 Thank you everyone so much for joining us here at New America and the Open Technology Institute for today's event. My name is Ross Shulman. I'm a senior policy technologist and a senior counsel at New America and the Open Technology Institute. The past few years have seen a great deal of focus and conversation about the large online platforms, regulators and legislators, both here in the US and around the world have begun investigating the power of these platforms and their impacts on online competition. And personal privacy among other subjects. Those efforts will take time and there's no telling what the outcomes will be in each country and for each platform. Regardless of where all those lawsuits and public common periods end up nothing much will change if we don't also address the fundamental aspects of the business model and infrastructure that led to that concentration in the first place. Put another way, even if you think that slicing a giant platform into little pieces is the right policy, it could simply reform itself or be replaced by another just like it. There have to be alternatives that are better, that are interoperable, that give people control over their data, including making it portable, and that can provide people with the services that they want. Fortunately, that alternative exists. In some ways, it has existed since the beginning of the Internet. In other cases, it's been built over the past 10 or 15 years. In the recent past, as privacy insurances, harassment and threats to free expression online have shaken many people's trust in the dominant online platforms, it's grown into a movement. Often referred to as decentralization. It's a loose group of people developing infrastructure, building tools and studying the economics and sociology of technology that doesn't rely on a single central entity to operate. We're going to meet some of those people today, and as well as the tools that they're building. But first, I thought it would be helpful to give 30,000 foot view of what exactly decentralization is and how it's different from the type of services that most of us use on the Internet every day. In most cases, a service that you or I might interact with would be a product from a given company. Let's say it's a simple calendar service. You visit the services website, you make entries, invite some other people, make the meeting as short as possible if you're a good person, and so forth. In all these cases, you're communicating with servers elsewhere on the Internet that belong to this calendar company. That company takes the information from you stores it on their servers, sends invitations to others on your behalf, and otherwise makes the service operate as desired. The company has access to your personal information, which because we don't have comprehensive privacy laws in the US, it might use in whatever way it sees fit. All of your appointments and contacts are also tied to this company and would be at risk if the company decided to terminate the service or if it goes out of business. With this power over your information, the company can take steps to shoulder out any competitors by denying them the ability to interact with any of its users capitalizing on its own network effect. This is the centralized model and it's become ubiquitous over the few decades since the Internet gained a large scale public user base. The best alternative looks different. A decentralized calendar app would use commonly defined open protocols for representing events, inviting guests, and limiting meeting lengths. Those protocols, they're really just contracts between computers. They say that, you know, Computer XYZ shall be available at the following address to receive event invitations so long as said invitations are completed in the form described by NX2 subsection B, blah, blah, blah, blah. As long as both of the computers in question are following the same protocol, they know that they'll be able to exchange whatever information they need in order to make something happen. Additionally, a new calendar competitor can show up and as long as it follows the protocol as well can immediately begin interacting with the whole ecosystem and competing on an even playing field. The other big difference in the decentralized approach is that instead of the event creator and the invitees both talking to a third server that captures their personal information, the two or rather their devices can talk directly to one another. The protocol approach back in the early days of the Internet because almost nobody had a personal computer connected directly to the Internet and turned on all of the time. In high school, I had to use an actual modem hooked up to a telephone line that my sister couldn't pick up to call her friends when she wanted to do so. Today, on the other hand, we each carry at least one device like that in our pockets every day. It's on all the time and it's connected to the Internet 24 seven. The possibilities that this fact alone opens up are incredible and will change how each of us interacts, both with the Internet and with the other seven billion people on Earth. Decentralization can still be an amorphous concept and a bit hard to describe, however, that's why I've got some help here with me today. Seeing something in action is always more enlightening than hearing it described. So our guests today are going to demonstrate some of the decentralized tools that exist today that they work on, build on top of and have explored. First up today, we're going to have Eileen Wagner and Chris McKelvie from an organization called Simply Secure. They recently undertook an extensive survey of the decentralized product ecosystem with the intention of analyzing the user experiences of these tools and providing advice at how their different features can translate into different interface designs. In the course of their work, they interacted with the bulk of the projects working in this space, and they're going to do some scene setting for ourselves today. Secondly, we're going to turn to Paul Frazi, who is the founder of the Beaker Browser project and a leading architect of peer-to-peer web applications. Beaker Browser is a new web browser that looks just like Firefox or Chrome, but has some decentralized tools under the hood that give it some fun and interesting features that we're going to find out about. And then finally, we'll have Amandine Lapip, who is the COO and co-founder of Elements, which is a company that's building the next generation of communication platforms. The protocol underlying element is called matrix, and it combines the real-time chat nature of Slack or Microsoft's teams, for example, with the privacy of keeping your data on a server that you control, combined with end-to-end encryption, and the power to reach any person on any other server around the world, just like you can with email. I'm in the interest of full disclosure that both Amandine and I serve as guardians, also known as members of the board of directors, of the nonprofit foundation that oversees the evolution of the matrix standard, and I promise I'm not getting a kickback here. So with all of that said, I just will get us started and hand it over to Carissa and Eileen to talk about their work. Thank you so much, Ross. That was a really nice intro. We're going to cover a little bit about the context of decentralization and specifically talk about standardization today and why it's so important. So we've come to a point where the internet is essentially a public utility. The internet is required for elementary school students in public schools. Anyone with $100 can get a smartphone and get access to the largest library assembled in human history. This is a huge opportunity for society and democracy. But just like we've seen with the printing press, there are questions that were still very early in the development of the internet. Who controls the cables on which this information travels? You can get real-time information about what's happening, but who is giving you that information? Who edits the news? Who controls what is accessible and what is not accessible? The key point here is that five major companies control most of the internet traffic, our connection and interaction with history and with each other in the present, and therefore also how we collaboratively create our features and our politics. Is this how we want roads and bridges to be of the internet? So when you think about the internet, very much the first thought really is code that we have that is deeply, deeply and hard coded in the internet is the law that we all have to follow. This is famously made clear by Larry Lessig. And if we think about it, like the mechanisms in which we do social control is in fact embedded in the code, then what kind of social messaging are we sending out here? Because if we do look at the code, we find that freedom of expression, privacy and universal access to knowledge are not embedded in the web. As Carissa just pointed out, all of these