ETHDay - Andrej Berlin - First principles for DAO design

[Music] so i actually find it quite remarkable to be here in amsterdam or in the netherlands in a country where the stock exchange emerged and the first bubble um and now talking about like new ways of organization and businesses it's really exciting i think also for the first his first time in human history we're really able to create organizations that are fully permission as autonomous infinitely scalable and also have a business model behind it and i so the title is like obviously slightly arbitrary arbitrary but on purpose i focus on design because the approach to an organization is actually very similar to actually designing something because you're listening to people and then you're creating an infrastructure that supports what they want and at this stage i also want to kind of thank everybody at deepwork who has been part of the studio that i built over the last three years because they were the active participants in a system that was able to build that a lot of what i'm going to be talking about is based on and it wouldn't have been possible otherwise before i go into a lot of details on on how to do that i actually want to highlight something quite interesting that i find about the history of work which is that before the middle ages work was seen as something that's a fundamental part of human life so people would just do it and through the middle ages middle ages chris at least in europe christian chauvinist decided that doing work is something that's like that makes you a good person so only if you do good work you're actually becoming like a an important part of society or like a well-functioning part of society because you would the prophets that you would receive you would donate to church and the church would um distribute that across the poor people so god finds that great so you're a good person now um then that kind of concept translated through industrialization into uh factories so that except the profits were not donated to the church but actually been handled by the factory owners as as they wished and so that led to a huge economic growth and also to this idea that new economists picked up on to think that it's good that people feel part of a society even though they think that doing work is a disutility and it's like it's tedious and they don't like it because through that work they're able to afford leisure and that's kind of this balancing it out so it's quite interesting that this shift kind of happened and i think um over the last couple of years we've seen the kind of slowly an interest in getting back to that and like making it a substantial part of life so the obviously the reasons why we had economic growth were still very um very positive in a way so we were able to centralize things like power and authority or compensation signal sensing quality control like that was the only way that the economy could reach that stage but now we're kind of getting to the centralization and i want to kind of also be very careful about the term decentralization because i don't think that's a panacea for all the problems centralization is like a and actually like a process of decentralizing these very specific aspects of of work using novel tools and norms that are accessible to us now as we kind of leave the centralized era and uh have web3 tools so um there's this kind of two and i'm gonna kind of walk through that now so there's these two um i would say extremes on on either side so on one hand you currently have traditional organizations which have a business model and they protect that business model by writing up job posts job descriptions and making it as promising as possible to find the perfect candidate and you have applicants who write their cvs and try to apply try to kind of create the best image of themselves but we we know that that's not a game that that works out well because both just continuously get better at lying to each other um there is rarely really a focus on the substantial like relationship between um employees and and workers people kind of feel that i think whereas on the other extreme um we have online communities which are like you know reddit or forums which base which are based on this model of discovery of content so people are just interested in what they really are passionate about and then they join these communities start communicating start exchanging knowledge learn and also accumulate these these portable profiles that now also show what they what they are knowledgeable but knowledgeable about and what they're interested in i think what we're looking at in now is like kind of the the combination of both where we're able to see or create this flexibility of um independent individuals in what they want to work on and combine that with financial and emotional stability that we would normally get from a from a job and i think i call this like discovery of work model because people would discover what they really want to work on we just need to frame that and i think we need three things to to get there which is uh we need to be able to get people to join an organization at any point in time at any at any skill level um we need to be fully transparent about what the work consists of and the financial compensation and we need to have portable profiles and portable reputation data that people can actually own themselves but before we get there i think we need to actually redefine what work really means so uh henry ford invented the eight hour work day which i think a lot of people here in the room but also just generally a lot of kind of privileged people were able to ex exit from and and try to create new models and we know that that's not sustainable the alternative was basically our tracking so kind of quantifying how much work how much time you would spend on work but that's also not really ideal because if you really love the work that you do then you don't want to track hours and if you don't love the work you do then you count more hours than you than than would in reality represent the work that you did um and so deep work started as an organization um by a deep work as a design studio and um we started as a purely project-based organization so we would actually take something that would normally be seen as a service like a design service and we would be very very specific about what we do so we wouldn't like in a traditional design studio you would go to them and you'd be like i need like all of these things and the design studio will go yeah cool like we we have the people we just do it and then it takes an indefinite amount of time and you spend money and you have feedback loops and then you end up wherever um we ran these design sprints which are exactly like seven days long we could actually say it starts here it ends here this is the amount of days this is the amount of hours that it takes the amount of hours it takes you the amount of hours it takes us and this is the exact deliverable and that was that was not good for like everybody but it was very specific to to a niche audience and a very specific deliverable and so now our entire organization is set up like that where regardless of whether it's an external uh project