Episode 8: What? When? and How? A Practical guide to Decentralization with special guest Julian Zawistowski

This week we are joined by Julian Zawstowski, a decentralized systems visionary and entrepreneur, and founder of the Golem Foundation. Located in Warsaw, Poland, Julian and his colleagues at the Golem Foundation are engaged in a number of projects that democratize access to computing, breakdown corporate data silos, and strengthen the position of individual users vis-à-vis service providers.
 
 Julian and his colleagues at the Golem Foundation are cutting edge Decentralists advocating for a new proof of use model call the User Defined Organization. Their first practical application, Wildland, is a distributed file sharing protocol currently in Proof of Concept.

Please join us on this episode of The Decentralists as we explore making decentralization a practical reality for everyone.

Transcript

The Decentralists: What? When? and How? A Practical guide to Decentralization with special guest Julian Zawistowski

Length: 51:05

Henry: Hey, everyone. It's Henry, Mike and Geoff of the Decentralize. And we've got a very special episode for you this week. It's called what, when and how a practical guide to decentralization with a very special guest, Julian Zawistowski. Julian is a Decentralized systems visionary, entrepreneur and founder of the Golem Foundation, located in Warsaw, Poland. Julian and his colleagues at the Golem Foundation are engaged in a number of projects that democratize access to computing, dismantle corporate data silos and strengthen the position of individual users, Visa V service providers. And economist by trade, 

Julian is fascinated with Blockchain technology, as well as the sociology and economics of decentralized systems. Now in a past life, Julian was a consultant and researcher in public policy, and the co-founder of the Institute for structural research. Julian and his colleagues at the Golem Foundation are cutting edge decentralists, advocating for a new proof of use model called The User Defined Organization. And the first practical application is wild land, a distributed file sharing protocol currently in proof of concept. Julian, welcome to the Decentralists.

Julian: Thank you for inviting me, I'm excited about our discussion today.

Henry: It's a pleasure. Well, we have to start with a very fundamental question what turns an economist indeed a public policy researcher like you into a decentralized? How did it all start?

Julian: Part of economics is the science on how people coordinate themselves, how people coordinate their actions, to achieve results to do something. And like a prime example of that is a market so market economy and that market economy can be more competitive, less competitive, more regulated, less regulated, one of the edge example with central plant economy, when you have in a communist when you have a central planner. But this is fascinating to understand, why things happens and for the first time, I came across blockchain technology that was curium a few years back, I realized that although there is a lot of technical aspects here and fascinating to some people, cryptography, but in principle, this is economics and technology, because what makes a system like blockchain to work is that you have some rules, you have some incentives, and then you have everything in motion without any central authority. This is pretty fascinating stuff, because I was always very pro freedom person. But at the same time, I think every economist, I know that there is some scope of regulation; there are some rules that are necessary for the systems to work. And I find the concept that we can build some rules and incentives into the fabric of things and we can offset some extent state or like centralized organizations with that kind of mechanism design. I find that concept, simply fascinating.

Mike: So, what you're saying is what you've got with decentralization is, is an economic engine that does not necessarily have a central point of control. Is it fair to say that that's kind of how you would have transitioned from say, studying centrally planned economies, which, I think you could argue, Julian that it wasn't just the old communist governments and economies that were centrally planned? I mean, the FTC, and the Federal Reserve and other kind of the EU central bank and stuff like this, they basically act as central planners for our economies anyway, right?

Julian: Of course, but they haven't failed yet.

Henry: Failed too much.

Julian: Well, comparing to the failure that was achieved by an Eastern Block.

Mike: So, do you think that if the centrally planned economies had switched over, they would have survived in some form? You could argue that I guess the Chinese economy might be an example.

Julian: Yes, I think that's exactly what happened there so, they allowed enough of a free market to sustain their economy. But you know what, I think China, is not the example I would use to support the decentralization case, I rather think that even we can move even beyond markets, because you can use like a backhand is design approach, also, for things where you don't have market set that's at least my belief.

Mike: Right.