issues that we're seeing right now are actually direct manifestations of how the internet as code has been written. And so from that perspective, Carissa and I have, as Ross already mentioned, talked to a group of technologists who are working exactly in this intersection of technology, code, and then law and design. So very much thinking about how do we make the internet work to serve the public interest? How do we make emerging technologies relevant to real human interactions? And how do we strengthen democracy around the world by thinking about access to a free, open and equitable internet? And it is in that kind of line of work that we met a group of technologists working in decentralization. The way to think about it is sort of a mixture between the people who are hacking networks and systems in their garages, but also people who really want to radically reimagine how the internet should function and how it should work for people. And in that intersection, we're thinking about, well, what kind of values do we want to encode? What kind of belief systems and ideologies do we want to embed in the code that runs the internet? And of course that in part is a complete technical question, right? We have standards organizations such as the ITF, the ICANN, the IEEE, W3C. There are all sorts of technical entities working at multiple layers in the entire stack and across different nations to think about how the internet should work. But that's not the same as asking how should the internet work from a values perspective? How should the internet work when we are thinking about transparency, choice, privacy and ownership? And those are things that are standards that have to be set by democratic institutions. And that's not something that the standards organizations such as the ITF can decide on. And so today I think what we want to do is kind of give you a couple of ideas around how these values could be manifested in such a network. And especially you want to focus on consumer choice and agency as two values that I think are not being talked enough about. So imagine you are a small business owner and you want to advertise your business on the internet. And you have a variety of social networks to choose from. However, you find yourself quickly that you actually have to sign on to each and every single one of them to be able to be, you know, discoverable on those networks. Why is that? Well, if I am signing on to Facebook, Facebook posts can only be read on Facebook, not anywhere else. Facebook posts can also only be written on Facebook. In fact, you have to be a Facebook user to read or write most of the Facebook posts that you can find. So really, there's that kind of vendor lock-in that makes it impossible for me to post something on Facebook and then automatically share it in the other networks. And this sort of network effect, people often talk about it as some sort of, oh, well, it just happens to be that if most of the people joined this network, then people are staying here. Well, it's also weaponized by platforms because once you have users locked into your platform and you want to keep them there, that's good for your business, of course. So our status quo is that we have a variety, a few centralized platforms controlling most of the data and the information on the internet. And then as a small business owner, I'm just wondering, well, is there anything I can use that I can have as a single entity that I have full control over that I can communicate with others with? Well, the good news is there's still email. There's still email. Email is not dead. In fact, it's growing. We have over 4 billion email users worldwide. 306 billion emails are sent per day. And yeah, it's a growing network. And that's a single thing that business owners can have and advertise and say, well, this is how you can reach me. What's so great about email? Well, first of all, email makes it possible for users from one network to contact users from another network. We can communicate. There's some sort of interoperability built in there. It's also possible for me as a consumer to choose between different email providers. And in fact, that kind of choice versus competition for better design and user experience. People choose Gmail because it's very convenient, not because it has a great logo. So email is a really great example of a decentralized technology, which is really what we're here to talk about today. And Ross already gave a wonderful introduction. So I'm not going to add much more to that. But I think I would be amiss if I talked about decentralization without showing you the really infamous network graph here. Centralization means one node. Many people connect to it a single point of failure. Decentralization means many, many centers. People can connect to it and choose between different nodes. So what we're saying here is decentralization is a network architecture that avoids reliance on a single party. And that's interesting, both from a political insecurity point of view, but also from an economic point of view. In the case of technology, you can think about centralization as Facebook's model. So Facebook has control over a single entity, a server or rather a server network that hosts all of your data and does all the services for you. If you want to talk to your friend, you have to install Facebook, your friend has to install Facebook, and you can both talk to the Facebook central entity. What email does differently is you can choose your own provider. So let's say I'm choosing yahoo.com and Carissa is choosing gmail.com. We each sign on to our own providers. We only host data with our own providers. But somehow magically, those providers can talk to each other based on the email protocol. And that central idea is super important when we think about not just email communication, but all kinds of other domains that we operate in these days. So since we're a design shop here, let me just think out loud and show you one possible way this could look like. Let's say I sign on to a service called Network Buddy. Network Buddy aggregates all social media posts from all kinds of different networks. So I have my friend, Doris Day, who is posting something on Facebook. I can see that. But then, you know, right beneath that, I see Oscar Peterson has posted some photos on Instagram beneath that. Hey, Doris Day is also signing on to this fictitious network called deja vu. Very cool. And then I see Twitter, Benny Goodman is up to no good again. Right. So there's a, there's a way for me to think about my social network independent from the actual platforms. And that's kind of a central idea here. And what's great about this way of thinking about data and network is also, it's pretty agnostic as to, you know, how you display and how you want to call it that information. Because really, what I can also do is I can have a second vendor called tiles tiles is showing the same kind of data, but it's showing the data in a much more, you know, sleeker kind of more polished way I kind of like the design of tiles it's, you know, the same people I can see the same posts However, oh man, beautiful tiles. So I'm going to pay $5 a month just to tiles, so that tiles can collate that information for me in a much better way. That's wonderful. And that's only made possible because we have a standard for social networks. Now, that standard doesn't only have to appear in social networks, it can appear in other places. In fact, we've seen some of that already with right sharing. So right sharing. In this case is, you know, displayed as Uber and lift, and it's on Google Google maps, which is kind of cool I can choose between Uber and left, but that also depends on Uber having an agreement with Google Maps and lift having an agreement with Google Maps. It doesn't quite is, you know, achieving the same kinds of openness that we want. I wonder if there is a really, really open protocol that I can use to, you know, do my right sharing. Well, that idea isn't new. This is a RSS RSS is a protocol and open protocol that has been proposed and widely used for social posts from podcasts to blog posts and whatnot, and is still kind of sort of used for various services. It's also not new for, you know, let's say if you ever had an ICQ account that's also decentralized technologies. And nowadays we have even more and bigger things exciting things such as the matrix protocol that we're going to talk about in a moment. But basically decentralization is not dead. But it could use some improvements. That was great. Why isn't everything decentralized, you might be asking as a viewer. Well, we'll talk about two different things. I'm sure there's lots of reasons why decentralized technologies have a hard time getting adoption. But today I'm going to talk about two things that maybe you in the audience could do something about. So the first is data ownership. So why is it so hard to compete? Well, as Eileen was talking about, you can't easily move from one provider to another. And in terms