is uh has in its orbit token holders delegates

voters uh service providers Foundation different individual contributors uh and what have you and this entity in the middle that is a dow is like a highly organized or a decently organized entity that can make things happen and you can often see in different like proposals for example or different articles say oh the Dow decided or the Dow will be in charge of but that not the reality this is not what a dow looks like the reality is something more like this all those things token holders delegates Pro even protocols and Builders um independent contributors all of them working towards a common goal that is the Dow so it's not all those people are not external to

the DA and they do not interact with it they are the da so whenever we talk about something let's say we are in a dow and we talk about something happening in the Dow we can replace the Dow with we and it works just as well and even more so like this is what Dows look like and even more specifically we getting even more specific this is kind of a more accurate representation where in its subgroup we have even smaller groups within them so we have like delegates it's not just a group of like all the delegates in different DS are not a well fit and organized group together that they decide on things there are discussions happening there are arguments happening everyone has different oppos views and same with all

the other entities so in reality Dow is the sum of all those little bubbles that you see uh trying to work together and organize and that begs the question of how do we organize do we all have a say in everything does a delegate or token Holder have a say in what the service providers do do the service providers have a say in what the delegates decide uh does everyone has an equitable voice in all the decisions should we stick to own circles or is it more of okay let's just assign authority over uh to some entity or some person either through like over the entire Dow or just to specific things and let them do their

own thing this is kind of like what um my talk is more about like figuring out this sweet balance between where we delegate authority to more like centralized uh decision- making versus where we make it more open and decentralized now borrowing from Network work um architectures we have this uh graphs of what centralized decentralized and distributed systems look like now I think a lot of people when thinking about Dows they're thinking about even though like decentralized autonomous organization like decentralization isn't the name of the Dow they're thinking more about distributed systems they're thinking about these clusters that are working together in unison all together all with everyone right whereas in

reality it's more about the second picture or the second graph which is more decentralized you don't have to not have any sort of centralization in a decentralized system it's just a matter of not having a single point of failure and as my my colleague Kristoff pointed out earlier the graph is slightly wrong simply because the dot in the middle could be perceived as a single point of failure so you can imagine some some other lines connecting everything together almost like a like a distributed system but not so much as it is so what are some challenges of getting there uh of being decentralized and working in a way uh that we don't have centralized control over decision- making some include just lack of

overarching goals right and this is kind of like a big one what that means is basically we can have service providers we can have delegates we can have different groups of people in a dow that each are working towards something that is to the benefit of the Dow arguably uh or can be perceived as such but they are not working with each other they're not working against necessarily but they're not working towards a common goal they do not have a shared vision of what they want to achieve so what ends up happening is everyone's kind of like moving around blindly trying to figure it out and that introduces delays introduces problems challenges and you know it it it creates um a lot of friction and that is kind of like what makes things inefficient all those

things together they add more time they add more uh effort uh they add more friction to everything that needs to happen in the Dow so that's kind of like why the more we decentralize without having these things uh more tied down the more inefficient things get so what is even like decentralization in a dow it's about opening up the doors to more participation it's about creating Democratic processes it's about equ equality it's about um not relying into a single party or entity for decision- making and the problem with with that is that it's not a one- siiz Feats all right we

cannot have decentralization for everything for every aspect of a dow and even within the same aspect for different parts of the process or of the initiative that we might have you can have different levels of decentralization um at different points so for example if we are generating ideas on how to solve a problem you can decentralize that process you can say say let's open up create a collaborative process of ideation and everyone can submit their own ideas they can uh propose something and it's open to everyone to participate but when it comes to ratifying that decision when it comes to actually taking the decision uh depending on what we're trying to decide on it might require more expertise

expertise that is not present in everyone and it's also about the time required to make those decisions right if I ask you right here um let's say after after the event we'll go for dinner right and I show up and present two places like a Thai restaurant and KFC I don't know this talk is not sponsor by the way um so if I just ask you you can basically decide right you can vote let's say we have a vote and everyone decides and it's an easy decision to make now if I wanted to make it more specific if we wanted all to eat like at a Thai restaurant but let's figure out the best Thai restaurant to eat in in this area now that like gets a little bit Trier now we have to think about which is the best restroom we need

to go on Google and start searching reviews or whatever in other words it takes time and it takes a bit of effort to inform our decision you can still vote blindly and say yeah whatever the group decides but if you want to make a more informed decision you have to make effort if that takes I don't know like five minutes to go over you know a couple restaurants around the area now imagine how much time it takes if we have to decide on how like what a dow Northstar like a protocols Dow should have as a Northstar and its Mission or how we can can div diversify the treasury imagine the amount of time and effort it takes for someone who doesn't have experience doesn't have context to get all that so they can make a decision if that was like a prerequisite to becoming a delegate and participating in

governance uh I think a lot of people wouldn't you know un wouldn't necessarily take up the responsibility of doing that so that's kind of like why we do not necessarily need and want to have everything be more decentralized in a dow and some level centralization is actually good and helps us be more efficient in the way we do things so now I want to do like a little bit of audience engagement and I will show some things some parts or some I don't know how to call them but I'll show some things on the screen and I want you to raise your hand if you think or rather just raise your hand and you can like express your opinion and where on the spectrum that I'll show on this spectrum so that that that thing fall on and you

