Talking Pears with Guy Swann

Hello, and welcome to 2026, and welcome to Freedom Tech Friday. Freedom Tech Friday is a weekly live and interactive show hosted on the Ungovernable Misfits, X, Nosta, YouTube, Rumble, and Twitch feeds.

We go live for one hour every Friday at 9AM eastern or 2PM UK time, and you can, of course, catch up later on the Ungovernable Misfits podcast feed. On Freedom Tech Friday, we like to cover the latest news and trends for anything relating to Freedom Technologies.

That could be anything from Bitcoin, Monero, encrypted messengers, privacy tools, and everything in between. Essentially, if there's a news item tool or topic that can help you take back some control in today's digital Panopticon,

we want to talk about it. Managed q and a, I'm head of customer experience at Foundation where we build Bitcoin focused sovereignty tools. As always, I am also joined by my good friends Max, the head honcho at the ungovernable Misfits Empire, and Seth who is now the COO at CakeWallet.

Feed and Set Friday is a live and interactive show, and you can help steer the conversation by commenting live, pre submitting your questions via socials, boosting the show via podcasting two point o, sending in your tips, or just sharing the show with your friends.

Top support for our final show of 2025 with Kian Rodriguez comes from Late Stage Huddle, who sent in 6,006 sats and just said, hashtag the pardon samurai. We could sign that message, of course. Thank you for your support, Late Stage Huddle.

We do have a guest that I'm gonna bring in shortly. But without further ado, Max, Seth, welcome to 2026. How's going, guys? I just realized that I haven't allowed Seth into the chat. Yeah. I was just about to say.

Here he is. I'm not gonna play Seth right now. I'll do it. Still in I said here he is, but I haven't I'll add him in. One second. There he Welcome, my wee guy. How are we doing? Howdy howdy.

Me and Guy look similar enough. I feel like he could he could he could be me. We got the fancy bikes that are pull it off. Similar headphones. We both have beards. Like, I don't know. It could work. It could work.

There you go. You've got your next time you wanna go on holiday, Seth, you've got you standing. Exactly. Exactly. Max, are you well? Very well, mate. Very well. Enjoying my coffee from my new coffee machine, so never been better.

Still a decaf, I hope. No. No. I'm I'm it's creeping up. Two a day now. And I knew you would. I knew you has me drinking Fiat Storm Guard today. Oh, man. It's alright from time to time. Now I've got this new coffee machine for Christmas, so I'm I'm a happy, happy boy. Espresso or coffee?

It's Q knows more about the machine than me, but it does espresso.

Nice. It does everything. It does all this stuff. I'm very, very jealous. But check the tapes, listeners. On the last Bitcoin brief, I did call that this was gonna happen, that it the caffeine would slowly start to creep back in. Here we are, nine days into 2026, and he's already capitulated.

So You're a wise old robot. I'll I'll take the victory lap on that one. Anyway,

well, welcome to Bats of Freedom Tech Friday, everybody. Welcome to 2026. Hope you had a good Christmas break. I know I did. It's filled with lots of merriment, family times, and probably a little bit too much alcohol, but here we are. Today, we are diving into something a little bit different,

technology that can potentially allow us to build apps that work without central servers, no cloud, no accounts, and no middlemen. Holepunch and Pear are the stack that bring apps like Keats and Pear Pass to life.

To guide us through this new, I guess, frontier, we've got none other than Guy Swan, host of the excellent Bitcoin Audible podcast. And the most vocal advocate that I'm aware of of the the kind of hole punch slash pair and keep ecosystem.

So, Guy, welcome to the show. Pleasure to have you with us. How are you? I'm doing great. I'm doing great. I appreciate you having me on, man. Good. Yeah. It's a it's a pleasure to have you here. But just to just to quickly set the scene,

other than being an advocate for the the the all of the aforementioned stuff that we're gonna dive into, do you have any actual, like, working relationship with the guys?

Sort of. I and I've been, like, kind of, I guess you could say part of all of their groups and conversations and stuff for two and some change years now. And I hope to be able to show them PearDrive when when we kind of get that to a state that I think it's that I would be confident enough to say, listen, if you would like to invest.

I think I have a very serious project that's, you know, built on the pair stack. But other than that, not really. Like, we did the the pair report and, like, I had them on a couple of times when when I had that show running for a short period, but I spread myself too thin. So I I all my ancillary or secondary shows, I just kinda cut them.

Understood. But which I still talk about the pair stack on the show all the time. It's just not the pair report anymore. But so no. Not not really. Not really. But I do work with them and talk with them all the time. They have invited me to speak at the plan b conferences,

which, you know, is that whole crew. You know, it's Tethr and then because they they help fund a bunch of the development on this. So, yeah, that's weird answer, but not not not really, but yes.

That's fine. I I just wanted to clarify because, obviously, like I said Yeah. I'm not getting, like, paid by like, I don't have, like, a salary from them or Yeah. Something like Yeah. Yeah. Yeah. Okay. And I definitely wanna dive into into Pear Drive later in the show once we've kinda set the scene a little bit. But Yep.

I guess to to kind of get us started, what what problem is the kind of whole punch ecosystem trying to trying to solve? Like, why why does it even exist in the first place? You know, I I think about this a lot, I think it really just boils down to censorship.

Like, privacy and surveillance and, you know, control over your media and stuff all come with that, but it's really kind of like you have to solve those problems to ensure you can be censored.

And so, like, if I if I wanted to put, like, one thing on it, it's it's about changing the structure of how we deal with data and how we connect and how we identify each other in such a way that we are the ones who control

the network, the content, and the connection and not someone else. Because we have found ourselves in a place where the Internet does the opposite of that, and that was never the intention.

Yeah. I I mean, it seems like a very a topic that's very tangential to Bitcoin in in its values and in its properties. And, again, we're we're gonna dive into those. But, yeah, from from my kind of early research, like, this week in building up for the show, it seems to be kind of removing middlemen, reducing the risk of of centralization

and censorship. So, yeah, again, very tangential to Bitcoin. So I guess let's just dive right in. Like, at at a high level, like, what is is Whole Punch? Is it is it a company? Is it a protocol suite? Is it kind of a a philosophy?

Like because that was something that kinda confused me because there's lots of there's lots of names and things that are all used in the same sort of conversation. I got a little bit confused with, like, whole bunch, pair, key, like The naming convention is the naming convention is bonkers. Yes.

And it's because it's a very, very long established project or a very long project. It's, like, 12 years old or eleven years old or something like that. And it's gone through many iterations to get to where it is today, and it started out with an entirely different name and kind of intention of how it was gonna be sold or or or positioned

for, you know, users. And so it was the hyper stack back then. So there's, like, hyper core, hyper drive, hyper b, hyperswarm, like, of these different things. But, really, what it is is a it's a protocol stack. It's a networking protocol stack.

So rather than, like, one simple piece of networking, it's like, okay. How do you deal with how do you deal with sending information out in a distributed way, and how do you deal with conflicts? How do you deal with ordering of that information? How do you deal with establishing and holding

intermittent connections? Like, how do I connect it? Like, if we're all three of us are connected, how do I establish that if I lose connection to one that the stream doesn't drop? You know? Like, there's, like, a bunch of different primitives that have to be sorted out to make this work reliably.

But I I honestly think this was kind of, like, the whole concept of the Internet. Right? Like, that's that's what TCP the Internet itself is just how do you make a reliable network on unreliable hardware is because centralized is the opposite of that is how do you you have you have one thing, and if that thing goes down,

then everybody loses their connections. The Internet was intended to be the solution to that. And this is basically taking it to a new level. And now it's basically referred to as the pair stack. I like that better. I think it's a much better way to to refer to it, and that's how I refer to it. Hole punch itself is both a method.

It's it's a it's an idea of, like, how this is possible, but then it's also a company. It's also the company that's doing all of the development. So they are hole punch inc, I think. And I think sometimes they refer to the set of tools that they have developed as, like, the hole punch

stack or something like that. But I I think it's easier to just think of it as, like, the pair stack, which is a set of protocols and and hole punch the company.

That's that's how I think about it. It's not a true Bitcoiner created tool if it doesn't have a lot of confusing names that no one can deal with. It's like if you don't do that, you've you've gone wrong somewhere. So

Alright. So so I appreciate that one. That that makes it much clearer for me. So we've kind of used the term, like, decentralized network and and, like, the obviously,

I'm immediately drawn to the analogy of, okay. Well, how does this compare to the Bitcoin network, which is, you know, the a decentralized network that we're all very familiar with on Freedom Tech Friday? Like, what analogies can we draw between the two types of network? Are they similar, or or is that kind of a name only?

I wanna kinda clarify that before we dive into kind of the the building blocks of what actually makes this work because it almost seems like it's too good to be true. Right? Yeah. Yeah. So in the context, like, I think it makes much more sense to compare it to BitTorrent because Bitcoin

like, Bitcoin is the BitTorrent of money, really. But there is one thing that Bitcoin has that decentralization generally doesn't need, and it's a global consensus. And in fact, oftentimes, you don't want global consensus on everything. There there's a lot of drawbacks,

and there's an enormous organizational problem or coordination problem, and you have to have everybody understand the same state of a thing. This is exactly why Bitcoin is so,

restrictive in its use case and its scope is because the consensus problem is so unbelievably difficult that as soon as you let it do too much and let it grow too big, the consensus itself breaks down. The the reason that that consensus is valuable breaks down.

So in the context of general decentralization, you don't have consensus over the BitTorrent network. Right? When you're, like, file sharing, you just you you have, like, 20 people over here sharing this movie, and you have 2,000 people over here sharing this other

file or podcast or whatever it is. Doesn't doesn't really matter. It's just it's just a data distribution network. And they don't have to care what the other people are doing. And everybody else can go offline, and all their stuff still works. This is much more like BitTorrent in that way.

Whereas, BitTorrent also just had an enormous amount of drawbacks. Like like, there was huge benefits, but it was also very dated. It was super clunky. There were a lot of restrictions in how you could do stuff. As soon as you, quote, unquote, made a torrent for data

for a piece of data, it was trapped. Like, you couldn't do anything with it. Like, if everybody like, let's say, for instance, like, if I had, like, my podcast and I publish an my my feed or whatever as a as a torrent.

Well, as soon as I wanna add another episode, that old torrent doesn't work anymore.

Like like, I have to I have to break it and make a whole new one to add a new episode to it. And now if I had 10,000 people who were downloading and listening and seeding that podcast to other people, well, they're all gone now. I just started the turnover, and I have to go find those 10,000 people again and get them to reconnect to me.

And so the alternative would be I have to make a torrent for every single episode, but then I have to figure out how to get it to them. And I don't have a torrent. Like like, the network is too fragmented and and kind of isolated.

There was no update mechanism, so to speak. And so the idea of the pair stack is if you built BitTorrent today and you actually wanted it to kind of have the capability

of just doing everything that we're used to on the Internet, have a website, update some stuff, or I can change this file, and there's just a record of the file being changed. There's a new update. There's a new episode in this feed. There's

there's, you know, new information about this profile from this individual. How would you do it? And so the stack, the the quote unquote hyper stack, the the the protocol that they build is basically a combination of, like, okay. What would be the main primitives that you would need and how would you solve each individual problem?

And then you put them together like Lego blocks and you can build an application that just kinda delivers a website and but it does it in a in a in on the back end in a BitTorrent way, but on the front end, you can just make it feel like the web.

So it's it it's a way to basically solve the problem of of the shortcuts we built to get here with the way the web and social and all of this stuff works in a new stack that does it basically without the Band Aids.

Like, we hacked it together. Let's fix it. With the similarities with Bitcoin, on my little bit of reading with all of this stuff, I was reading up on their pair pass. I guess one thing that I would say is similar is, like, personal responsibility by not relying on these centralized servers.

If you don't back up or you don't have things on multiple devices, it seems as though you're fucked. Yeah. For for better or for worse, you are responsible. Yeah. Yeah. It's very much like your keys, your data sort of thing.

Yep. However, it doesn't it also has the benefit, like and and maybe we'll get into this in, like, a separate question or framing, but there's nothing

like, there's a lot of, oh, there are no servers, but there's no reason that this has to shun servers. It's really just there's no centralization by default.

So but you can easily leverage it so that someone else has a backup of your stuff, and you can build it so you you can pay for it just like you would on a normal web. You just don't have all of these enormous frictions that you would otherwise have around doing that or the difficulty in providing that service. Like, I mean,

how many people really know how to set up a server and then host data storage for somebody else? Like, the the complexity and pain in the ass. And most everybody's just gonna rent from AWS to then forward that service to some other user because it's just it's kind of a nightmare to do. I mean, it varies sort of low, but it ain't that low.

And that is exactly why we have found ourselves on a whole bunch of central well, one of many reasons we found ourselves on a whole bunch of centralized platforms today.

Alright. I I wanna kind of this is the question I've I've wanted the answer to the most about the whole stack, and it it gravitates towards generally how does it work. Right? I networking is is probably my least favorite part of interacting with computers. I just don't understand it properly.

With that said, I probably never speak to audience. I have a very love hate relationship with it. Yeah. Indeed. Indeed. Now I I know enough to be dangerous. Right? I know that, you know, connecting peer to peer over tour is is latent, but it's easy. There's no port forwarding and all that jargon.

How does the Pairstack kind of fix that without I I I understand needing Tor or any port forwarding. Like, how can my computer talk to your computer without all of the the usual or traditional hurdles that we normally run up against when we try and do that sort of stuff.

Okay. So without getting too clunky about it or or too in in the weeds, it it essentially does it very similarly to the way BitTorrent and just any general DHT, any general peer to peer network would.

But but even like, there's even things like, like, bit Bitcoin, You are not a listening node if unless you, like, pour forward and you actually make it so that you are accessible on the web. And so that's still, a problem. There's a lot of there's if you think about it in comparison to BitTorrent

is you have cedars and leechers. You have people who just download, you have people who upload. Well, most nodes on the Bitcoin network are just downloaders. They're downloaders, they're securing it for themselves, but they actually can't handle requests from other people because of this really frustrating networking issue.

But, essentially, everybody at their in their home has, like, a really big painful firewall. And it's like, okay. Well, how do you if I want to get through it, how do I get through it? And so the hole punch mechanism that they have built,

the the easy way is kind of if if we don't have, like, a broadcasted method of, like, okay. This is the exact port we're using and nobody knows what port we're using and there's 64,000 whatever ports, how could you how could you make it so that we could agree on a port without without communication?

Like, without actually being able to identify exactly what the other one is using. And they take advantage of an of a fascinating little statistics trick called the birthday problem. And so have you ever heard the thing that,

you know, there's three hundred sixty five days in a year. Right? And so it's like, how many people do you need in a room for two people to have the exact same birthday?

And the number is something like '29. I can't can't remember exactly what it is, but that seems crazy. That seems like very tiny in consideration to this three hundred sixty five days. And the reason is, though, is because you don't care what day it is. Right? You don't care if the birthday is January 1.

If you had one person that says this person's March 12, how many people do you need in a room to to have another person whose birthday is March 12? Well, you need, like, you know, what what's half of March? January, January January.

And so but if you don't care if it's March 12, if you don't care which person you're matching the birthday with, you only need, like, '29 or '30.

And then there's going to be two people who have the same birthday on one of those three hundred sixty five days. And so they do the same thing with ports. It's like, okay. Well, we're gonna we're gonna ping on random ports or common ports or whatever,

and I'm gonna look for your response on this one, and we're gonna keep trying them until we just happen to have an an attempted handshake over the same one. And then we're going to essentially allow across through that port,

and then boom, you can establish connection. And shockingly, you can actually do that even though there's 64,000 ports on a computer or on a network. You can do this in just a like, a couple seconds, like, one or two seconds. It like, it's very, very fast because computers are fast.

And so they've basically created this method and and unless there are, like, insanely restrictive firewalls and, like, communication standards or whatever, there every once in a while, there's, like, a super, like, VPN behind a VPN thing that, like, it's, like, you're not on a very friendly network. This is not gonna work very well.

But, otherwise, it just it just connects. And it's kinda wild that it works that way, but it's not got the horrible latency of Tor because it's ClearNet. It is encrypted because you're actually identifying each other by your key.

So it it like, all communication is literally through your public key. So it's like, you know, sending a Bitcoin transaction is you don't have to worry that if they locked it to your key, you don't else is gonna get the Bitcoin.

Same thing with the communication in the PearStack. And and it just it just works. It's fast, super responsive, and super quick. And it's kinda wild because not much does that on the Internet.

And is that a one time thing with each specific peer, like that kind of, you know, where we we choose random ports and so we connect. And then once, you know, once my computer's found Guy's computer, that's it. We it's kind of locked in? Or is it is that, like, dynamic thing that repeats itself?

In unless IP unless you just don't get a like, unless the IP address and stuff changes. But if you change the IP address like, I did testing with PearDrive after we updated or I can't I can't remember what the update was. It was driving me crazy.

I think it was something with how we were using HyperB. But, I was like, okay. Like, how do we deal with how do I make sure, like, I'm sorting out, like, local network changes and IP address changes?

And, like, if you like, I I would disconnect, have it stop in the middle of a file transfer, disconnect, log in to my VPN, reconnect, boom, starts right back up where it left off. I disconnected.

I, literally shut my computer, opened it back up, reconnected, continued where it left off. Then I, then I specifically, renewed my lease with the with my router

so that I had a new IP address, and I went from, you know, one seventy two to two forty eight or something like that. And boom. Reconnected and picked up where it left off. So it it I I don't know the intricacies of all the little details of how they have done this, and the I think this would be the

HyperSwarm part of it. I'm not even sure which networking piece they this would be in, but the their mechanism just it just works. It just works really, really well. And it just refreshes if you have a IP address change.

And but otherwise, the data is saved, it just continues to try the the old thing, obviously, to be efficient. Like, it wouldn't it wouldn't make sense to do the complicated part over and over again if the computer's still there.

Yeah. I think maybe something to compare it to for people who are familiar with Tailscale. I know under the hood, it works differently than Tailscale, but a similar idea of allowing you to connect two devices peer to peer with NAT traversal in a way that has essentially no impact on speed.

Like you said, not you're not losing speed like you're doing a Tor connection or something, which I know a lot of people, especially in the past, have used Tor for this type of p two p app, which is good from the sense of it's very simple, but bad in the sense of the latency is just so bad comparatively

to something like a Tailscale or a whole bunch. So I for think those that are familiar with tail scale, that's a good comparison as well to think of it in a similar way, different under the hood, but similar way. Yeah.

It's basically tail scale without, like, a server doing assisting in the connection. Yeah. Yep. Yeah. No coordinator. Mhmm. What does identity look like on on the when we're using this stack? Like, do I have, like, a singular identity?

Is it different across my different devices? Do I have anything to back up? Like, what what does that look like for for a user? So this is this is actually totally up to the user. So right now, the way it's done is every instance creates a new public key, but everything is public private key pairs. Everything.

So but then the network, it's like, the the they're referred to as topics, but, essentially, the the group or the I guess you can think about it like a a live torrent that you're connecting to has its own, like, topic key.

So you can access the the network from a bunch of different devices even though your devices have different keys as long as you have key to the the conversation, so to speak, online. And and that can be public or just you privately hand it around to people so that they can join it.

But everything everything is public private key pairs, and every device, like, instance that you boot this up creates a new key pair. However, if you want a persistent kind of, like, global identity, you would create a key pair, and then you would generate subkeys from that, which is also kind of a simple

application that you could do in this, in this whole stack. But you can basically create it however you want. So it's like you have public private key pairs. How do you want to manage this?

Do you want to have a global identity? Well, you can build your app this way. Do you want to just have it where every device is just a device and everybody connects and we're just kinda handling things by the conversation keys? Well, then you don't have to do any of that.

So it's really just up to the builder because the the protocol is kind of agnostic to whether you want an identity or not. You have a public private key pair that can be an identity and is a device, or you can just refresh it on your device over and over and over again. It doesn't really matter.

It's really just how do you build what do you wanna build with the app? Just like the web. Like, you can use a website without users. You know? You can you can create one that has accounts, create one that doesn't everybody's anonymous, Fortune.

So, essentially, you do the the same thing here. It's up to the builder how they wanna use the protocol. But if you want a global identity and you want a a basically a backup, you would do a you would basically have a seed phrase just like you would with Bitcoin.

And everything would be connected to that that master key. And if the data was still available out there on other people's devices, like other conversations, and you put that master key back in, you would be reconnected to those conversations, redownload the messages, data, whatever it is.

Yeah. Kind of, circling back a little bit to the the peer to peer approach. I I think something that's always jumped out at me and I've always kind of wondered about I actually, like, I don't necessarily always

want my apps to be peer to peer. I actually Of kind of a big fan of the Yep. The end to end encrypted, but centralized server infrastructure approach that Signal, NotesNook, Proton, etcetera, a lot of really good tools in the space use.

And I think there's a lot of benefit of, like, leveraging the the privacy and censorship resistance that you get from the intent encryption with the centralized server infra that gives generally a better user experience.

Like, why in the pair stack was it seen as necessary to go p two p only? Like, that is what we wanna do. Like you said, I know you you could build something that has some sort of server infra as well, but the whole approach from the ground up is built to be peer to peer. Like, why is that seen as necessary as opposed to

centralized server info that because of the end to end encryption, because of the privacy that's built in, can't actually see anything and thus cannot perform any targeted censorship? Obviously, they can't do just complete white list or black list censorship, but that's a whole a whole another side of things.

I think that's actually a misunderstanding because of how everybody, like, math and crew and everybody who's building it talks about it and basically how they they kind of, like, philosophize on what their goal is and specifically how key is built and all of that stuff.

But I think it's really much simpler is that there's actually no centralized servers aren't shunned in this protocol. There's there's nothing wrong with centralized servers are still totally possible. It's just that if you can build the entirely peer to peer version and the entirely blind forward and the entirely distributed data version,

well, then you can just take the benefit of centralized when the centralized provides a benefit. Like, from the context of, like, PearDrive is like, I have no problem with centralized servers either as long as I am not subject

to them. You know? As long as I am not dependent on them and they don't own my data and they don't own my ID and they don't own my network connections, well, then centralized servers are great. They're just service providers.

You know? Like, imagine if somebody could facilitate your package going from a to b, but then they never actually owned the package. And if they tried to open it up and steal your stuff, your package just disappeared from them and you still had it and you could send it to somebody else. Well, then I don't really care who does it.

And I don't care that it's centralized either because it's still mine.

So, like, actually, a really great way to think about it is Bitcoin. So when I use Phoenix Wallet or

Nunchuck or whatever, I'm using a centralized service for my node unless I connect my own personal node. I'm using Nunchuck's

node to determine where my Bitcoin is and my transaction history and all of that stuff. And maybe I connect over Tor. We have, like, a private thing so that like, I do that so that they don't know exactly what I'm requesting

or use Nutrino filters or whatever. But they're a centralized service, and they're providing me with that service. But if I don't want to use them, don't have to. They're just there to make it easy.

Just like you said, it's a much better user experience if it just plugs in and it just works. Well, this is the exact same, is you can build and it still sometimes has a lot of drawbacks and it's still much more difficult to do it this way.

You can build totally distributed, totally peer to peer and everything, but you can also just have somebody set up a server and do the NAT traversal, and everybody connects to that from your service.

But the big thing is that if you still build things as topics, you still have it so that the network connections and the, keys, the identities are, you know, built into a hyper swarm. What you're doing is you're just helping to make sure that the network is always there. The data is always there. And if you go down,

everybody can actually still connect. Everybody can actually still do the peer to peer way. Everybody can still get it the peer to peer way. And if somebody if not all the data is there because you were everybody was leveraging you, well, then some people lose their data, but until you come back online.

But it's just centralized is totally possible. And and that's how I think about it with PearDrive is that I would love everybody to be able to peer to peer sync and just easily connect and download stuff between their devices. But there's also nothing wrong with

Pear Drive or you or anybody else putting up 16 terabytes on the network and saying, can I can I pay you or can you pay me if you wanna pay me $2 in Bitcoin or whatever, I'll I'll let you have a 100 gigabytes on my computer and offer I'll offer it as a centralized service?

That there's nothing wrong with that. Absolutely nothing wrong with that. So Yeah. It's it's super helpful to realize that it it can be a mix and match thing. And I think especially your point at the beginning of, like, having the option and the functionality already built out to go with the hardcore, peer to peer only

way gives a lot of freedom to developers and users to choose, like, what's best for you. If you are super hardcore and you never wanna touch a central server and you can choose not to, that's pretty awesome. And if you want the best user experience and you're willing to sacrifice

something for that, you could use the central server info that someone is providing. That's a that's a good a good good way to think about it, and I had not realized that from the way I've heard people talk about this this deck before. It's helpful. It's definitely confusing because I think the diehards the philosophical diehards don't

wanna wanna sell it in this. Like, it's all peer to peer. It's all decentralized, and that everything's distributed. But it's like there's no reason for that. Like, the benefit is that, like, now you can just host something with NAT traversal and easy connection.

And, like, I can easily I do it all the time. I'm I'm doing it right now. I host a few services on my Linux that are totally centralized, and I have all my other devices connect to them and use them. I do, like, a bunch of AI stuff. I do an LLM because that thing's got, like, a hell of a GPU that my other devices don't have.

And I just log in and use it. And I do it over wholesale, which is which is tail scale, but the pair version. And that's it. It's a it's a centralized server that I'm just hosting myself, but I can let you log in to it too. I could just give you my key and you could log in to Olama on my computer, and you could just use it to, you know,

chat GPT, whatever you wanted. But I don't want you to do that because I'm I'm using my GPU. But but, yeah, it could be centralized or whatever. It's it's it's totally up to you. Nice. We've sort a lot of protocol stuff. I wanna kinda bring it to life from a a user's perspective.

We've kind of danced around the fact of, you know, there's some some popular apps that have emerged, built upon the PairStack, The most popular among which would obviously be Keap, the messenger app. We've also got Pearpass and Peardrive, which I wanna dive into as well.

With with Keap, like, we we've kinda spelled out the the obvious that, you know, there's no centralized servers. It's it's peer to peer. I I presume it's encrypted. When I've been listening to to other podcasts where you and and some other guys have been chatting around,

something that was interesting to me was around the the unique scaling aspects that the PearStack offers, particularly around, like, message rooms and file sharing and things like that. Could you maybe just dive in a little bit more and kind of put some meat on those bone for us as to, like, what the the benefits are there?

Okay. So there's a if you go the kind of, like, distributed data route and everything is a topic so so you can have it where well, actually, by default, when you create something, it's created as as your key. And so, like, you're the owner,

and then you can constantly update it. It's really much it's very much like a feed. Like, when you subscribe to a podcast feed, like, in Spotify or whatever, you can't edit the Spotify pod you can't edit the podcast feed.

And if somebody else connects to it, they're seeing the exact same thing. It's owned by the host. Right? It's owned by the person who made the feed. And then you can connect to a whole bunch of different feeds, and you can do whatever you want. And those feeds can go wherever you want. You can do them through Apple podcast

app. You can do them through Castbox. You can do them through Fountain. You can do it through Spotify. It doesn't matter. It's the same same feed everywhere. Those are just windows into looking at them.

Well, PairStack works very similar to this conceptually. And but there is an issue is if your host goes down wherever you are keeping it on the the Internet, like mine is Castos right now, if Castos went down, my podcast would be gone.

Even though people listen to it on Apple Podcast, Fountain, Spotify, Castbox, whatever, the source is down. And and what's funny is that, like, tons of those people, you download the episode. Like, it's there on your phone to to listen to. Like, it might be in 20 it might be on 20,000 different devices,

but because the central place is gone, the whole thing is nobody else can download it. And then there is the hug of death. You have a website on a really small you have a really small website, like a blog, and, like, you have an article that just, like, boom. It hits home. It goes viral.

You know, you got a digital ocean, smallest droplet that you can afford at $10 a month or something. And,

a million people go to your website trying to read your article, and the thing's down after the first 10,000 try to connect. It's just dead, and you're scrambling. You're like, how do I spend $800 from AWS to get a bigger server? Oh my god. This is such bullshit. And everybody who wants while your content

is in its height and it is most desirable and everybody wants to see it because it's news, you nobody can get to it. It's it's it's dead in the water, and you're scrambling to try to keep it back up.

Those two things are problems of centralization. Like, just because of the architecture of how the data is built, and horribly inefficient, it's horribly fragile. If my podcast goes down in the pair sense, everybody has joined a topic. If I, as my host, go offline or Castos goes offline, it's still out there

on, you know, 20,000 different devices.

And if you can't connect to me, it just connects to somebody else and downloads from them. So somebody else listening to the podcast will feed the podcast to you. Like, it just ex just like BitTorner. Right? If 10 cedars go offline, but there's still 900 cedars left, you just download it from somebody else.

And because of this, the scaling properties are incredible, especially for something like a just a simple website with an article. If a million people are trying to download it, that's fantastic for your up uptime.

That just means that it's in a million different places and a million people are providing it. Your your website, your tool, your whatever gets faster and more responsive, and it's easier to download. Like, a a a movie with 10,000 seaters is, like, the best thing ever. It's like because now I know I'm gonna get it at insane

speeds, and I'm gonna be able to watch it live if I want to. But if there's only three seaters, well, then now we're all bottlenecking on just those couple of computers trying to download the movie. But every new person that gets it makes it faster.

So this is this is this is actually utilizing the beautiful nature of the Internet and how great it is in in a in a way that's self reinforcing. So popular content is just more available, and it doesn't it doesn't cost you.

Like, you I don't I don't have to figure out how to get, like, a AWS server to handle 1,000,000 users. In fact, don't even have to be online. I can literally go to sleep, cut off my computers and everything, and the the content can live by itself.

But the big thing is that, like, you especially

what I think the the right target is for something like Pear Drive is startup costs, man. Like, you think of, like, a Nostr client or whatever. Like, some people some people are hosting these relays and thousand dollars a month just in traffic on a small network. Like, most are most are awesome, but, you know, a 100,000 users,

daily active users or whatever, that's not a big network. That is not much. And this is eating into the ability for people to run nodes or run relays.

And that's a huge problem. Startups can't afford that, especially in today's environment. Well, what if you could just host some what if you what if you could do something with zero server costs?

I mean, other than just, like, having something online at your house or something. Like, like, I used a, I think it was one of Cali's.

Crap. I don't remember. It was a it a somebody was hosting, a lightning service, and they were literally doing it from, a a dual, like, Raspberry Pi thing in their home. And, like, I was I used it for it's like a tipping service that I used for, like, a year.

And, like, you could literally do that with any service or or tool or or data, and you don't actually have to handle all of the users. The users can help the other users in a peer to peer way. And that's just I think that's just incredibly powerful. Like, you could you could make a game where you don't have to have, like, a game platform.

You don't I guess that's one of the things we're attempting to do with the Pear Drive core engine just to show, like, the breadth of what you can do with it. Yeah. That'll probably be the first thing that we build with it just because I think when people hear, like, oh,

file sharing or file syncing, they're like, oh, it's definitely just another Nextcloud. It's like, well, that's that's what it looks like if you don't think about what you could actually do with this thing. File sharing and data syncing is the Internet.

Like, it it's everything. Can you can you dive in a little bit to Pairedrive? Sure. Like, I I understand the sort of the basics of, like, these are almost like applications being built on a whole bunch a whole punch network or protocol, but how do these differ from what else is out there?

Okay. So the idea really is for all these various tools. Like, our ours isn't really different. We're just I'm just trying to tackle a a very specific problem in a very targeted way. I hope I do it well.

But the the idea is, like, Keyt specifically is a chat app. Right? And one of the things that I I feel like I have felt like for a long time they should have just done file syncing first because a chat app is ten ten thousand times harder to to get right and to get all of the organization and the ordering of the conversations and stuff.

There's way more involved. Whereas, I felt like the kind of lowest the lowest hanging fruit was get files to sync between my devices. And so that's what we've been focusing on. And

it it it's utterly simple. Like, you know, how difficult how difficult is it to move files that are, you know, more than 10 megabytes from one device to another, especially if they're not on the same platform. Right? Is it sometimes I get the benefit of Apple, iCloud,

or AirDrop, but I swear to god that shit works half the time. It it breaks as often as it as it works. And it's like Bluetooth. Like, swear to god. I've yesterday yesterday, I tried to move a video from my phone because I still don't have Pear Drive on my phone.

But I tried to move it from my phone to my MacBook, and I'm just and I'm and there it is. I select it, and it does this little spinning wheel, and it says, can't connect the device. It's like, why?

Why? Fucking why? I don't get it. It drives me crazy.

Like, this has literally been the bane of my existence. I have so many data from one computer to another. I lost a whole another I lost a whole day yesterday because I ran out of space on my computer again. You deal with a lot of media, everything's 20 gigabytes. It just it's just a pain in my ass.

But PearDrive is like, how could you make stuff seamless across devices, and how can you leverage that device that has a lot of storage to get the most out of the device that has a little storage.

And iCloud really tries to do that, but it does it in such annoying way and in such a frustrating way and in an unreliable way, and and in a totally centralized way, which I don't like either.

Mhmm. So PearDrive is like, how do you fix that? How do you make it so that when I boot something up and I got all my photos, it just syncs it just syncs to my Linux. Like like, if I just have it open, it's it's it's putting it on my Linux machine. And if I delete anything from my phone or my MacBook,

I still see it. I still see the thumbnail and I still connect to my Linux to bring it up. And I can I can even play it directly off of my Linux machine? And if I want it locally to edit with or to share with somebody or whatever, I just hit download.

But, like, how how could you actually make that experience just feel like make it visually aware of where the data is and make it feel like the data just exist everywhere at once. Mhmm. That's that's what we're trying to do. But then in addition, is that if I add you as a friend, you know, you can't I I can't send a a one gigabyte

movie file. Like like, I we tried to we recorded a hold a lot Can on Telegram. The other night. It'll be super compressed if it's a video. Yeah. But and it's also, like, you know, again, if Telegram goes down or Telegram decides that, you know, everybody knows the story of this Telegram founder and what happened with him being

arrested for, you know, letting people have private conversations. So I I guess he's not arrested now because we are still having private conversations. Probably not. Mhmm. But,

again, you're dependent on a centralized platform. But, you know, if I just wanna share we go a trip together. It's like, you know, Christmas twenty twenty five. And, like, I I just scan a QR and I add my family and all my family or my friends or whatever. It's like a group text. You know? It's like, well, why can't I just

why can't we just group photo share? Why can't we just group group share stuff on my Linux machine? Why can't why can't I group host a service? And, like, there is this, like, huge gap in how the Internet works. And file sharing there's this, like, big black space where, like, file sharing doesn't work.

And there is also this mode of interaction where everything is social or everything is, like, publicly social or everything is, like, super just direct. And there's this whole missing piece, I feel like, of

of of just file sharing. I can't I can't tell how many times I'm trying to get stuff to my designer or my producer or whatever. And I'm like, okay. We'll try to do it on mega. We'll try to do it on iCloud or whatever. And it's just scattered. It's just a mess, and it constantly

is. I have to upload it, and then he's like, I there's nothing in the folder. It's like, it's still uploading. Give it give it, you know, ten minutes. Because everything's bottlenecked by the 10,000 other people using their service right now.

And then they have to download it. But, like, what if he could just look at it on my we connect directly, and the fact that it's on my computer is his access to it. Would that I'm I'm very, very nontechnical, so this might be a stupid question. But when you're talking about this file sharing between

friends or people that you're working with, let's say that I shared a video with you. You're in a a group chat, and it's one of my podcast videos or something like that. Mhmm. And then I get mugged. I go to London, so, obviously, I get mugged. And then I lose my phone. I I I lose my laptop.

And they're the two devices that I had that content on, and I have no backups. Does that mean that now that I've shared it with you, you have it on your device because you've downloaded it, and therefore, I could get it back?

Yes. Yeah. If somebody has the somebody has the content still, like, you shared it with me and I downloaded it, well, then when you booted yours back up, I could just give you the key that you gave me back, and you could connect right back to it and then get it back from me.

So the and, like, if you had a third of the information, I had a third of the information, and Seth had a third of the information, And Mhmm. One of us, you know, lost one of our backups or whatever. Like, we can still boot back up, and all the information is there even though neither none of us has the

none of none of us has the full amount. Right? But between the three of us, we have the full amount. So is and that's just as long as you have the key or as long as somebody can still give you the key.

So in, like, a network situation, I can just give you the key. Mhmm. But if it's an encrypted thing, obviously, I could give you I could give you the encrypted blob. But if you lost the decryption key, that's we're both screwed.

Mhmm. Okay. So in an ideal world, like, if someone's running this and relying on this for for everything, if this thing builds out, you'd maybe consider, like you'd have, a home server, and then you'd have your other, like, your phone, your laptop, whatever, your other devices, and you're storing most of your stuff on your server.

But if if that went down and, god forbid, your other devices went as well, you could most likely get a lot of the content back anyway without relying on, like, an iCloud or of these things. That's pretty cool. I like that. But

then in addition, you could also just wrap an iCloud like service into it as well where you can just say, back up everything that's on my server, and you could do it in a specifically blind way

where Like a Yeah. It's just an encrypted blob on their computer. All they know is that there's a 180 gigabytes here. They can't they have no idea what it is. They they have a key that gives them access to the not not the content, but just to the communication layer.

And so they have it backed up. And if I lose my server, I can still just have a network backup, and all my other devices can boot back up, decrypt it, look at it, and it's like nothing happened.

But that's a service. That would be a centralized service that you would do in addition to the fact that you could just have a home server that held it all. And you can also have your brother host a home server. Like, I'm I intend to when when I get my full on thing set up, my brother's gonna host a copy in a different country,

and he's gonna have I'm gonna send him hard drives if I have to, and we're just gonna have it in two different places. And we'll host it ourselves, and I'll have his backup, and he'll have mine. Why not? We're we're we trust each other. We he would be in my my friends and contacts list. Why would you not leverage

other people and your friends to to have a backup for free? Tit for tat. I back up yours. You back up mine. Yeah. Definitely definitely love that concept.

I I I think one thing I wanna kinda circle back to, I know Q has a couple of quick listener questions that we can answer as well. But, like, for the person who is wanting to use this stuff today, like, how close are we to this being a boring, it just works

everyday thing for a normal person? I know a lot of it can we can get kinda get stuck in theory, but I'm curious, like, where the state of the actual end user apps are today in the Pear stack.

There's still there's still friction. Key works exceptionally well, but they change and update stuff so much that it's it's got a bug every once in a while. Like like, one of my rooms yesterday

just, like, stopped loading. There was, like, some sort of a rendering bug. And then, like, I hit math up or whatever, and he was like, oh, I think I think we found it, and they pushed an update, like, within an hour or two hours or something, and it was back up and running. But they they are doing stuff a lot,

and because of that, it gets a little bit buggy, but I'm constantly I use this as my daily driver. Like, this is, like, my main work thing. Like, all my all my business stuff, all my work chats and everything, and then tons of, obviously, of just, like, fun chats, like the Dank Bank, our our meme group,

is is is on key. And this is this is my main one that's not, you know, Twitter or Twitter Telegram. Twitter, Telegram, and Key are it. So it's in in my opinion, it's ready aside from the fact that all of the end user tools haven't been built.

The protocol is ready. The protocol is user friendly. Like, PearDrive would be user friendly if this just wasn't, like, a huge mess of realizing we built it kind of wrong

and then rebuilding it again and then realizing there was something at the bottom that should've been the that we should've built first and then built the other stuff on top of it and then rebuilding it again. Like, this this has been the process of

it it's a little confusing when you first work with it. It's got a lot of different work moving pieces. And, you know, if you don't realize which Lego blocks should be on the bottom, you figure it out when you're halfway done with your tower. And then you're like, son of a bitch, you have to start over.

So that's been the story of BearDrive. And also realizing there's some there's some oddities in how it treats things. Like, when when you share a file into the network, like, as a as a hyperdrive, you actually your computer actually makes a complete copy of it into what's called a core store

so that the the files are ready to be delivered. But that means that you lose if you want to share a 100 gigabytes worth of stuff, it takes up 200 gigabytes worth of space. Mhmm. And that was basically a nonstarter for us when we were trying to figure out how to do this. So I

so again, that would we built it the first way, and then we realized that that's how that's how it works. And then we're like, okay. We have to rebuild it again. Then we have come up with a new method.

And so now what we've got is there's a hyper blob system, essentially. So we we use the network to create a communication layer, and then we we we send around we we sync a list, just the list, and then and identify everything by hashes just like BitTorna does.

So you can have a name you can name your movie matrix nineteen ninety nine or whatever. Mine could be the matrix dot m p four, but it doesn't doesn't really matter. We'd know it's the same file. We can do whatever we want locally different.

But when when we're syncing that, rather than creating an entire core store for the file to sync, we actually create a an empty, quote, unquote, HyperBlobs, like, package. And then I just start feeding pieces into it, you download the pieces So that we've essentially eliminated a very frustrating and difficult problem.

But, again, we've had to rebuild this thing four times, I think. I think I think this would be the fifth maybe. I don't I don't know. But there there's still some some things to do to to get the user experience of Pear Drive where I want it, and I have high standards because I hate shit that doesn't work.

So but when I am even in the messy, kinda clunky state that it is right now, I'm using the the proof of concept version, which is a command line tool. And but when I yesterday, when I was dumping stuff from my MacBook, I moved it with Pear Drive,

because it's just the easiest way to move 60 gigabytes from one computer to another right now. So it's getting there. It's just been a very, very painful and expensive process to to find out how not to build it.

Yeah. Things are a lot simpler when it's centralized and doing things in a a more normal, more common way, but that doesn't mean it's not the right the right approach. We'll definitely follow-up Specifically, when a 100,000 people have already built a version of it and you can just kinda do it the way they did it, this is fresh.

Right? Mhmm. Like, Kite has just been built and Kite's not even fully open sourced. All the base and, like, protocol and everything is open sourced, but kind of the the front facing, how they organize and and

align things is not like like, the the the front app is not open source. Pearpass is fully open source, but it's been out for, like, two weeks or something like that. And Warefam is open source, but it's also been I had been out, like, a week. Like, there's just not any examples. So I'm I'm going by we're by docs and stuff. And so, like,

you know, when you're the first one in the woods, you have to hack a lot of weeds Oh, yeah. And cut down trees. And that's really kinda thing, the big problem. But after you've got, like, 10 good apps built on this, I think it's gonna be so much easier.

And one of the core things with PearDrive is our PearDrive engine or PearDrive core is going to be a simple import and then have a have a basic, like, web RPC type interface so that, oh, you wanna do something with files? Well, here's kind of your primitives and just use the commands. We want amateur developers

doing very simple things to just not have to think about the peer to peer stuff at all. And so there's two pieces of this. There's the app and then there's the core. And we are trying to make the core so that other people can build stuff with Pear Drive

and not have to think about the shit at all. So that that's also a reason why we have to build it very, very specifically, and we have to keep rebuilding it when something isn't right because it doesn't we we could band aid it and just make the app work,

but we can't do that and and then offer it to other people to build. And then they'd be like, why is this super weird thing happening or these extra resources being used? It's like, yeah. Sure. Just modify core and do something else with it. No. That needs to be solid and reliable and very easy to understand. So, hopefully hopefully,

we can be a very important part of making this incredibly user friendly and easy to build with. That's that's a big part of the mission and why we spent so much time on it. So

Awesome. Well, I I'm definitely looking forward to to getting my hands on it. We are at time. I just do I do wanna give you two very quick fire listener questions that were pre submitted.

And it's almost like you can see the back end show notes here because you've already kind of touched on both of them, but we have a quick question from from Schrodinger on Noster who asks about the the open source nature of the code.

I know you briefly touched on it there, and I also wanna caveat it by saying that you're only working on one app on this network, and you you might not have any further information as to the kind of open open source

nature of of, like, key to etcetera. But is there anything else you have to add on that one or any any alpha that you have? It's just Kite. It's just Kite is not open source. Everything else is open source. So, you know, if if that's like a big thing for trust, which completely understandable,

I would just wait. Granted, they've been on the we're gonna open source it. And the reason when I talk to Matt or the crew or whatever, they basically say we're making so many breaking changes, and we want to kinda have primitive so that other people can build chat apps like Kite.

And so their reasoning has been, like, if we release something and then make a breaking change, none of this is compatible anymore. And now 20 people who built an app with it have to have to rebuild, you know, have to retool.

And so that's been they don't feel key is done. It still it still says beta up in the top right. And I think that's what they're waiting on, but, I mean, you know, it's been it's been quite a while. I wouldn't hold my breath.

Yeah. And maybe they don't open source it. I don't I don't know. So if that's a big sticking point, then, you know, just don't worry don't worry about Keap. I think it's cool and it's great, but everything else is open source. Pear Drive is gonna be totally open source too.

Pear Pass, which they've also built, is totally open source. Warefam, I believe, is open source. Wholesale is totally open source. So there are open source examples now.

Kite is literally the one that they're still working on that isn't. So Okay. Cool. I'm glad you mentioned Warefam because the the final listener submit question from Tam Tam Bam on Noster

asked what when is Warefam going to get an Android release, if you're aware of that? And maybe just a quick thirty second elevator pitch on what the hell Warefam is as well would be be useful.

Okay. So Find My is giant centralized and Apple has every data point on everywhere you've been and that sucks. Where fam is, let me just share my location directly with you, and please

you know, granted Apple can still track you if you're logged in and, you know, all that stuff. But but my I should just be able to control who sees my location. That's what Where fam And it's dirt simple. It was actually part of a hackathon that they invited me to be one of the judges for, and I absolutely

chose it as, like, the, oh, shit. This is great. He was the winner of the hackathon. I can't remember what the reward was, but it's like it's dumb simple. It's like, you should bring up a map, you share your friends,

and then you can see where they are. You can, you know, say turn it on and turn it off. And so, like, you're in a museum together and you wanna split up. And it's like, where you guys are?

Very, very simple conceptually, very, very simple from an app. I have zero

idea when an Android release is. I had been out of contact, and I kept searching for it. It's like because I literally, when he showed it to me in the hackathon, I was like, literally, dude, I will buy the I will spend $5 and buy this app right now if this is available because this is just cool and it's simple.

Love single purpose apps. But it was not until very, very recently that it finally popped up. And I was like, hey. And downloaded it, and got to mess with it for a minute yesterday. So Android, not a clue.

Not a clue. I have not even talked to the guy since the hackathon.

Alright. Okay. Well, Guy, this has been absolutely wonderful. We are a little bit overtime. I just wanna say thank you for for sharing your your your knowledge with us. It's been very useful for me, and I'm sure for the listeners as well. It's been a Oh, yeah. A great albeit very, very fast hour.

Thank you for joining us, Max. Thank you for for for for welcoming us back into 2026 and for everybody to stopping by in the live chat. But a couple of boosts over on Nostice, so thank you for everybody in the Zapdot stream for sharing us with some sats.

And, yeah, thank you again, guys. And we will be back at the exact same time next week for more Freedom Tech Chat. Have a good one. Good hanging out, dudes. Bye. Later you go.

Thank you for listening to Freedom Tech Friday. To everyone who boosted, asked questions, and participated in the show, we appreciate you all. Make sure to join us next week on Friday at 9AM EST and 2PM London.

Thanks to Seth, Max, and q for keeping it ungovernable. And thank you to Cake Wallet Foundation and my Nim Box for keeping the Ungovernable Misfits going. Make sure to check out ungovernablemisfits.com to see mister Crown's incredible skills and artwork.

Listen to the other shows in the feed to hear Kareem's world class editing skills. Thanks to expatriotic for keeping us up to date with Boost's XMR chats and sending in topics. John, great name and great guy, never change and never stop keeping us up to date with mining news or

continuing to grow the mesh to Dell. Finally, a big thanks to the unsung hero, our Canadian overlord Jordan, for trying to keep the ungovernable in check and for the endless work he puts in behind the scenes.

We love you all. Stay ungovernable.