of data ownership, that means you don't actually own the data that you create on Facebook, legally Facebook owns that data. So case in point, email versus Facebook, Instagram, or WhatsApp. If I write an email on a server that is my email server, then I own all my emails. Whereas on Facebook, if I post the post, it is not mine. This also extends to contact lists and this kind of thing, which further locks people into a service for a long time as you grow your contact list. And it is not owned by you. How are you supposed to export it? This has become better in recent years, while these platforms have been forced to provide export capabilities, but it still isn't great. And another reason why this isn't great is the legal aspect. So even if you export stuff, how do you actually get that into another service? Well, there's something called the Computer Fraud and Abuse Act, which you might have heard of. It's 1986 federal law and it punishes you if you violate the terms of service. Any website can put anything they want in their terms of service. So it's a pretty broad stroke. It enables widespread cease and desist across the internet of competitors. There's a case you can look up if you're a lawyer interested in this MDY industries versus Blizzard, which sets a precedent around scraping websites and using that as a competitor. But there's another precedent as well. It was set last year, which is HIQ versus LinkedIn. And this set that automatic scraping of publicly accessible data does not violate the CFAA. And the judge explicitly said they wanted to prevent information monopolies that harm the public interest. And they wanted to maintain that users retain ownership over their profiles. So this is a huge ruling. It obviously is still in the Ninth Circuit, so it still, you know, could be appealed. I think there's a lot more to do in the legal side. I'm not a lawyer, but I think if you're interested in this and you are a lawyer, please help. Next slide. Standardization. So this is more of a technical thing, but I think it also has some legal implications. There's monopolistic behavior, and we've been talking about this all year. Platforms exclude competitors, and they prevent entry to competitors. They have really tight vertical integration from the device all the way up to the apps. Case in point, app stores. They're kind of like railroads. Think of Apple. They own the device, they own the app store, and they have their own apps which compete with you on the app store. So if you're a small business owner trying to create an app and put it on the app store, you have to compete with Apple to put it on the app store. And they're the ones that tell you if or if or if not, you can actually have that app on the app store. And then if you finally get it on the app store, they take 30% of everything. So this is a very, very difficult situation for folks. I think it has worked so far because this technology is so new. We're still in a very open and new age of technology. The internet has only been around for a few decades. So I think we do need to see some regulation step in to make it easier for folks to enter the market. Yeah, so imagine a world where, and I think Ross talked about this, imagine a world where you could only call people on Verizon if you had a Verizon phone. It just doesn't make sense. So we really need to think about this more in terms of standardizing all of the data that flows around the internet on different platforms. So what's the path forward? Talked a little bit about trains. Trains are really amazing opportunity to look into the most recent history of how very fast moving technology was regulated and standardized. And we still have this all over the world. We don't have standards for web. And we can start in the US. I think it's a great thing that folks are looking at this in the US because this is where these platforms are headquartered. This is where a lot of this technology came from. So I think it's really, really important for us to do something about it. We're really interested in this, you know, intersection between technology, law and design, and we're going to be working on this intersection for the next year or so. Our work is titled decentralization off the shelf, and we're standardizing design patterns for decentralization and creating resources for policymakers and users and technologists. Thanks so much. I think we can take questions now and you can find us on decent patterns that XYZ can sign up for our mailing list and join our matrix channel. Awesome. Thank you so much, Eileen. And Chris, that was very interesting as a reminder to anyone who does have questions. Please feel free to drop those in the Q&A section. That's a little button at the bottom of the zoom window. And if you have any of those, we will put them over to our panelists. Also, if you are interested in engaging in the conversation, check out the hashtag decentralized tech on Twitter where we're all following and can take questions from there as well if people would rather do that. On that note, I'm going to get to get us started with some some demos and Paul, if we can turn to you to talk about Beaker browser. Yeah, let's do it. Thank you all for those awesome talks you're invited from now on a time after present. You can start things off. So, because you did that about as well as I could ask. So, before I get started, I want to give a small introduction to some of the technology that I'm going to be demonstrating so that you have a little bit of context about what you're seeing. So taking it from the top my name is Paul Frazee I work on an application called the Beaker browser, which is a peer to peer web browser. Our goal with Beaker is to commoditize Facebook and Twitter. We want to say that publishing and communication are too important to be features within a couple of applications. Instead, we want publishing and communication to simply be part of infrastructure that many applications can use as part of an open market. So let's talk about what it's going to require to build that kind of infrastructure. Historically, as everybody has mentioned before, internet applications have been using a server based model which we often refer to as the cloud. So in this model when somebody wants to create an application they'll create a server and all users will connect through that server. This means that the software and the data are stored on that server and all interactions between users are intermediated through the server on their behalf. In Beaker browser we have effectively inverted this, this model, rather than connecting through a server beaker users connect directly to each other and what we call peer to peer networking. This means that the software and the data are stored on the users devices interactions occur directly between the users. The effect of this design is that it greatly simplifies hosting, whereas setting up a server is a highly technical process in the beaker browser any device can begin hosting a website at the press of a button. You could think of the peer to peer networking as a way to democratize online publishing, similar to how desktop printers democratize printing on paper. Rather than having to pay for expensive servers anyone with a device will be able to self publish effectively every device is becoming its own cloud. The important feature of this whole system is that users retain control of their data on their device as a result, there's social graph and their personal information is available for new applications to leverage it's effectively open by default. Let's dive into a demo here right to my screen. Okay, so what we're looking at right here is the beaker browser 1.0 release candidate this is not some of the stuff we're going to be seeing here is not actually released yet we're basically two, three weeks away from putting this out, which is we're all pretty excited about. You'll see it looks a lot like a typical browser but there are some interesting features that come along with it that you wouldn't normally get. I'm going to start with some of our social applications. This right here is beaker social. You'll notice this looks almost exactly like any social media application that you're familiar with I have a nice little timeline of social media posts people posting pictures and things like that. We have comments and conversations here we're talking about this premise of social networking and how it's similar to the indie web in this model. Here we're talking about debating whether or not we should call it subscribing or following. Like any social network we have notifications, as well as search. The key to all of us to remember is that this is working entirely in a peer to peer fashion. The software that I'm running on my page right now is running entirely on my device. And when I'm connecting to people I'm actually connecting to their devices to get their posts and their images and everything else. So there is no server involved in here. We have a couple of other applications. We