No guarantee of the accuracy of the conversation content. This transcription is for non-native English speaker who have a hard time to listen English, so here is the transcription of the conversation. Clarion can greatly helping the EOS community. I highly encourage the EOS community watch this video, or read the transcription.
The original discussion as below:
Daniel Larimer - ClarionOS Project - The Cryptowriter Podcast #3
Brandon Lovejoy (0:44): Today, we want to talk about your new project Clarion, or Clarion OS, give us the bird eye’s view. What is Clarion?
Daniel Larimer: Clarion OS is an attempt to re-enable free speech, to allow anyone in the world to communicate with anyone else who want to listen to them. Imagine we have a magical, technological box that would allow you to broadcast a radio signal that anyone else in the world could subscribe to and hear what you have to say. And you can use that same box to subscribe to as many other people as you want to hear what they have to say.
Now that would be direct person-to-person communication and that’s sort of the idealized form of what I’m hoping to create with Clarion. Instead of a magic box that violating the laws of physics as we know them. It would be done through repeaters. It relay messages through your friends and your family. So you’d establish direct connections to their computers, to their phones, and it will relay from them to everyone else.
So you could imagine as a friend to friend network that allows you to communicate to millions of people without having to rely on centralized services whether it’s Twitter, or Facebook, or Youtube. I think the main thing that makes it different from other “decentralized system” like Mastodon, or Matrix, what not is that it’s not federated. And I can go into more details on what that means.
Brandon Lovejoy (2:36): Yeah, that’s what my next question. So a lot of people when they learned you were working on a new project. They thought, well why don’t you work on this project with all these other people they are doing thing “similar projects” in the space, such as Mastodon and Diaspora, or what have you. What makes it different? Explain federated and non-federated for us.
Daniel Larimer: Federated system the most widely known example of that is email. Email in theory is decentralized, you can email anyone in the world from any email address, and you can host you email server, and that’s in theory decentralize, because anyone can get their own email.
But in reality, you don’t own your email address. You have an account with a service whether it’s Google, or Apple, or Microsoft, or Protonmail, or whatever. And if that service goes down, your identity and you email disappears with it. It’s a similar process with Mastodon, and Matrix, and alike, is you’ve got an identity, but it’s tightly coupled with a particular service provider. And there is only so many service providers out there. And if they don’t want to talk to each other, then your messages actually don’t relay. So you’ve got a Republican Mastodon and a Democrat Mastodon, and they decide they don’t like each other, they won’t connect and the two user bases can’t communicate.
So of these systems have ways that in theory, you can export your account information, and import it somewhere else, but you lose a lot of information, you lose a lot of your followers and your routing information. So it’s no really truly decentralized system that allow you own your identity. Because your identity is controlled by private key, you are the only one that can sign messages, and you can send your messages through any number of these channels. You can have service providers, like a Google or Apple, that will relay your messages for you and you can use multiple of them simultaneously for one identity.
But you don’t have to even use a service provider. You can do it entirely with your friends and family. Anyone in your family that really good with computers, they want to run a node in their computer. They can do that. And the idea of Clarion is that eventually you can download an app. You can double click it, and just run on your computer and when your computer’s on, its facilitating messages from the network, and when it goes off. That’s not a problem, because everyone else who was relying on your computer actually has multiple other alternatives for relaying messages.
So that’s the difference between federated model and the decentralized model that presented by Clarion.
Brandon Lovejoy (5:29): Is it a bit like the approach that Bittorrent takes in terms of content propagating through the network. How is it that, with friends and family hosting nodes, how is that you ensure that your stuff is always online if you’re not always online?
Daniel Larimer: There’s big difference between Bittorrent and Clarion. With Bittorent, you’re randomly connecting to other peers that want the same content. The good news about that is you’re only getting content that you’re interested in. The bad news is you’re connecting to people you’re not necessary trust. It’s different than other protocols like IPFS. With IPFS, you’ve got to host your own node and pay someone to host your content. Even though it’s universally addressable, you get the same file name anywhere. If your node is not online, and there’s no one else who’s downloaded it with a node online. Then your content’s not going to be available and there’s a lot of overhead to maintain that availability.