know something should be more kind of like decentralized or more centralized and I'll I'll do that simply because I want to see that even with just a hunch without having to discuss anything between us we all have a notion of the things um that need to be more decentralized versus more centralized so for example a grant program where in this spectrum oh how do I go back okay a gr program where in the Spectrum does do you think grant program should fall anyone okay more decentralized more decentralized okay yeah less decentralized sorry less decentralized what

about setting Direction like setting a vision for a dow more centralized more decentralized more decentralized okay operations that's a good one operations over the Dow like less or more less decentralized control of the treasury and protocol less decentralized or more more decentralized oh that's divided the room a little bit yeah that's a valid point I mean like control in the sense of making decisions about it less a little bit of both yeah that's

storn and lastly a post on Twitter like let's say we have a a dow account should be decentralized or should be centralized should we decentralized do we do have vote on Twitter posts I don't know okay so this is my take it's not really my opinion it's more about my understanding of what the average opinion would look like now the reason for that is post on Twitter right you can delegate that authority over to someone you can hire as a you can elect someone to control a Twitter account be it a person or a group we don't have to vote on every single post right imagine us having to sit on days and I don't know like Saturday or the week and

create a Content calendar and we vote on all the different posts we're going to go through the week that's mental right I don't want to do that uh operations we can have some degree of decentralization and have some oversight from the Dow delegates but in order for something to operate uh and operate efficiently where efficiently means uh in a timely manner and with removing friction so something works we can be a little bit far uh left on the Spectrum then grants program is kind of like more in the Middle where you still need someone to run it uh be that a group or a person or even like hopefully we can go more into Automation and have like smart contracts around their own grants programs but you still need to also have control as a doubt both in direction of what the things

that we want to fund are maybe RFP processes and general like oversight and then more the more we go to the right of more decentralization you can have control of the treasury and the protocol the reason why I didn't put it like on the far right fully decentralized is that not everyone is here is uh is fit to make decisions uh including myself uh or you know in any doubt is fit to make decisions on both like Financial or treasury like I'm not a technical person you don't want me making decisions about protocol upgrades right uh I thankfully have people to rely to to look and make a decision about um propos changes in different protocols but me and myself like like I do not how know how to read code I don't know how to write code so

probably I want the best person on my own to assess technical upgrades similar for treasury right I do have some Financial background but I wouldn't want to be the person responsible for let's say diversifying two billion doll treasury I not fit for that so what I am good at is probably understanding the challenges and the things that we want to achieve with those things and delegating the responsibility to someone who's fit for that and have them do it and then lastly setting Direction it's about figuring out like the collective mission of what we want to do as a DA figuring out the north star that guides us now I discussed that internally with with my colleague and we we've done that

in the past uh Kristoff and I know he has different views and I also I'm very torn on the matter should that like should this Vision be created from like crowd Source from contributors in adal or should it come from some should someone step up and say all right this is the vision for the Dow this is what we need to look at and if if it's someone then who should that someone be should it be just a contributor should it be uh the foundation or a labs this is kind of like something that I think most people would feel that it's decentralized but personally I'm torn and again this is kind of like a reflection of my understanding of the average opinion about those things not necessarily my opinion on these things although it's very close to to what I would do so so lastly like

decentralization is not the way it's the end goal right if we just say that we'll decentralize everything that doesn't make us a dow it's a a state of things that we want to reach at we can do that and we can be a DA if we have some sort of centralization or some parts that are centralized within the Dow we can have a grant program that someone is responsible for let's say I'm a you know imagine a word where I was actually a good person to manage a grand proog program and you funded me I don't know with a million or 2 million and I was in charge of running the program people would apply to me I would be the one making the decision just me no one else if I do that and I actually fund the things that matter uh which we have

collectively decided um you know then that's the reason why grants program for me are kind of like more in the middle so if I do that I'm the one like you know a central point of control for that grant program but that's not a bad thing you can you know I'm I'm responsible for it there is ownership you can hold me uh hold me accountable for its impact but it's still a dow right if if the Dow wants like if delegates one they can clb back the funds they can you know fire me but this is the difficult part this is kind of like the part where for me it's what we need to figure out this this bridge between you know in in a traditional company let's bring it that way in a traditional company or

setting let's say in lb right we have Kristoff he the governance leader L be I'm working alongside him but also like hierarchical you can say that I'm below him so he's he's responsible for my performance right in a way if I don't do things he can fire me in a dow if we elect someone then we don't know who's responsible for their performance so if I'm elected as a grants council member or the grants program manager or whatever and the Dow you know we all decide that yeah sinkers will probably be a good person to handle some money and they fund me now who's responsible for keeping me accountable is it everyone is it no one is it myself cu

the the answer the Dao doesn't really work the Dow as I mentioned in the in the beginning is just a collective of different people contributors service providers everyone so this is kind of like where decentralization faces some challenges because in traditional settings you have you know a central control over everything like there's a CEO or the founder of a company and eventually all lead up to him but in a Down by default even if you do have some centralized part there is no overarching Central authority to make decisions like be the ultimate voice of Doom for different things so this is kind of like where we need to find a good balance of how do we make make make it so that we take initiatives or aspects of it down

