Dec. 6, 2022

CD82: Bitcoin Lightning Privacy Research + Browser Based Mutiny Wallet with Ben Carman, Paul Miller, and Tony

The player is loading ...
Citadel Dispatch

support dispatch: https://geyser.fund/project/citadel

EPISODE: 82

BLOCK: 766044

PRICE: 5906 sats per dollar

TOPICS: lightning privacy tradeoffs and concerns, building a lightning wallet that works in any browser, app store limitations, auto update tradeoffs, implications of artificial intelligence

GUESTS: @benthecarman, @futurepaul, tony

twitch: https://twitch.tv/citadeldispatch

youtube: https://www.youtube.com/@citadeldispatch

bitcointv: https://bitcointv.com/video-channels/citadeldispatch/videos

podcast: https://www.podpage.com/citadeldispatch

telegram: https://t.me/citadeldispatch

stream sats to the show: https://www.fountain.fm/

join the chat: https://matrix.to/#/#citadeldispatch:bitcoin.kyoto

Transcript
Unknown:

Happy Bitcoin

Monday, freaks.

It's your boy, Odell, here for another Ciel Dispatch.

Ciel Dispatch is a live interactive

show

focused on Bitcoin and Freedom Tech.

Try

and come live to you guys every week. Sometimes

it's longer in between. Sometimes we have multiple

dispatches

per week. This week, actually, we might have another dispatch this later in this week. I'm not really sure. We have a great,

great group in town in Nashville

this week for the beef initiative, Tennessee beef initiative. So

we might have a rancher focused,

CIL dispatch

in the next few days.

I'm not really sure on the planning on that, but we will see.

You guys know the deal.

Dispatch

has no ads, has no sponsors.

We have no paywalls.

We're funded purely by you guys,

with Sats.

The easiest way to fund the show with Bitcoin with Sats

is podcasting 2 point o apps, such as Fountain Podcasts,

Breeze,

Echo ln,

Podverse dot f m,

and many other podcasting, you know, enabled apps. Podcasting 2.0 is very similar to your normal podcasting experience.

You simply download the app. You search Siddle dispatch. You press that subscribe button.

You choose how many sats per minute you think the show is worth to you, and then that value gets streamed directly to the Citadel dispatch note.

You can also do something called boostograms.

Boostograms are flat,

you know, lump sum payments

that you send and you attach a message to,

which is really cool. It's like a comment section that everyone can read that has value attached to it, which is really awesome.

Also, I think 2 weeks ago,

we added dispatch

to geyser.fund,

which is a Kickstarter like platform

that uses Bitcoin as the standard.

Cool thing about geyser.fund

is that

it uses

I I can use my own node with it. So I use the Citadel dispatch node, I hook it into geyser.fund.

It acts as a live leader board, so you can see

the stats streaming in minute by minute just like I can on my note.

You can see the boostograms come in with the messages.

You can also go to that page, and you can donate

directly there instead of using a podcasting 2 point o app, whether that's on chain or lightning. You can include a message. You can include a name. You can even link your Twitter account if you wanna be on the top of the leaderboard

with your Twitter account, or you can do it anonymously.

Now

I wanna thank you, Freaks, because first of all,

you got dispatched to the top of the fountain podcast leaderboards. Right now, it's sitting at number 2, but it was at number 1 for a little bit, which is pretty impressive considering Adam Curry,

is is the competition there, and he's he's the father of podcasting 2.0. So thank you for that. And then the second thing is you guys happen to break geyser.fund.

They have never seen so many streaming payments coming in

to their platform before.

So right now, if you go to

yep. Ben is telling me that Adam Curry is also the father of podcasting 1 point o. That is correct.

So if you go to satsperminute.com,

which is the redirect to the geyser.funpage,

you'll notice right now

that it's actually not loading correctly.

They're working on it. They think they'll have a fix out

by midweek,

but that's thanks to you freaks streaming so many sats per minute

to Ciel dispatch. So thank you.

One good thing to note is,

yes, geyser dot fund is down because they can't display it, but that's just a read only website.

All the stats are still coming to the note. The payments are still coming through. The total is still updating.

So I appreciate all your freaks who continue to support the show. Like, you you guys are amazing. Couldn't do without

you. Super special seeing the sads flow into the node,

and

it's great it's great seeing seeing the haters proven wrong because a lot of people think value for value can't work. A lot of people think

that the ad model is the only way.

I think they're

wrong. I think you guys are proving them wrong,

and we're gonna do everything with dispatch going forward as radically transparent as possible, so you will be able to track it live on that page.

I know it's a bear market.

If you don't have sats to spare,

easiest way to support the show is

to subscribe to it in your favorite platform. We're on YouTube. We're on Twitch. We're on every single podcast app. We're on Rumble. We're on Bitcoin TV.

Subscribe to it, review,

listen to the show, share it with your friends and family. And of course, last but not least, join us in the live chat. That's what makes the show truly unique.

In an age of AI

and fake news everywhere where you don't know what to believe and what's real,

Dispatch is a live, unedited, raw show where we have

this interactive live chat that's just broadcast to the world. You guys are all part of the show together with me. So I like to think I'm not the only host. I like to think

there's thousands of hosts, and that's all you guys who who come in week in, week out and join us, in the live chat. Reminder for the live chat, you can join either through

YouTube or Twitch. That's what's on the right hand side of the screen. In the middle of the screen is our matrix chat. You can find a link to that and link to everything else at sail dispatch.com.

Now,

last but not least, I'm gonna read the top boosts

from last week.

We have BTC NIM with 50,069

SATs

saying 50 k boosts are the on air sat standard. I'm happy to have stirred that pop. The freaks are making it true.

Chad Farrell with 50,000 sats. 50 k club, great grip. We have Eric 99 with 50,000 sets. Stay humble, stack sets. Great advice, Eric. We have Thydicus with 50,000 sets. Let the chaos continue.

Appreciate all the quality content, Odell. And we have CA Danner with 33,333

sats saying, I wish I could be at Bitcoin Park for the Tennessee beef initiative, but my paintings are there at Bitcoin Park instead, and I'm there in spirit. Shout out to Texas Slim and KNC cattle.

Proud that I saved a friend from Block 5. Proud of you too, CA.

And, yeah, just a reminder, we do have the beef initiative this week, Tennessee beef initiative,

on Thursday at Bitcoin Park. Wednesday is a meetup open to the public for free. If you're in the area, come down to Bitcoin Park. It'll be a fun time.

On Saturday, I'm speaking at the US Cattlemen's Association.

Gonna be surrounded by a bunch of cowboy hats.

Gonna be completely out of place, but I'm gonna

show them Bitcoin

and sovereignty.

Well, not sovereignty. I mean, they live sovereignty, but

I'm gonna show them Bitcoin at least and and and help and help them set up their wallets.

Unknown:

Are you gonna wear a cowboy hat?

Unknown:

No. No, Paul. I am not gonna wear a cowboy hat. I'm not gonna wear a cowboy hat. I'm not gonna wear jeans.

I'm not gonna try and pretend to be

a rancher. I'm just gonna be a humble Bitcoiner

answering their questions.

So with all that said, I want to introduce our guests.

We have

3 rider Dies that have been on the show many times,

basically built dispatch on the back of their friendships,

so I'm excited for them to join us again.

We will be talking about Bitcoin Lightning Privacy Research and their new browser based mutiny wallet.

1st guest I'd like to introduce has already introduced himself. We have Paul Miller here. How's it going,

Unknown:

Paul? Hello.

Things are good.

Unknown:

Thanks for having me. You're the most iconic hello. I love it.

Gets me every time. We have

we have,

Tony Tony g,

famous Tony of of lightning privacy work. How's it how's it going, Tony?

Unknown:

I'm good. It's always it's always good to be on here talking talking about lightning privacy and lightning

things. Thanks for bringing us on.

Unknown:

My pleasure. And we have Ben the Carmen, one of the best Bens in the Bitcoin space. How's it going, Ben?

Unknown:

Good. Good to be here with actual audio this time, hopefully.

Unknown:

Yeah. Ben's building if if if you freaks don't remember, Ben joined us,

for the

lightning development talk we had,

and, he could barely he could barely contribute because his audio wasn't working well. He's building a along with Paul and Tony, he's building a browser based

lightning wallet, but

he can't figure out

browser based voice calls. So we're working through it with him. Because Firefox is a shitcoin.

If you we're not gonna start with that. Okay. So let's start with before we get let's let's start with the exciting shit. Let's let's start let's start with Mutiny Wallet.

What is Mutiny Wallet? Why should people care?

Let's let's go over the basics.

Unknown:

Yeah. No. I'm I'm gonna start off, but

it's it's it's a browser based lightning wall. And and what's kind of exciting about it is we're using LDK

and importing it into WASM and having the entire node run straight from, like, the browser

window. So you load up a link. You load up right now, it's rekkles.muni

wallet dotcom. Definitely don't put all your funds there yet.

But you load up that link, and it's like a live lightning node running. You can connect peer to any node on the network like normal. You can open channels to them. You can do it all in, like, a noncustodial way,

that actually lives right there in your web browser. And, you know, if you can still, like, close your web browsers, restart computer and stuff, and launch it back up again, and all your channel state

is, saved there in in your browser storage, and, you know, you're kinda good to go. So, like, we're it works as a a mobile wallet. It works as, a PWA, so you can kind of a progressive web app, you can kind of install it

on your phone, you know, quote, unquote install it by just, like, loading the web page.

But besides that, it's just, like, we're kind of experimenting with having, like, the most

you know, putting the putting it in the web browser and having it available to, like, anyone in the world,

by just clicking a link.

Unknown:

So I have I pulled it up on the screen.

These are my 48154

sats that are are currently stuck in your wallet.

Because, also, as soon as Tony sent me the link,

Unknown:

I I loaded it up with SATS before he told me not to put too much money in it. See, I was typing the message explaining what you're you're like, woah, dude. Is this a is this a note in my browser?

And and I'm, like, typing out the explanation of it, and then you're like, oh, wow. I just put funds on there. And I'm like, damn. I was literally typing don't put

double funds on there.