have a link, which is our sort of credit clone. It's actually what you're sharing or your bookmarks so as a sort of fun sort of feature you can bookmark something and market public and it'll just show up in this feed right here. And you can have your comments and your outputs and your downloads as you would expect for a Reddit like experience. Finally we have a blog reader, which if anybody here is familiar with RSS this should look pretty familiar. One of the nice things about this is that it actually has social features, unlike RSS historically so you can do things like commenting and uploading on the blog post. So we're able to have that open system like people always want with RSS but without having the kind of downsides of the limitations of RSS. And included in that is also site discovery which is using the social graph to suggest people that you might want to follow and see their blog post as well. So those are some of the high level applications and I thought it was important to start there to demonstrate what the system is capable of doing. But I mentioned before that this trivializes creating new websites so why don't I demonstrate that. So I'm going to create a new website using beaker. Now normally if you wanted to make a website on the web you would go through a pretty long process you'd have to rent some server space from one of the many providers like AWS or Google Cloud. You'd set up a Linux instance install some software configure DNS, even a pretty technically sophisticated user is going to spend 30 minutes to an hour getting that done and if you're not technically sophisticated you're probably not going to do it. In beaker I open up that menu and press new site and I name it my cool new website. And just like that I've been dumped into an editor to begin working on the site. There's a default rendering when there's no HTML but I'm going to go ahead and create some HTML zoom in a little bit on that. And there you have it my HTML is getting refreshed on the page right away. So we have the editing experience, which you know this is somewhat technical if you like to do a HTML programming. This is for you if not you don't actually have to make websites but it's important to have this, because this is why you're able to build new applications and things with basically no overhead, no, no cost and if you're, especially somebody that's new to building websites. It's a really nice tool to not have to deal with any of the servers to not have to do with the command line you just press buttons and you're right into your HTML. If I were to share this link up here, you can see it's a pretty long link but it's kind of like Google Docs one of those share links and if I were to share that with somebody their computer would connect directly to my computer to access this website. That's how the peer to peer network operates. So website development is nice. It's important for publishing, but the probably my personal interest comes in whenever you're starting to get the applications development and being able to build applications which leverage each other's data. Right. It's an open system it's an open network we sometimes call an on the walled garden. So let's demonstrate that I've got a code snippet that I'm going to paste in here. And what I want to demonstrate is that it is incredibly easy to start building social applications in the system. So this code snippet is 40 lines long all told it's not going to be a very beautiful application, but it is going to be an interesting application it's going to go through my social graph everybody that I follow. And it's going to get the latest post that they've made on their feed and list them all out. So I'm going to save and it'll refresh. The next thing it's going to do is ask me to log in right this is important. Any kind of platform it's going to have to ask for permission to get this kind of information so here it's asking me. Okay, do you want to sign in as Paul phrasey yes I do. You want to let it read all of your public profile data. That's fine with me so I'll hit okay. And just like that the application runs. And so now here we're looking at there's Tim Caswell's latest post turns out also his first post hello world. So here I am kale playing with some markdown. That's basically rich text in the post so you can do things like balding and headers. Here's top Robbins testing out the 1.0 RC. Hopefully this is all not too hard to see I forgot to zoom in for you. So again, the, the overhead of building this application is incredibly small compared to basically any other platform that you would potentially build social applications on. If you were to build something on Twitter you'd have to sign up for an API key, you'd have to be dealing with various permissions in terms of what you're allowed to access at any given moment. That's not the case in this system. The only people you have to ask is the user, if you can have permission to access their data and then from there you're ready to go. So let's see. I've got how much time do I have left Ross. I, you know the other five or five to seven minutes. Again, I suppose just for that was the parts that I definitely wanted to hit. But for the rest of the demo I suppose it might be interesting to show a little bit about how the data works. We focus really really hard on making sure that the data is visible to users so that it's somewhat easy to understand how the system is working. Traditionally in a platform like this you might have some kind of a complicated database model which would not be easy to explore you have to be technical to see your data. We decided to focus really heavily on just using files. And the goal of just using files is that people understand files they understand how to explore files. And so it gives us sort of an intuitive way to be working with your personal information. So we're looking at my personal website here, which you can argue is like my profile it's where I do all my posting and commenting and things like that. And if I scroll down past my intro there you can see all my recent posts here's a comment I made here's a post I made asking about the subscriber follow term. You can see my blog posts over here and who I subscribe to over here. This is a nice rendering of all my information but let's take a look at the actual data so I'm going to open up a little sidebar over here for the files explorer. And now we're looking at my personal information and the form of files. It works just like websites traditionally do you just have a set of folders and files. But if I wanted to find all of my let's say blog posts, I'm going to go into the blog folder. And here they are just a collection of markdown files. And so the most recent posts I made with social media and game rule collapse, which is that file right there. I'll double click on it. And so now I'm viewing it it's being shown in the files explorer over here and actually on my website over here. And I can just pop open the editor to start working on it if I were so inclined. And there's a little bit of metadata that's put on the different files. In the case of a blog post you'll just have a title that makes it nice and queryable from the outside. Let's look at some of my other files. We have let's see subscriptions. And these are all the special kind of file we have called a dot go to a dot go to file is kind of like a link. So if you're thinking in HTML you have that a tag for linking the things will dot go to is like the file equivalent of an a tag. And it's basically a way to just point it URLs through the file system. And so anybody that I subscribe to go to file gets quit in my subscriptions folder. And that indicates that I'm following them. It's a very simple system if I deleted one of those dot go to files I would stop following them by adding new one I'd be following somebody new. It's simply by adding and removing and editing these files that we're able to build all these applications. One last thing I'll show is the comments. These names are particularly interesting to look at, but they provide basically all the information you need for our commenting system so I'll double click on one at random here and here I am commenting and you can see on the right side. Here's the content of the comment being rendered on my website. And there's the actual file. And this system while it is fairly simple is capable of handling everything that we need for these social applications commenting voting posting everything that we do is handled simply through files. And again we think that understandability is a really key factor and empowering users to have control over their system. This is what makes it possible for non technical people to feel empowered in a open system like we have designed here. And that's just one other thing because I think it's fun. This is a terminal program that we have, which I'm working on right now so we don't know how strong it is going to be but it's just kind of fun to be able to browse around in your websites using a