we decentralize them just enough so they are efficient but they are still accountable and they are still like the people running those initiatives responsible for the outcome of their initiatives that would be all for my talk uh you can scan this to go to the governance page of l2b and learn more about a little bit more about what we do in governance and I've included my telegram below if anyone wants to reach out uh I know a lot of of you here already uh and I talk with you but if anyone wants to reach out and talk more about governance you can find me there that's all and thank you so we have time for a

question what do you think yeah sure who has a question anyone yeah I love it Icebreaker yes okay I can start uh not a question actually but I would like to argue with sinas uh when he own mentioned that like I don't agree with him about the setting Direction being on the more decentralized uh you know part of the spectrum uh and the reason why I don't believe on that is that it all bows down to the to the thing that sing has mentioned of how much time do we spend on actually you know being able to make that decision and sin mentioned like the

decision of hey let's pick up a restaurant that we will all go after this event if we try to debate that probably you know all of us would would have different opinions different you know uh priorities in terms of what we want to eat like SAS will be always like chicken nuggets like you know doesn't matter that we're in Bangkok like chicken nuggets are fine uh it would be hard for us to to agree even on such a simple thing like where do we want to go to eat and mostly because like before we even would be able to discuss it we first would need to do some research like research like what what are other our options even you know know what are restaurant what restaurants are available how far are they from where

where we live and stuff like that imagine how how much information do you need to have to even be able to discuss setting direction for example for a protocol how much context do you need to have how much background in terms of like what happened in the past like what are the different players in the ecosystem who is your decision affecting and how how much of that you need to have in order to be even able to discuss it and then you know we expect let's say you know th thousands but realistically like let's say 50 people spend literally tens or hundreds of hours researching like being capable of actually taking

part in this discussion it's it's insane to expect that therefore we are having those discussions where people simply you know having some break during the the day sitting on the toilet they jump on the for and say hey and I think we should do that you know I think we should dump all our Eve and do something you know it costs really really little to to to write a post it's just like five minutes but actually this this person's opinion is probably not as valuable as someone who spent last five years working on this protocol this person is probably a bit more capable of coming

a decision so without about the setting Direction my opinion is that coming up to a decision is something that should be less decentralized what should be decentralized is ratifying this decision yeah valid point I know we have discussed it uh internally and I agree nice and there was another question right there I think there given that we want things to be more decentralized what about voter apathy

isn't it always the same darn people that keep voting again and again doesn't that make it less decentralized at the end of the day yes and no um so voter apathy comes in two forms one is voter apathy from delegates like delegates that already have voting power that do not exercise their voting power and they do not participate but then also apathy comes from token holders that do not red delegate their tokens away from inactive ones now this is kind of like has been uh something that has been talked about and discussed and trying to be you know find solutions to that and do like de redelegation weeks and uh subsidize the gas cost for redelegation things like that but ultimately if you think about it um if we go back to kind of like this

slide of what the Dow looks like you can imagine like the token holders as the people feeding the delegates right with with their voting power now the delegates can create a lot of things uh within the Dow but it's very difficult to create a system or enforce something that makes token holders delegate their voting power at the end of the day it's up to token holders theoretically you know we could argue even that ultimately they are the ones that can decide what happens in a dow right like for example l2b has a lot of voting power in different protocols but that comes from people delegating to us right so yes we have power and we vote on different things but if the

people who are delegating to us just have different views and they are willing to express those views they can take away the voting power from us and suddenly we don't have as much say so but there's no you know there's no mechanism uh at least up until now to enforce such a behavior so that's the kind of like I think the bigger problem is about apathy from token holders not so much from delegates like we're seeing a lot of um increase in participation from delegates um I have an interesting question is does decentralization necessarily mean good governance or are we compromising on good governance for decentralization that's to say I've seen people kind of equate

decentralization with why we want blockchains to be decentralized but those are a bunch of nodes coming to a deterministic end answer and decentralization means more honest notes will come to the right answer but there's no right answer in a lot of governance decisions so I'm curious what you think about like how do you see decentralization being different than Dows and how should that be balanced against maybe the more effective governance um potential do you think decentralization and good governance actually do have correlation um I think the like I would answer with a question of what does good governance mean like even if in a you know in a like let's say in a country if you ask the poor people uh good governance means increasing the taxes to the you know for the rich if you ask the

rich people it's like no we need more subsidies for businesses whatever so you know good governance is something subjective I would say that governance in general and decentralization can work very well together it's just a matter of figuring out to what degree that's kind of exactly what I'm talking about like balancing them you can have more like decentralization in governance where it's more public and and seeks opinions and views from a lot of different people and you can have certain points at which you don't want that you want more like domain experts to make uh a decision themselves without necessarily looking to get you know as Christo said like someone from their toilet just posting

on the form so this is kind of like my view on that is you you can have good governance an effective governance whatever that might mean to you know each person subjectively uh and have decentralization at the same time thank you