On The Brink with Castle Island - Galen Wolfe-Pauly (Tlon) on Urbit and the Architecture of the Internet (EP.176)

Episode Date: February 8, 2021

Galen Wolfe-Pauly, cofounder of Tlon, joins the show to give us an update on Urbit in light of the politicization of internet infrastructure. In this episode:  The purpose of Urbit How the power of internet oligopolies has been recently revealed in the last few months Does the internet of today resemble the internet of 1994? The problem with having a single ToS/legal system for major internet platforms What's new with Urbit since our last Urbit episode in 2019 How hosting makes Urbit much more user friendly Urbit's governance system and how it contrasts from the feudal model on Twitter Galen's political taxonomy for Urbit - and where galaxies and stars fit in to this How internet platforms are like living in Vegas Whether users literally own their handles on internet platforms The politicization of internet infrastructure How resistant is Urbit itself to deplatforming at the infrastructure level Whether there is any irony in Urbit offering hosting for users Galen's view on data portability legislation The near term roadmap for Urbit The distinction between Tlon and Urbit How to get started with Urbit Apply to Urbit hosting here.

Hello and welcome back to On the Brink. So in November 2019, I went to the Toulon offices in San Francisco, and I sat down with Christian Langallis and Logan Allen, and they gave me a full explainer on Erbit, basically from scratch, which is actually one of our most popular episodes of all time. And I've always wanted to revisit the Erbit project and get an understanding of its roadmap and its viability beyond just the 101 level explainer. As the cracks in the internet began to show through really last month, as we saw AWS politicizing its service and some pretty coordinated deplatformings at the infrastructure level, I thought the time was nigh to revisit the Erbit project. And I've also been following it pretty closely, and it's made some really

great strides forward in terms of usability, and you don't have to use the command line to use it anymore, which is great. So we brought Galen Wolf Poly on the show, who's the co-founder of Tlawn, to talk about the state of Erbit and what's new in the last 18 months or so. This is a more general conversation about the nature of internet platforms, the politicization of internet infrastructure, and, of course, developments with Erbit itself. Now, if you'd like to try out host-house, they're rolling it out in a closed beta. You can use the link in the show notes to inform the folks at Talon

that you're interested in hosting and hopefully we can get you on the hosted platform ASAP. If you do get on Erbit, you can find me at Tickbin, Wixub. So without further ado, let's dive into the episode. Brought down by bad mortgage investments,

Lehman, which has 25,000 employees, will be liquidated. The federal government loans American International Group, AIG, 80, $25 billion. This is a different kind of market, and the Fed is asleep. The federal government is stepping it to stabilize Fannie Mae and Freddie Mac, the two mortgage

giants that have been threatened by the housing crisis. The Bank of England has pumped 75 billion pounds more into Britain's ailing economy with a new round of concentrated easing. You print a couple trillion dollars, and all of a sudden, people start to worry. So out of this worry, we have something called the Bitcoin. Bitcoin. Well, Galen, welcome to On the Brink. This is our second Herbert episode.

Thanks for coming on. thanks for having me and I'm happy that Erbit is interesting enough to have two episodes. Yeah, we don't often do, you know, repeat episodes on the same topic, but the first one we did with Christian and Logan was very popular. So that was, to this day, remains one of our most popular episodes. That's great. And a lot has changed with Erbit. There's a lot of new stuff to talk about. That first episode was in November 2019. So it's been a well over a year. year now. And I am now using Erbit in a hosted manner, which is great. I mean, some people might say that's contradictory and against the original vision of Erbit or whatever, but it's very convenient.

So there's a lot of new stuff to talk about. It's true. Yeah. Should we start at the top, or should we dive right into the difference? I wish I could remember better what they had talked about the first round. I mean, I'll direct people to the first episode, which is really meant to be a full explainer of the Erbit system from scratch, which probably can't be done in an hour. So I don't know how we managed to do it the first time around. But what about a kind of three-minute, you know, high level to shorter. Yeah, let's just do the various basic, most basic, basic, basic basics. So, yeah, before we jump into details, Erbit is two things, technically.

So there's an identity system and an operating system, say what we call an overlay OS. So it's basically just a virtual machine on the OS side. But let's start with that. So there are two sort of fundamental problems with the internet. I don't own my identity or my network address. Those things we don't sort of generally treat as property in the current system. And then the server stack has become so complicated that for the most part, you run a company if you're going to build an app of any kind.

So we address this problem by building two components. The identity system is basically a global PKI where the idea is to get individuals to own their name and network address as a simple pseudonym that is also a crypto wallet. And you can send packets to if you use that name to boot a node on the network of urban OS instances.

So these instances are each little virtual machines running on servers, you could run them on your phone or your laptop or on any cloud server. They create their own network, and the idea is to build sort of a single, very simple self-contained system that can store your data, run applications, and just let you communicate and collaborate in similar ways to what we use basically consumer SaaS for. But the idea is to actually, you know, build a platform that individuals can own and explore what's possible with without, you know, needing a third party. So for a developer, this means in the future you could develop software for Erbit, just ship it to the network, individuals run it themselves. The idea from just an individual user standpoint is hopefully something similar to kind of like a Western version of WeChat.

So my Erbit notes, stores everything, runs everything I need. I log into one place where I can chat with people, book a restaurant, pay that restaurant, pay my friends, whatever, engage in communication and commerce. freely in a way that you can never do in this kind of like fragmented SaaS world. I think that was less than three minutes. Yeah, those are obviously more than you can. No, that's very concise. You can run for a long time. Anyway, that's the basic lay of the land.

What would you say is a more revealing question to answer about orbit, the purpose of the network or what it actually allows you to do? Both important question. I mean, I feel like the purpose is to get people control over they're computing again. Everything we do with computers with network connections involves some company. You know, Bitcoin is obviously an exception to that. And so I always think of Erbit as this kind of like natural complement to any

blockchain-based system. They're kind of spiritually connected. I mean, what you can do with Erbit today, chat, share links, write with a group of people. honestly like so landscape which is the kind of main little app that we've built on top that people use that yeah is what people use in a hosted environment as it gets more mature i feel like just the ability to talk to people with no third party in the middle and collaborate on you know working together so we use this as the company and and for the broader our Erbit community. It's very simple, but you can't. There's nothing quite like it because everything else is incentivized to keep you around or tell you something. Right. Does it ever strike you as,

I don't know, kind of a little bit of entertaining that Erbit has had all of this work put into it. And then when it comes to describe what you can currently do with Erbit, a lot of it has just, you know, the typical response is while you can chat with your friends. Yeah, it's hilarious. It sounds like the funny thing is like the amount of work that it takes to allow you to just to chat with your friends where you're, you know, and on a platform or a network where you're not worried about that thing disappearing, no one's spying on you. you know the interface isn't sort of trying to keep you around all the time it sounds like such a simple thing

but it's actually like a frustratingly hard and and kind of big problem yeah there are many things that erbit will do in the future but uh i think people actually overlook like i mean it's kind of similar to i feel like this is you know it's like digital money is deceptively difficult too yeah like what's that expression if you want to build an apple pie from scratch you first have to invent the universe yeah basically that is the history of this project is that it's you're you're like oh be nice if people could compute freely together oh wait we have to reinvent most of computing or quite a lot of it in order to do that yeah well we've certainly gotten a very direct look into the risks of concentrating, you know, consumer SaaS with a small handful of internet oligopolis

in the last couple weeks. Regardless of, you know, the morality of the situation, it is disconcerting to see how few companies wield so much power in terms of our internet footprints. So I think the Erbit project is more urgent than ever. I mean, it's been urgent, but it's sort of especially urgent now. Yeah, it's such a strange time because I agree. I mean, I would have put that five years ago. And it's interesting to see, I mean, we live in such a strange and fascinating and incredible period of history, honestly. Like, the internet itself is so significant in terms of changing the way that we form communities and communicate with one another. but yeah it's also deeply flawed but for such a long time no one was really

most people just not either aware of or willing to admit the depth of those problems and then only in the last couple of weeks or months I mean very short time it's like everyone is aware of this all of a sudden yeah and we're actually and erbit has matured a lot like urbitt is retirement earlier but for me the problem yeah years ago or as I've been working on this thing that I'd be able to argue, I think, relatively compellingly, hey, the system doesn't work. We should rebuild it. And most of the time, you can get someone to agree with that, but they're like, okay, sure, but like, what can I, what can I do? Like, what can I use as an alternative? For the most part, I just, you know, Erb, some sort of like, well, Erb's not really ready yet, you know,

if you're a nerd, you want to play with it, go for it. And now it's like, well, actually, we're pretty close. And, you know, we use it ourselves. And if you're willing to just host yourself like yeah I mean people start their own communities on orbit all the time now yeah so it's a weird it seems like this shift in you know going from centralization to decentralization like I think it's actually totally possible whereas honestly if you if you if you'd ask I mean I probably wouldn't assess publicly but if you'd ask me honestly some five years in that sort of five year about five years ago even like 2017 I was like I don't know I mean I think it's the right thing to do but can we actually pull this off I don't know and I think I would have said that about basically anybody

in this space. And right now I'm like, oh, this is going to happen. It's amazing. So before we dive in, I have one more question for you. So, you know, I've become rather pessimistic about this thing we call the internet. Like if you think back to the way it was described in 1994 kind of thing, the internet that we sort of have today, the like consumer internet doesn't resemble that at all. It's like very, very siloed on a application basis. on a jurisdictional, geographic basis, you have different internets. And it almost seems like there isn't a single internet the way that people sort of envisioned it in the 90s. So would you say you're optimistic or sort of pessimistic over the fate of this thing that we call the internet?

Well, the internet's such a slippery word. Because if you mean the internet in terms of the core protocol suite, not to be too much of a nerd about it, but like, you know, I don't know. HDP, TCP, TCP, DNS. That's not going away. But I think that the internet, as most people experience it, is basically like TV. And I also think that's not going away. But I think the original dream of the internet being this, you know, basically a set of open protocols for people to communicate, collaborate, figure out how to coordinate themselves.

I think that in a way, like, that's coming back. And that was something. to get out of that protocol suite, but that protocol suite actually wasn't designed for billions of people to be freely connected. It works fine if everybody's like, you know, well, or a small community of hackers, basically, are running their own notes. But those protocols are just not designed for the way that we use the internet today. Yeah, I kind of think about it like a, you know, a city where you've got, you know, the internet of like the early 2000s was a city, a lot of like grimy alleyways where you know like people could hang out without too much

scrutiny and you know you had like 4chan and like similar such gathering places for various malcontents and then the internet of today is like rapidly changing into a place that's brightly lit everywhere um with no like dirty holes anywhere and it's sort of very sanitized and clean and it seems like there's like a certain loss of culture from undergoing that process. Does that analogy kind of make sense? Yeah, I think that's broadly true. But I don't know if it's, I think to me, the way I think about it's like, if you, if you think about a physical world, so when we travel to different cities, even within a single

country, and it doesn't matter, which country, really, like the cities always reflect the people that live there, even neighborhoods. And they are very different, even in the way that they're built or the way that they're managed. And I think of definitely sort of like the American experiment or the American system of government, and you can find that all over the West, is being mostly designed to preserve those, preserve and almost encourage those differences. It's like we've got all these different kinds of people. We want them to be able to live together and share infrastructure,

and we're going to try to figure out how to make that work. The problem is that in the digital world, we never. never like that charter in a way like never really existed. And so I just think the centralization of power among these, what are effectively media companies that we use to communicate with each other, Twitter, Facebook, whatever, it's a total accident. And these people have ended up in this position of power just because they were, they were actually, I think, very earnestly excited in building stuff that worked.

And I think they're sort of still, they still feel that way. the problem is that they're building it as sort of in their own image in the culture that they believe in and that they want to participate in. And the reality is there are all kinds of different people out there who want to live differently and think differently. And the software that they build can't be adapted. It can't be adapted at all. It's completely monolithic. And that's true at a technical level and it's true at an experiential level.

And then we intuitively know that it sucks basically because the result is that everyone is kind of the same. Like the internet is just this, yeah, it's like Disneyland. I mean, Disneyland's fine, but if the whole world's like Disneyland. I'm constantly struck by the fact that these large internet oligoplies or platforms or whatever you want to call them aspire to have a single, all-encompassing terms of service. And like, you know, you can just interchange terms of service with legal system and it's the same thing, right? Yeah. And then I think about how the actual world works and you have, you know, hundreds of countries and, you know,

unto thousands of administrative subdivisions and people have local political units that are relevant to them and people sort of self-associate into their local set of laws and norms and political modes of governance that are totally distinct from each other and there is no single global law and we've always rejected that and yet there's this like utopian idea that pervades the governance of these internet networks that somehow you can jam everybody into a single legal setting and have it work. And it just doesn't seem to work, but they keep trying. Well, they have no choice. Like, it's, I mean, they run physical infrastructure and they run companies that have, you know, like, that exist within a single jurisdiction for the most part,

or like, you know, are primarily governed by one jurisdiction. It's sort of like, that's just the byproduct of centralization. There's no way, you're never going to change that so long as there's one company running the server side. It's just the most natural thing in the world. Like I agree that it's stupid or it just feels weird. It's not very, it doesn't suit the people that depend on this stuff. But like, yeah, it's only, it's controlled by this tiny minority of people.

I mean, it's going to reflect their values. We don't even question it, though. It's not even something that people think about that often. Like, you don't think, huh, it's weird that. Facebook is trying to unite four billion people under a single standard of rules. Zoom out for a second and it's like that's the craziest thing ever. It's completely insane. It's always been insane.

Yeah. It's totally nuts. I don't know. It was nuts to me when I feel like when I got a Facebook account in probably like 2004 or whatever. I mean, it just seems, I don't know. I grew up on the internet where I actually was, you know, writing perfectly. on a computer that I controlled and that's like this incredible vehicle for creativity. And I think even conventional or ordinary creativity, I don't think you need to be a programmer

to sort of like experience the magic or sort of like the general purpose computing is like an incredible, incredibly powerful thing that everyone should have access to. And like nothing about the current internet is about providing. people with general purpose computing or communication tools. It's just completely fucked like sort of from the outset. I don't know. I like I agree that it's weird that people, it's not a commonplace. But that's kind of the interesting thing about the present is that it's sort of becoming

a commonplace like, oh, this is a serious problem. And which makes me very, yeah, that makes me very optimistic, I guess. A year ago, like at the time you guys recorded this last episode, right? Like a year ago, it was so normal for me to have a conversation about Erbit. And people were like, oh, yeah, that's a good idea. But it's probably, like, how are you going to convince people to believe in this? And my position is always like, we're going to have to kind of convince the world that this is the right thing. And now I'm like, oh, like, it's obvious.

Like, you don't have to convince anyone. Yeah, I remember on that episode asking Christian and Logan whether Irby was computational Esperanto. and that it would be a great, you know, elegant system that no one would ever see the need to use. But that's, as you say, that seems like a less relevant question today. I mean, there are plenty of, like, Urban is technically unusual because it's, I mean, because it's solving this problem that most people have ignored for a really long time, which is like, how do you get a bunch of people to compute the same way? Like, if you have all these nodes out on the network, they somehow have to share a model

of computing, which is generally not how most conventional software has been built, right? Because there's like one server, well, that one server enforces the model of computing. So there's a bunch of everything that's kind of weird technically about Erbitt comes from this single, you know, approach, which is actually not that different from building a blockchain, right? Yeah. You're like, hey, we just have to have one way that we define the state transition from this state to my next state. So anyway, I think it's like now, yeah, now it's basically defensive We built this Esperanto, but it's solving a very specific problem that is, yeah, I think

it's quite urgent. Yeah. So let's talk about what is new with Erbit in the last kind of 13, 14 months since we last talked about it on on the brink. So you guys had kind of, we were joking, like it was kind of like the moment where you break the champagne bottle against the side of the ship. So you had an OS1 launch. You now have hosting.

So tell us about what those things mean. Yeah, so the challenge for, it's kind of similar what I was saying before. Like, I've always wanted to be able to, you know, just offer someone an alternative, something that has the, you know, exact same level of approachable usability as any conventional app, but that gives you the sort of freedom and flexibility that one would get when running their own server. So the first challenge we set ourselves was like, all right, we have to actually stop using this stuff ourselves. Like just for daily communication, coordination at the company and within the community,

we got to build something that we can just depend on. We have to homestead this thing. And so that was probably, we experimented with a bunch of different stuff. And then by that when you talk to Christian Logan, so yeah, like about a year and a half ago, we were sort of honing in on what we really want is something where we can bring together a few different modes of communicating, which ended up being for now, and we'll probably expand a little bit, but basically chat, sharing links, writing notes, and having long-form discussion, and bring those all together in one interface that's centered around groups. So it's like build a

group, and then that group shares channels, which can be of any of those types. That was kind of like, well, that's landscape. So landscape is like the thing that you use through a browser that makes orbit this whole new system approachable and understandable at the level of a web app. We announced that, yeah, I think in March, it's right in the beginning of COVID. It was sort of strange timing. It was still pretty rough, honestly. It was very classic MVP. I think we had adapted ourselves to using it, but it was somewhat painful to use at the

beginning. It's gotten a lot better over the course of the year. At this point, it's actually, I think, quite good. it lacks two major things that are on the way pretty soon, should be just mobile app and push notification. So you can actually just use this thing as a messenger. So a messenger view actually, it's like I love, you know, I use signal all the time.

I think telegram is pretty good while they sort of trade off usability, but you're still always using a third party there. So the idea that I can have something in my pocket that just connects to something I own, and I can use that for communication and something I'm looking forward to a lot. And the way that'll be made easy for most people to use is through hosting. So you can actually go to Talon.io, sign up, get onboarded pretty quickly and just get going.

So yeah, the barrier to entry is on its way down very quickly. Yeah, you've gotten a preview of that. We've kind of tried to onboard some friends. We're sort of continuously scrambling to make this thing better. But yeah, it's coming along really nicely. I'd say in the next quarter or so it'll be at the point where we'll be more aggressively bringing people on whereas now it's I'd say it's still

basically like kind of you know you know late stage beta the waiting list we're onboarding people slowly yeah I don't want to flex but I am on urban hosting it's great works great been using it so thank you the only problem is I can't get your attention through orbit yeah but yeah that's true

I do need a way to be notified when you text me on there. We're getting there. I think it might be instructive to talk about the governance mode on her. But, you know, so obviously when something happens on Twitter that people don't like, you sort of report them to the commissar and then, you know, Twitter decides whether they should just obliterate their account or not. So that's one model of governance, right? which is sort of futile in nature.

What is it like on orbit? What is it like if there is some entity, an orbit, planet, star, galaxy, that you sort of dislike or you find to me malevolent, how do you disassociate with them, given that it's like a pretty decentralized network? Yeah. So you're making me think I should probably lay a little bit more groundwork.

So your urban idea is this synthetic, name. It's like a automatically generated pseudonym. We issue these, these names are finite, so they have value, which by the idea is anyway, that because they cost something, the network is civil resistant by default. Not only do you have to buy an address, but you need what we call a sponsor. So a sponsor basically is just this, you know, sort of binary, hey, I'm comfortable wrapping packets for this person. It's not really a full-blown reputation. It's just like, you know, this is someone I'm willing to support being on the network. And so let's see. So these addresses are, I always struggle to like not give too much detail. The address space is a whole system that's actually deployed to the

blockchain and, you know, is self-governing and self-updating. And we could get into the details if you think it's relevant. Well, we covered it in the last show, so I'll refer people to the last show for the address-based discussion. Got it. I guess the main thing to understand is it as it relates to how the network is governed is that like there's a pretty clear distinction between, you know, what goes on at the network level or the urban ID level versus like what goes on at the community level. So when you're issued this address, you own it with a key. It's yours. Like, It's yours like a, you know, like Bitcoin is yours or whatever. But you need a sponsor.

So your sponsor can reject you at the network level, which we expect would be something like someone is not comfortable for legal reasons, basically routing packets for you or helping you with peer discovery. Your packets are all encrypted. So, you know, that's something that's kind of like an escalated issue. It's like your ISP banning you. And so in general, you'd assume that that level of like someone being de-platformed or kicked off of orbit is pretty rare.

Sponsorship is two-way, and people can move. So if you have an urban ID and you decide that actually your sponsor is slow or, I don't know, they're suspicious for some reason, whatever, you can move to a new one, and it can go the other way where your sponsor can reject you. Anyway, to assume that's relatively rare, but it's a good backstop for the network overall, right? If you have bad actors, you do want to be able to potentially blacklist them or just have, you know, there's no one, no one, no one, no one sponsor them. So then when it comes to, you know, actual day to day, like how do you do content moderation? Yeah, like we were talking about

before, it's a little bit more similar to federated systems like mastodon. That's basically up to the community. And those, each community is totally independent. And everything on the network is peer to peer once peers discover one another. So if you decide to start a group and invite all your friends to it and, you know, you can basically, set whatever rules you like and decide to who gets to stay and who you want to ban. And we don't even know that you exist, basically, which is not that different than like a, you know, signal, group chat or something like that, if that makes sense. So when you think about the different political units on orbit, I know the culture is still

developing and you probably won't know for some time what the emergent qualities are like, Do you think of the political and administrative units in terms of galaxy, stars, and planets, or is there a different taxonomy that you use to sort of think about it? Yeah, I mean, that's pretty rhizomatic, to borrow a phrase, or whatever. So I'm trying to think of the outer space hierarchy, so galaxy stars, planets, is basically just, the infrastructure. And you could imagine planets and stars sort of rebalancing themselves to the most efficient infrastructure nodes. Most people shouldn't notice or care about that, and it shouldn't reflect the communities that are built because communities are built peer-to-peer.

And so those rich and complicated like to use the word archipelago, right? It's like this is just a bunch of, you know, completely ad hoc connections between the people on the network that self-govern and self-police and, you know, to do whatever they like. I do think it's not that dissimilar from what you'd find. I mean, I'm sure if you look at a map of like telegram groups or something. Yeah. It's more really not that different from that. Yeah. Yeah.

And it's funny that we have to describe this as kind of an alternative to the default system, whereas this is a more organic and natural system. fundamentally. I mean, it more closely maps onto how humans actually work. Right. I mean, that's pretty much how we think about what we do, right? It's like, you just want to let, you want to build a network or build the tools that let people decide how they're going to organize themselves and certainly want those tools to be, you know, exceptionally high quality and long lasting, although let's be honest, they probably wouldn't be adopted as tools if they weren't. And what people do with them is, yeah, build their own image of the image of their communities.

So it's like if you wonder what Urban is supposed to look like, it's like, well, it's supposed to look like the real world, I guess. Yeah, as opposed to these kind of artificial, you know, uni node networks where there is like effectively a single dictator in charge of everything, which seems incredibly archaic and old fashion. It makes me think of Babylon or something with the king, you know, handing down decisions and pointing to like a code scrawled in. you know, a clay tablet somewhere.

I mean, it's totally insane. Like, Panopticon, not far off. I often think of it like, and ALDI also uses like, living on these platforms is like going to Vegas. I'm not a big fan of Vegas,

so, you know, maybe some of people are. But just the, if you had to live your whole life in a hotel in Vegas, uh, meaning you can't modify your room, like everything is climate controlled. You know, like you just, it doesn't, it's totally determined by the entities there to basically make money on you. And like, yeah, maybe it's fun to go to Vegas for a while, but like, you don't want to live in Vegas.

Wow, that's a great analogy. Yeah, I will confess I'm not a fan of Vegas either. I found it hellish when I was there. It's really tough. I mean, it's just hard on your body, you know, like you don't feel good because it's, it's not. I think it's just because you're being, it's so tightly controlled, you know, like it doesn't. It's like Vegas if Vegas was a prison. That's kind of how I look at like the modern internet because you're stuck.

Like you have no alternative. It's like Vegas that you cannot leave. So Galen, I've developed this like hypothesis that I've been told by lawyers is wrong from a legal perspective, but I'm sort of undeterred, which is basically that users of social media literally own their content in a kind of physical sense. So it's their property. They own their handles and they own everything they produce. That's their commercial labor has commercial and social value. And they should have the equivalent of digital squatters rights. And now, of course, I'm aware of the fact that the terms of service don't see it that way. But I'm curious as to

whether you think there's any merit to that theory or normative opinion. I mean, stepping out of, I mean, from a technical standpoint, Erbit very much treats your data like your personal property and makes sure that your identity and network address are as much your personal property as possible. It's an interesting question because I'm not sure what the precedent is, you know, like is speech property? I mean, I suppose if you write it, it's like, I was just trying,

I don't, I'm not a, I shouldn't. even dabble into this because I wouldn't understand that. I don't have like immediate background. The analogy I would give would be people on the frontier that, you know, I know this is also like an idealized analogy because like, you know, was there ever truly an uninhabited Western frontier? Like, of course, no. Like, you know, people had to displace the Native Americans. But, you know, imagine for a second there weren't any and you know you stake your claim out in the frontier and you build your homestead. Even if that land was owned by, you know, Lord Delaware, 3,000 miles away in England, you know, once you've sort of under the Lockean theory, once you've mixed your labor with the land,

you maybe have a pretty legitimate claim to that land. And so at that point, you know, you can sort of claim that it's yours. So that's what I refer to when I say like, you know, digital squatters rides or digital homesteading. Yeah. I mean, the problem is that you did it in Vegas and the guy who runs a hotel like makes so much money on that hotel that he's like no fucking way that you're getting squatters right yeah yeah i don't think it's like it's just a question of who's in power it's not a question of whether you're right so i probably i'd agree with you in theory obviously right because like erbit wants to provide those affordses but i'm not sure if you just think of just the power dynamics and how much money is involved i'm like that one is not uh yeah you're gonna

lose that fight. Yeah. Yeah. So, well, the other big topic of the day is obviously, you know, what I would refer to is the politicization of internet infrastructure. And it's not just applications. Like applications was, you know, somewhat perturbing to see politicize. Like, okay, you know, don't like Twitter. Well, okay, build your own Twitter. But then it gets more perturbing when it goes deeper down the stack, in my view, to the more essential, you know, systems of the internet hosting, DNS, DDoS protection. To me, there's something a little different about AWS, you know, saying, hey, like, Parlor can't use our service, as opposed to Twitter saying, hey, we're not going to give you a microphone.

Do you kind of share that intuition? It's right. I don't fall AWS in this instance because I think I understand how they're acting. They're just like, we don't want to deal with this. And they have every right to say that. Yeah. I, yeah, my position is I think probably what you might expect, which is like, yeah, well, if you had protocols for all this stuff, then you wouldn't have this problem. You know, like, hosts shouldn't be put in this position because, anyway, I don't know if you want to dive down that rabbit hole. But I think, to me, it seems like, maybe you're getting a, I tend to be much more of a like, I'm like, well, if you have a problem like this, then you didn't build it right. You know, it's like it's a consequence of something going on about like the architecture thing.

It's not about like this kind of high level, you know, when you're seeing like, it's the same thing you're building software, right? Like you see really a lot of trivial bugs. Well, there's something architecturally unsound about it probably. Sure. So I feel like this is like, you know, yeah, the whole system is pretty, it's. It's pretty weird. That's because applications are so complicated. They're so big and complicated.

Anyway, go ahead. Yeah. No, no, no. No, you're right. So I guess the natural follow-up is how resistant is Erbit to any internet providers at the service level, you know, trying to de-platform orbit for whatever reason. Maybe they listen to this podcast and they don't like it. You know, how resistant is it at the hosting level? if you imagine the consolation of users running their own orbits.

Yeah, yeah. Yeah, we can try and, it's like doing a security audit or something. You know, you're also trying to produce the manual for an attacker or something. So, yeah, I think that orbit is designed to be able to be run by anybody relatively technical. It's actually not that hard. You can just follow some instructions. We sort of, I think for a long time, we were off. and actually probably more, for any of the listeners who, like, you know, looked at our

insulation instructions, they were like, you might not want to try that. It was actually more because the system itself was still young. You can practically just copy and paste commands into the command line to Pluto and Erbit. It's quite easy by comparison to most software. So the only reason that matters is that, yes, everybody could run their own node. So when it comes to, like, exposure to hosts, that I think is relatively easy. to mitigate, meaning if we were, so it's important to draw the distinction, I work for TWA, the company that works on this thing, and Erbit itself is owned by lots of people and it's just this protocol. Toulon provides hosting, so we have some exposure to the infrastructure

that we use, the fact that we're a U.S. company, I mean, we have all kinds of exposure, probably. So Tuan is vaporized by whatever, by our hosts and, you know, a concerted effort by hosts and state actors and so on, can Erbit survive? Well, yeah, everybody can go and host themselves. It's not that difficult. The Erbit community is very motivated to make this as easy as possible, even in terms of running your own hardware.

It's a pretty enthusiastic community. People who run Erbit on Raspberry Pies and want that to be as easy as it can be. I tend to think then, though, that the likely outcome is that there's a mix of people who are hosted and people who are self-hosted. And ideally that there's a balance there because that balance provides almost this kind of herd of unity for the whole platform.

Meaning basically that, like, Erbit should be, you know, as risky as signal or telegram. It's a, sure, like, we provide people good privacy and durability guarantees. There are pitfalls to that. But overall, it's similar to what people get in the real world. And so it has sort of precedent for why I think it's like, you know, not something that needs that is kind of like, exactly at odds with the state or something like that.

And the fact that there are mixed ways for people to, you know, keep this thing online is just like generally good for the robustness of the network overall. So anyway, yeah, I think the likely outcome is that you have people, you have lots of people doing both self-hosting, hosting, hosting in bunkers in Iceland or whatever, you know. And there's a whole sort of spectrum of like,

you know, depending on what type of person you are, you have a more or less resilient urban hosting setup. It's very easy to host one in your closet if you are. That's exactly analogous to Bitcoin. You have the capacity to run your own note if you like. But you can also, now there's a bunch of hosted tools, aka custodians and exchanges out there that people do take advantage of 20 to 30% of all

bitcoins at least are held with these third parties. that's it? Well, I would have to be more. That was my last count, but you can probably get, you can probably find some more if you were to be more thorough. But I mean, I was encouraged when I did that analysis, but there's no way. Yeah, it's hard for me to identify by looking at an arbitrary coin on the blockchain, whether it's hosted or not. So yeah. Yeah, I think I like, like one good indicator is that this is a. pretty lively debate at the company, within the community,

like whether or not hosting is a good thing because it's, and way before any of this current, like, event stuff is going on, because it's an obvious risk. And, yeah, there are all kinds of interesting solutions,

but I think the interesting, the important thing is that that debate exists and that debate exists in the Bitcoin community. And it's like, it's actually just like it's a legitimate debate. And people should care about it and think about it. then that should push them to come up with lots of different solutions that they can offer back to the people that use this thing and giving them all the more choice, which is great. So anyway, yeah, I think that that will drive Urb to be quite resilient.

Yeah, my stance on the question in a Bitcoin context has always been that custodians and effectively Bitcoin banks are completely acceptable, especially if there's a diversity of service providers with the caveat that it has to be easy to take physical delivery of your bitcoins and with. so that you can move to a different service provider if they're misbehaving, or you can just completely move to your own self-custody, if necessary. And that's the distinction between Bitcoin and like gold, for instance, because it's very hard, you know, to take ownership of bars of gold. So I would say with Erbit, I would see it analogously, as long as it's cheap and not too difficult or cumbersome to run your own instance or your own Erbit, the existence of hosting isn't an issue as long as you still have that sort of withdraw exit option.

Yeah. Yeah, that reminds me. It's actually worth specifying this. So if we host you, we promise, and I think we'll actually try and make this as much of a legally defensively prompts as possible that will give you your orbit event log bat. Like, we'll give you data back. And when we give you your data back, this isn't like that useless zip file that Facebook gives you.

Like we actually give you the entire thing. You can run it yourself. It's really easy to do. And so the cost of exit is pretty low. The other thing is that you may, maybe you become suspicious of us. And you decide you want to basically revoke our ability to host you. Well, you actually hold the key to your identity. So you could effectively just re-key and lock us out and like shut your orbit down.

like the one that's running in a hosted environment. So yeah, it is similar. In some ways, the sort of ability to lock someone out, and you'd keep the identity, of course, you'd lose your data. Yeah. Makes it slightly better. Maybe it's just different than Bitcoin, but it's similar. I mean, Urban and Bitcoin have a lot of things in common.

What do you make of some people are proposing data portability legislation for social media firms, so basically turning them into, you know, kind of open API systems such that they would be legally obligated to make it, to deliver your data in like a standard format should you want to leave. Does that seem like a viable fix to you or not really? I mean, if I, if there was any incentive on it within those companies to actually abide by that legislation, like there has to be some internal incentive to make that actually. work well, right? Like, it has to actually, you know, like, do these companies care if their customers

are unhappy? Not really. Right. I mean, not really. Like, so, so really, you really think that's going to happen? That just seems incredibly unlikely to me, but I will happily, you know, have a drink with any of the proponents of that in five or 10 years or whatever. We can talk about how it played out, and maybe it'll work with that, you know, but it seems unlikely to me. So, before we go, tell us a little bit about the sort of near and medium term future of orbit. You know, we were talking about hardware a little bit before we were talking about the distinction between Toulon and Erbit.org. So touch on that briefly before we wrap up.

Yeah. Yeah, yeah. So I think the main thing like I've been sort of repeating over and over, right, is just you want to get to the point where Erbit is viable as sort of social software for people to start new communities or some companies probably and basically. And basically just like for people to start knowing what it feels like to actually control their computing in this very simple and straightforward way. So that's like hosting plus their mobile experience plus notifications. And that'll happen in the early part of this year.

And then in through the end of the year, I think, you know, Klon's main goal is just get people on here and get them to actually, you know, experience what it's like to sort of be free from the digital gulag or whatever the internet is swiftly turning into. In doing that, yeah, Talon. So Talon is a company that we founded quite some time ago now to work on this project. And as Talon starts to get into the services business, providing hosting, probably, you know, working directly with communities to get them using Erbit. We do want Erbit itself to be a platform that people can extend, experiment with, build software for, you know, it's not just meant for social software. You could use it for all kinds of things. And so to that end, we've been working to get erbit.org, this basically sort of like internal foundation that we've always earmarked a bunch of address space for and wanted to exist.

That's getting up and off the ground to really just support, you know, growing the developer community, getting people to understand the system and build on top of it. And so, yeah, I think we'll try and get that actually, you know, sort of out the door independent this year. And there's actually kind of a lot going on there already, just in terms of like we were run a really active grants program and there's quite a lot of just like developer focused events and community stuff happening there. So I expect that to ramp up quite a bit too. A big thing on that front is really just getting Irv to the point where we can do really good software distribution

over the network itself, which today is possible to experiment with certainly. And people do run a little bit of their party software. And I hope by the end of the year will be relatively easy. So I guess the main thing there is that, you know, our hope for developers in the future is that you have an idea for a weekend project. Great. Do it. When you're done with it, just ship it to the entire network and never do DevOps or think about it ever again. Just let people use it. So irvitt.org's mandate broad listening that's on one hand is that make that possible, make it easy for people to develop on top and experiment with Irvitt itself.

And then, yeah, like we were talking about before, make sure that this self-hosted. alternative for running your own orbit is as easy as possible potentially even yeah providing the resources at least for people to build their own rigs which is a sort of something that community seems to really love which is cool or just making sure that that's you know as easy as possibly can be so yeah it'll be a big year

and I mean we tend to think you know orbit obviously is this pretty ambitious long-term I'm obviously pretty, I'll be doing this forever. You know, we have a long road ahead of us, but we also don't. People often ask me sort of like, okay, what's going to happen to urban five years? And I'm like, you know, the funny thing about this project is like we've always pretty much planned like six months out. We know like in the very long term, you're like global computing platform, like this is how people communicate, collaborate to business. You want to actually like just replace networks computing with a new computing platform.

and then it's like, well, what do we have to, you know, what's the step? What are the next two, three steps? So after that, I don't know. Other exciting and interesting things. But the general trend is back to the land computing. Yeah, yeah, exactly. It's like 1968 again. So what is the, so for people listening who are now curious about orbit,

what is the action item that you would give them beyond? So, you know, how do they get involved? Yeah, so two, there's sort of two paths. If you're somewhat technical, you want to play with this thing and see what is actually going on under the hood. Go to erbit.org. There are installation and setups instructions there, and there's, you know, pretty extensive documentation, tutorials, and so on.

If you're not technical and you just think this sounds cool and you want to use it, go to tlawn.io, which is, I suppose we can put a, you know, T-L-O-N-I-O-O-O-D-O. That's where we're offering hosting. Got a wait list there now, but we're trying to onboard people as quickly as we can. And tell us a little bit about what you want to do with this thing so we can figure out kind of like how to get you on most effectively. You can also find us on Twitter.

That's pretty much the only old world social media that we use. Honestly, otherwise, you just find us on Ervitt itself. Yeah. I'm going to self-shill right now. I am tick bin-wifin wick sub. but that's that's tick bin with a why uh in the bin so hit me up hit me up on orbit folks there you go that's the the best way to get in touch with nick so why wouldn't you just you got to get on the network the most self-sovereign way to find me that's right you want to get

the most premium encrypted packets to nick carter i'm gonna i'm gonna make sure to check my DMs on there more than once a week, for sure now, which there are many, apparently. There's always something in there when I look. But yes, I can attest to the fact that it does work. And I think I was able to get live on it with the command line the first time, the way you describe copy and paste. But I found that difficult, I will admit. It's kind of, it's not.

I mean, yeah, you requires a bit of, well, it's better if you run it on a VM somewhere. like hosting is the, I mean, now we also have pretty good, even if you're hosted, like the terminal is not bad. So if you want to experiment with the programming language and even call the APIs, like you can do a lot of that just being hosted. So I think that will be the default on ramp. I would love for us to, we don't yet have like a free immediate tier. The next thing for hosting will be, it'll be amazing when we can just say like try this now, spin up a node for you and get it going. it's not quite there yet. Yeah.

I'm looking forward to like the same way of physical Bitcoin nodes. I'm looking forward to a physical urban node, which also doubles as an email server. And maybe it does my cloud storage for me too. Local cloud storage. Yeah. That sounds like a weird contradiction, but that would be great. There's so many pitfalls of this.

I meant to mention this earlier, actually. Someone at the company it came up this recently, I really like this. I think the way that should happen is basically that you have a, sure you have a you know Kasa style whatever like a little box a little black cube that's like you know milled out of obsidian or whatever and that thing that's what you do most of your computing on there are all kinds of problems running boxes at home people always course over this just like dynamic IPs are complicated ISPs don't like it whatever anyway what your host should provide I think is encrypted backups so you run this thing at home but you have a host that's

always you're always sending up an encrypted backup to them. So that in the case that whatever your house burns down, you know, the thing disappears for whatever reason, you have a backup similar. I like that, actually. I think that's kind of an interesting, like, middle of the road hosting option. And I think we've actually thought about trying to do this.

I think we could do it. Yeah. And if you pair it with multi-sig, that encrypted backup can be kind of very, very resilient to a number of blocks. Yeah, you can do this. some kind of like social recovery stuff for the IT too. Yeah. Yeah.

Well, Galen, this has been great. Thank you for coming on. It's been a long time coming. And yeah. Yeah, my pleasure. Where can people find you on Erbitt? Oh, I'm, so people seem to be finding me, which is weird. Is it a secret?

Or is it not a secret? Oh, it's not a secret. Okay. Yeah, Rathmel Ropdile. I don't know where. people seem to be finding that name, but it's out there somewhere. And that's literally the only place that you can find me. I don't use social media otherwise.

But yeah, thanks for having you, man. It was fun. Okay, well, you heard it go harass Gailen on Urban itself. This has been great. We'll bring you back in a year's time and see how things changed. That sounds great.