It's still incredibly I mean, it works, and and we're definitely using it to test. And there's definitely issues,

that we have to work through.

We're we've been building this as part of the bolt off fund hackathon,

and we got it to the point where I said, well,

why not, like, have it on test or on mainnet and just, like, start playing around with it?

So it's still super early. There's there's a lot of things that we need to do to get it performant and and a lot safer than it is now and

work through the bugs where people can't withdraw.

But besides

that, it's like a super early experiment that we're excited to keep progressing on.

Unknown:

So, first off, I'm in settings.

I love I love the UX of write down these words or you'll die,

for the seed words. I think it really

it hits hard. People understand what that means.

So props on on that UX.

So let's just walk through this real quick. I basically this is a new wallet

that I just loaded.

See, now I'm in transactions

view.

So it was it was confirmed, you know, maybe

45 minutes before the show, so it's like a user just signed in and loaded it up.

To use it, I need to manually create my own lightning channel. Right?

Unknown:

Right. Currently at least.

Unknown:

Yeah. We're we're kinda we kinda went, like, depth first of, like I mean, we this is basically, you know, entire

lightning implementation, and we knew we needed all of that functionality. So you need to be able to connect to peers. You need to also be able to manage on chain funds.

You need to be able to create channels and close channels. And so we wanted to make sure, like, everything worked, and we built UI to exercise all of that. And I'm still in the kind of the glow

of, oh, man. We got it to work. And how cool is that?

Next is

or soon, maybe not

immediately next, but soon is

we need to zoom out and figure out how to make this, you know,

just absolutely

usable because it it is kind of a a a play for

making a really accessible wallet. We're cutting out some of the steps that exist right now for getting your own lightning node. So we want it to be very accessible. Hopefully, can,

either, you know, hide some of those things. Maybe those are, like, advanced user things,

automate some of the node management stuff, you know, getting that initial liquidity. Those kinds of things, hopefully, are not things that the first time user has to worry about. But, like, right now, the UI just basically

it's a it's a node administrator.

Unknown:

So a couple of things.

Unknown:

The the cool thing is, like, we

we did it, like, the most, like, hard way possible. We're, like, you know, it's not like Moon where it's using, like, submarine swaps and, like, fake lightning or,

like, using, like, hosted channels or anything. Like, it's all, like,

raw like, a real lighting node with, like, full functionality.

So now that we have that base of, like, we can do everything, we can, like, then, like, okay, if we wanna make, like, the super user friendly UX,

we can cut stuff out and, like, make the, you know, moon version of it or something or, you know, as in the current state, it could be it could be, like, the power users, like, full control lightning node. Like, because we did it this way, we now have, like, the ability to build anything on top of it because we have, like, full functionality now.

Unknown:

It's a strong foundation.

Unknown:

Yeah. Exactly.

Unknown:

Wait. So there is this is literally an LDK node that's running in a browser window.

Unknown:

Yep. And, and BDK as well.

Unknown:

Yep. The main thing that doesn't actually happen in the browser is it it's,

it doesn't run a Bitcoin full node for

you. So it talks to,

like, the block stream API

for block data, but you can

we've got a a pull request right now for being able to configure that URL so you point it at whatever electro serve you you want.

There is a there's a proxy

because the web browser doesn't natively speak like the low level TCP IP

that you'd wanna talk to a

a node.

Typically, it just the basically, the only way we can

communicate out of it is via WebSockets.

So we talked so we have this proxy

that the browser talks to the proxy, and then the proxy communicates,

to nodes on the lightning network.

It's nothing there's not a lot of, like, fancy logic in there. It's it's just a real dumb,

proxy. And so that is also something that will be self hostable.

And, potentially, like, we might make some sort of self hosting bundle. Like, if you want to be more sovereign,

with mutiny, you know, you could host the website yourself,

so you know that we're not swapping out the JavaScript on you. You could host,

Electris. You could,

host the proxy. I Feel like there's one other piece I'm forgetting that we also configure.

Unknown:

No. That's Yeah. I mean, I think the yeah. Just the domain itself, you can point it to your own domain. You could be like Odell's wallet.com

if you wanted to,

you know, potentially

Odellrug.com?

Yeah.

Rugpool.com.

But but, yeah, the the the proxy is and and then also storage eventually as well.

We wanna

since we're doing on this hard mode and we're really hitting, you know, probably, like, performance issues and reliability issues, keeping

data in the web browser. Right? Like, like, if you just, like,

clear your browser data, then you would, like, you know, delete you'd wipe your note, and that's that's not that's not great.

So,

there we already have this, save state button. So, like, at any point in time, you can click save state, and that backs up all of your your channel data, all of your wallet data,

your UTXOs, like, all of the information that it needs to run that it actually loads up the first time it runs. And then you can go to a new browser window and import that, and then boom. It should be exactly like it was before,

and you should be good to go. Now you'll still have to you know, it's not like a one and done. It's not like just, like, backup now,

and then you can you can use it anytime in the future. You do kinda need this, the latest state.

So there's there's all kinds of, like,

you know, storage issues that we're gonna have to solve as as we go forward, but also being,

doing them in a way where it's always, like, self hostable. There's always an option.

And and and, really, what we're talking about here is, like, the networking layer, it's it's like Paul said, it's, like, it's dumb as fuck. It's, like, taking some some packets of data that it's incoming and and and stream it outgoing to to some other node on the network and vice versa.

The storage is gonna be that way too, like end to end encrypted storage. Like, if you wanna host

a simple key value store,

then then you can host that any way you want it, on your Raspberry Pi or or server or, you know, whatever you want. Or use use ours by default,

and it's all in an encrypted, or will be. We haven't built that feature yet.

And then, you know, point your yeah. I mean, in theory, you can use any, you know,

WebSocket to TCP proxy on the Internet.

We just kinda, like, built our own and and open sourced it. So,

yeah, eventually, you can host the entire stack. We wanna be able to come up with, like, documentation on that or one click buttons of, like, hey. Here. Deploy it yourself.

You can occasionally update it so that you're at the latest state, but other than that, it should be, like, your own personalized

wall. You can even, like you know, we're we're even thinking about, like, you know, having a lot more Tor integrations as as well where you can host it behind, like, a Tor endpoint, and then you can access it via Tor.

It's kinda cool that the wallet,

already works on the Tor browser itself. So you can load it up in the Tor browser and have this, like, node. And let's say you just wanna use it for that session. You know, you just wanna do some, like, Bitcoin wallet things and then throw it away whenever you're done with it,

that's totally possible. We even have, like, multi node support inside of it,

although we have to kind of expose it to the front end to, like, kinda make that easier. Right now, I think you just hit new node, and it generates a new node. But we need, you know, a lot more, like, node management stuff first.

Unknown:

Right. So

first of all, we have rider die freak, Cyclops, in the matrix comments saying, one dude's expensive lightning lessons,

the saga

the saga continues. That's that is my

that is very true. That is my experience with lightning. I just Yep.

Unknown:

People have already put way more than that on there despite us warning them, and they're having issues

currently. And I'm I'm like, I'm sorry.

Like,

I'm trying to help, but it's it's still,

Unknown:

Do you have the warning when you first load it up now? I don't think you had it originally, but think I saw it for a please don't trust it with money you don't wanna lose.

Unknown:

Yep. I think I think we had to add that warning because people already put their life savings on there first.

Unknown:

I have a couple other things.

The other thing is,

if you're watching live, you might see that the matrix chat, like, it says click here to show picture.

It's because I set up my matrix on my side to make sure that I do that when I'm live streaming because we had the one when the richest rapper of all time,

got arrested,

the Bitfinex Hacker.

You guys were posting all those lewd photos of her, and I I got very I I got very concerned about you guys having photo privileges in live chat. So

photos are viewed before they go on the live stream now.

So when when we're talking about the save when we're talking about

the save state button,

The save state button is really the important part of the backup, not the seed words. Right? The seed this the unchanged seed words are included in the state anyway, right, in the state file?

Unknown:

Yeah. That that's correct.

You know, if you wanna make sure that you your your your on chain funds are, like, always good, you know, definitely back up. You can back that up once, the words.

But Do you think your warning

Unknown:

your warning's a little bit

your warning's a little bit confusing then because save these words or you'll die. Well, you'll still die if you don't save the state.

Unknown:

You won't die. You'll just lose your lighting node.

Unknown:

Right.

Unknown:

Well, just consider that. I you know, if you want me to submit a pull request, I can submit a pull request, but No. That's good. That's a good point.

Or an issue.

Unknown:

Yeah. We'll probably soon have to add, like, the end to end encrypted storage

and Yeah. Sooner than later.

Unknown:

Yeah. We definitely kind of conceptually in our heads is, like, we're going towards this storage thing just because, you know, as much as as power the, like, the browser's, like, really powerful, and it just definitely does have some

some some limitations. Like, it the the it would be nice if the browser had, like, a, like, a built in secure enclave the way,

native mobile has right now. That that would be nice,

for for storing secrets.

And,

but then, yep, this persistence

is is is usually, like it's kinda like best effort. And so something like this is going to

need a like a, yeah, a back end server,

that it that you can be confident in is storing the state for anything,

Unknown:

like, other than what Tony said of, like, that ephemeral. I'm gonna launch this, use it real quick, and then destroy it. But, yeah, if you if you wanna be So how are how do you looking at that? Is that gonna be, like, you give me an account number, and if I put the account number in, it decrypts the file it, like, looks at the file and decrypts it or something? Or is it gonna be a username and password? Like, how are you thinking about that?

Unknown:

It's it's probably gonna be something like that. We'll we'll we'll have to play around with auth options. I definitely like

it we talked about l one URL off for a little bit, but then we're like, is it wouldn't it be weird if people were trying to log in to a lighting node with a lighting node?

Unknown:

What about just using the seed words as

Unknown:

the like, just everything could be encrypted on the seed words, and the seed words can also be used to That's what I would like to do because, like, if we do anything else for, like, username and password or something like that,

then we're just trying to solve the problem of, like, how do we abstract a private key to a username and password?

That gets extremely hard, and, you know, that's what

basically entire Bitcoin world is trying to figure out already.