Geoff: Although I think what's interesting is, when I was at university in the late 1980s, just before the fall of the Berlin Wall, I was just finishing university as the Berlin Wall fell, so might be a little bit of a clue to my age. But I think what we were taught back then, was that free market economies were tightly intertwined with the notion of democracy, and personal freedom, and that you couldn't have one without the other. And I think some of the examples, particularly in Asia, China, the obvious one, but also places like Vietnam, have perhaps demonstrated that fundamentally, perhaps people are happy if they are just making money and putting bread on the table. And perhaps they are less concerned with freedom, personal freedom less concerned with Big Brother then they might have been provided they can buy a Tesla and have a big screen TV and they have sufficient freedom. So whereas, when I was in university, I would have been told that that was never the case and that was impossible, and you can argue that many of China's largest businesses still have a very large state ownership component. But nevertheless, there's a lot more free market activity than then one might have thought was possible.

Julian: Yes, that's true; I'm not an expert on China. But I think that what we are witnessing right now might be a reversal of that trend. So as far as I know, China is closing up at the moment and we'll see where it goes, maybe turn out that this is possible only in the short to medium term, but not long term.

Geoff: And even in the United States, you're seeing that and this is spreading elsewhere, US aren’t unique, but even the United States, you're seeing that. And perhaps it's always been this case, is making money and making billions of dollars is sort of more important than anything. And what the politics in the White House are Visa V Air quotes fascism or not, is somewhat irrelevant as long as the economic engine is being allowed to surge. And even that, to some degree would be contrary to what we used to think was the case.

Mike: Right. Alright, so I have another question Julian, let's throw this one out there. How do you define decentralization?

Julian: That's funny, because that's somehow is related to what we discussed a second before, because I think ultimately, this is about freedom. So, when you look at the system, like whatever system it is, like political system, like a software, social platform, freely whatever, like a financial platform. It is important to understand who has power, who is in control, and the holder rated that it is impossible to have a free market economy without democracy is about property rights. So, you can be like a dispossessed of everything you have by the tyrant and you have nowhere to appeal and this is why like a free market economy needs ownership and owner’s property rights need democracy, [cross-talking 09:46] approval flow. And I think this is pretty much the same with the decentralized systems, we have problem with at the moment so, if Facebook or Twitter or Google or whoever decides to the plot from you and has your social graph and have your data.

Mike: Happened to us, we got kicked out a Twitter three years ago.

Henry: Two years ago. 

Mike: Yes, no idea why, still no idea why.

Julian: Are you still out of Twitter? 

Mike: Yes, that's crazy.

Henry: So Julian, do you mean, you are very interested in decentralization? Because there is a proof of ownership, is that what you mean?

Julian: No, what I mean, is that the system is which is decentralized is a system where it is controlled by a very few, and that the centralized system is or you have real control of what you have, or what you're using, or what you're doing. So as using as a comparators and using as an example, for example, your data so, if you store your data in a in a centralized system, then whoever operate that system, can they get this. 

Henry: Well, they own it. 

Julian: Turn it off, and they actually own the data. Now, if you are using a decentralized system, I think the minimum requirement to call that system, the centralized is that you have sovereignty over your data. So I would use that I call that like a pragmatic this decentralization, I would use, like the pragmatic definition of decentralization, that the system is decentralized when a user or a person has reasonable scope of control over things, because I don't also think that this is binary. So this is not like that the system sees are centralized or decentralized, you can have systems that are more decentralized or more centralized.

Mike: But don't decentralized systems still operate under a rule set? 

Julian: Yes, of course. 

Mike: If all the different blockchain out there, they have different views on what it what it means to participate, and the role that you play, right, whether it's processing the transactions, or whatever the case may be, and then you have actually just even the overarching, kind of, what is the system there to do? And so, you know what I mean, to me that is, at some point, even every decentralized system started as a centralized kind of concept around some rule set or work that needed to be done, but you just needed more people to do it than you had.