But with Clarion, when you send a message, you push it to your friends and family. You don’t have to pay them, they’re your friends. You can extend them some space on your computer and you trust them to not upload illegal content or anything that want on your computer. Or they wouldn’t be your friend or family. And if you don’t trust your friends and your family, then you don’t invite them to actually be your friend and family on Clarion even if they are in real life.
So you push it out to your friends and family, and if they know anyone who’s interested in following you, they will relay it to their friends that are interested in following you and so on and so forth throughout the network. So assuming you have more than one friend in the world. You always have people that you can start propagating your messages to. And if you are not online and you want to see (let’s say) a received private message, people can send the private message to your friend and when you come online, you’ll download it from your friends. In that way, there’s always ability to communicate with someone even if their phone’s offline for a while, or their laptop’s lack of power. You don’t need a horizon internet connection.
Brandon Lovejoy (7:55): Cool, so it uses a friend of a friend, sort of network topology. Let’s say Elon Mask wants to use this and I want to connect to Elon Mask. How do I follow what he’s doing?
Daniel Larimer: Hopefully you’ve got friends who have friends who are friends of Elon Mask. There’s Six Degree of Kevin Bacon, it’s a whole game. If you actually follow your social network on average. It’s 6 hops between you and anyone else, some people might be one hop or two hops. So if you have someone who’s a celebrity like Elon Mask. That’s gonna lot of people that are interested in following him. So all you have to do is know someone who’s following him, who knows somebody who’s following him, who know someone who’s his friend.
The bigger celebrity they are, the more likely that people will gossip about what they say. So Elon says “Funding’s secured”, and he pushes that to all his friends. All his friends say they have a lots of friends that know that they’re friends with Elon and they want to follow Elon so they say “Hey, Elon said funding secured”, and then just propagates through the network.
Any messages propagate based upon interactions, positive interactions. So if your friends like something that they see someone else say “You’ll see it too even if you’re following someone, how would you discover Elon in the first place?” Well, your friend likes, when he says “funding secured” and the next thing you see that your friend liked it. And then you say “Oh, who do you like? You like the message from Elon Mask. Who’s Elon? Maybe I follow him.” You follow him, now your friend knows “Hey, anything else you see from Elon? Send it my way”. And then anything Elon likes, or why not start propagating your way?
Brandon Lovejoy (10:11): So, in a nutshell, the difference between a friend and somebody you follow is just the directness of the connection, and the mutual trust?
Daniel Larimer: So what I call a friend is more like a user. There’s somebody that you’ve given an account on your computer which means they have permission to store content on your computer, to use your computer as their server if you will. So, if you’re doing this like a business, Google and Apple might be your “friends” (assuming they are running on Clarion server). And anytime you send a message you push it to them. Anytime someone wants to send a message they can give it to either Google or Apple, then you’ll get it. But they’re basically giving permission to utilize resources on your computer, and it doesn’t have to be mutual.
Someone might friend you, but you might not friend them back, which means you can their computers but they can’t use yours and vice versa. So that it’s just really about who do you trust to publish and who are you going to allow to utilize your computer as “an always on node” as a temporary storage of their entire account information.
Brandon Lovejoy (11:36): Awesome, one other thing I heard earlier we were talking is that Clarion is uses private key encryption similar to blockchain in order to secure your identity and allow you to sign a transactions. So only you can sign your posts. How do you foresee Clarion solving the problem of usability when we’ve seen key management being a kind of a problem in a blockchain base. So how do you wrap this into social in a way that’s easy for your uncle to use.
Daniel Larimer: When you download the app, is gonna generate a password for you, as you to store it in a password manager. Initially the app is just a website. You find somebody that’s hosting Clarion interface, you load it on your phone. I’ll just save it right directly to your password manager on your phone. From there, you can upgrade it to a progressive web app which will be installed on the home screen of your phone, or on your computer, and that will give you longer term storage and alike.
But your password (also your private key) would still be in your password manager, where it should be relatively secure assuming you’re got good security password manager which you should, and people should be familiar with that. From that point forward, you can sign your messages and everything, is good.
What happens if your key is compromised? Or you forget your key? You have to create a new account, a new key. And have it endorsed by your friends and family as this is the same person and then people can set up automatically following your new account. You can import all your information from your old account because you know you download that from other people that know about it. And you’re back up and running with a new key and a new identity. And it’s also care about your friends and family. This identity here is not based upon your government IDs, it’s based upon relationships and it’s based upon cryptographic as its attestations of those relationships.
Brandon Lovejoy (14:08): So if a key is compromised or if a password is lost. In the way that Clarion is set up using a sort of trusted network of friends. Wouldn’t it be ideally suited to create sort of a multi-sign password recovery setup, or is that just too complex, and keep it as easy as possible.
Daniel Larimer: At the end of the day, password recovery means somebody has to have your password. You can do multi-sign on an account to control authorization of who’s allowed to publish to an account. And that’s a way of securing your account. But from a usability perspective, that’s not the most practical. But your account recovery, the updating of your account could be done through multi-sig and say "I’m changing my new key, my friends and family have signed off on the change, and now it’s updated”. And if anyone suspects that account is compromised if their key is stolen. They can use their key to sign their “relocation” and propagate it to say "Hey, that key is dead, don’t trust it anyone, and then the hacker is unable to broadcast and your friends and family can now endorse a new key for you.
Brandon Lovejoy (15:25): You mentioned Progressive Web Apps, tell us what Progressive Web App is for those who don’t know, and why that’s crucial to Clarion and what led you down that path?
Progressive Web Apps have a little bit of additional power over in the browser. They’ve got their own storage. They’ve a little bit more trusted. But fundamentally they’re an app written in web technologies. It doesn’t need to go through an App Store. And all working on all devices so you can use the exact same interface in your browser and then progressively upgrade it to a native experience which means you can use it offline. You no longer need to have connection to the server.
So let’s say that you initially have your account through let’s say clarion.com, you can install the app. You’ve got clarion.com on your phone. And let’s say clarion.com goes down, you can still use your app and it will through websockets it will connect to other peers, other domains, other full nodes that are able to reach, and so you’ll still get your messages, you’ll still be able to receive even though the website that originally hosted the app is no longer available.
Brandon Lovejoy (17:05): A lot of people wondering what could be some points of synergy between Clarion and EOS?
Daniel Larimer: EOS has a major challenge. How do you create new accounts on EOS? Because you need to know someone with an account, you need to exchange public keys, and they need to pay to create an account on the network. That’s really bottleneck for signing up to EOS.
With Clarion, you don’t need to have a blockchain you have to register your account with. So you download the app, create an account and it’s already set up to invite friends and to kind of exchange this key information, which means that we can with the push of a button create an EOS account for a friend on the Clarion network. Because it uses the exact same keys that you would use for your Clarion account. And then the Clarion progressive web app can connect to EOS API endpoints to token transfers, to register global unique names. So that you don’t have whatever pseudonym or label that you’ve assigned to your friends and family.
So you can start to integrate those types of services with Clarion as a purely optional. The EOS blockchain is not mandatory, but there is a lot of synergy to EOS in having integration with Clarion. It can solve a lot of the usability problem with EOS, and as people adopt Clarion as an alternative to Twitter and Facebook. They are automatically introduced to EOS. So it’s a win-win situation.
Brandon Lovejoy (18:46): A social platform that dovetails into EOS sphere but it wouldn’t be limited to EOS necessarily either. Clarion is fundamentally doesn’t have its own token, correct?
Daniel Larimer: It doesn’t have a token. It’s not a blockchain, see that the challenge with blockchain that it requires everyone to process every message, and that doesn’t scale. It doesn’t scale in a decentralized way that we would like in a logically decentralized way. It might be politically decentralized using token holder votes. It might be geographically decentralized by having many nodes on many continents. But logically there is a single thread of events with a particular order, and you’re either in it or you’re not in it.
In that sense, whether it’s Bitcoin, or Ethereum, or EOS, you’re logically centralized. But with Clarion, you and I can be using Clarion, and two other people can be using Clarion. They might have zero connectivity between them, but we’re both using Clarion, and if one of those were to connect. It would seamlessly integrate. Whereas with the blockchain, you might start off with one chain, and the other chain split. But after the chain split, if there’s any conflicts, that can’t be merged back in.
With Clarion, there’s never any conflict, it’s just a gossip protocol combined with an algorithm that’s run on your device to take in all that gossip, and present an experience to the user of who said what when, who likes what, and then on device local AI and algorithms to make recommendations to you. So you’re not leaking data to the cloud and no one knows what you’re actually seeing unless you choose to broadcast the things you like to your friends and family. So it’s being logically decentralized, is fundamental for lower level than even blockchain. Because the purpose of blockchain is to be a tool to help a group of people reach consensus. And then they reach consensus on the order of events and a deterministic way of interpreting those events.
The lower level of consensus is how do I communicate? How do I find people? How do I have a discussion? How do we make sure that we can even agree on which blockchain to use, which token to use. All the subjective messiness of the world is achieved through communicate. And if all communication has to be funneled through a centralized provider. Then all of the sudden, say EOS and your tweet is banned. You cannot talk about ‘EOS’ on Twitter. I’m not saying that’s happened yet, but it’d be very easy. Use the word ‘blockchain’, ban. All of the sudden you can’t discuss the tools you need to reach consensus if you don’t have a tool like Clarion at the base layer.
Brandon Lovejoy (21:55): Cool, I’d like to get in a little bit more to the philosophical underpinnings of a lot of this, but real quick one question that I just got on in the Clarion Telegram group, which relates to what we’re talking about now, but maybe in a way I don’t fully understand, say if I were a Clarion user that wanted to leverage blockchain tech. What will be the incentives and/or efficiencies gained by using EOS mainnet versus say a Proof of Burn chain as Dan has mentioned?
Daniel Larimer: Part of Clarion phase 2 would be to introduce a new kind of blockchain I call it ‘Proof of Burn’, but it’s not proof of burn in a traditional way. The idea is that, instead of having the possible block reward, like you might have a Bitcoin where every block creates new coins. In order to produce block, you have to burn coins or burn tokens, and whichever blockchain, whichever fork of the chain has burned the most tokens is the best fork.
The principle there is that if a blockchain is viewed as a radio station. It has got only so much bandwidth, only so much time on the air. You’ve got to figure out are you going to allocate it to this ad or that ad or this song. And you either have a DJ that’s centralized, is saying “alright, you’re allowed or not”, or proof of work is you burned a lot of electricity, so your song goes, or some kind of lottery event that makes that there, or DPOS, I already got 21 DJs. They’re each taking turns, figuring out who gets to include songs on the air, in a kind of a time sharing way. But with Proof of Burn, it’s who’s ever willing to spend the most, because it’s scarce resource, and you want it to go to the person who wants it. And who do you want to benefit from it, all the people who aren’t broadcasting, all the other people who are refraining from speaking should benefit when someone is speaking on the public channel.
So it’s more decentralized than proof of work because the only way that you can keep speaking forever, is to keep buying tokens from the market, pushing the price up, which means you have to have a lot of money to keep it up. Eventually if you ran out of money, somebody else gets to go on the air and start talking. And so censorship can’t last forever. There’s not the waste that you see with proof of work. There’s not a need to have a technological infrastructure, capability, access to ASICS, access to cheap power. If you’ve got an economic resources, you can get on the air, broadcast a single block, get off the air. But Bitcoin, you can’t buy a bunch of mining equipment, mine a single block, and then leave. It’s a long-term commitment to get your money back, and to get your message out there.
With Clarion, the goal is to allow everyone to run it on your own computers. So instead of having one chain processing 10,000 transactions per second, you have a thousand chains doing 10 transactions per second, or 1 transactions per second. And then you selectively pick which chain you want to follow. If you look at your typical telegram channel. There’s only about one message a second. If it’s going any faster than that, you can’t follow it anymore. So we set it up, so that you can do many-to-many broadcasts on public channels with spam prevention but still being decentralized so people can run these blockchains on their home computer.
The downside to this Proof of Burn consensus algorithm is, that’s slower. It’s more like proof of work, where you’ve got one block every 15 seconds to a minute, instead of half second blocks like you have on EOS.
So the Clarion UI would integrate with EOS through public API node provided by the block producers. And if those API nodes are not there, then those UI elements would not function. Whereas a Clarion Proof of Burn chain, you just synchronize to the Proof of Burn blockchain that you want to listen to. And then you have a full node and your UI can interact with it without requiring any API nodes, or any other infrastructure. And then you get the blocks and everything are just propagating across the friend of friend network. So it’s basically you are following the blockchain.
Brandon Lovejoy (26:48): Under a Proof of Burn scenario, what sort of token emission method is there? How do tokens come into existence and are we talking about multiple Proof of Burn chains?
Daniel Larimer: Like I say earlier thousands of chain potentially, you have to have initial tokens somehow. Somebody is given 100% of the tokens and then they distribute them how they want. They could do an airdrop on all EOS holders. They could do an airdrop on all BTC holders. Whatever community that they want to give the power to communicate on this particular channel. I’m using channel like a blockchain as a radio station. So you figure out who you want to have the power to communicate on it, and then they can sell and distribute access rights to that particular channel. And then the more people that are subscribed to the channel, the more value there is to broadcast on that channel. It really closes the economic utility and value of the token. These would be blockchains that aren’t really opinionated on the content. All options about the validity of transactions on smart contracts would be implemented upper layer. The only thing the blockchain would be reaching consensus on is aside from their own token balances, would be the data that is to be included in the block. And the order in which it was received.
Brandon Lovejoy (28:16): Cool, you’re already looking ahead to Clarion 2.0 or the second version of Clarion.
Daniel Larimer: I think there is a lot of value that EOS token holders could get by funding such a project and then air dropping on EOS token holders. The cost of building that would be relatively small compared to the value of that blockchain, even if it was only valued at 100 million. It’s only going to cost millions to produce. So there’s big multiples for the creation of these chains, assuming we get the developers and resources necessary to make it happen.
Brandon Lovejoy (29::00): Why is Clarion so important?
Daniel Larimer: The freedom of speech is on the decline. All around the world and the past year, we’ve really seen it in action and very blatant in obvious ways. So much so that people not just the platform from one platform. But there would from all platforms simultaneously, and their hosting provider, and the App Stores. Which is a level of censorship that was inconceivable. A couple years ago, no company would dare to risk their reputation to censor people solely based on political believes, or opinions about medicine, scientific research and inquiry.
But now we’re seeing that the big tech companies can’t be trusted. And the only way that you can secure your freedom of speech is to make sure your freedom of speech does not depend upon anyone else. Which get on the philosophy I outline in my book – More Equal Animals, which is free from moreequalanimals.com. But in that philosophy is independence, you can’t create freedom of speech as a service. You can only create tools that allow you to speak more freely. A ham radio is a tool for freedom of speech. A microphone is a tool for freedom of speech. A service might help you to reach people but now you got a third party between you and the people you want to reach. And they have the ability and incentive to control who you can talk to and when, and what you can talk about.
So it’s about my philosophy right now is creating tools that people can use. Clarion is a tool, not a service. It’s not something a Steemit, or Hive, or Voice. Where there’s a company that is providing the service of web hosting for you and if that company goes down. You don’t have access anymore. Now Steemit is a little bit better and Hive is little bit better because you’ve got multiple parties providing services. But you still have one blockchain with one governance on the blockchain. And if the blockchain governance kick out, that you no longer have the ability to use that service. So that’s the key is taking responsibility maximizing individual independence by giving them tools that make them more productive. Just like giving someone a hammer, a knife, a computer, those tools make them more independent, makes them more free. Giving somebody software as a services, makes them dependent, they don’t get they don’t get freedom.
Brandon Lovejoy (31:58): We’ve touch a lot of different aspects of Clarion. Is there anything else that is on your mind that you’ve been thinking about that you want to share to the community?
Daniel Larimer: I’m super excited to be engaging with the EOS community and working on new governance processes that can utilize Clarion to help reach consensus about the direction to take EOS network. So I expect there’s going to be a lot more to talk about in the near future and thank you for having me today.
Brandon Lovejoy (32:40): In the not too distant future, we can look forward to having a more in-depth conversation about EOS and community governance and such thing if you’re into that.
Daniel Larimer: Absolutely, I’m put a lot of work on boiling the abstract ideas presented in More Equal Animals to concrete implementation as smart contracts on EOS. So I’ll be releasing a paper on that shortly and then we can do another video to introduce those concepts to the world.