or an internal project and this is what's actually applicable to any organization you just scope out what a project would consist of so you'd look at like what's the work what's the recurring work what are the things that people take responsibility of on a recurring basis you would start at one point you would end at another point you would say like this is the exact deliverable and it could also be like two weeks or like one month of work or something it's still a project because you specify exactly what goes into that project and what's the hours spent on it so then you calc like technically you want to take this further and like projectize your work you just look at like okay what are the hours that i spent you compare these hours to a market rate and then you have the value and the the cost of that project and then you can go to either a client or a stakeholder and be like is that actually the work that you would pay for is that a deliverable that you would pay this specific price and that way you can figure out if that's if your work is valuable and so the in practice the shorter negotiation the more the closer you are at the at the correct like market estimation of the value of the work um the longer it is the more there's a chance that you need to either like scope down or extend it a little bit um different conversation but that's the idea and essentially like the this projectizing helps you to mirror talk like basically create a meritocratic compensation because now you just have this deliverable and you know exactly the value of the work that the market is willing to pay and it also solves another couple of problems or like it actually helps having more frequent feedback loops so rather than having feedback loops after half a year here you have like every like every two weeks you can have a chance to improve internally you can improve the projects you can improve the people and you offer new people the ability to actually be introduced in the organization in a lot more frequent basis just at the start of every project at deep work we first like hired people in a full-time position and then ended up um realizing that the market fluctuates too much because we were focused on web three and it was like okay we don't have any work now for three months but we still have this guy um and um and it was not it was not good for anyone so um by projectizing we were basically able to say okay look we know you know it's projects we know it's project if there's no work you just like find work somewhere else because you're a freelancer and we are able to sustain our business model now over time like it becomes more and more predictable how much work we would get and so but actually let's let's actually look at um the um the actual onboarding and introduction into uh into the work because i think that's also kind of the recruitment of people um because you can take these projects and roles and go even deeper and describe it in a lot more granular detail so you could go into like this project consists of these specific roles and these individual tasks so that and in deep work we actually created like a documentation on git books where we would have very very specific description of what the work requires and that's not just like to set the quality standard for the entire organization um because like good organizations also do that um but now that also acts as an attractor for new people to actually just discover the work that they would love on one side and on our internal side we also look at like this is the quality standard this is you know it it becomes more and more like it becomes less about the kind of deceiving each other and trying to get the best deal or the best candidate um and it becomes more and more about like is that actually the relationship that i want from this on on either side and progressive organizations also use this kind of documentation to establish a mentor and mentee relationship so you know as someone would come in obviously like people mostly need some kind of training before they start working and in our case you can create this like this relationship between a mentor and a student um that's that's really that's really tight because everybody because with with the project-based work the the cost of firing someone is zero and everybody knows that so guess what people do they they strengthen the social network the students make better predictions about what they really can deliver upon or deliver on and and get better at being being like self-driven in their learning and the mentors get better at their own skills because they get to teach them at a regular basis they learn how they learn recruitment they so an actual kind of social skill they learn communicating properly they learn mentorship there are all these like things in soft skills that we considered as an external thing in normal organizations and now it's becoming like this like the systemic property of this uh like organizational infrastructure so it's like benefiting everybody to create this like very honest social network and with web3 tools we also have the possibility to to take that even further because now we can like we can because we have these kind of tokens and co-ownership of organizations we can um essentially give people the ability to also make long-term predictions so they would obviously get better at finding new candidates that fit the organization because they might know someone in their immediate social network who is like can be introduced because they know that they will be better for the entire organization because that's valuable for them we also have this concept of uh a recruitment bonus so as uh as um um as mentors and students split up the work like you know you you have your role and you split it up um the the project budget can be split up in relationship to the actual like task splits so now you you can kind of um fragmentize the work while keeping it still meritocratic which is also kind of cool and then if you have extremely high market demand for the actual work you would increase your prices so the prices of your service would basically would automatically increase like you would manually increase them in order to keep with that keep up with that pressure but that profit can automatically go to the people who are recruiting so that you incentivize those to actually scale the organization at the same time as it's needed um in on the on the open market and so the this whole idea of these these kind of um composable roles offers a lot of opportunity to like play with that data because it's public um or like public kind of for the first time in in history we can we can make that public without actually destroying uh an organization because the setting up the infrastructure for our organization is now free it's just data it's like that's the documentation you can read it you can copy paste the designs to you if any one of you wants to create a new design studio go for it copy paste everything and then the the the result of that is that you're just like saturating the market or like you're actually extending the market because you wouldn't be able to to copy exactly the same market you wouldn't be able to steal that that market of one