Geoff: I think the challenge is that the enforcement of that rule set is opaque. So you don't actually know if that's occurring. So if I make a contract with Microsoft to say, okay, I'm going to give you all my health data and you sign this contract that you are going to keep all that health data in the United States and it's not going to migrate into your data centrein Venezuela. They'll sign that contract, but how you audit that how you prove it? Is almost impossible and how do you find out that's happened will typically you'd find out there was a breach, the data was breached. And then that triggers the clauses and all the lawsuits, and then the lawyers become millionaires over the next five years. But you're still back to square one so, I do think that environment of rules is important; Google has this miles long privacy policy about what they'll do if you hack in them all your pictures and this kind of thing. But you have no way to know whether that is being enforced in the backend until something bad has happened and then it's too late.

Julian: And I love this medical data example, because it illustrates, but the real problem is that you use third party as a custodian of your data. And that's where the real problem is, if this is for example, stored in this centralized or decentralized way. So, one of the reasons they quite things are centralized and why we have such problems with that, when it comes to data or software in general, is that using custodian is very convenient.

Mike: So do you believe then, Julian, that when you talk about decentralization, okay, so I like to talk about decentralization, it's a broad topic. And different people understand it from different ways, a lot of times people's heads immediately go to blockchain or whatever the case may be. But when I talk about decentralization, to me the most fundamental building block of any decentralized system is the individual user of that system. And the, I guess, ownership and control they maintain over their data within that system, whatever it is. Would you say that's a fair definition of decentralization?

Julian: Yes, that resonates with me.

Mike: Because if you look at the evils without now hey, I mean, like it's, you could argue that Google and Facebook and YouTube and Twitter and Weibo, and all the Chinese alternatives, they just they wouldn't exist, there would be nothing with them, if there wasn't for the users that were on them. I read a an article yesterday that was talking about kind of the irony of social media, in that all of these providers like Facebook, Twitter, YouTube, and all this stuff, they have all of this data and they make all of this money off of this data, yet they create none of it. It's all created by the users.

Julian: Yes. And like we are coming here to why it is like that and I think it is widely accepted now that what is broken is the economic model of the software, or internet services that they are in theory there for free. Sounds like 20 years ago, but everything is based on data collection and advertisement, at the beginning, I think everyone liked that but it turns out that this is sometimes not that good.

Mike: So, do we want to decentralize everything?

Julian: I think the technical sense, it is impossible to decentralize everything. But then if you take that decentralization is about more freedom and control by the user, then I think, yes, we can decentralize everything, at least we can have more decentralization are more freedom, more control for the user, or for the people in everything we are doing. And that is true for software development, and for technology. But I think we also should consider that, for example, for our societies for our political system and so on.

Geoff: I think and people have heard me harp on this before, both on this podcast and elsewhere. I think part of it is for that paradigm shift to occur two things need to happen. I think the first is that people need to accept that they will have to sacrifice a little bit of convenience. And if they want to move to decentralization, perhaps things won't be as automatic, as we say, as they might be used to. And then the second thing is that they have to accept that they might have to start paying for some things. And you touched on this a little bit, Julian with these free things. Now, they are paying for them today, they're paying for them with their personal data, with their identity with their profile, something of which they have no idea.

Julian: With the freedom ultimately I would say.

Geoff: Yes, exactly. And when I look at downloading an app, and the app is "free", air quotes "free" again, my first question is, how is this company that's making this app paying the bills? Now, if it's something like VLC player or Audacity well, Audacity is a bad example now, but fits an open source solution, and it's just a bunch of guys in ponytails that are doing this altruistically, then that's great. But in in many cases, if you just download some game on your phone that's free, that gets an update every three weeks and so on. How you're paying for that and how they're keeping the lights on and paying their developers is an interesting question? So, I think giving up a little bit of convenience and realizing that you'll have to sprinkle a little bit of your hard earned cash around are two key pieces that can't be overlooked.