terminal. So for the nerds in the crowd, you may like that. Nice. Thanks. So I think that's a wrap on that. That's great that's awesome I love, I love this Paul I'm looking forward to the to the 1.0 release of beaker so I can play around with all this, all this fun stuff. I think you had a couple questions during your talk Paul and I think one that I want to put to you and then I think another that I'd love to put to Eileen and Chris excuse me. And so someone named Roro in the zoom comments asked whether it would be possible to use existing social media as a way to store data with with keys that you control locally. You can overlay privacy and security on top of existing infrastructure and, and I think that sounds a little bit like what you're actually building in beaker except obviously peer to peer. I'm wondering if you have any thoughts on that. Yeah, it sounds a little bit like what Roro is describing is things like publishing on Twitter encrypted posts, where you're sharing the keys through the peer to peer network and then anybody that has the keys would be able to decrypt the posts perhaps things like that. It's been a little bit hesitant to sort of attach on to the existing social network for some of the reasons that Corissa mentioned there's a little bit of legal hazard but still getting sorted out and we really don't want to fuzz with that. It's also kind of a pain they intentionally engineered their application so that they're hard to mess with from the outside which you know right I understand. The one thing that we've done that's kind of close to that is we've introduced this idea of a comments pain so that if I'm looking at jump on my Twitter here and hopefully nothing too embarrassing. And if I click on the comments pain we actually have our own in network comments that we can have on any page on the internet. So if you wanted to have discussion that's not on Twitter but about a tweet you could do that through the network. And that's really as far as we've gone with this we haven't gotten into any kind of extracting data out of the existing services or leveraging them. Because it's a little bit of a legal murky area and I think we can do a pretty good job just building entirely new applications on the peer peer network. Great. One more for you Paul actually that just came in Emma Herman asks, will content be moderated how does moderation work in in the system that you have developed and then also what about advertising. Okay, so the primary way that we think about this is that our first goal is to unbundle all the functions of social media. So, this means the interface the algorithms the moderation the data storage all of these pieces need to be separated out so that they can be basically automated upon by the market. That means moderation is really a combination of algorithm design as well as actual sort of people work, you know sitting there and actually looking at things and marking them as appropriate or not. And are in what we would like to do with that is to open that up to basically an open market solution where we say we're going to have a couple of nice tools baked in for you to be able to do things like block lists. But what we would like to see is that there are people. I mean, in my ideal world there would be companies similar to news organizations that do this sort of moderation. I don't know if that's feasible but it seems like having a sort of an open market around moderation and creating spaces of content which people find to be appropriate might be an interesting model to go with. We'll see about that. Nonetheless, the point is that the system remains flexible on that design and people can find what works for them. We do not have any top down control over the moderation. Regarding advertising we do not currently have anything baked into our software that would provide advertising. I think there's an interesting discussion to be had about whether or not there should be ways for content creators to publish ads. At the end of the day, this is an open system if there is a client that you use like a Twitter clone that you're using that has advertisements being placed in it and you don't like that you can actually open up the source and fork it and change and pull out those advertisements so to a degree. This is going to be kind of a dialogue between the users and the publishers as to whether or not they're going to be willing to tolerate ads. Certainly I think there's a positive story to be had around a form of advertising. It's not tracking based but it's instead channel based, which is sort of like what Duck Duck does with their advertising trying to create a more ethical form of advertising. Again, whether or not users would tolerate even that is sort of an interesting question, but that's the situation. I'll just enter on that. There are some services that are experimenting similar to how Eileen was showing with Google Maps where they have Lyft and Uber but they have partnerships. There's some news organizations that are partnering with like RSS like readers and people pay $3 and to remove all the ads on all of that. So by having decentralization and having this you open up the possibility that a company could be created on it or people could choose to just use ads. But at least right now it's really hard to find that choice. Ross, you're muted Ross. Thanks. Thanks, Paul. And thank you, Chris. I have one more question from Brian Alexander who posted this on the YouTube. And this one, you know, for Eileen perhaps or or Chris. I noticed people have a hard time rocking mastodon that is understanding us around. It's still very technical. How do we get past that hurdle. I think that's sort of like question number one for for the work that you all are doing. Yeah, indeed. I mean this is the sort of thing we've been grappling for some time now, grappling with this is very much a question around how do you introduce new technologies to people. You know, a mastodon is of course one of the federated social networks that is very popular in decentralization. I think there's a, there's the biggest hurdle is that you have to introduce a lot of these concepts before you can even put any kind of interface in front of people. I mean what we have done today here is, you know, talk a little lot about like how networks work how platforms work, but really the, I think the biggest question for me is, do you when you design something like that, do you adhere to known and familiar interfaces do you really try to make user experience as intuitive and fluid as you can, or do you want to make it something completely new and exciting. You know, I have beaker browser I am so excited it's so new and exciting and you have never seen anything kind of like that before and you really want to jump in. But also the question is, you know, what I have to explain like how a website is built. Is that the right way to introduce a new service to people. I'm not saying there's a there's one solution to the problem but I think there's definitely a lot of work to be done also from this community in making sure that things are accessible and easy to use, and we meet people where they are. Thank you Eileen and thank you so much for the work that you and Chris are doing. It's, I think, actually one of the most important pieces of the decentralization puzzle is indeed the user interface and the user experience. So, with that, why don't we turn it over to Amundine to talk a little bit about matrix and element. And yeah, I, yeah, I agree with Eileen it. There is a big, a big work of education to be done here because it's a new project we're trying to introduce to people so that's part of the big interesting work everyone is doing so it's really interesting. So, yeah, thank you, Ross, for your intro. Just going to show my screen quickly. All right. Okay, so there's been good introduction so far on what we're trying to do what decentralization is open standards, etc. So, I'm going to just quickly summarize again with different words on all this. So we're, we're clear on what we talk about. On our side, as Ross say, I'm the founder of the matrix foundation where I'm working with Ross, and also CEO and founder of Elements, a company which is hiring the coaching of building who is building matrix. And basically building apps on top of matrix and trying to promote the ecosystem. So our focus is very much on real time communication today. Yes. So as we said today you have no choice as a user to trust big companies with your data if you want to talk to someone you have to install the same app that this person is using. You cannot open up to others unless you're actually using the same provider. And it's, it's a massive, massive trust you're putting into these companies with all your conversation or your personal data. The good news is that email had the same problem 30 years ago, and it's been solved. It's been solved because a pragmatic open standard was formalized, and it did gain new instruction because people don't want to be trapped in these silos when there is a better open alternative. So today we have very big networks when you look at the likes of WhatsApp or Facebook, they're huge. So it feel like a network but it's still very closed. So the next step is very much to provide an alternative which is as useful as big as these, but actually with the value add of data ownership and control and build of an open ecosystem. So just to recap a bit in terms of open standards. What does it bring the data ownership as a user I can choose who I trust with my data. If I'm not happy with this provider, I can move to another provider have data portability, we're not locked with one vendor from an enterprise point of view as well. There is a healthy competitive marketplace. If one app is better than the other, it will get more traction so it encourage every company to build the best app they can, rather than just take all the users into one big network with potentially not very, let's say, addictive, addictive usability and this kind of thing. It also drives innovation rather than monopolies. And of course, like Carissa and Eileen were saying it drives the consumer choice and their agency, you can fit the app which actually fit your purpose, your expertise, and yet you can still talk to your, to your the rest of your network. The best example I have is talking to my grandmother. She likes a very simple app, she can barely use a tablet while I'm kind of for user here so I'd like to be able to tweak all the buttons on my app and configure it as much as possible. So back to Matrix itself, Matrix is an open network for secure decentralized and real-time communication. While today, one of the biggest usage it's used for is interoperable chat and voice over IP, it can be used for any type of data, including any type of social media data or virtual reality, IoT, it's just one big fabric where you can put anything you want and actually control where this data is stored. There's another way to put it, if you want to simplify, it's an open standard for open secure decentralized alternatives to WhatsApp and Slack and etc. So how it looks like today, lots of stuff on this slide, but it's a quick overview of the ecosystem, more than 20 million addressable users, more than 50,000 deployments, hundreds of projects built on top of it, French governments, the Armed Forces in Germany, schools in Germany as well, and projects like Wikimedia, Mozilla, KDE actually using it to communicate. And Matrix, as the open standard, is governed by the non-for-profit Matrix Author Foundation. That's the important part, we're building an open ecosystem, so this standard needs to be neutral and not be protected from any for-profit company, including Elements. So how it looks like. So this one is an example of a deployment that could happen in government, for example, because as you could see, governments really like the idea of actually being able to deploy their own communication services internally with each ministry controlling their own deployment. And it means each of them has the control, they can apply the security level they want. The Ministry of Culture, for example, won't have the same anti-virus as the Ministry of Army, but they can still talk to one another. And that's for government, but it can be anyone, it can be companies, it can be individuals, it can be associations. The other interesting bit about Matrix is that it has very simple APIs, which means you can build bridges to other applications. So it's not like we're trying to replace everything that exists, but Matrix is trying to bring together all the existing networks, have bridges to Slack, RocketChat or WhatsApp, and we just want a way for people to be able to communicate with one another, even if some apps are very specialized into one domain, like collaboration for Slack, for example. It's not because I use Slack that I can talk to someone who is doing something like just a one-to-one messenger discussion is not necessarily collaborating on the desktop. So we see Matrix a bit as the communication layer of the web. So in terms of how it can look like in practice, once you have this big network on top of which you can have several applications developed and running used by a whole bunch of different people, I can show you what ours look like. It's the most many and I'll show you later on others, but for example, looking at Elements, it's an alternative to Slack or WhatsApp, except it's given the, it's built on top of Matrix, it means it's decentralized. I choose the server where I want to use and register my account and to an encrypted by default, and it can interoperate with any other Matrix application out there. So features that go through the app itself, but as you can see, it's group group discussion. So for example, this is a public chat room, it have more than 6000 people into it. And looking into these people. Well, I'm on the Matrix.org server and so is Matthew here, but looking at Travis, he's using his own server. And we have Eric as well using his own server, gki.re, et cetera, et cetera, et cetera. So now more than I think 500 servers into this very specific chat room, which is also bridge to IRC for those who know the big communication network that IRC is. So in terms of feature, well, you see chat can have URL previews, you can also see who have actually read the messages, and I can react to a message if I want to. If I go to a different room, I can also have widget actually embedded straight into my chat room. So trading view Bitcoin versus USD. Well, I could set up a video conferencing going in the corner. I'm not sure how it interacts with my zoom ongoing so I may not go into that direction and sending of images all along. And same thing you can explore your chat room and look at the various images that were sent. In there. And actually be able to share the room with others. So the interesting bit is that this app here is my own window into the Matrix network. So I have a whole bunch of chat rooms, thousands and thousands of chat room, thousands of thousands of people discussions. And you see you can configure whether you want to see the messages from the chat room or not, but I can filter down. So if I look at this specific community, which has disappeared, we go. It's filtered down to one, a smaller view of the entire matrix ecosystem here. So, another interesting view is looking at the, this one, which is the decentralized web, submit chat room, a lot of people, including Wendy here are actually bridge prompt slack. So, if I look at this one. I can filter by one day because I know she is we go. You can see that her ID is very special and she's actually chatting into this room directly from chat from slack on her side, and it's completely transparent for her, whoever is replying from matrix or whoever is reply directly from chat. So, that's a quick overview of the of the chat here, I wanted to show you encryption. The private chat rooms are encrypted, and then you can also verify the various, the various people who are who you are talking to to make sure you're actually talking to the right person, thanks to small things like for example, emoji, you compare emoji to make sure that the person you are talking to is actually the right person, and you can verify the device level, making sure that this new phone which is turning up on your friends account is actually has been actually been connected by this friend. So, element is a good example of how basically how an app which is built on the decentralized open standard can be as simple as any other centralized app and bringing all the new features. Some people may think that hey, this decentralization is limited to a very geeky public but no we're now in a state where anyone could use the centralized application and the user experience is made for people who are used to the slack and the use of this work. So that was for elements which is one, as I was saying, of the matrix applications, and as you can see there are tens of others. So these are the most advanced one, and there are a lot of, I think they're about almost 100 other matrix applications down out there at the various stage of usability but these here are usable are at least in beta, often in production already, and they have been developed by the community by different parties in the ecosystem. So the app itself is one side. And then, as we were saying, if you're using a decentralized application, you have to choose the provider you trust with your data. And that's where Austin provider comes in. And at Element, we do have our own version, which is element matrix services, which also provides the app store where you can pick up the widgets you actually install on the, on the within within your chat room. But once if you don't want to trust us with your own data, well, you have a choice to potentially go find someone else and there are other matrix hosting providers out there, or just actually run it yourself. As all the governments I was mentioning earlier, basically you can get your own equivalent to slack with complete control over where the data is stored, it can be on premise on your own hardware, it can be on the public secure cloud, and it can be still connected to a wider network without limits, like you may. It's not because I'm working in a government or in a company that I don't want to communicate with someone outside. And why do we have these limits with chat. So that's what we're trying to open with matrix. And in terms of what comes next. Today, as you could see earlier, the, we're using servers to store the data that we're working on peer to peer. And we do have the beta version of peer to peer elements should be coming in a few days. Fingers crossed. And the next one is also decentralized reputation. We were talking about moderation earlier, and what we're trying to build is very much a world where as a user, as part of an open network like matrix. I can say, I can create the things I want to see I can create the people I want to talk to. We want to build something where you can see the entire world and dial up and down the level of things you believe in. Like kittens, you want to dialogue the people talking about kittens. If you like, if you don't like dogs, then you want to die down. People talking about dogs. You can ask, like Carissa was saying, as a potentially, how do you say, not a not a news media, a media company you may want to create your own reputation list and say hey, I do align with this sort of thinking, and you can sell this. This list to users so that they can curate their own world based on the publication or the celebrities, they're lying with. So that's very much the kind of thing we're working on to help this navigation into the big open network. So, basically, as a conclusion. I strongly believe that open standards are solving most of the antitrust issues that we can see these days. And it's not a matter of forcing what's up and Microsoft to change their current protocols implement everything on matrix is hey guys. If you're unhappy about Microsoft dragging all their office users into teams. Why don't you all bridge to something open like matrix for example, then the best app will win. And the decentralized alternatives are bringing both the usability and the data ownership to the users. So that's very much what we're trying to build her. I'm not sure there were other ideas or questions. Thank you so much I'm in Dean that's that was wonderful. Did get a question or two well, while you were chatting. I think one of the interesting ones that we've sort of talked around a little bit, but but we'd love to talk to here from everyone on this question is from Brian Alexander who left it on YouTube said. The reason that the centralized giant succeed is because of their immense user bases. How do we convince those legions to switch to decentralized projects which have, you know, smaller crowds right now and this is this is sort of the network effects question is, you know, if everybody is is on, you know what's up or something like that how do we, how do we build the network on the decentralized tools. Open that for everyone. We have two ways of addressing this on our side. First, when we built matrix, one of the key thing was have very simple API's so that anyone should was able to actually build chat apps on top of it, because we believe that one of the reason it's not the only one. And the reason why there are so many silos of chat is that everyone thinks chat is simple. Yeah, it's just sending a message and receiving a message. Actually, if you want group if we want end to an encryption if you want history, that starts to be a bit more complicated. And that's when you start getting into your own rabbit hole. And the incentivated incentivation of actually monetizing all of this. So matrix very simple API, no need to be a chat expert to build messaging apps and hey, you create your app, the way you want it and it can communicate with the rest of the network out there. And so one of the ideas actually get all the long take as many people as possible building on top of the ecosystem, and then you raise the level of the water, and then you have the big giants, the big, which are out there and yes they're very, very big, but at some point users may realize that if they want this app to actually talk to customer service or their estate agent, or any sort of use case, they don't have to change up anymore they have one on their phone and they can talk to the their insurance their estate agent and anyone say, why is what's not linked to that. And maybe one day that will help. And one way of doing it is by building apps which user interface wise are super easy, and then attract them by the fact that the app is fun the app is pretty the app is easy to use. And then they start dragging the attracting the network and dragging their friends into it for what the app itself is, and oh they get privacy by default. That's how we can, we can grow the network as well. Yeah, I would just echo what amandine said there. It's if we want to win in the marketplace we just need to make good products is kind of really what it comes down to and what you're describing is exactly that. There's could be things like a government action changes the nature of the market, due to regulation could be something like a big event occurs that freaks everybody out about privacy and causes a migration those things could happen but generally speaking I think the best thing for each of these projects to do is just put out something that's actually really good and expect the growth period to take some time. And eventually, like Mending described the markets expectations will start to shift so that if the features that are inside of these different sort of decentralized systems are not present. People will start to be like well that's a little odd. And that's how you end up changing consumer tastes. That's really true we do need to create good products and that's what Eileen and I are really interested in focusing on over the next year. I think our long term worry, and I think everyone should be worried is the way that these platforms buy up competitors. So this is another thing we have to consider at least from the legal side. You're talking about the long game of having a good product and hopefully over time, people will notice it's a good product and move that requires a lot of capital and a lot of money for you guys to just continue working on a product without any users. And that's difficult so I think a lot of you know at least this has happened before a lot of apps get bought may they give up and they get bought by these companies. So what we have to think about is how we can make it easier for people to compete and how can we get easier for people to move from Web 2.0 to Web 3.0 as some people call it but this bridge doesn't really exist right now because if you were to create a bridge to WhatsApp they'd probably sue you. Yeah, you have to. Sorry, just to answer on the WhatsApp thing. Users could run them run it themselves where some exist, we do not run it to people for the users because as you say, legally, it doesn't work. And that's also why building on top of this big open standard means that anyone building on top of it benefits from the network the small apps don't have to actually gather the network and you benefit from what is out there. And there are some people out there who actually can benefit like a looking for the privacy like the governments etc they're looking for the privacy and they're attracting their users they can stop to communicate with their citizens etc and this big brings not sorry increases the network as well. And I agree like if you have a product that serves all of these different interests it could serve both, you know, users civil civil society users and governments and corporations. I do think though that I'm calling in from Berlin, you know there is a large community of like open source vegans who would never use anything that's you know surveillance capitalism and whatnot. But that number has a has a limit like it's not it's not the majority of the users out there and at some point we're going to reach the threshold of people who are willing to, you know, boycott Facebook. So I think what what Paul has said about we just need to build better products that is definitely like a better approach. But I also think that, ultimately, it is a it is a legal question, because I don't expect that we can build better products and tech giants wouldn't fight back. There's definitely there's going to be resistance and the best protection we can get is legal protection to say, you know, people users legally own the data they post on these platforms. And if you want to port them or if you want to build a client on top of them and bridge between all of these different services. You know that's legal and that's okay and that's ultimately the way you move from, you know, silo platforms into into an interoperable web. So, yeah, I think it's as much a legal question as anything else. Absolutely. And if people are interested in asking further questions, hit up that Q&A button at the bottom of the zoom screen if you are on zoom or feel free to leave a comment on the YouTube stream. We are checking that as well or tweet at us using the decentralized tech hashtag. I do have one more question that came in for folks and this one's a little bit sort of more blue skyish but I'm curious what people think and insist really for for any of y'all. Chris McCrae on the zoom call here asks, how does decentralization connect with the future of he says AI for good and I would even broaden that just to AI. More generally is, you know, one of the great sort of concerns that a lot of people have is the nature of artificial intelligence is that it sort of, it needs a whole ton of data as an input in order to develop the sort of connections that it develops. What does decentralization mean for AI and or vice versa. I think MND and you talked a little bit about, you know, dialing up kitties and down dogs. I think that there's a little, there's a lot of trust in AI that you have to have any user, like as a user. So if I'm using Twitter, I have to trust Twitter's AI platform. I think, you know, AI for good is, at least as far as I understand it's kind of trying to figure out how to tweak that and use that algorithm in a way that you trust it, maybe for different domains. I think that this is possible with decentralization. I do think it requires, if it's going to require a server, you have to trust the people that are that are running that you're sort of delegating that AI to someone else. And that that should also be possible. Why do I why can't I use the Eileen filter on my Twitter. Why do I have to use the Twitter filter, for example, the Twitter recommendation algorithm. So I think that there's, you know, a lot we can do in terms of allowing users to choose which algorithms they have. But yeah, I think at least technically maybe someone else can speak to decentralization and AI. You know, one of the questions that often comes up is people will ask if we are pushing for a world in which only peer to peer communication occurs. And my answer is always no. What we should be producing is a hybrid of the opportunities. There are just situations where a server based model is going to be a more appropriate choice for whatever you're trying to do. And then is that you actually have the possibility of the peer to peer system of the decentralized system in the stack so that you can actually choose something that's appropriate for the task. The AI question is somewhat similar. There are certain AI tasks which will only work if you're able to aggregate masses of data across large groups of people, and if people want to participate in that they should be able to and some people are willing to make that trade off. So a number of situations where that trade off is not acceptable. I think questions about any sort of voice command devices like Alexa, open up some questions about, you know, I don't really want that information moving out of my home network. That's those questions are going to become even more important as we get into IOT as we get the AR and VR and AI will be a significant part of all those systems. And so, what I think is important is that we actually have a tool set which allows for personal private AI systems to be created at this stage we don't have do not have the infrastructure for entirely local networked applications, which is what peer to peer enables. And so we're basically playing with half the cards missing from the deck. So, yeah, there are going to be some AI applications which just don't work with decentralization because they depend upon aggregation of data, and that's going to be a question of informed consent by people that are participating in it. But if you can do an AI application in a totally private environment, when privacy would be important, you should do that and we should make sure those tools are there. Yeah, I think this upsend decentralization opens up really a whole new marketplace as we were saying earlier of AI algorithm, whether you and who you trust, which person I go which persons algorithm are you going to to actually deploy on your server on your server are you going to deploy it etc etc. So if anything, it's going to help on the privacy. And I can just add to that. I work at simply secure which is the design nonprofit and the single most popular interface that we have been designing in 2020 has been data donation workflows. There is a lot of people out there who would like to collect data from people to reverse engineer things to, you know, do analysis on inflation influence campaigns. And the, the, there is, there is a population out there that is willing to donate data. And, in fact, if anything, as you have all said this is only good for informed consent and privacy that we have this sort of model to go by. Absolutely. Another example is the, the Mozilla voice projects where where Mozilla is collecting volunteer submitted texts, people reading text, as a, as a means to train a voice sort of AI, you know, voice recognition and voice text to speech analysis. In our in our last kind of few minutes here, there's a one more question from Sharon Bradford Franklin, one of our one of my fine colleagues who asks sort of the million dollar question which is, you know, one of the problems with with the big platforms that we're trying to do here is their business model after all it's they're monetizing our data, our attention or engagements for the decentralized models that we're talking about what is the business model how do you support the operation of the tools that you're talking about. Let's take on the example we have on our side. We provide so while the app is out there and is open source. We provide hosting. Anyone can go and set up their own matrix hosting platform. And then moving forward, reputation list and app stores for the matrix ecosystem are also models that that could be used so each application can have their own value add on top of the baseline use case that provide. Yeah, I think that you guys at matrix everyone at matrix and element are really pioneering the business model for this. I think WordPress automatic lots of open source. You know, you could think of email even right like how do people make money on email. So there is a market that is demonstrated for people to have platform service or have you host their, their servers for you. And that then funds the development of the protocol or the underlying software. So that, you know, that shows that I think that model is possible. I think there could be other more other interesting models that people have talked about everything from cryptocurrencies to, you know, actually paying for the product if it's such a good product. Why don't you pay for it, and then you won't have ads. So, you know, there are other models but I do think that, you know, falling in the footsteps of email is a good first bet. You know, that the entire concept of, of asking about a business model sort of presupposes the answer that there has to be a business model. And I think that, you know, that's not necessarily the case, you know, part of why we are doing this is, or why decentralization is interesting is because we are taking out those intermediaries and allowing people to connect directly with one another. And, and when you take away the intermediary, you take away the need for somebody out there to make money off of the transaction of me, you know, sending my mom a picture of her grandson. You know, I think, aside from the transport of the bits which we pay for because we, you know, all of us sort of have an ISP that we shell out monthly too. But, you know, maybe there doesn't need to be a business model for me sending a note to my mom. Yeah, I'd like to just piggyback on the, the public infrastructure ISP situation and the way you mentioned that there's all growing movement that say that the internet is a public should be a public utility as well. So, you know, we could expand this pretty far if we wanted to. And in thought, I do think that starting with ISPs and thinking about, you know, access to information as a right just as like transportation we build roads we don't think about the business model of roads, for example, so. All right. Well, on that note, I will, unless there's any more questions it doesn't look like there are. I will say thank you so much to Paul, Eileen, Amandine, and Carissa thank you all so much for joining us. Thanks also to the New America events folks, and particularly McKinley who's been helping us out on the back end shuffling shuffling questions and feeding them to me and all of this could not happen without those five folks so thank you to them as well. Thank you to our audience it's been great having you with us and we hope that you'll stay in touch with each of us individually and feel free to keep using the hashtag decentralized tech to continue this conversation after after we close up here. Thanks very much. Thank you.