organization because again the infrastructure is free um compared to previous organizations but you would actually just like a lot faster be able to expand in your own market and actually kind of saturate a completely new vertical and so and you can kind of compose these roles into whatever you want like we have roles of very very different skills like you know user research but also front end like it really doesn't matter like take any kind of role in every job you can com like componentize that you can build like your hierarchies your dominance hierarchies with that three tools again you can assign them specific uh roles and uh sorry specific governance votes uh so governance weights so that every role has a specific kind of um um well influence on the organization um you can have uh what do you call it like um conditional requirements for people to just kind of go along these like trees and and expand and it generally just creates this environment where you have an anti-fragile organization by having always a mentor-mentee relationship always being as close to the market as possible as you need to scale you can also kind of take a roll and then programmatically extend it if that's actually needed so if the market demands something more from one particular role and needs adjustment then you just like you create that and then you update that across all the people who own that role and then you could also have the the possibility for retroactive compensation because now you have roles that are very specific in terms of their value you know what their value is to a specific organization you can like two people or three people come together create a new project you scope it out like i explained earlier and then you know the exact cost so if that cost is actually reasonable to an organization you show the value you can be like okay cool i'll pay you exactly that cost because it's been quantified so that ownership of these roles leads to the last part that i want to kind of touch on which is the portable reputation that's been so far in traditional organizations locked in because of competitive uh reason so it gives it gives it gives the organization's competitive advantage uh keeping them secret as well as um the it keeps the privacy of people safe like keeps the it it preserves people privacy with their data about their work um we know that the alternative is like linkedin or personal portfolios but even there it's also like semi-ideal because people usually write about themselves and again they optimize for the best but i think the actual kind of um the the roles that someone has an organization that is also like especially if they're transparent that is actually then a skill but it's also verified because you worked at that organization the organization you know the value of the organization you know the value of the project it has a brand you work there so that's your skill now as you've worked on you collected that and you found like you you attached that to your personal profiles which will make possible with like web 3 wallets and integrations and you know all the all the cool 3 tech that we're working on um and so the the biggest challenges now become like how do we actually gather all that reputation data that is that is around uh and how do we kind of um sell like how do we make that accessible to organizations um like like these things already exist like reputation data is already already being gathered there are really cool reputation tools around and we also have these um the um things we have organizations uh getting the um like basically finding that data but i think what's really important to keep in mind here is that we're also the the individuals who have to be who have to own that data and that's also something that we can only do now with the web3 technology is that we can actually like aggravate that data but also and make sure that it's visible that it's transparent but at the same time also give people the the possibility to decide who they want to show it to because it's theirs and and um and they have to govern about it so and that whole thing opens up like a huge potential for completely new markets where you can have like the discovery of skills you have curation of roles and skills you have you can create predictions like you will be able to create create predict we're not there yet um we will be able to create prediction markets so you know like imagine having a skill that is on your profile and then you have an application that shows you hey maybe you look into this thing because that's on your trajectory on your perfect trajectory of development or like compensation or whatever so i think as we kind of uh seen from this kind of very rigid financial stability uh and emotional stability of traditional organizations as that's kind of falling apart and leads humanity to stagnation i think these this transparency over work the the portable profiles and projectizing it and and the feedback loops they create this environment where every individual develops the develops the growth mindset um by nature as as again as a systemic property of the whole environment this whole organizational infrastructure um and also makes the whole organization a lot more resilient so we have the ability now to kind of have this individual like the the the freedom of the individual and the independent individual to find work like i guess once that structure is more prominent across more organizations once that's more saturated or like much more accessible um individuals can actually be free and what they choose and what they discover and at the same time like develop this financial stability through the growth mindset or like through the ability to actually find the work that's also being compensated at the value of the the work that they that they do so um it's pretty it's pretty exciting to me to think about it because the we're kind of not looking for organizations anymore that are just purely driven by by growth or like being being faster than anyone else but we can actually focus on the singularity of solving a particular problem on a global scale and um what i find particularly nice is that we are almost like these designers now of the organizations that we build at the same time as we are the participants of these decentralized organizations that's me thank you [Applause] thank you well there's there's one very particular question that i'd like to ask you and it's is it possible to replicate for software development the deep work model used that you've developed for design how applicable do you think it is to software development yeah it's easy i mean we we also kind of do that or i would say it it kind of emerges as a as a consequence of it because um again i think think about software development not as like this whole thing of like you're doing everything but like focus in on this one particular like problem that you would that we would solve so like for example as a front-end developer you're just like of course you could take all the libraries into consideration but like you know focus in on like either a subset or like focus on what you can like predict how long it will take you and projectize that that's the value of what you can do [Music]