Julian: Yes, for sure. I would like to more things changes I've always heard like, no matter what the tenant is about So, I think that there has to be, like, critical mass to flip the switch and we do not have apparently enough of that critical mass yet, of course, people are already learning all the time about new things, it started, I think, with the Cambridge Analytical, where a lot of people realize that something funky is happening. And I think we are accumulating now, the snow, literally society, intent we are going toward that critical mass. And we will get there at some point sooner or later, hopefully, sooner than later. But I think there is more to that I think the other problem is that this is some kind of like, escape from Freedom situation, like it from style. And that can be seen in a very simplified aspect like, for example, what you mentioned, convenience, having nice UX, having your password recovery, which means that someone knows your password effectively or can know your past password so this is thick and went for it. But then I think if we want to look for like deeper, philosophical in that, I think that there is a lot of people for whom it is okay, that some centralized, AI superpower will not only hold the private keys, but will also tell them like where to go for dinner, what to read, and ultimately, who to marry. And this is a little bit like a... 

Mike: Who to vote for. 

Julian: Yes, who to vote for, this is like a hierarchy style dystopia but this is convenient in a way, this is like an ultimate escape from freedom. And I think that at some point, we will have that conflict, between those who value freedom and those who value convenience. And this goes much deeper than only password recovery advice UX.

Mike: Yes. So you bring up an interesting point, and this is something that you kind of have to sit and reflect on for a few minutes. But you had back in 2019, and December 2019 Jack Dorsey put out a tweet, it's his own platform and it was this long rant about how Twitter needed to decentralize because they could no longer operate under the kind of the regulatory regime and the kind of the threat of huge fines and everything that came with owning the user's data. And then he put it further into context and he said, realistically, what social media is at, they've built themselves into recommendation algorithms. Okay, so Twitter recommends tweets to you, YouTube recommends videos, Netflix, Facebook, whatever. And they use that based on your metadata, so when you come through, and you start clicking through movies on Netflix, they profile those movies and say, anybody who downloaded the Guns of Navarone will watch these other war movies or whatever the case may be okay. 

Julian: Great movie. 

Mike: Yes, it was. But it's funny, because so on the one hand, what that indicates is they don't need to know who you are. They just need to know what your preferences are, in order to run the engine, okay. So to me, that means all this time, they haven't needed to know it and the second thing is I don't see a difference fundamentally from say in effect, social media is the new television or the new phone, it's just a broadcast medium these guys are basically now just backbone, kind of almost bandwidth providers, you could argue that you use mobile, SMS or mobile or whatever. But the truth of the matter is when you picked up your phone, back in the old days, they could charge you long distance because you were using their infrastructure to communicate from one person to another. Whereas, now you're using that same infrastructure, but you're not using any of the platform to do it. And so I wonder about this idea that all of our data has to be owned and controlled inside the same silo as the services get provided is just a big red herring that we've all bought into and it's completely unnecessary.

Geoff: I'm not sure that we've bought into it, though Mike, well, you and I have not bought into it. But buying into it implies that you listen to a sales pitch, and you agreed with it, and you put your money down. I think when Netflix recommends the Dirty Dozen to you, because you've watched the Guns of Navarone. People just think, that's neat and they click on that they don't realize that they've sort of bought into anything. They're just sort of accepting what they're seeing, and when Julian brought up Cambridge Analytical, I don't think, until Cambridge Analytical, and then all the other things that fell after that many people were thinking about what goes on with Facebook, what goes on with Twitter, how they're that and the 2016 election, hammered home for them, the different silos that we're all living in, thanks to social media, but to say that people have bought into it suggests that they've been told all about it, they've agreed, and they've said, yes, that's what I want. 

Mike: Well, or that they've had a choice.

Geoff: Right. Now, they may loosely want Netflix, when they press a button, what else can I watch for it to not come up with a bunch of 19th century rom coms if they want to watch more movies. So perhaps they've bought in in that regard but I would suggest the vast majority don't necessarily know what's going on there.

Julian: What they do to us is, in a way, you can be called like, a curating content for us, so like what is our news feed. And that's what happens with the data presented to us, always through the like a modern age like newspaper in a way, that's the same. So the problem with the monopolistic social media is that there is no real choice. So if you are on Facebook, that you are subscribing to the Facebook content, curation and if you think of it, like in fact, censorship is a way of curating content. 

Mike: That's interesting.

Julian: And in a way, we all need someone to curate content for us to effectively censor some content out from our newsfeed because there is so much content, there is so much happening that someone has to limit this flow of information to something we as human beings can manage. So of course, the problem is not that the content is curated, or even censored, because it has to be done. The problem is that we are subscribing we all of us are subscribing to one or two very few sensors and those sensors are not allowing us to see how and why they are censoring some of the things we just in fact manipulation because if I decide whether I buy a New York Times or Washington Post or whatever else newspaper I want to buy, or I'm watching CNN or Fox. This is my conscious decision I can somehow suspect what is there and content curation, okay, what they send out from my newsfeed, I don't know that about Facebook or Twitter. So we need control over our data and we need so called competitive compatibility or even enforced competitive compatibility to change social platforms into something they should be, which they are not at the moment. So platforms that do curate content, but we'd compete between each another in how they curate content and if they allow us to influence how they curate content okay. It is then, we do not have one cruise demo but we can decide to eat attend that we want to subscribe in a way.

Mike: So, that's based on and I hear you on that your analogy between, say, The New York Times or The Washington Post, or the Guardian The Telegraph for whatever kind of left and right wing, kind of traditional journalism. But those outfits were all run by some editor or press Baron, they used to call them who had a certain viewpoint and made sure that that he or she hired the people with shared viewpoints. And so you knew if you picked up the Guardian, you were going to get kind of maybe more of a left wing progressive viewpoint in the journalism and if you pick up the telegraph, it was the opposite. And so, I guess the question that I have here, is the problem with centralization on the internet, the fact that there are profits driven human beings making those decisions? If what you truly did is you took social media, and I'm going to take us into the Terminator land here for a second. And you basically removed the Mark Zuckerberg the Jack Dorsey's, the surrender period, P Chais and Tim Cook's, out of the equation, and had all of the, I guess, decision making on the centralized platforms done completely by an artificial intelligence so think Skynet, or something like this. Could you could you envision that those platforms would be healthier places than they are now? Or would they be even worse? 

Julian: Yes, I think that the fact that the platforms are profit driven, or rather, that they are not acting in the interest of the users, but they are acting in the interest of their shareholders. And as it happens, those interests are unfortunate, not aligned is the base problem here. I don't think AI would be a solution here, perhaps it could be a solution, but maybe AI will have its own agenda, and maybe this agent that will not be well aligned with humanity agenda.

Henry: Yes, someone would have to program the AI well, there is that?

Julian: Yes, exactly. And then this could be kind of blank books, and whatever. Anyway, when it comes to platform management, I think what is important is to create a situation where the platform has to act in the interest of the users. And that's where our concept of user defined organization comes from.

Mike: So tell us more about that. What is a user defined organization?

Julian: When we try to understand what's wrong with the organization's right now, so, the software producing organizations, let's call it like that, we'll thought that this is pretty obvious with the legacy models, legacy organization. So, I think we have discussed already what happens when the word products are for free, when the products are not for free, and people are paying for them situation is much better, but it's also an ideal because still like the organization's is profit driven, then that may lead to situations that users like, for example, vendor locking, in the case of this social media, as we discussed before, this is probably huge problem. So paying for products that does not necessarily solve all the problems. When you look at the centralized scene, then there is a lot of very interesting discussions and experimentation of will the decentralized organizations, what we think is problematic there very often is that one of the basic models is so called like a governance token we're just tradable asset. And when you do governance with a token that represents some kind of voting power or equity, and then you have two major problems. First is that as this token is tradable, people can sell it people can also buy it, so that gives different vectors of economic attack on the platform. 

So, you could, for example, they can borrow a lot of tokens take a governance decision that will influence the price this way or another and then make a profit from shorting it. Then another problem is that this is against situation where token holders, token holders interest is not aligned with the users interest, because they are token holders and not necessarily users. And that comes back to like, in fact, replicating the situation of the traditional equity based business with completely different technology that is used to implement that, but with very similar in fact incentives. So, what we thought that could be a nice concept is to somehow provide users with the proof that they are using whatever service is provided or whatever organization that and use that proof of use it as a governance token. So in case of Wildland, and this is pretty simple, if you have a marketplace like we have with Wildland, where this use it is simply derived from transactions you make on Wildland. So, the concept that we are going to implement here is that is as the Wildland is used for data you need to buy storage for your data, you use Wildland marketplace for that. And whenever you do transactions on Wildland marketplace, the gap fraction of that transaction is used to generate that token, which is proof of use it is attributed to the user that started the transaction. And this is a very important part this is non-transferable, so this is non-speculative talking. We just attribute it to that user and only that user and then there are a number of other problems you can have with Decentralized Governance, but this is step towards situation where the incentives and interests of users and of the platform itself are aligned.

Mike: So what you're saying Julian, is in a decentralized system, so let's pick the ones that people understand, like you think about Bitcoin, okay. Bitcoin has an incentive, which is the Bitcoin token for the blockchain and that token is proof of work, I believe. So, I have a bunch of computers, I process the transaction, I get a Bitcoin and then you have a theory in which is kind of proof of steak, I have the biggest number of computers working processing the transaction, or whatever the case may be. And so what you're saying is, is that the user defined organization would be say, for want of a better term, a blockchain or some kind of decentralized service, where the users of the service so Wildland is and then we want to get into that in a second is kind of a storage focused solution. And so the idea is, the more I store, the more I transfer, the more I use it, the more of these tokens iron. 

Julian: Yes, and the more... 

Mike: And that gives me more of a say in say, a future decision as to what the platform is going to do?

Julian: Yes, exactly. It is hard to compare that to proof of usage, the proof of stake or proof of work, in a way that proof of work and proof of stake are like a mechanism to secure the network, okay, they are like a consensus mechanism in a way. They don't have any direct impact on governance, while proof of use it is not used to secure any network anyway, because there is no network. In fact, there is only an organization that is defined by the set of contracts on Ethereum but this is a governance solution. So, you could envision, that for example, Ethereum could decide and I'm free styling right now that if you do transactions on the Ethereum blockchain that use some gas, then if you converted that gas, or if you meet some proof of use token for Ethereum, that would be based on the amount of gas you burn, to execute your transaction so that would be what we do in Wildland.

Mike: Right. I guess one of the things that has always dogged or at least in recent years has dogged a lot of these kind of crypto reward token based systems. Okay, so there's been, I remember back, geez, it must have been 2014 2015 and I was looking at these brand new things called ICOs, Initial Coin Offerings. Where you had people setting up blockchain based systems where they would create a token, they'd call it the bingo token, or whatever it was on the bingo network and that's what you needed to use in order to use these networks. And they pre-sold these tokens to people as a way to fund the business. And then you had this kind of almost, we had this governance model, I don't know if you remember, but back and over the last couple years, they've had I think, forks of both Ethereum and Bitcoin that threatened to destroy the blockchain because nobody could agree on the future direction of the chain. And so what do you do? Because what you're describing as a user defined organization governance model is you're describing something similar to just being a shareholder. if I'm a shareholder of a company and I have enough shares, then when the annual report comes around, and they say it's time to vote, my vote could influence the direction of that company theoretically.

Julian: Yes, so, I think that there is what we propose this is different for a number of reasons, and like I think two most important like first as I said, this is not non-transferable, or non-tradable. So this is a very special equity, then you cannot monetize on by selling it that's the first case and the second is, if you go deeper into how exactly the mechanics is going to work there. The generation of the proof of use it will have some costs, so, the idea is here that whenever that transaction on Wildland marketplace happens, and there is some fee subtracted from the transaction, and that fee is then divided into two parts. So one part is like just a fund that can be used by the organization for the future development, whatever the government decides. And the other is burned; by the way, it is burned by buying DLM so, the Golem network token. And this is because Wildland is a spinoff of Golem network of Gollum project that they started back in 2015. So, you're still funded from that source and that person is burned so there is the economic cost of generating proof of users. And then the magic is about fine tuning or the parameters in a way that this is a difficult to use proof of use it for your own benefit, I think we are able, like this still in a early experimentation phase, but I think we are able to fine tune the parameters in a way that there is no way of attacking the organization that way. And this exactly, because generating proof of use it has a real cost. So it is in fact impossible to use that mechanics to siphon funds out of organization at the cost lower. So there is no way of siphoning funds out of organization in a way that you will benefit.

Mike: Right. 

Geoff: I also think where the shareholder analogy falls down is if you buy 1000 shares of IBM and you've invested in the company, maybe you're hoping for some dividend checks or to sell those shares in two years and see a capital gain. But you're not expected to contribute to IBM in any meaningful way. And certainly, there's the example of the shareholder meeting that you talked about but for many corporations, they just assume their shareholders stayed out of the meeting and stopped asking difficult questions or tried to change the board or whatever else it happens to be. So, you're a shareholder to some degree but you're also expected to contribute, which I think is a key difference compared to owning a piece of a company a very small fractional piece of a company or even a large piece of a company.

Mike: Julian, what advice would you give to all of these budding decentralist out there that have an idea or that think there's something wrong, or whatever the case may be? What would be your advice to somebody who wants to approach you know, stick a toe in the water? In terms of, can this process be decentralized? How can I decentralize my own experience? How can I decentralize the services that I offer? What would you advise people to do? How do they look at it, how do they start?

Julian: Yes, I would advise them to ask not how, question first, but rather to ask why question first. So, why that needs to be done? Okay. Why I think this is a problem? What's the problem that I'm solving? And I think this is the most important thing, because in our space, it happens every so often, that people are trying to decentralize something. Because they think that they have an idea how to technically implement some kind of decentralized protocol or decentralized something, usually with some decentralized piece of technology. And I think this is exactly how this should not be done.

Mike: I'm right with you on that one. 

Julian: I think we need first to identify, what's the problem? Why we want to do that? And then we move on. So I think this is number one, have to take a very good answer to the why question on a personal level, so it doesn't have to be, the part of the piece of the project but you have to know that for yourself, if you going on. And then the second advice, I think, is to put users freedom in the centre okay, this is very important distinction to putting user in the centre, because I think that the usual mantra for companies is to put user in the centre, which is not exactly putting users freedom or users control or user sovereignty in the centre. And this is what I think, decentralization is all about so, if you don't want to do that you're not decentralizing things.

Mike: Yes. It's one of these things, Julian, where it's funny, a lot of people in the decentralized world that I talk with, it's almost like it's decentralization for decentralization sake. It's not necessarily why are they doing it? 

Julian: To solve a problem? 

Mike: Well, yes, and I think that most of us would agree that the kind of the most outward evidence of say the problem with decentralized systems is the effect it has on the average use. It's been proven to radicalize people, people make mistakes, when they communicate that end up with some poor backpacker getting lynched in a village somewhere, because somebody says, hey, this guy, and they don't think about, it's all just, let's get as many messages out there, let's transmit this, let's do this, let's track lettuce from the factory to the end or whatever, but they don't understand what the benefit of the decentralization is actually doing for the ultimate end user of that solution. And decentralization is something that that needs to actually accomplish a purpose. Otherwise, you can understand, there's a reason why our planet has evolved from say, loose tribal structures into these kind of organized political institutions, where bureaucracies are created that can help things keep going like hospitals, and roads and things and you could argue those are the centralized organizations. 

And the theory is, is that those organizations, if you want them to be in the interests of the citizens, those organizations have to be owned and run by the government, or at least has some kind of citizen led government oversight, if that makes sense. And where is it seems like the difference where the social media surveillance capital world went was just the Wild West Television Without Borders, no commercials and then all sudden the commercial showed up and all this other kind of thing, it was just, there were no boundaries on the internet when it started. And so there was no need to focus on users, it was all about just getting as many users as possible into it and now they're in this problem where they are not thinking about the users benefits at the beginning and what would be a good experience for their users is what got us into this problem in the first place.

Henry: Yes, you're absolutely right, Mike. 

Geoff: Precisely. 

Henry: Julian, I have to thank you for joining us here on the decentralization. You know it to me it was fascinating, because we have been speaking about decentralization for so long now. But to hear your perspective, which is fascinating, because it's slightly different from ours, but we learned a heck of a lot. So Julian, thank you so much for joining us.

Julian: It was pleasure. Thank you, guys.

Mike: Thank you, Julian.

Geoff: Thank you. Bye for now.