So if we just did, like, the seed phrase, then, yeah, we just drive a key from

that seed and then, you know, encrypt your stuff from there and it'd be fairly simple. But

yeah. I mean, it's a hard problem, like, I mean, everyone in the world is trying to solve. How do you make private keys easy to use?

Unknown:

But the browser has a really cool new standard,

like, web authin,

and it is

it's it's it's public private key cryptography.

So instead of, you know, giving them your email address and a password, and then they, like the service hashes the password and stores that, and then you

authenticate against that.

You just prove to them that you hold a private key, and then they they give you challenges to sign when you wanna, like, sign in.

And so that's being built into the to the web platform, and that's really exciting.

But

that cryptography stuff isn't exposed at all. So that all all that could only do,

authentication. It can't do

encryption. So you couldn't encrypt

something,

to that,

that private key for for for web auth and for instance.

So it does sadly, it's, like, close. It's really interesting, but it it it, it doesn't quite quite solve this. But I do think that's an exciting thing in the web, generally.

The other the other thought we've we've we've thought about is is there are potential because this is a web thing, like,

doing something kind of,

like, web like, more traditional web auth. It maybe has the potential for,

like,

integrating with other services.

So, like, you'd you know, you you're you're you're on some site and you wanna pay and you, like, authenticate it with

mutiny or something like that. Like, this is just, you know, random ideas we've been kicking around.

So like, you know, like, you know, and you, like, authenticate something with GitHub, and then you give it privileges

just to do a certain thing.

I don't know. There there's there's some there's some other thoughts there, but, yeah, I I kinda like the idea of driving,

you know, the basic,

off from from your private key, if we could pull it off.

Unknown:

Yeah. I mean, I just like the simplicity of just save the seed words and then just reenter them if you're on a new computer and everything just loads.

Unknown:

Yeah. Exactly.

Unknown:

So, I mean

well, let's talk about, like okay. So it's fucking awesome that we have a a full LDK node in

a browser,

like a proper lightning node in a browser.

Last time,

we I mean, we've discussed we've essentially discussed

this wallet

maybe 3 or 4 times on dispatch. It always has different names.

I guess last time it was called private lightning wallet or something.

Unknown:

Yeah. I'll give you I'll give you a quick history.

Yeah. Go for it. Wait. When,

to

Tony, Ben, and I were at Pleb Lab at the old location in Austin

in this, like, really foreboding

office building

surrounded by defense contractors.

And,

and we just decided,

wouldn't it be cool if we made a wallet? And we started thinking about,

what that would be? And I I forget which was the chicken or the egg, but that coincided with this

wasabi grant for researching a private lightning

wallet,

which we applied for. And I'm sure we'll get into that in a in a while.

But we, we just kinda came up with a document of a bunch of ideas of what we would want to do in a wallet. And kinda realizing that, like, there's so many things that, like, we want to play around with in Bitcoin and in Lightning.

Things that are interesting to us, experiences that we wish would exist. And, like, the the wall, it was, like, kind of the place to be able to deliver

those kinda new different,

experiences and capabilities.

So

we had thought about that then. Then,

I don't know how much longer,

later, but, there was a hackathon that, Tony and I were in.

Ben was there as well doing a different project at trans TransLND.

But,

Tony and I did PLN, private lightning

net network wallet. So that was a mobile wallet, and the idea was it's, you know, the privacy by default,

which we then came on your show to talk about

and kind of explain kind of our thoughts around how to to make a a a private

lightning

wall. And then,

I I forget. It was maybe a couple of months ago.

Tony was trying to

sign up for an Apple

developer account,

and

it turns out he's on some sort of list.

Unknown:

And He's on any lists.

Unknown:

He's on one of the bad lists.

I don't know all the details of that. That just depends your perspective. Yeah. Yeah. Yeah.

Yeah. Good list. Yeah. But to to to Hold on. Cool.

Unknown:

No. I mean, that really changed my, like, entire philosophy. I'm, like, doing things legitly. You know, I have a company

set up, legal entity. I'm doing all these things, and I just get hit with, oh, you've been approved. Congratulations. I'm like, awesome. Let's start, you know, putting mutiny and putting it on the App Store and, like, the, you know, the old mutiny,

the way we were going about it back then in the summer. And then I, all of a sudden, like, I get some emails saying that I'm on some sanction list.

And I'm like, what the fuck? This is like, what sanction list am I on? This is absolutely insane. And I started going back and forth with them, giving them every piece of information beyond what I've already given them.

And and I was just that whole week, I was like, fuck this.

Fuck Apple. We're not gonna keep going with this, like, you know,

this world where they just, like, completely dominate the app stores. They decide what goes on. They decide what tax they get from everyone. Let's just, like, go to the fucking web, and no one can fucking stop this shit.

And that really just, like, sets a direction for for Mutiny Webb, and that was really only, like, 2 months ago that that happened.

Unknown:

Yeah. I remember you texted me. When that happened.

I think I just responded all caps, LMFAO.

Unknown:

I used to do something.

Unknown:

Yeah. Because they sent me a link where I could, like, search if I was sanctioned. I didn't know that link never existed, so I just sent it to you just for so you had that link. And I think you said, like, I am not clicking that link.

Unknown:

I was like, don't drag me into this, Tony.

I'm already on enough of the good lists.

Unknown:

Yeah. Sorry, Paz. Didn't mean to interrupt there. Oh, you I found the message.

Unknown:

I found the message. Tony sends me a message. Find out if you're a sanction.

Https.trade.gov

/datavisualizationcsldashsearch.

Unknown:

Like, I am not clicking that one. Do you do you show up out of a chart or something?

Unknown:

It's most most most I don't even show up there. Yeah.

Unknown:

Oh, so that's yes. It's

yeah. And, obviously, there's been some drama recently with the App Store.

It's tough, and it it it's one of those things where it's like Apple,

you know, ultimately decide and we see Bitcoin companies,

Bitcoin apps go through this often where Apple all of a sudden decides, like, we don't we don't like this all of a sudden.

They can block updates even so you you can have a app on there for multiple years, and all of a sudden you decide to do a new feature,

and it's,

and you can't. And so you have to

complain or cry or don't do it.

And that I mean, that's kind of a

a shitty place to be as cool as the the platforms are.

But, yeah, when Tony

told me about that, and I just really started thinking about it, I got hyped on this web idea. Just partly just

see if we could.

And I feel like I'm I'm missing some details. I don't even know

when when Ben got involved. But when we started writing Mutiny Webb, Ben wanted to to help.

Yeah. Maybe you can explain

Unknown:

how you got involved. I don't totally remember. I remember just, like, texting Tony at night talking about, like, hacking on

different project ideas. He's like, you should help with the dude. And he was like, alright. And we showed up at Blub Lab one day.

He told me you accepted my, my resume,

and now we have a lightning on the wall in the browser. I don't know. I don't totally remember the total story, but, yeah, there was I promised him we would do DLCs,

Unknown:

and

and he didn't,

we didn't do DLCs.

Unknown:

Yeah. So I pulled up

Unknown:

we had Bitcoin and Clyde challenge me to click the link live on air. So I I pulled up if you're watching, this was the website you sent me. Yeah. The consolidated

Unknown:

screening list. They said I was either on list this list or another government's

sanction list, and I was just like, well, I'm not on this one. So I don't know what government I pissed off, but I'm showing up on some list.

Unknown:

I'm glad Who did you tell me?

Unknown:

Me.

Unknown:

Ben Carven on this. Congrats, Ben.

Unknown:

So, anyway, the the reason you bring this up is because by doing

browser based you see I have Namecheap on another tab is because I tried to buy rugpull.com

when you brought it up. It was bought in 2020. I did the same thing.

Yeah. Now I'm thinking

Tony, did you, Tony, did you ever figure out who bought mutiny.gay?

Unknown:

Yeah.

This this gay barber in New York.

Unknown:

He texted me the other day. He was like, Matt, did you buy mutiny wallet dot gay or mutiny dot gay? And I was like, no. I wish I did, but it wasn't me. It was bought in 2022. So I was like, one of my friends did this. It had to have been.

Unknown:

But yeah.

Unknown:

So so the the whole reason you went down this tangent is because by going to the browser, you skip all these app stores. You don't have to deal with the app stores. You don't have to deal with them potentially asking for this 30% cut on in app purchases or whatever.

You don't have to deal with any of that censorship. And, also,

as Ben has been tweeting a lot about it,

the just the ability to just straight up update everyone's instance right away,

sounds sounds incredibly useful in terms of iteration.

Unknown:

Useful and also scary. I mean, I think the very first thing rug pull technology.

The very first thing I think you'd or maybe the second or third thing you texted me was was like, wow. That's great if you want a rug pull. Like, you can instantly update the app and take everyone's funds. And,

you know, that's

it that's not obviously the purpose of it, and the fast iteration

is a huge plus. I mean, you have talked about,

you know, everyone's just when when a security vulnerability happens on lightning, which seems to happen every month now,

you know, everyone goes and rushes to their umbrella node if they if they even care enough, and they go and hit the download now button. And then they Yeah. They press the big blue update button, and they don't check what they're doing anyway.

Yeah. So this is definitely not like the wall we want even when we go live. We don't want everyone storing their life savings on here still. But, like, something that's like a walking around wallet money,

we think it's, like, perfectly

acceptable to to have that. And, of course, you know, we can do a sweeping update. Let's say we just stole all of our users, like,

walking around spending money, that could end up being a lot if in in in,

aggregate. But, you know, we're we've been thinking about, okay. How do we do the most that we can to ensure that, you know, at any point in time, people can check,

the the the the WASM,

you know, binaries or or or WASM files that are,

that get downloaded from the website. So we thought about, like, maybe having some third party watcher that just,

like, calls into

mutinywallet.com

and pulls down,

the Wasm binaries and just sees if that's, like, the

Unknown:

built?

Unknown:

Yeah. Binary watchers.com

or something. Binarywatch.org,

Unknown:

I think.

Unknown:

Right.

So it's it's definitely

an an area that we don't see too much of. Like, who is verifying website,

source code? I mean, you could right click then see all the source code right there, but it's not right. Verifying anything. Yeah. I mean, there's, like, dozens of us. There's at least a process,

to verifying, like, real binaries and execute those for the for

for,

the computer. And then even, you know, for Android, you see it a lot more where you can actually verify the app download from from the App Store.

But we don't really see that much for individual websites. So we're gonna we're gonna probably have to come up with something unique and custom to be able to just do the best effort in,

hey. We we verify that this source code is from us. It is what we downloaded on our phone or, sorry, built on our computers and and and reproduced

and deployed that. And at any point in time, if there is mismatches or or something, then we can, you know, alert people. But

for the time being yeah.

Unknown:

I mean, when the or something. Like, we don't we don't trust Apple and Google,

like, the 4 of us, and most of my listeners

like, most dispatch most of the dispatch audience doesn't either,

but that is something that,

like, the general public appreciates Google and Apple for. Right? Like, that's one of their, I mean, that's one of the the positive arguments for App Store model in general is that they verify everything for you.

Unknown:

They don't they don't verify

that the code was what the source code originally was or, like, an open source project. They don't verify that that matches. Verify that it's signed by the maintainer's key or whatever. Right. Right. Yeah.

Unknown:

Thing is that that key is normally just, like, good of actions, so it's not,

you know, it's it can still be easily

coerced. Like, you know, if someone gets gets gets into your GitHub account or

if Microsoft goes rogue, you're still screwed.

Unknown:

But I guess this is what you're kind of preempting by saying that anyone can run their their you want it so that anyone can easily run their own instance.

Yep. Yeah. So it's like My my idea is if you run it if I run an instance, like, Odell's wallet,

then

I like, I'm in charge of manually pushing updates that you guys have. Right? And then people are, like, kinda trusting me as a middleman proxy or whatever.

Unknown:

Yeah. What one way to think of it is is you

so the the all the Rust code gets compiled to to WASM web assembly.

It's called the native binary target for the web.

And so if you imagine, we're we're

we're toiling away in the rest minds, and we push a great new functionality.

We build that binary. We publish it to, like, GitHub releases. We we sign our release, so you know it's from us.

And you can, you know, you have a a decent idea what, you know, what your code it was built from,

and you can have other people verify it as well. And then let's say you you could be self hosting your own MeetMeWall in, like, potentially, like, a replet.

Like, you know, really easy spun up environment where you clone the project into there, And you just have a dependency

that that when the page fires up, it pulls in this this WASM blob from somewhere, either locally

or, you know, from from, like, like, a node package or from GitHub directly, something like that. So you'd be in control then where you're not that blob that you're pulling in

is,

is not gonna change unless you change it yourself.

And so maybe there's,

hopefully, ways to, you know, make an even nicer self hosting experience that can make it really clear

and maybe similar to the existing

node implementations of of how updates work.

But there there's definitely a lot of potential there, but, yeah, not not a lot of this stuff exists right now for web projects.

Unknown:

Yeah. I think something that's cool is, like, the,

like, buried a self hosting isn't as big as, like, like, an Umbrel node or something. Like, if you wanna self host Umbrel, you, you know, you buy, like, your Raspberry Pi and you'd, like, terabyte of storage and, you know, it can be, like, couple $100

versus with this, like, all you need to do is distribute,

you know, like, 10 megabytes of files

through a website. So it's like like, you know, I think we're just hosting it off for sale. It's, you know, someone can sign up for free and do. Or if you really wanna go, like, the, you know, full self host route of, like,

posting the website yourself, like,

hosting

a super minimal website isn't extremely hard. Like, you can sign up for a shitty VPS or even, like, do it off your own laptop if you really wanted to.

Unknown:

Well and and the dream would be kinda this is something I've kinda wanted in in Bitcoin in general is the kind of the idea of, like, progressive enhancement. Like, you start out

as a fully

or not it's not fully custodial, but, like, you're you're relying on us a lot for a lot of things. 1st,

you know, hosting the website,

storing your state,

giving you block data,

proxying

you to the rest of the network.

And you can, you know, start using MeetMe wallet, and then you can start hosting some of those things yourself or all of those things yourself and have kind of an uninterrupted

experience. The

same it'll be the same wallet, the same UI, the same even the same channels,

but you can move them over to something more self sovereign.

Whereas, you know, kind of the jump right now from a custodial lightning wallet or or even something like Moon, you know, to a self sovereign stack. It's everything's different. And and you gotta

you gotta buy raspberry pi, which is hard.

Unknown:

Yeah. And and you could even go from, like, mutiny wallet.com

and export state and go to adele

wallet.com

and import state and then boom.

You know, as long as you're kind of at the same version of the other one, you, you know, you kinda stay up to date on odellwallet.com,

you should be able to just, like, import that across. So you could like Paul said, you could start out even, you know, using using uni wallet dot com and then,

you know, start pulling in your own source data for blocks and start using your own TCP proxy. And then, eventually,

you know, have, you know, your wallet dot com and and export state, and then you're good to go.

Unknown:

So we have so Trinity in the comments

asking

about doing this for an on chain wallet,

and he goes deep and he says, like,

having,

you know, Sparrow wallet functionality. But, I mean, right with with Whirlpool and post make spending tools and collaborate transactions.

I mean, right now, it is a on chain wallet. It's a simple on chain wallet by default

to begin with. I mean, it it is almost it's easier to use as an on chain wallet than it is as a lightning wallet right now. Right? Because a lightning wallet, you still have to open a channel manually and

Unknown:

go through the process. Yeah. I'm gonna take the mixing side of it.

Unknown:

Yeah. I mean, just quick point is, like, every lightning wallet is an on chain wallet. It needs to be. So, yeah, we we definitely have, like we use BDK. We have, like, full

support there. And then, like, on the mixing side,

we

we would like to integrate

Vortex into there somehow because,

the nice thing is, like, the way I built Vortex was, like,

everything is just like a WebSocket

or, like, a HTTP request.

So

we could

theoretically just build the entire vortex protocol into the browser as well and, like, have a full

you know, you just leave your web page open and your on chain funds are mixing. So is Vortex a separate project? Or

Yeah. That's building 2 separate self custody lightning wallets right now? Well, Vortex doesn't have its own wallet. It's like it just uses LND's wallet or c lightning or whatever node you're using. So it's really just like a

kinda just like a protocol

on top of, like, a wallet. So

now that Unity has its own wallet, we could just, you know, do the protocol on top of it, and it would work.

Unknown:

I mean, that makes sense to me. I mean, you should combine the 2.

Unknown:

Yeah.

Need to actually release vortex to do that, but,

getting close. But yeah.

Unknown:

So I wanna go back I wanna go back to the updating thing

because, I mean, I was at I saw you boys,

in Austin

couple weeks ago. I don't know. Time just blends together now.

It's like when I started the episode. I don't know if we're gonna have another episode this week or you're gonna go on a holiday break. I don't know. I'll keep you freaks updated on that front. But I was in Austin at some point recently, and we were at bid devs together.

And I made my comment, which is a comment I've made on dispatch a lot lately, which is,

we've had 2 l and d bugs,

that required users to update their nodes ASAP.

It's fundamentally

different than the ethos of Onchain Bitcoin

where

users are encouraged to be slow to update,

and not update to the immediate release right away.

If you can't verify it yourself, at least give it some time out in the wild before you go and rush to update.

But with lightning, we have these hot wallets that are connected to the Internet

and iteration

and building on top of them is happening really quickly, and there's moments where you have to update quickly.

And we touched on this earlier in this episode, you know, people will just rush to their umbral nodes or their raspy blitzes or their start nines, and they'll press or their nodules and and press that big blue

update button. It won't check. So my argument was,

maybe some of these node packages should just have an opt in, you know, not by default auto update feature.

Ryan Gentry,

oh, I guess I guess I'm not supposed to say who says things. Okay. Well, I guess we'll talk about other things, and then I'll bring it back up later. Somebody.

Yeah. Yeah. Some I well, I can't talk about it now.

He wasn't he wasn't even there. I don't know what you're talking about. Chatham House rules. Chatham House rules. You're not supposed to say who says what.

Okay. So let's go to the next topic, and and I'll circle back later

discreetly.

Unknown:

Sorry, man. This is a clever interaction.

Unknown:

But okay. So let's talk about okay. So we have Unity Wallet. You set it up.

What what are you guys the the real pain points for light the actual

real lightning users is,

they they wanna be able to send a lightning they just wanna be able to send and receive Bitcoin. At the end of the day, they just wanna be able to send and receive Bitcoin. I know when you guys originally talked about private lightning wallet,

we weren't able to

you guys just had received off by default.

It seems like you've enabled receive in the most recent update. Can we receive on mutiny?

Unknown:

You can, but it doesn't always work well.

So, like, the way we did it was, like,

for like, we wanna have kind of the ethos of, like, what Peel originally had where,

like, if you have a channel,

it has its own lightning node associated with it. So you don't,

like, have, like, one node with, like, 10 channels, and all those channels are associated with, like, a single identity.

So we have, like, this thing where we every channel creates a new node. At least that's the idea.

But for receiving,

it would suck to, like, have to say, like, I'm receiving through this node. So we did this thing that LDK has called a phantom invoice,

which basically lets you, like, create a fake note behind all of your nodes and receive through all of them.

The problem is, like,

it's not very well supported,

it seems, and there's a couple bugs or something. So our receipt doesn't totally work, but it it kinda does.

Unknown:

But you literally just added the receipt, like, what, like, yesterday or something?

Unknown:

We we've had it, like, I guess, for,

like, a week or or 2 ago, we added receive.

It's just it's just like, yeah, like like Ben said, like, Phantom

Invoices. I mean, that's that's kinda how Cash App is set up too. They have, like, multiple highly performant nodes,

and they create Phantom Invoices, kinda replicate that. There might need to be some updates we have to do on the LDK side to, like, make it reliable.

And then, also, like, when services probe, it's kind of a problem as well, and so we kinda gotta solve,

the probing problem,

it you know, because some services won't send to you unless they can probe to you first.

So there there's a few things that we kinda have like, Phantom Invoices works really well for Cash App. But in this context, we might have to,

you know, maybe, like, upstream some changes to l to k or or ask them to do some of that to get it to work properly. But I think for the immediate term, we're gonna disable that. We like, maybe a few hours ago, I got mute need a mutiny connection's working where, you know, you can have a web browser instance,

actually do a connection

to each other and talk through,

WebSockets

to each other, and then it it looks like it's a it's a new node. Like, it looks like it's a real node. And that

that to me has actually been, like, super phenomenal.

Like, the idea that you can have 2 wallet users

talk to each other

through you know, you're going through a website, a proxy, but all of that works really well, and you can pay the other user and and and receive. We we don't have any graphing

data currently, so, like, we don't pull in,

the lightning network graphing data. So we can't

currently pay people that aren't your direct channel partners,

but we'll we'll try to probably do that soon. Which is, to be clear, super not private.

Unknown:

But it's just where we are are technically.

Like, we're we're we're we're we're definitely gonna drive back in that privacy direction, but we're we've been mostly focused on getting

getting this all to work. But, like, to to answer your your earlier question about just, you know, that that user experience of being able to send and receive,

Tony and I work at, Voltage, and Voltage is sponsoring,

us to basically halftime work on open source, and our project is Mutiny.

And, but in our other half of time, we're we're we're looking at and working on some some stuff to to help,

basically, projects like Mutiny to be able to really easily have,

inbound liquidity,

so that you'd be you'd be able

to fire up a fresh wallet,

have no on chain funds on there,

create a lightning invoice,

and that that first,

payment to that lightning invoice,

would be,

would it result in a channel open to your mutiny wallet node

and would give you all the inbound liquidity you need

to receive that very first payment. So that, like, boots on. How Phoenix works?

Yeah. Basically, very similar to to Breeze or Phoenix.

Just cover and spin on it. But yeah.

Unknown:

And so voltage would be the liquidity provider?

Unknown:

Yeah. In this case.

Unknown:

Is that gonna be open source?

Unknown:

I think so. I'm pretty sure.

Unknown:

Yeah. At least all the client side stuff.

Unknown:

Well so, I mean, let's just talk

let's just, like, think of think this through on air. I mean, it sounds like I mean, I got the whole team here.

What like, if you went to if you went to reckless.munitywallet.com,

it should just there should be a button that's, like, easy mode, and then there should be a button,

you know, pro mode.

Then easy mode should just

be very basic.

Receive and send.

No no channel management or anything like that. And then pro mode, you should be able to, like, choose

choose who you open channels to and have, like, intricate

management.

Unknown:

Yeah. I mean, that's a good question. Like, we we definitely want to

make user experiences

as easy and simple as possible.

So where do you draw a line on, like, you know, having simple buttons that that everyone can easily use versus

hiding some things versus, like, allowing those users to use it.

I mean, I'm a power user myself, so I would probably, like, prefer to have this tab that says connect peers. I can manually connect to whoever I want in a tab that says open channel. So I can do anything I want there. It's it's just like, you know, how do you provide that that UX while still having that simplicity and the

you know, Paul's the UX designer here, so he's he's gotta figure that out.

Unknown:

For for better or worse.

But, yeah, I think yeah. My my vision is is that yeah. It's not it's not an option of easy mode or hard mode. It's just it's just easy mode.

And then if you're Tony, you click a more button and maybe you click it a couple more times, and you you you'll you'll and then it it will

hopefully unfold into all the functionality

and full control.

But that it can

I I think it could be pretty effectively hidden

from the basic beginner user so they're not freaked out about everything that's going on?

But, again, like, I I really like that idea of progressive enhancement. Like, when you're ready, what you know, after you've watched 5 different YouTube videos about what lightning channels are, you start

you start to let you know what? I think I'm kinda getting this. Does my, wait, does my lightning wall have channels? And then you dig in. They're like, yeah. It does. Here they all here they all are, and here's how they you know? Like, I I really like that idea of

helping

Bitcoiners,

level up. You you I don't wanna just, like, you know, spoon feed somebody and keep them in kinda like a like a like a state of babyhood forever. Like, you know, the the by learning how this thing actually works, you can be much more effective

in using it,

Unknown:

and help other people along as well. So So the I mean, that brings me to another thing, and it kind of ties into the fact that we left the trinity hanging in the comments.

If Craig Wright

listens or watches this dispatch

and he sees

that

Trinity

recommended

or not recommended,

was going through the thought process of a progressive web app to access Sparrow, he would lose his shit. He just not,

like, did I say Craig Wright? Goddamn

it. Oh. Fucking I'm like MBK over here. I've said his name a million times. I'm okay. If if Craig Raw if Craig Raw

he hates web browsers.

So

if he's not gonna build anything that's that's web browser centric. But Yeah. And maybe I misunderstand the question, but,

Unknown:

asked, you know,

a back end to run multiple wallets, mutiny, sparrow, whatever, all through a single self hosted progressive web app. I mean, there's nothing that we've done that it makes

all wallets progressive

webappable.

We

but, I mean, there's BDK,

and we were we're leaning on BDK,

you know, for our wallet,

and that's being compiled to WASM. So, theoretically, if somebody wants to make, you know, a browser experience based on BDK,

they can do that.

Maybe I'm missing something.

Unknown:

The reason I was thinking the reason I was thinking of Craig Wright was because you said your vision earlier, and I was gonna make a joke about,

mutiny wallet PV, Paul's vision.

Unknown:

That's right. He would do that one day. I'd believe it.

Unknown:

So so but I was gonna bring it back to so the the question becomes

so Craig

Raw

believes very strongly

in pushing the user to best practices. And I know,

like, you and Tony in the beginning of Private Lightning Wallet, like, that was also the case. Right? It was like

the defaults were gonna be strictly focused on privacy

regardless of user experience.

Unknown:

And then we sold out to big big browser.

Unknown:

It's a hard balance to find.

Yeah. I think user friendly and also default search.

Unknown:

Yeah. I think right now, we're we're just so focused on getting a node running in the browser and just exposing all of its functionality. I think we still want to,

drive in that direction. That said, like, yeah, I think if

if you really know what you're doing,

I don't know. The browser is is interesting and a little bit of a dangerous place,

and, you know, you're loading up a URL, you know, you're kinda trusting.

I mean, I I really don't like when people say Bitcoin is trustless, and usually, you know, they'll correct themselves pretty quickly as they trust minimize. But, you know you know, you're

a lot of us are trusting,

software,

in a lot of ways that are are pretty hard to audit. I think the web browser does add a couple more layers

there that, you know,

potentially a desktop

app

can avoid. And, you know, like, theoretically, we could we could build mutiny to to as a, like, a cell like a single binary desktop app that we could deliver to people as well to have some get get a little bit outside of the the the browser,

fears.

But but, yeah, it it it's definitely there's definitely a trade off to the browser. I think in some sense, for what most people are actually able to verify,

they

what they're running in the browser is about as safe as the the binaries they run on, you know, directly on their computer,

Unknown:

But I'm not a total expert on that. So what do you think about people that say that having auto updates

gives the developers increased liability?

Unknown:

What kind of liability?

Unknown:

Yeah. I I know it. I, I I'm not a lawyer,

but the

I I I don't I don't I don't find that to be true,

that pushing updates

gives liability, but I yeah. Again, I'm not I'm not a lawyer.

It it

I'm trying to figure out how to say it properly, but

I I think

what what additional liability is there besides just a user clicking the update button and, oh, oops. It downloaded, like, the wrong thing.

Or it picked it picked the version that

the developer wanted

anyways,

and not the version that it should have. Right? Like, we I think you had this in our dispatch a a while back talking about

node managers or node node implementations,

like node boxes.

Unknown:

You know, how do we Node node roundtable.

Unknown:

Yeah. And and, you know, whether or not one of them like, what if one of them decided that they wanted to put you know, during the fork wars, they wanted to put, you know, Bitcoin Cash on there instead.

You know, it's it's still the developer making that choice

whether the user hits the update button or not. So I don't buy that liability argument at all.

Unknown:

And you can also make the argument that, like,

we have

maybe, like, less liability because now or maybe not less, but, you know, it's easier to defer because,

like, say, in the App Store world, if we say, like, we introduced a bug

or, like, you know, we're just, like, using l and d and the l and debug is now in our code.

If we need to push an update,

well, it could be, like, a week or 2 before Apple approves our build for the user to be able to update their their app and fix it versus

with this, we can, you know, deploy an update in an hour, you know, within 5 minutes if we really need to. And,

so we're able to, like, you know,

maybe use it for bad. We can also use it, like, for much a lot of good as well and being able to just, like, fix issues immediately and not have, like, users waiting for, you know, days for an update.

Unknown:

And and the spot answer in the comments brings up a good point. Like, you know, if you do provide an auto update button,

maybe if you wanna avoid

liability, you just have, like you click this, I agree not to sue the developer if this auto update makes me lose funds or something.

Unknown:

You know, that that should cover it as well. Yeah. It was like the BlockFi terms and conditions. It's like we will

lend your money to high risk traders

by default.

It's not our fault.

Unknown:

It's not our fault.

Unknown:

Okay.

So should we talk about let's talk about lightningprivacy.com.

You guys have a great domain. You did a bunch of

research there

on lightning privacy.

What's what's the big takeaway?

Unknown:

Big takeaway is that

the lightning

privacy stuff that

Citadel Dispatch episode 21 talked about about 60 episodes ago, 60 one episodes ago. Great dude I had on that show.

Yeah. Yeah. I wonder what happened to him.

RIP. RIP.

But

it

we didn't we we didn't get we really didn't get anywhere since then, I mean, to be honest.

Unknown:

Yes. You have better diagrams now.

Unknown:

Yeah. Paul well, yeah, we we brought on Paul to do the diagrams.

The the best thing that happened to Lightning privacy since

since 21.

Unknown:

My main takeaway is Lightning's fucked. Don't spend time building a lightning wallet, and,

only use on chain.

Unknown:

Yeah. That'll be 2 lightning wallet.

Unknown:

And, like, not

to not to shell our own bags.

It's so hard, but,

like, the fact that, you you know, Tony mentioned mute need to mute new channels. Like,

there so there's a lot of the lightningprivacy.com

is is basically the result of a bunch of,

weekly

calls that we did,

with, Evan and,

Evan Klutis

and,

Max Hillbrand,

and some some special guests,

some various domain experts,

to learn about

future,

things coming to the lightning

network potentially

that could improve privacy. And so we explain them,

from, like, the, you know, the the

routing. The big thing there is PTLCs,

coin joints, big thing there is, Ben Carmen,

and,

blinded paths, and trampoline routing.

And a lot of this is just kind of in I don't it's just kind of in slow implementation

and consensus

limbo.

But, you know, there is a theoretical I don't wanna overpromise, but there's a kind of a theoretical world where, you know, we have our own node implementation now. You know? There's

there's,

because LDK is a library. It's not it's not the node. We built the node.

And so if we wanna get some functionality,

at least between 2 mutant users,

we could do it,

which is I I don't know. I feel really liberated by that. But, you know, for the most part, a lot of this stuff is sort of waiting on,

implementation,

and then widespread adoption

to be to be truly useful.

Unknown:

Do you think we're gonna see as, like, a splintering of the Lightning Network where

people just get tired of waiting, and they're like, we don't care if we're

compatible with other Lightning implementations. We're just gonna build this thing that works with our custom

Unknown:

Well, I mean, that already existed in the implementation.

Like,

whatchamacallit?

Phoenix. I forget which one he uses. Black is it trampoline or lighted paths?

Unknown:

It uses a lightning. Trampoline, but it's I mean, they have their own lightning implementation. Right? Yeah.

Exactly. Phoenix is the front end for a clear and async maintains both. Right. When you have Phoenix, you have, like, this one channel with,

Unknown:

with async,

and,

and they are able to have, you know, some some interesting,

techniques there that are not widely available on lightning, but it doesn't matter. It just matters what you and your channel partner want to and can support.

So it does matter, you know, at some point for some of the stuff that's bigger and more multi hop that requires more of the network to adopt. But there are some things that can just be implemented

just peer to peer.

Unknown:

Do you guys have first oh, go on. I worry about the PTLC.

Unknown:

Like, if there's gonna be a PTLC split

at any point in the future.

I I think it was after tab conf or during tab conf.

Someone told me that, you know, the l and d team, they've they've done some research or they they have some knowledge that, you know, there was this misconception maybe that, like, PTLC channels wouldn't work with HTLC channels. So you would have, like, a split or, like, a new lightning network, quote unquote.

But it it seems like there there could be a pathway where that's interoperable.

But, you know,

you know, there's a lot you know, after all these years, like, there's a lot of things that are possible,

but it just depends on someone to build it. And if they don't build it, then it's not it's not gonna happen. So, yeah, I I do wonder if there's gonna be a PTL split at some point.

Unknown:

Yeah. I don't think, like, I think that's a lot of just FUD because I I think the, like, PTLCs are probably one of the biggest

privacy upgrades for lightning. Like, if you read Tony's routing analysis thing, it kind of gets into it. But that would solve a lot of the issues of just, like, you know so right now if you do a payment across the network as the same hash, that's very obvious, like, it's the same payment going across the network.

Whereas, PTLCs would fix that, and

I I don't think that would be a huge problem because, yeah, like, you you just said, you can just like, a Taproot channel could use HTLCs.

Like,

the implementation LND is working on right now for Taproot channels is gonna be a a HTLCs only, and then they'll later add in PTLCs

to it.

So I think it'll always be able to do, like, both. It would the the hard part would just be

switching to get everyone to use PTLCs because for a while, like, you know, it'll be, like, only a small subset of the network supports PTLCs.

So no one will use them until it's, like,

everyone can use them. So it'll be hard to, like, kind of flip that switch and, like, you know, you might have reduced payment reliability.

And, but it should be doable.

Unknown:

Is it possible to FUD

a protocol that uses the self described meme reckless?

Unknown:

Exactly. I thought that was just just starting out.

Unknown:

Are we not still reckless? I mean, it's reckless That's true. Wallet.com.

Unknown:

No. We're we're reckless because we're new. And also, actually, Core Lightning just launched a plug in manager called reckless.

Unknown:

So, yeah, I guess it's you're right. We still do I mean, because what? FUD stands for fear, uncertainty, and doubt, and that's basically the definition of reckless. Right?

Unknown:

Which is the definition of the lack of FUD. What?

And then that's the definition of lightning.

Unknown:

Do you do you guys have a response to Chips Ahoy in the comments, or are you gonna ignore it?

Unknown:

Yeah. I mean, I mean, I think we even talked about it a little while back,

whether or not working with Wasabi was a good move or not. And,

like, our our stance is we we we do this lightning privacy stuff anyways. I mean, again, like, for almost 2 years now, like, doing lightning privacy stuff. And if someone is willing to give us a Bitcoin to do it,

that's I mean,

this the stuff that I'm gonna do anyway. So Right. Gave you guys a Bitcoin for the lightning privacy research. Right? Right. Which back then when we started, that was fantastic. We were like, fuck. Yeah. That's that's, of course, we'll take that.

And then we over the last, like, month or 2, we barely hit a wall, and, you know, it's been slow to do any work on it. And I was like, wow. That yeah. Because that one Bitcoin's not shit anymore. No. I'm it's all all jokes. Bitcoin is 1 Bitcoin. I know. I know. I know. All jokes aside, it's just a joke. I I am appreciative of that 1 Bitcoin

amongst all of us. But, I mean, yeah, it's it's it it is what it is. We're we're independent,

researchers in this regards, and and we accepted funds for someone to to sponsor our research. So that's

that's kind of if if some other, you know, privacy implementation that cares about lightning, which there's not many,

wants to do it, then it wants to sponsor us, then we'll be more than happy to

Unknown:

accept their status. My idea, Tony.

What what was what was that? I told Tony that he should title his research breaking the Lightning Network's privacy and then just fund it from Monero's chills.

Unknown:

Yeah. Like, like, right well, actually,

it's a Monero

Unknown:

it's a Monero nonprofit that they're going through to give us these funds, so that makes it even spicy. I asked I asked chat GPT how to use lightning privacy, and one of the responses was use Monero.

Unknown:

Brutal. Fucking AI. Well, we can talk about AI for a little bit, but before we do that, let's,

what so Trinity has a good question. How far out are we realistically from blinded paths being available

in a lightning implementation?

Unknown:

It's pretty close and declared. I know the codes are already there for, like, the cryptography part of it.

And I think they have, like, they might have gotten merged. I can look, but they had PRs

about it.

You know, it's

the the hard part is, like, you know, once Eclair has it, then

we need, like if you wanna be able to pay an Eclair node that has 1 in past and, like, you know, maybe then your l and d node needs to be able to understand it. But I think, like, eclair is close to having it. And It seems like something it's not like goal twelve or

people wanna drag their feet and not do it. It seems like everyone's on board with it,

but

I don't know. I would guess, like, by summer next year, it would be, like, the latest that, like, you see it.

Unknown:

Yeah. I think the the thing is, like, both in order to have a good UX around blinded paths,

you kinda need bolt 12, and it's you know, l and d is

at least well, an independent a new,

independent developer is working on

Bolt 12 stuff.

An ex Lending Labs employee

is working on it.

So they may have it. And I think in order to get the privacy benefits of bolt 12

so

in order to get better privacy benefits of route blinding as well as extension to that,

we really need

all, you know, all major implementations supporting it or at least a majority of the lighting network supporting it, which currently the majority of the lighting network is l and d. So it's it's not a very privacy

enhancing solution to just have CLN support,

blind and passive vault dwell. We're just declare. We need kinda need more implementation. Like a clair

Unknown:

LDK

and

Unknown:

Yeah. Coordinate wouldn't be enough. You really need l and d to get the privacy benefits

of it. We need l and d to to support it as well.

Unknown:

And more importantly, I mean, if you're showing a bolt 12 invoice, you need

pretty much all wallets to pay it. Right?

Unknown:

Yeah. Yeah. Otherwise, we're to read it.

If yeah. Yeah. No one can pay it, then why would you display it? And we've seen, like, the inverse where LND added their version of AMP,

Unknown:

but only they support it. And even though they're, like, 85 percent of the network, no one uses AMP invoices because,

you know, because, like, a a phoenix node couldn't pay it so that no one would display it. So, like, you really kinda need a 100% adoption.

And,

yeah. But it seems like everyone's moving towards this.

Unknown:

There's has got another good question about,

will blinded pass cause more issues with payment reliability?

And I don't know the specifics on that. Maybe you guys can speak to that. But I will say that a lot of these, if you look at these diagrams,

which is my primary contribution

Unknown:

By the way, I love the how it

how it is, how it could be. How it be How it usually be. How it could be. It's great.

Unknown:

Thank you.

But, yeah, if you look at them a lot like this one, you know, this has how many 1, 2, 3, 4, like, 5 hops.

And so you imagine, like,

with a lot of these, we're imagining, like, a bad guy, like, multiple points in a route.

And we're trying to, you know, fix that problem so that a bad guy at multiple points in the route can't,

you know, figure, figure out too much information about you, about who's sending or who's receiving.

But in reality,

as far as I understand, a lot of lightning happens with, like, 1 or 2 hop routes right now.

And,

Unknown:

and so Yeah. And everyone's doing, like, one hop payments.

Unknown:

And that's where the reliability's

at right now.

And

so

we we just just getting longer routes being a regular occurrence at lightning would be a,

a move forward in privacy. It's my So how do we do that?

Unknown:

I don't I have no idea. I mean, I I mean, we clearly don't do that by offering Unity wallets an easy voltage channel.

Unknown:

Yeah.

I I I think it will come from me. For 1, like, I I want more visibility

into,

the payments that are being made on lightning,

exposing that to the end users. So, you know, having an advanced tab, I would. If I made a wallet and I had to put an advanced tab in there, I would wanna see the payments,

being made and where they went to,

because that's that's the thing. Like, you don't everyone just hit send, and they pay whatever lightning fee that they their wallet told them it's gonna be, and and it went to something has, like

Unknown:

Zeus lets me choose, like, a max fee.

Unknown:

Oh, right. Right. Right. Yeah. Some some have that at least, or they have default swim max fee built in somewhere.

But but, yeah, it's I think it was the lightning payments panel at Bitcoin 21,

with Rockstar.

Alright. Sorry. 20 yeah. 22. 22 already.

Unknown:

It's almost 23. It was this year.

Yeah. Hard to hard to believe.

Unknown:

And they were talking about it with with Andre,

from Zebedee about how, you know, from their experiences running large lightning,

payments for, you know, consumers,

being, you know, very popular lightning,

wallets.

They were saying, like, after after one hop, the payment reliability drops down, like, 80%, then it's from, like, you know, 60% after that, and it keeps going down and down.

So I I would I would say if we want better privacy on lightning and we're willing to get it,

we have to probably have more

low time preference

in the fact that we're not always gonna be able to find a route instantly and maybe have to sit there for 30 seconds or 60 seconds while, you know, longer routes are being taken.

And I I would hope that that would be something that, you know, people can configure in the future. Like, hey. You know, I I think,

there's this idea of, like, you know, pick 3. You you know, you want speed, reliability, or security, or privacy

or speed,

security,

or privacy,

and, you know, pick pick 2 of those. Right? Well, you know, if we could say, like, hey. If you're willing to not have speed

and you want more privacy, you know, wait up to 60 seconds, and we'll try to we'll try to pick the most

private route that we can.

You know, it it it it and uncle thinks is saying, like, that sounds counterintuitive to the point of lightning network being, you know, instant and fast and, you know, free, quote, unquote free.

I would say, well, you know, the Lightning Network can be a really good privacy tool

if you want it to be a really good privacy tool.

If you don't want it to be a privacy tool and you just want it it to be a fast tool,

then then, you know, you're giving up some privacy to do that probably if you're if you're trying to maximize for that. So I think lightning is whatever you want it to be,

to to kinda say that. But, if you want it to be really private, it can be really private. And and not that we're, like, sacrificing on, like, you know, stripping out all the hardcore privacy options by default with Muni.

I think there's a beauty in the way that Muni can be ran,

that could allow someone that cares about the privacy options to get it all

while still allowing users that, you know, you know, maybe care about privacy a little bit or, you know, want a good UX to also use it.

And when they want to care about privacy, they also have that option available to them as well.

Unknown:

But you can't force more more hops, right, on the wallet level?

Unknown:

Mhmm. Yeah. You can. If if it's available. Like, you know, if if there's not, like, a hop to the the destination, then there's nothing you can do about that. But you could, like, you could just do some sporadic route yourself that makes, like, 0 sense to do in order to achieve better privacy. Like, if if I have a direct channel with you, Matt, and and your node, but I choose to just go through Ben's node and Ben's Because the sender's choosing the route.

Unknown:

Yes.

So but just no existing wallets right now, like, have a toggle that's like, I want

to prioritize more hops.

Unknown:

That's correct. Yeah. Hopefully, we can add something like that to our wallet.

Unknown:

This graphic is something we kind of an idea we came up with at the Austin Bitcoin Design Club when we were talking about some of this stuff. So you could imagine trying to kinda get,

users

more comfortable with this idea of they're searching for a good route.

This is just to show, you know, like, what, you know, what's happening while because wallets are typically, like, probing even right when you pull it up, and they're definitely optimizing for like, before you even click confirm, I think some wallets start probing,

and they're definitely optimizing for payment reliability. But, yeah, if you maybe if you got some dial,

some privacy dial you could crank up. And, like, I'm willing to sit here for a while

and come up with some really weird shitty routes to get a little bit more privacy.

And and I'm guessing,

correct me if I'm wrong, you could also that might be another way to opt in to some of these features that aren't fully supported

yet, like, like trampoline or something like that. Like, hey. Like, try let's let's try to find a trampoline route first and give that a shot. Even if it's poorly supported, maybe we'll get lucky.

Unknown:

Yeah. I think, like, it kind of is a default. Like, the more privacy you try to get, the less reliable payments we'll get. So

it's gonna be at the end of the day, like yeah. It's gonna be trade offs and then, like, up to the user and probably even more so ultimately up to the wall developer, and then, you know, the user will pick

the features that they want. So,

Unknown:

Well, if you leave it up to the user, they're just gonna use a closed source custodial wallet like they're currently using.

Unknown:

Yeah. You gotta you gotta believe.

Some people do the hard things. Some people some there's some try out still try hards out here trying to do trying to do better.

Unknown:

Yeah. But if there's, like, 10000 of us, 5th like, that is the difference between building a tool for, like, 15,000 people versus building a tool for a 1,000,000,000.

Unknown:

Well and then that's why we're kinda building a tool that, like, billions of people can access from any browser in the world. So that's kind of, like, our mentality on that.

Unknown:

Do you think users

are do you think so, like, one thing that was interesting that, Craig Raw did with Sparrow

is he's like, look. I know people aren't gonna necessarily use their own nodes.

So if they use an Electrum node,

I'm gonna force them with, like, 4 4 recommended choices that I've, you know, that he's vetted,

that have good reputation, but, obviously, you're trusting those Electrum nodes,

with your privacy and the verification of the chain.

I think he's got MZ. He's got Blockstream there.

He's got a few others that you can, like, choose

from in, like, a drop down menu when you first start the wallet if you don't run your own node. I could kinda see the same thing with routing nodes.

You know, if if you're just a regular user and you're not really thinking about privacy, okay. Fine. But, like, we're only gonna open channels to

these four routing nodes that we think are more likely to

not spy on you or log your information.

Unknown:

This is something I've meant to talk to Tony and Paul about of, like because, like, right now our our wallet gets

all the data from blockchain dot info, and we do we are adding the option to configure it, but, like,

it's like a trade off for, like

you know, I have friends that work at blockchain, but I still don't exactly

trust Blockstream to not use that information maliciously or maybe I I completely trust them, but then daddy government knocks on their door and, you know, it forces them to start logging and give That's a bigger issue. Yeah. So

the alternative could be,

well, why don't we run it then, and then

we can

and then we know, well, I trust myself. Like, I'll you know, maybe the user doesn't, but, like, at that point, you know, we can be we'll know explicitly that, like, our user data is not being shared in a bad

way according to, like, my morals or whatever. But the problem there is then

well, if we start running it, well, then now we know

exactly who is a mutiny

user and who is not versus with Blockchain. Info.

It's not just the people using

mutiny Everyone using that. Yeah. Exactly. So you have a different anonymity set. So and it's the same thing as well you're talking about with, like, routing nodes where, like, if we set up just like the new new routing node that everyone connects to, now we know, like, same like, with the same way like a clear works, you know, pretty much every Phoenix user versus, like,

Unknown:

the not what I'm saying because that's not what Craig Raw does. Like, Craig Intention doesn't run his own server.

So it it would be instead of using a mutiny routing node, it'd be like these 5 routing nodes we think are the least likely to be spying on you, so we open channels there. But but we don't operate them.

Unknown:

Yeah. So you can do that. The problem really kinda though is,

like, you can say, okay, open a channel to this person, but, like, receives get very hard. So you kinda need, like, an open protocol, like, standard for, like, receiving or, like, kind of an LSP protocol kind of thing, which

doesn't really exist. I mean, Tony and Paul at Voltage are working on something.

That's not like a standard protocol. It's more just like a

I don't wanna say voltage product, but kind of. And, you know, likely, they'll be the only ones running it at launch.

So

it doesn't totally solve the problem by just saying, like, oh, this open channel to, like, bit refill,

async, and, you know, the Bitcoin company. Like, you know, you need to and you still need, like, liquidity services, which

Unknown:

is a harder problem. Yeah. The receive is the hard part. Yeah. Exactly.

Unknown:

Yeah. There's there's,

there's an LSP working group at least, and,

you know, I think they're

they're wanting to kinda build, the APIs for that, like, any node sorry. Any any node could be that is an LSP could could offer liquidity services to to other people. So there's,

you know, whether or not,

that's what we pulled it. So we're we're we're still early in development as well. So, you know, TBD on some of that. But, I think it's a fair point to at least bring it up, but there is, like, a working group at least for, LSPs to kinda, like, make it more

generic. So that that could be something to be positive about.

Unknown:

Ben, you you had a really good tweet early on on Odell, unless you wanted to say something else. Well, I was just gonna say just real quick. I mean, if you go back to SIL dispatch 21, which was our first lightning, like, really focused on lightning privacy discussion on dispatch.

One of the things that we kept talking about is if you wanna use Lightning privately,

you really wanna just keep spinning up new nodes often and frequently and wiping them and creating new nodes. And the fact that you can easily do that in a web browser now with Unity Wallet,

that alone for power users makes it way, way, way easier to interact with lightning privately.

Unknown:

Yeah. I mean, the the the vision that I had back then with, the ability to to quickly spin up those nodes and to, you know, I I kinda almost forgot about that aspect by itself. But,

let's say you have one UTXO and you wanna, like, open up the Tor browser, spin it up,

and, you know, open a channel with some node, deplete the funds by, like, pushing it through the network and then closing that channel,

and then not having any you know, basically, not having any left in that channel,

closing that channel. You're you're basically just, like,

taking that UTXO,

sending it across lightning,

and then all of a and then you just close it you close the tab. You don't have to, like, set up a new Umbro node. You don't have to, like, figure out how to put, like, multiple LND nodes on your Raspberry Pi. You don't have to do any of that. You just open a tab, and at the very end, you close channel, clear state. Boom. You just you just push a UTXO through the light network and, like, no one you know, if you have enough hops and stuff, then, like, no one knows where it came from or or where it went.

Unknown:

And to answer uncle thinks in the comments, node or channel, we're talking about nodes

because

every channel that you open on a given node, they should be considered connected to each other.

Yeah.

And any UTXO used to fund any of those channels should be considered

connected to each other. Any when you send Bitcoin into the easiest way to think about it is if you send Bitcoin into a lightning node,

any Bitcoin you've sent into that lightning node is connected to any other Bitcoin you send into that lightning node. Yeah. Almost consider always the case, but assume that.

Unknown:

Yeah. Almost consider, like, if you spun up if you reuse the Bitcoin address, like, an on chain address, like, that's basically what a normal lightning node is

is a reused

address.

And we're trying to also make it so that, you know, you can spin up many nodes from even in a single session,

like a single tab.

We we still have a lot of work to, like, make that a clean UX,

but, like, so far, that kinda works. So

happy with

that.

Unknown:

Yeah. Like, the base of Unity supports that. Just our

UI doesn't really support it yet.

Unknown:

Awesome, Buys.

I think I cut someone off. Paul, did I cut you off? You were gonna say something? Oh, I was gonna say something else. Like, Ben was talking about,

you know, using,

you know, getting your own block source to someone else. You had an interesting tweet today, Ben, that,

maybe you wanna dive into more about using

someone else's

implementation for fees is is kinda dangerous, like, using someone else's mempool for fees. I don't know if you wanna

dive into that at all, but we def at Muni, we definitely use Wizz's mempool for fees. So,

if you wanna elaborate on that.

Unknown:

Yeah. I mean, that that's kind of, like,

indicative of, like, all the RBF

debate stuff.

But I think I'm gonna write an article about my thoughts on this because it seems

I think about it differently than most people. But,

kind of the idea is, like,

the the point of why we want mempool for BF is so

it's my reasoning is that we I want my mempool to be as close as possible to the next block that's gonna be mined

and, like, therefore,

the

miner's mempools.

And to do that, full RBF makes it more like like, you know, a lot of the high few transactions, so most likely that's what miners will mine.

And the reason I want my mempool to look like a miner's mempool is because

when I estimate a fee for my transaction,

I want it

to be correctly estimated where I you know, if if fees are once after buy, it doesn't really matter. Like, it's gonna get, like, a little bit higher. Irony of we have the mem we have Wiz's mempool up, and it's just, like,

one separate byte, next block. Boom. Boom. Who needs the estimation? Just pay 1 set. The last two blocks are empty. Like, not entirely relevant right now. But, like, if you remember in, like, end of 2017,

when the when, you know, people were paying a 1,000 SaaS per byte and, like, transactions being dropped out of mempool or even, like, I think it was, like, last year during the China ban

and, like, fees were, like, you know, a 100 SaaS per byte.

You need, like, you know, for your, like, you know, depositing to your cold storage, like, it doesn't really matter, like, if it takes 2 days or, like, a week to get in because it's, like, going from your wallet to your wallet. But it becomes extremely,

critical for security for things like lightning

or other layer 2 stuff where, you know, if my forced closed transactions fee rate,

is wrong,

then I could lose my entire lightning channel's balance if my counterparty tries to cheat me.

So

being able to estimate

fees correctly is, like, critical to Bitcoin success, I think, because if you can't do that, then lightning doesn't work, and then then we're, you know, stuck with on chain or some custodial solution, and that's not really ideal. So

I think that's important. And,

like, to go with that, like, we we need to, like, solve

fee estimation properly. And,

you know, I think people have, like, woken up to the fact with that from, like, the China ban in 2017 of, like, we can't just say, like, just do once after buy it, but it's not Well,

Unknown:

that's where we're placed by fee really comes in handy. Right? Because you're able to,

in a sustained high fee environment, which we're not in,

and, you know, I have major pie in my face,

the last time fees were up. So I'm

very trigger shy about talking about sustained high fee environments, but a lot of us expect that to happen at some point in the future. In a sustained high fee environment, what you wanna be able to do is you wanna pay

the lowest conceivable

fee that you think will get confirmed in your time frame.

And the only way you can comfortably do that is if you can increase the fee later

if you completely estimated wrong

and fees run out of control. Like, we saw in situations where,

you know, you you check you either check wiz's mempool@menpool.space,

you check your own self hosted mempool,

and you're like, oh, if I pay 17 sats per byte, it'll confirm in 3 blocks.

And then next thing you know, like, a bunch of high time preference traders all try and use the chain at the same time,

and you fall back, like, 8, 10, 15, 20 blocks,

and you're stuck. You're stuck in this situation. So then what happens?

And and because you can't replace by fee, because you can't increase that fee very easily,

you end up in a situation where the next time you send a transaction,

even though it says, oh, you can just pay 17 sats per byte, you'll get in the next block. You just wanna cover your ass, so you pay 42 sats per byte. And then it becomes a circular feedback loop where everyone's overpaying their fees by 2a half times or whatever because they're afraid that the the fee market's gonna run away from them. So if you really want efficient fee selection, like, you need an easy way for people to

update their fee.

Unknown:

Yeah. And, like, I think important too is, like, you can't always update your fee. Like, but if it's your own transaction, sure, you just you can always just replace it by resigning the second one. But if it's like a lightning channel force close, it's a 2 of 2 multistix. So I need my counterparty to sign it, which isn't always reliable because of, you know, they could be offline or trying to screw me over or something.

Unknown:

So That's why you're forced closing. Right? Otherwise, it's not forced close.

Unknown:

So

yeah. You'd have, like, you know, you're forced closing because they won't sign for you. So, like, you need that transactions fee rate to be good. Otherwise, you could be, like, completely fucked. And,

you can't do that without having an accurate mempool. And, like, we see, like, in other chains, like, Ethereum, like, they don't even have a mempool anymore. It's just like an API to Infura and stuff. So, like, it's

not, like, you know, on the censorship resistant aspect, it's terrible. But, like, for security aspects of, like, actually making a sound layer too, it would then be well, just asking Infura what fee rate to pay instead of, like, actually being able to self host it. And, like, then it could come down to, like, okay, if Infura blocks your IP, you can never be able to get an actual fee rate again, and that would be, like, you know, terrible for your security and everything. So I think,

like, being able to have a sound man pool is, like, critical to Bitcoin's security. And a lot of people are just saying, like,

oh, you know, it was RVF was a thing before so we need to keep it. But it's like, I think it's really bad for the future of Bitcoin if we kept it the old way.

Unknown:

I mean, I wouldn't say it's, like, bad for the I mean, I guess yeah.

I mean, most people are are just gonna use other people's mempools,

but it's important that we we have our own mempools.

I mean, I think that's what I we had,

we had Sergei on dispatch,

and his business, Bitrefill,

you know, uses 0 confirmation transactions. He's, you know, one of

one of the individuals that is strongly against the the recent Bitcoin core change,

which is the ability for users to opt in and flag,

RBF by default. So they only they always are willing to accept a higher fee version of a transaction into their own mempool.

But, yeah, I mean, I I think at the end of the day, regardless of what Bitcoin Core did,

like, miners in a sustained high fee environment are gonna figure out a way to accept a higher transaction

fee. And if they're doing it in an opaque way, then that's worse than it for every other participant on the network because they have no idea what's going on. It's just like all these off band,

like, out of band things that are happening that that my node is just not familiar with, that has no idea what's going on, just completely clueless to the real economy that's happening.

It's just not the situation you wanna be in. Yeah. Like, if you remember our last,

Unknown:

dispatch with Barack where he said he he had to email his transaction. You on that one? Yeah. Yeah. And he or he It was hard to hear you.

Fuck you.

Yeah. He emailed it to f two pool. Yeah. It's like that was never in my mempool. So, like, you know, if everyone was emailing

transactions, now, like, you can never estimate fees again without having access to f two pool's email client, which is, like, you know, I was getting them. Yeah.

So, like, that's

if we don't, like,

move the mempool towards pushing to correct fee, like, you know, getting the most accurate representation of the next block,

then, like, the end game is, you know, we're all emailing our transactions, and that's,

Unknown:

you know, I mean, that's all I get. Yeah. Yeah. But, like, at least we're going to a web API or something and just Yeah. Fucking submitting them out of band and then just saying, like, a lightning invoice. Or even in 2017, we saw people doing it with credit cards.

There was, like, there was mining pools that would accept a credit card payment to double spend a transaction.

Unknown:

I think those still exist too.

Unknown:

Ridiculous.

A future finance. Pushtx.com

Unknown:

or something.

Unknown:

Yeah.

You guys wanna talk about AI real quick?

Unknown:

Yeah. I know I know Paul would love to. But you you do you mean my new best friend?

Unknown:

Do you have, like, conversations with with with the chat the chat AI? All the time. What's it called? Chat GPT?

Chat GPT.

So what are your thoughts on it? Is it a positive?

Unknown:

I I yeah. I just love it. I mean, I've been doing a lot of it with code stuff.

So just, like, just being kinda kinda lazy and,

just having it write code for me that I usually can

pretty easily copy and paste and modify just a little bit, or I just

the the the thing I really like about it, and I feel like I had the sensation, so I I've I've been talking to chat GPT all the day. And so you you ask it something, and they're like, okay. Now can you code me example of that in Rust?

Okay. Can you change this? And and it it it it's this really great iterative process,

and I really like this

for me. I mean, I'm relatively, you know,

I guess what you say, a verbal person. I used to be a writer. Right? So that's it's a very comfortable environment for me to be writing these plain text prompts.

And then I, like, I jumped over to

to Amazon,

and then I made a search for,

for a nap pod.

And it was all these

pod nap pods for babies.

But I I want an adult nap pod, and I immediately had this sensation, like,

because I've just been spending all this time with the the AI.

I I wanna like, but for adults, like, that's how I would've refined that that search. And I'd have a prompt right right there, and I would've been able to refine it and conversationally,

like, hone in on what I'm looking for. But with chat or with traditional search, like on Amazon or Google,

you append to the original query or you write a new query, and you're kinda, like, blowing away all the context each time. So I think there's something really, really powerful about this sort of,

conversational

contextual,

way of

of discovering something, and and it's because it's kind of ultimately,

it it it really is search. You're basically searching Reddit and Stack Overflow. Flow. It's like Google Search 2.0. Right? Yeah. You're yeah. You're looking for what what do normies think about this thing? What is what's, you know, common, like, development? Like, sometimes even, like, if I'm searching for some rusts, you know, asking it to write me some It's like what Ask Jeeves pretended it was gonna be.

Yeah.

Yeah. And then and then there there's potential there. The the wild thing some some people are doing these crazy experiments where they

they, like, wire it up to repl it. So, like, you can ask it to write you a, like, a a Python function

that queries a certain API

and then, you know, manipulates the data that it receives back from the API. Right? But it it it can't run it for you. You'd have to copy and paste that somewhere. