Below is a transcript of the talk that Lucius Gregory Meredith gave in front of a captive audience at Ecole 42 on April 12, 2022.
Intro
Well, thank you very much for coming, I really want to want to thank Ecole 42 for having us here. And I’d like to thank Raphael, and dappy, for arranging this. It’s a pleasure to be able to address you.
My name is Greg Meredith. My full name is Lucius Gregory Meredith. It’s a funny name. Lucius is the Latin for light. Gregory is middle French, I believe for vigilant. And Meredith is Welsh, for watcher of the sea. So my name means lighthouse. So if you, if you if you see me online, sometimes you’ll see that I’m known as @leithaus.
Right, so I’m here to talk a little bit about RChain Cooperative.
RChain is Cooperatively Owned. One Member One Vote.
There’s a lot of technical innovation in RChain. Mostly, what I will tell you is common sense. We’re proud of both the common sense and the technical developments. But one of the things that I’m personally most proud of is that we are a cooperative.
So cooperative means one member, one vote. Even though we are a blockchain company, the governance is not about how many tokens you hold, the governance is about your participation. And we’re very happy with that. Because over the last five years that we’ve been in development, the membership has provided a kind of collective intelligence. And without that collective intelligence, we would have gone off the rails many times. So we’re very grateful for that. And we really appreciate the participation. And maybe I’ll get to some themes at the end of my talk, in which we talk about how that participation becomes more and more meaningful as the years go by.
Lesson One: URI’s, Not Addresses
0:01:58
All right. So with that….I used an overused meme. But again, apologies that I’m not giving the talk in French, you would not want me to do that. But you know, I think you can probably all read it. But you know, I’m going to change the world for the better, right? For the better, right? Yeah.
So there are three lessons that the blockchain seems to have missed from web 2.0. And they’re all common sense. And it’s like things that we’ve known for decades now. And and we should not forget, it seems like, as human beings, we have this tendency to forget things that we’ve learned. I know I’m doing that, especially as my hair gets grayer.
So the three lessons are very straightforward, right? So we’ve known forever, for a very long time now, everybody who is online knows that URIs are the way to address resources on the internet. This is a very straightforward thing, but there’s actually a logic and reason behind it. And so path structured names make a lot of sense when you’re talking about getting to resources that are organized in a graph or a tree. Right, so the resources are going to be plopped down and locations either in the graph or the tree. And, and so a path in the graph for the tree, makes sense as a way to address the location in the graph for the tree. Right? This is very, very common sense. There’s not a lot of, you know, brainpower it takes to to get this, but it also has to do with the fact that if we’re going to assemble large systems like the internet, we have to assemble them piece by piece we have to assemble them into in terms of components, right. So if you think about the internet itself, it didn’t suddenly burst into being. What really happened was that there were lots of little intranets, right, and those intranets became glued together because the value of connectivity overcame the worry about security.
So that’s how we have to assemble large scale systems. And by the way, it’s not just how we have to do it, this is the way nature does it, right? If you look outside, right, there isn’t a king tree that’s telling all the other trees how to organize themselves. The trees are organized in networks and the networks are connected by mycelial growth. And it’s actually through these mycelial networks that they share resources. So it’s very much like a kind of internet, you know.
So these principles of organization, they’re all throughout nature, and you only have to open your eyes to see them all around you all the time.
So URI’s not addresses.
Lesson Two – Search is Greater Than Tokenization
0:04:50
Number two, search is greater than tokenization. So I’ll talk more about this as we go through. But think about what was the real Progenitor, maybe many of you are not old enough to have seen the transition, but the transition to a global economy was really all about search.
When you could search for people, you could search for goods, you could search for services, you could search for reviews, you could search for analysis. That’s what made the global economy spring up. And, yes, the social media amplified that transition, but it was really ubiquitous search.
And we’ll talk a little bit about what that means in the culture of NFTs later in the talk. But really search is what brought about the global economy.
Lesson Three – Advertisement is Bigger Than Paywall
0:05:42 And advertisement is bigger than paywall.
So how many of you who are using your phone right now are using free apps? Yeah, yeah. So and those free apps are subsidized by sponsored content, right? So your uses, your use depends upon that. So we’ll talk about all of these points.
Compositional Solutions Need Hierarchical Structured Names
Okay, so I’ve kind of I’ve already touched on, on what’s going on with URIs. So compositional solutions need hierarchical structured names. And I’ll touch on this again at the end. But I do want to point out that in Rholang, I took this to the extreme. So with the rho calculus, which is at the heart of Rholang, and RSpace, which is the way RChain scales, the idea is not just paths, but actual programs.
So in Rholang, programs are names.
So it’s even more amazing and dynamic. And if you think about how it works in chemistry, or in biology, when a molecule impinges upon the surface of a cell, there’s quite a bit of a computation that’s going on before there’s any sort of binding that happens. And so I wanted the rho calculus to be able to model those kinds of phenomena. So in rholang, what you’ll see is just that the programs have to match structurally.
So rholang is organized over two simple principles, senders and receivers.
It’s very much a message passing paradigm, and senders and receivers, they send and receive over channels. And in rholang the little mind twist, the little insight is that channels themselves are programs. And so at the first layer of of work in this area, you just check that the programs have the same structure. But if you go read the rho calculus paper, you’ll see that we propose another rule for communication, which we call the annihilation rule. And basically, the program’s have to eat each other up until there’s no more computation left. But in that rule, the channels are actually running computation against each other to see if there’s a match. Yes, of course. Yes, please. I love questions.
Questioner:
You have a solution for URIs on blockchains?
Lucius Gregory Meredith:
Indeed, that’s what I’m describing right now.
Questioner:
Are there other solutions? Or
Lucius Gregory Meredith:
No, all the other chains that I’m aware of, they use flat addresses. And that’s a mistake. And you can see with the diagram kind of the point I was trying to make is that just in the setup, right?
The Internet didn’t come into being poof, overnight. Right? Lots of little intranets. And that’s going to be the same with blockchains. Right? So as blockchains get to the point where they’re scalable enough, they will be sharded. Right? They’ll be organized in a hierarchical fashion. Right? And so if I have just one moment, let me finish that point. And then so so if I have a bunch of shards that are assembled in a tree or a graph, if I have a resource down here, yes, then I need a path. From the entry point to the resource is very obvious. Yes? And also, when I start to have interoperability between networks when I aggregate networks, and also if I have a resource down here, and I’m entering in up here, I need a path. It’s very simple. Yeah. It’s not very complicated. It’s just I as I promised you, it’s all common sense. Please. Oh, sorry, you had the question first, and then yes,
Questioner:
yeah, in some way URI or just like a readable layer on top of IP address and some, like stuff that human don’t read and don’t write. So we could imagine something like DNS, for example, that will just add a readable layer on top of addresses.
Lucius Gregory Meredith
Yeah, but if you start with flat address space, you’re not compositional. You see what I’m saying? So if I start with Shaw, right, then that’s flat. I can’t I can’t combine, but it’s not about human readability. It’s about composition. It’s not about what the humans can see. I’m talking about for all computational agents. Right, that the point is that with with, with IP addresses, you have a structured name. The name is structured, right? And that’s the point names should be structured. And that structure, at a minimum, is going to be a path. And our argument is that better than a path, you make it a program. And that’s that’s kind of the pin, that’s the ultimate, that’s the pinnacle of computational complexity for things that are going to address resources. Yes. Yes. Great question.
So I tried to give a metaphor, right. So the metaphor was an intranet. Right? So in RChain, the network is sharded, in the sense that there are resources concentrated here that, if you don’t ever step out of this network, then you don’t have to worry about going up to here for consensus, or other kinds of computational resources. Right? If you have a transaction that touches resources here, and resources here, then you’ll have to go up to here to get the consensus to work. Yeah. Does that make sense?
Right, so you can tell from the color of my hair, you know, I can speak with experience. Back in the 80s, we worried about this kind of thing. So for example, we were planning to be able to do composite transactions, like if you go traveling, what do you want to do you want to book an Airbnb and an airplane, and maybe some ground transport? Right? And so if you do that, maybe two out of three, you want to definitely be booked before you head out the door. Right? And so that’s like a kind of composite or nested transaction. Right? And so you can imagine that there’s a hospitality shard, that’s mostly focused on you know, hospitality, hotel, Airbnb, VRBO, that kind of stuff, right? And there’s a travel and logistics chart that’s mostly focused on those kinds of transactions. And that would be nice, right? Because if you’re not worried about hospitality, right, you’re a shipping company, right, then then you’re only interested in booking the freight. And so your transactions remain here. And so they don’t get mixed with a bunch of other transactions that are isolated. And so this runs faster. Does that make sense? Yes.
Questioner
Yeah. So it’s just to understand are these the bridges between blockchains that we’re talking about?
Lucius Gregory Meredith:
So I was just about to get to that, right. Yep. Excellent. Excellent. So RChains’ sharded architecture. It works for RChain networks, right. But it also extends to bridges for other networks. And that’s why I have Ethereum down here is because the same capabilities in terms of… if I have a deployment for a rholang contract that comes in here, that mentions resources here and here, right, the capabilities for accessing this at this level, they look the same. Right, just another shard, from from this contracts point of view. Yes.
Questioner:
Can you step back just a little bit and, and talk a bit about URIs versus addresses for the …(unintelligble)….
Lucius Gregory Meredith:
Yes, sure. So are you? Are you familiar with using Bitcoin or Ethereum?
Questioner:
Yeah.
Lucius Gregory Meredith:
Okay. So so you know that when you’re when you’re dealing with Bitcoin or Ethereum, you have an Ethereum address.
Questioner:
public or private?
Lucius Gregory Meredith:
Yeah. So if you’re sending to a smart contract that address the address for that smart contract. That’s public.
Questioner:
Right.
Lucius Gregory Meredith:
Yes, exactly. So regardless of whether we’re talking public or private, that address is drawn from a fixed address space, very large. But fixed, it is not compositional, you cannot see by looking at the address the structure of that space.
Questioner:
Right. So I think it’s, it’s purposeful.
Lucius Gregory Meredith:
No, no, it’s not. Right. The point is that, you don’t even have to go as far as homomorphic encryption, to know that if I were to structure a bunch of chains in a graph, I want to be able to see the path to know that I’m getting to the right resource. Right? What they’re talking about is: this graph is dynamic, I can pull this shard out, put another one in. Right? So I can add and remove shards, I can add or remove sub networks at any point in the graph. Right?
Lucius Gregory Meredith:
And the fixed flat address space is not going to work. And the reason it’s not going to work is: a) because these addresses will be separate from those and these addresses will be separate from those.
Questioner:
This is what we call URIs.
Lucius Gregory Meredith:
Yes. Past based resources. Yes, exactly. Okay, one more question, then I’m gonna move on, go for it.
Questioner:
I’m not sure I understand what the main net is there. Is that the Ethereum network? Or is that a smart contract of yours on the Ethereum mainnet?
Lucius Gregory Meredith:
Neither. This is this is this is our own layer one blockchain.
Questioner:
Okay. Those are all layer ones. Right?
Lucius Gregory Meredith:
This is like a layer one, but its subsidiary to this one, in the sense that it knows about how these resources are related to each other. Does that make sense?
Questioner:
Yeah, it does.
Lucius Gregory Meredith:
Awesome. Great questions. Very good. Audience. I love it. Thank you so much. Okay, so I think you kind of get the ideas about a compositional structure is how we scale and as a part of that compositional structure, we need naming schemes that match up to the compositional structure.
It’s common sense. And it’s how web 2.0 has been organizing the web for decades. Right? Probably many people in this room are younger than the development of URIs to develop to access resources on the internet. Yeah, so very ubiquitous technology.
There are potentials and capabilities in terms of search that we have yet unlocked.
Okay. So I kind of prefaced this point, by talking about how the global economy arose from search. And now what I want to do is to go a step farther. So there are potentials and capabilities in terms of search that we have yet unlocked. Obviously, I think many of you are probably aware of machine learning, and its capabilities for doing search. And, if you will permit me I will walk you through some more exciting capabilities for search as well, that are not machine learning based. But regardless of the means by which we enhance our search capabilities. It is a step backwards to put resources behind unsearchable things.
Right, so, Steve Ross Talbot, who’s sitting in the back. He is the CEO of RChain’s new subsidiary RChain Publishing, and it has very cleverly been running user groups with artists because one of the things that RChain Publishing is doing is launching NFT capabilities. And, what we have heard, and he will verify this, what we have heard from the artists over and over again is: “now I’ve minted my NFT. Now what?” “How do people find me?” Right?
And so the market solution right now is: hype campaigns. You run some giant hype campaign, you get a bunch of people on board, and that creates enough buzz through the social networks that people discover the NFTs that you’ve provided. But that’s not the way the Internet actually works. The Internet actually works through search. What you would like is for people who don’t know you – to find your work – people you have never met – to find your work.
So let me give you a simple example. So we can imagine that, you know, after an event like this, there might be a party, and there’s a DJ, and the DJ wants to plan their set. Right? So so she might, our DJ might might look at a very simple arc for the show, right? So she begins chill, she has a rise to a crescendo, some climax, and then she has some datum. All right, so very natural, simple arc. So what does that mean in terms of actual properties of music?
Well, it turns out that there are objective properties of music that we don’t have to argue about. Right? So beats per minute. No, like people might argue is Miles Davis is Bitches Brew, jazz or rock? Or maybe it’s something else, that that’s something we can debate about. But everyone can just pull out their metronomes. And check how many beats per minute is on the title track, right? And if there’s some wobble, you can detect that too, right? This is an objective property. We don’t have to worry about that. Well, how does that relate to our DJ? Well, chill means downtempo. Yeah, it’s very simple. crescendo means pounding fast, you know, many, many beats per minute, right? Again, so now we can begin to use these objective properties against digital assets that we can query for right now you can detect beats per minute, algorithmically, given a digital asset right now, no big machine learning to do that. Now there’s some fancier tech to detect key or keys in a given piece. But that’s also detectable these days.
Right. So down down at the bottom of her arc, she’s looking for ambiguous keys, ambiguous mode, so maybe some Mixo, maybe some Dorian, but not strong commitment to you know, like a major or a minor key, right? That’s gonna happen at the crescendo. And then it’s going to be on the chill side. Again, it might begin to get ambiguous. So these are all and we can also algorithmically detect instrumentation. So that’s objective, right? We’re not going to argue about whether there’s a trumpet in that track. We’re not going to argue about whether there’s rock drums or rock guitar in that track. Right. Okay.
So the point is that our DJ could issue standing queries against the network of resources like this one. And those standing queries, she could use to craft her show. Right. So now, the people who are producing music that the DJ is going to consume, they don’t have to personally know the DJ. They put their music NFTs, up on a search enabled chain. And the DJ does the search and finds the new music. It’s very simple, right? This is all common sense. There’s like no magic here. Right? And so if we are to go backwards to the way NFT’s are being done today.
So there’s a certificate minted on Ethereum if you’re going through Open Sea, so there’s a certificate method on Ethereum – and that’s not even the digital asset, right? So that part is not searchable. And then the asset is stored on something like IPFS, and that’s not searchable. So that’s a step backwards from the way the web works today. Because they don’t offer a transactional query mechanism.
It’s that simple? Yeah. So Rholang, despite the fact that there are six language constructs in the core of Rholang.
You will find that in addition to being a concurrent, smart contracting language, it is also a transactional query language. And that’s what allows us to roll out search.
But of course search wouldn’t be meaningful if you couldn’t also put the resources on chain. And that’s what we’re doing. RChain is scalable enough to put the digital assets directly on chain.
When we mint an NFT, and Steve can show you these demos, the NFT goes right onto chain. There’s no certificate that’s pointing to some other place. It’s on chain. And that becomes important for a lot of reasons, but especially for search. You had a question?
Questioner:
Yeah. When you did speak about a browser on chain. I have a question about Ujo. Do you know Ujo? Yeah. U G O. How do you ….?
Greg Meredith:
You? Oh, I thought you said YouTube. No, you’re saying something else?
Questioner:
No, Ujo, U J O …. it’s a first marketplace for artists.
Greg Meredith:
Oh, Ujo, yes. I know. I know. The founder. Yes.
Questioner:
And what about you? And Ujo? Are you the same??
Lucius Gregory Meredith:
We’re not. We’re not the same. No, they were at least originally. They were they were working with Ethereum. But Ethereum doesn’t scale. Right. So so we would we would love if they try out RChain? Yes, please.
Questioner:
My question is about, how do we make search? Because today? We store NFT with metadata? And so on. Yes. For instance, for music? Or do you? Even if the music is stored on chain? Yes. Then what do you charge for it? And know, for instance, that is techno or? I don’t know.
Greg Meredith:
Yes yes yes. So this was exactly the point I was trying to make. So tags like “this is techno” or “this is jazz”, or “this is hip hop” or whatever. That’s subjective information. So you could add that as metadata that you stored the package, and then that becomes searchable as well. But the point I was trying to make is that there are objective properties that you can pull straight out of the digital asset. Right. So I’ve got a I’ve got an actual wav file. It’s on chain. And I can go run an algorithm on it that says, What’s the beats per minute? Yeah, that’s the point. Yes, please.
Questioner:
I really don’t understand what’s the point of storing data on chain? Because it’s replicated. So it’s a huge consumption of energy.
Greg Meredith:
Oh, yeah. I love I love that. Okay, you have another point.
Questioner:
Second point. Is that as hard to have a domain name on your last slide on your URI system, as it is in the real internet? Or Is that easier?
Lucius Gregory Meredith:
That’s very easy. Yeah. So the second point is easy. Now, let’s talk about how that works in Open Sea and what’s happened.
So they, they don’t replicate the data. And they have lost many assets. And through their lack of replication, so for example, an $11 million album, the digital assets were lost. And there are hundreds of assets that have been lost, because they don’t replicate the data. So with respect to charging, what you do is you provide a rental model, just like Dropbox, and Google Drive, and all the other storage systems.
Questioner:
So most of the blockchain right now have been built with limitation in terms of scaling and stuff. So that’s pretty much why Ethereum is not storing all the metadata and all the images from NFT because it’s already hard to scale with the small data that we have, so we can’t have everything. So I really don’t haven’t seen how you could have all this stuff and still be scalable and still be accessible and cheap.
Lucius Gregory Meredith:
I love that question.
Questioner:
And still like I love decentralization. So either you have some huge machine so that you can run like Solana does or you have to find a magic trick and I’d love to know
Lucius Gregory Meredith:
Okay, so it turns out that it’s not a magic trick and again, it’s totally common sense. All right, so I would like you to imagine for yourself an eight lane freeway, right? So see it in your mind. And now imagine that most of the cars observe good lane discipline. So mostly they stay in their lane by and large, right? And so that eight lane freeway is going to get eight times the throughput of a single lane freeway. Yes? Okay, so now what happens if I funnel all those lanes down to one lane? We’ve all seen this right? You only you only have to look at satellite pictures of Paris, or Beijing? Or Shanghai or New York. Right? We know what happens. This is all common sense. Yeah. So with Ethereum, they have a single threaded VM.
So that means that even though most transactions are isolated, right? ….So it is certainly the case that if I pick out two people, one person in Seattle, and one person in Paris, one person is buying coffee, and the other person is buying baguette, right? Almost certainly. Those transactions are isolated. They’re not touching the same resources. So maybe we draw our sharp boundaries between Seattle and Paris, what do you think? So now I go just focus on Paris. And I find two people who are purchasing something at the same time, almost certainly, those transactions are isolated. So maybe I put my shard layer between North Paris and South Paris, or by divide Paris in two shards in quadrants, or where do I put the boundary? …..No, doesn’t work like that.
The right way to do this is you have a model of concurrent computation. And that model tells you when transactions are isolated. That’s what RChain does. So we can do many, many transactions all at once.
And the way that the consensus works is, the only thing you have to worry about is winners of races. So there are two simplest cases of races, you have two senders and one receiver. And you have two receivers and one sender. Yeah? Those are the only kinds of races and your consensus only has to make sure that all the validators in your network agree on the winners of the races. And now they can all run those transactions concurrently.
Questioner:
This is a really nice. I think Solana is doing pretty much the same thing in some way. And even them, acknowledge that they couldn’t store all the data themselves. So they are putting all the data on on Arweave, which is like another layer one blockchain, which is made for data storage. And Arweave is like, I don’t know how many data they store but like, no one can have like, a node and store all that much data.
Lucius Gregory Meredith:
That’s why you have shards, you don’t draw the shards because of transactional boundaries, you draw the shards because of resource boundaries. So down here, you might, for example, focus on audio data, and over here, you focus on video data, or whatever. That’s what sharding is for. Right? And, by the way, please, Steve, go ahead.
—–(gap in audio)—–
Lucius Gregory Meredith:
It is. There’s a simple way to think about it again, all common sense.
Do you want to start with a cash register and build a general purpose computer on top of the cash register? Or do you want to start with a general purpose computer and build a cash register on top? I think this is obvious.
Yes. Right. So Ethereum built a cash register – a payments network – and tried to build a general purpose compute system on top. But that’s not the right architecture, not if we’re going to build Web3.
So now I’m gonna move on to the next point. And again, this is completely common sense. Right? So how many people here use Gmail? Yeah, yeah. Do you click to pay anytime you use Gmail? How many people use Google Maps? Yeah, yeah. I think you see where I’m going with this. Yeah. Okay. So, what we’ve what we’ve learned with the blockchain, though, is if I have resources stored on chain, let me take a step back, a web 2.0 application, right?
Request comes in from the user, it hits their web server, the web server figures out, Hey, I gotta go to the database, or do some compute and then go to the database, and all of that back end processing. Right. That is, you know, it’s probably running on, you know, Amazon, AWS or Google or, and if not, you know, it’s running in their own server farm.
So if it’s running in the cloud, then they can settle up with the cloud at the end of the quarter, right? Yeah, we’ll pay our bill then, right? So you don’t have to charge the user right then and there. But as soon as you hit the chain, right, if I hit Ethereum, I gotta pick gas right then in there. Right? Because I can’t get it processed unless I’m paying the gass right then and there on that request. I don’t get to wait three months to settle up. Right.
So what that means is that for web 3.0 applications, you have only a handful of choices. Either you’re going to make the user experience miserable. Okay, pay, okay. Pay, okay, pay, right every time you go in and and hit I want to read the next email. Oh, you gotta pay every time you do your next tweet. Oh, you got to pay now. That’s ridiculous. Okay, then the other solution is they put up a paywall. So how many people, they hit the paywall, and they go, “I’ll find another way”? Yeah, yeah. Okay. So then what is left?
What’s left is sponsored content, right? And guess what, if the sponsored content can be delivered through a decentralized mechanism, then maybe just maybe we don’t have to have the sponsors or the advertising network spying on you as a user? You know, like, it’s kind of creepy to have people spying on your behavior in order to deliver you sponsored content. That’s kind of silly, why not just ask people? What would you like to see? Right? Like, if you’re interested in I don’t know, golf, or, you know, music, then you say that, and on top of that, you get to participate in the selling of your data. I think a lot of people would be happy to give up their data – for example, as a result of COVID, I think a lot of people would be happy to publish their health data to support research on on COVID, as long as they were given cryptographically strong guarantees that the data was anonymized. Yeah. And if you could make a few REV, or a few token off of that, you know, giving up your data, that would be more, the better, right? You’re helping out and you’re getting compensated. Right. So instead of having these backroom deals with Cambridge Analytica, people who just say, if you’re going to sell my data, I want a piece of it. Yeah, this is common sense, please.
Questioner:
Oh, there is a lot of data, which is thought by the ?. And I wanted to know, first, are there URLs for users use? Talked about smart contracts? I’m correct. And who is data shared is that by my own will, I will say my name is”x” and my age is “x”, for example, for the COVID example, my age, my sex, my medical history. Is that me writing the things or?
Lucius Gregory Meredith:
Thank you. That’s a great question. So so you’re getting right to the heart of what I’m trying to say. So let us imagine that down here is an address and at that address is a smart contract. And that smart contract is going to store data for someone who holds the private key to that data. Yeah, very simple. Okay, great. So then you can put whatever, if you hold the private key, you can put whatever data you want. And maybe just maybe, that smart contract has an arrangement with another smart contract, where a healthcare provider can upload data to your address with your permission. Right? So you walk up, you get a COVID test. And you click click a box that says yes, I would like you to store the data on my private wallet, my private personal info wallet. And so they, they submit it to your info wallet, and maybe even they agree to forget the data. Right? They did the test. They know who you are, but they’ve made a legal agreement with you. But we’re not going to remember this. Alright, so only you have that data. And now if you want to share it with someone else, You can share it with someone by giving them the private key or giving sharing with them some API to your data wallet. Does that make sense? Awesome. Okay, please. By the way, you guys are great. It is so great to get good questions. Thank you.
Questioner:
Isn’t that bad for the security of the blockchain to give private keys to advertisement companies? Because advertisement is dynamic, and it changes quite often. So they to interact like in a very dynamic way with the blockchain, which makes the blockchain not that isn’t. Let’s call it like not that secure, because it’s not only the responsibles of the blockchain, interact with it. And maybe with time, you could even take the advertisement service and integrate it into the blockchain. But then there is the risk that there is no decentralization by then. Because if you own the blockchain that has got the info, and you give me the advertisements, then this is not decentralized anymore.
Lucius Gregory Meredith:
I love your question. So now let me address that. Okay, great. So by comparison to the way the Brave browser does it, right, the Brave browser inserts sponsored content in resources that are delivered at the browser layer. Right. So there’s a there’s a problem with that solution, if you’re coming up through the mobile path, right, if I have enabled native mobile app, brave doesn’t get any piece of that pie. Yeah. So. So that’s, that’s a problem. Right? So now I’m going to talk about a solution that solves the Brave problem, and answers your question. Okay, so now imagine that my sends and receives, they add an extra parameter, right? So a request comes in from the user here, which then sends a query, which is a receive…down this way, right? And so the request has all the key authorization. Right? But if you don’t supply an advertising token, at this point, then it goes over to another source, and picks up advertising content, and interweaves it at the data layer, right? And then shoots it back up. So now, the two are separate. And if you change the resources here, that will just be reflected in the request. And it’s just as decentralized.
Does that make sense? Do people follow what I’m saying?
No, you don’t. Okay, so let’s imagine that over here, we have an advertising shard. Okay, so Starbucks has purchased an advertising campaign. Right? Okay. So they do one for spring, and they do another one for summer. And they do another one for fall and another one for winter? Right? So let’s say it’s spring in Paris. And so they published their advertising campaign to this shard. So now the user request comes in here, and no advertising token has been supplied. So that means that this request is going to get sponsored content with it. If you get enough token, then you don’t get any sponsored content, right.
So that gives you the ability to have a paywall, and no paywall otherwise, okay, great. So now it comes down here and says, Okay, I’ve got the data, right. But its only allowed to see the data that’s behind the private key it’s got. Signed by the user, right? Okay. So then it says, but we don’t have enough advertising token to deliver this data pure. So then you just go over here and you say, show me whatever campaign you got. Okay? So then the campaign data comes in, and is now delivered. So what’s here is both the user request and the advertisement.
Does that make sense? Okay, that’s completely decentralized. There’s no controller of that network. So that’s what I’m saying. There’s another shard just like this one, but it’s an advertising shard.
It’s on the blockchain. Exactly right. So we now have decentralized advertising infrastructure. All right? And that’s how we’re going to scale web 3.0. Right? Web 3.0 needs ads more than web 2.0. Because otherwise you hit the chain resources, and you have to pay gas right then and there. With this architecture you don’t.
Okay, so I’m just going to summarize now. And then there will be time for questions. Three points, right?
1 – URIs not addresses.
It’s what we knew. We knew this for decades.
2 – Search is greater than tokenization.
We cannot lose search, we cannot go backwards, we must go forward with search. Otherwise, if you go back to the overused meme at the beginning, we’re not making the world a better place.
And then 3 – Advertising is bigger than paywall.
I think everything I’ve said here is totally common sense. And it’s something we all already knew from web 2.0. All right. Now, any other questions?
Questioner:
I mean, there’s already web 2.0. All the services, your advertising, are very useful in order to, to have decentralization, but is decentralization so craved by the user, that they will switch to a different…. I mean, most of us even when we were interacting with with web 3, we’re still doing Google searches. So yeah, yeah, that kind of a revolution – is there enough incentive to change everything?
Lucius Gregory Meredith:
I love what you’re saying. And I think this is the crucial point. So one of the ways that I thought about this was when Edward Snowden came forward, and he revealed what the US government was doing with Project Prism, then a lot of their response was to make jokes about it on Facebook. I don’t think you’d quite getting it. Right. But what we are beginning to see is a change in public sentiment. So over the last three years, all of the big internet companies have been hauled in front of Congress, because the public sector is worried about the centralization. Right? And that I believe, is the beginnings of what is going to change that. But at the end of the day, you just have to deliver better solutions. And so what RChain is focused on is delivering better solutions, better search, better capabilities to locate resources. Right? Better advertisement. At the end of the day, if you could make money off of people reselling your personal data, wouldn’t that be better? Wouldn’t you jump to that instead? That seems to me the way to go about this. Just make it better. Yes,
Questioner:
Just wait, I have a question from Ravi. Okay, so how to get started with Blockchain? How to get started with Blockchain?
Lucius Gregory Meredith:
That is that’s actually a really good question. I don’t know, Ravi’s background. Like, for me, I had two paths in. One was: I’m a mathematician, and blockchain is a way for me to pay for my math habit, right? So there’s lots and lots of technology. And if you want to talk to RChain, we’re very happy to talk to… especially with with a with a community this smart, we’d love to talk to Ravi. The other way I got into blockchain, and this is what I was hoping to talk about at the end. And I’ll just give a little bit of flavor of this and maybe more as we go on. So two quick points, and I promise I’ll get to the meat of Ravi’s question. 1 – If you read the IPCC reports, it’s pretty clear we’re on target for 2.7 to 3 degrees Celsius temperature rise.
Lucius Gregory Meredith:
So unfortunately just passed 1.5 – we’re already at 1.2 – 1.5, the tropical zone around the planet becomes uninhabitable. So the conditions are such that the humidity and the temperature is so high that your sweat doesn’t cool you off. So you don’t know you’re dying. So now let’s just ask ourselves how many people live in the tropical zone around the planet? 40% of the world’s population today lives in the tropical zone. So we’re looking at about 2.8 billion people that are going to have to relocate. What timeframe are we talking about this? So by calculations from the IPCC report, it could happen in as soon as 15 years. So in 15 years, roughly 3 billion people are going to have to move somewhere else.
This is a calamity, the likes of which the human race has never seen before. And so there will be – you can count on it – even with all the genius solutions out there, and all the genius people working on things, there will be stress and pressure on the infrastructure. So we will find that the power grid, and the telecommunication grids amongst many other pieces of infrastructure will be stressed to breaking.
So what is the real purpose of decentralization? Its availability. What’s genius about Bitcoin, is that whether it’s a Russian gangster, or the Chinese government or the US government, or somebody else, if you shoot a Bitcoin node in the head, 10 more can spring up around the world to take its place. That’s what’s cool about decentralization. It’s about availability, it’s about robustness in the face of disaster.
And that’s why I’m building RChain, because we need a telecommunications structure that can integrate with a smart power grid that can handle the stress of what’s about to come.
Right? I’m building these tools for you guys. Because it’s your generation, that is going to have to bear the burden of the consequences that we set up for you. So another way into blockchain, Ravi, is to begin to think about what is a resilient and robust infrastructure that’s going to handle what’s coming. Next question,
Questioner:
So my question is about the speed of the transaction. Because normally, when you do a transaction in Bitcoin, for example, takes almost 10 minutes. So every time you want to do a search, and you want to get an ad, so is it going, how do you do to make it faster?
Lucius Gregory Meredith:
I love this question. This is the best question. Okay. So for decades, for about a half a century, how have we been scaling software systems? Do we just sit at our desks and wait for a flash of insight? No, that’s not how it works. The way you scale software systems is you throw hardware at it. That’s what you do, right? That’s why the cloud exists. Okay, so let’s check the existing chain architectures. Do they get faster or slower as you add our hardware? So as you add nodes to these networks, do they get faster or slower? They get slower?
Okay, so this is an anti scaling architecture. RChain, gets faster. It’s a scalable architecture. And by the way, I’d like to put this in context. Right. So what are the hardware trends? So I’m going to make one point about the hardware trends. And again, this is all common sense. The point is more physical threads per cubic centimetre at lower power consumption. So you don’t have to believe me you can go check, what is the main point behind the M1? More physical compute at low power consumption.
That was a watershed moment, because at that point, with the M1, we finally saw the death of the FAT instruction set architectures. M1 shows that the future is going to be reduced instruction set. Okay? And the reason is because reduced instruction set allows you to deliver more physical threads per cubic centimetre at lower power consumption.
Now, let’s go out from there. So the Tesla Dojo architecture, right, what is the point of this modular supercomputer architecture? More physical threads per cubic centimetre at lower power consumption. Let’s go out from there: optical computing. What do you get? You can tell me, right? That’s the trend, right. So if your software doesn’t eat physical threads and turn it into performance, you’re already dead, you just don’t know it.
So RChain, as near as I can tell, is the only blockchain architecture that eats physical threads and turns it into performance. We will eat as many physical threads as you can throw at us.
So the numbers that we got from benchmark tests in December, were roughly 1000 transactions per second per node per CPU. That means if you have 10 nodes in your network, each of which have 10 CPUs, that’s 10,000 transactions per second, if you have 20 nodes, each of which have 20 CPUs, that’s 20,000 transactions per second. Right? So we scaled linearly, but obviously, it starts to get pricey out at 100 CPUs per box, right? But that’s why we’re counting on the hardware trends. What the hardware trends will be delivering for us is cheaper, lower power consumption, physical threads. Next question.
Questioner:
Get one online from TomTom. So how easy or hard is it to make coding in Rholang? How much time does it take to learn that language and to create some first decentralized app?
Lucius Gregory Meredith:
So that’s also a great question. So I’ll give you two or three examples. So you can judge for yourself. And then I’ll give you some objective numbers. So example number one, RChain is a cooperative. And that means that at least once a year, in October, we all get together and we vote. And what the coop members vote on is they vote on who is in the board, and other items of business. So in 2020, a) we’re spread out all over the globe. So we have many, many members in China, we have members in the UK, we have members in Europe, we have members in South America, in the US – all over the globe. So getting us all together to vote is already a difficult process. And b) then there was COVID. And so I said to the membership, guess what guys, we’re not going to get together this year. But here’s an algorithm which will allow us to vote on chain. And the algorithm will give three properties.
One – you can vote from the safety of your own home, as long as you have an internet connection.
Two – no one will know your vote except for you. Unless you tell them.
Three – as soon as the polls are closed, you can go to the chain and get the results. So if the coop is telling you something different about the results, you know, something is squirrely. Okay, so I didn’t write the code. I said, here’s the algorithm, the community in six weeks built RVote with no one in charge. We’ve been using RVote to do all of our voting now for two years. And if you compare that to the US elections 2020, where 1000s and 1000s and 1000s of elderly people were standing out in line for hours in the middle of a global pandemic.
I think we could have saved the US government billions of dollars, maybe maybe, maybe hundreds of billions of dollars. So that’s just one example.
Number two: So we work with a South American cooperative called Fiqus. They’re a network of about 60, developer cooperatives, you guys might be interested in them, and they might be interested in you. So we had a member take the Zulip open source chat program, you know, probably you guys have used it. It’s like Discord only it’s open source. And like Discord to talk to a particular community, you gotta go to that server. So one of our members in an afternoon, ferreted out in the postgress schema, where the chat tables are being updated and added some triggers so that the chat data would go to RChain go on chain. And then that kind of, you know, he didn’t have any more time, so, Fiqus wanted to do some work for us. And I said, Okay, well, you know, here’s a little project, take the chain data that represents the chat data and put it into into a Zulip server. And so now we have a little demo of a federated Zulip so that you don’t have to hit the particular server. You can hit any server that’s connected to RChain. Right? So what that means is we now have a template for off the shelf, standard open source stacks to make them decentralized. That took the Fiqus guy three days starting from no blockchain dev at all.
Right? He had no experience writing blockchain. And he learned Rholang and the rest of RChain stack and wrote the code in three days. And it took our member half a day. So call it four days.
Okay, so that’s, that’s another example. And then if you go look at the core of Rholang, you will find six language constructs. So with a group, this smart, I’m pretty sure you can count to six.
So that’s my answer. Yes. Next question.
Questioner:
Yes, I have two different questions. The first one is, what about privacy? Like, do you have any privacy enabled stuff on RChain? And the second one is, where can we test any like, Do you have a main net? Or do we have a Live app that we can test and try it out? Absolutely.
Lucius Gregory Meredith 1:00:58
Yes. So we’ve been in main net for two years.
Lucius Gregory Meredith 1:01:03
There are lots of lots of apps you can try out, you can, you can if you have a group that you’re organizing, you can run RVote to decide something. Right? There’s RChain Publishing, which has a really cool piece of technology that they call RDrive. Steve, we’ll be demoing this tomorrow at the Paris blockchain Summit. Just to give you a quick summary, it allows you, on your local machine, to mount RChain as if it were a file system. So you can drag and drop resources onto chain using the local file browser.
Steve Ross Talbot:
And it should not cost you more than it takes, energy wise, to make a cup of tea. Right? Right now it does for many, but it doesn’t on RChain. With RDrive, and what we’ve been doing, what preceded it was we wanted to build a COVID passport. We built one, we built it. Raphael was involved in this – he’s in the audience. So we built it in like six days. And we showed this to lots of people in the medical industry, they loved it. The problem we had was as a decentralized organization, coming up against centralized organizations that wanted to give the work to their buddies. They didn’t care actually about the impact of the solution, and the decentralized nature. And we went further because we went from there, and we’ve actually got a design sketch which would take somebody maybe a matter of a week, and we’re happy to provide it if any of you join the coop or join RChain Publishing, you know, join RChain Publishing or the coop and sure we can sort out fees – they are very, very tiny anyway – but I reckon your students say probably get in for nothing. Join the coop and you can access all these resources, because there’s a design pattern for doing a refugee passport for the UN Refugee Council.
We can’t do it all. We’d love that to come from someone like this. So any of you guys could do that. And then you know, then you really you know, for the first time in my life 62 years, I know I only got 33 for the first time in my life actually think the technology that I’ve been doing all these years, actually has a chance of making a difference. And that’s a huge thing for me, at 62, I can tell you, it’s just that you guys are all young, you’ll be so much more able to make a difference than I can. So I invite you, that’s my challenge to you. Get on board and dream of what you can do to enable people to live a better life. There’s so many things we can do in the decentralized world. I’ll leave you with that thought.
Questioner:
Hi. I’m sorry, I don’t know much about the project. I mean, I’ve been checking really quickly that to github. So one of the main issues that have been relevant to all blockchains over the years is something that’s arrived when people get in basically. So today your project might have some success funds for you you get financing since 2018 it seems. So what happens if 1 billion people get on board? And what would you say to use Casper? Casper is a proof of stake. So it’s basically you vote by the tokens? If I’m correct, right?
Lucius Gregory Meredith:
I know a little bit about Casper.
Questioner:
So proof of stake have been criticized most often by Bitcoin communities because it’s basically based on who is the richer has the most power. And this is pretty hard to overcome on the proof of stake blockchains
Lucius Gregory Meredith:
So that’s not quite how our version of Casper works. Okay. Just a little little bit of history. So I worked with Vitalik And Vlad to do the first versions of Casper….
Questioner:
and the the thing is, so then, do you have tokens that have to be paid as fees, to of course, overcome the abuse and spam, Iike any other chain.
So if 1 billion people use it, how do you plan to make sure that all the users around the world, we’ll be able to keep up with the price of the token and the fees like Ethereum did when it was 10 cents? Remember, 2014 – 15 it was really cheap, right? But now, if you want to push your NFT, as we mentioned earlier, it can cost up to a couple 100 of 1000 bucks just to be published in Open Sea. So how do you make the difference on scaling your business to 1 billion people and not making sure you’re getting in the same situation of Vitalik, even if you have a huge success in the tech is great, right? Like, how are you going to overcome this?
Lucius Gregory Meredith:
So those are all very important questions, and certainly questions that we think about a lot.
First of all, because we’re linearly scalable, that means that we can roll out as much network infrastructure as necessary to match the transaction load. And we fully intend for that to be decentralized. And the the token necessary to stake those validators is spread more evenly based on our version of Casper. So that’s a technical answer. But with respect to the deeper question, this is why we support a sharded architecture.
You want to have lots of little sub networks, so that if you’re going to localized resources, you only have to go there, and you don’t have to go up the tree in order to get the consensus to work. So that has a localizing effect and a dampening effect on the rise of the of the token. And then also because the tokens in each shard, the staking tokens in each shard are independent. When you wire them into the tree, then they get there’s a relation, there’s a financial relationship between the staking token at the child shard and the staking token at the parent shard.
So what that so what that means is, you get what you have here in the world, lots of local economies with local currencies and exchanges, as you start to do transactions that take wider and wider scope. Does that make sense? Yeah, no, that’s also something that we thought a lot about. And the back of the envelope answer is very straightforward.
Steve Ross Talbot:
RChain isn’t just decentralized technology, RChain is decentralized off chain too as a co-operative. And this isn’t by accident, this is this is actually very, very, very deliberate. RChain Publishing is also a co-operative and separate. DAASL is also a co-operative and separate.
When you look at the development of shards, which has some notion of localization, may not be physical localization, it might be semantic localization. Those semantics have value to the people involved in the shard. If they’re members of that co-operative, they collectively, as part of the governance, help to set the economic model around the use of the shard. And that might include the price of the token, that might include whether the token is ever listed. It has to be that way. Because if you don’t do it that way, what you have is a group of bureaucrats who make those decisions on your behalf, rather than having a cooperative. RVote is not an accident. And RChat is not an accident. RChat is a mechanism to deliberate right that’s trusted. And RVote is a mechanism to support the rollout of many cooperatives who may wish to cooperate, based upon shards talking to shards.
Lucius Gregory Meredith:
Yes.
Steve Ross Talbot:
So it’s decentralized on chain, and decentralized off it and that decentralization off it is possibly the most important step for RChain. And I think for the rest of the blockchain community,
Lucius Gregory Meredith:
I agree. And that’s why at the top of the talk, what did I say? The thing I was most proud of was the co-operative structure. Right? And what we want to have happen is for there to be a European or French, RChain Cooperative, right? Or even even just a student RChain Cooperative, right? And the reason is because we must, from a bottoms up approach, we must begin to self organize. The consequences of what’s coming because of climate change are disastrous. You should all be very worried, you should be worried about your future, you should be worried about the future of your loved ones. And we must begin to organize.
This is not a joke. We must begin to organize. And part of the structure of the coop is to give people the tools to self organize. That’s what it’s there for.
So I think with that, I would like to close it out if that’s okay. And we can have we can have more conversations individually. Okay, so
Questioner:
I would think first Well, last last question from the top of the list. A question from Rita, when we are talking about advertisement, either centralized or decentralized, wouldn’t that require some sort of tracking in order to personalize ads to make it efficient or opening the door for internet censorship and monitoring? Well, doesn’t that go against what what free is about?
Lucius Gregory Meredith:
So we agree that surveillance advertising is definitely the wrong way to go. It’s creepy. It’s weird. You don’t want someone’s spying on you, to tell you what you want. So what we offer is a set of API’s that allow people to say what they want, as opposed to being spied upon, right. And what we what we offer is a set of API’s that allow them to participate in the resale of their data. That’s what it’s all about. So with that, I really want to thank every one of you fantastic crowd fantastic questions.
Moderator:
Thank you very much. Maybe if you want you can say hello to the camera Rob for students of ? or for like you, students, if you want to say hello to all of us award hours and thank you, bye bye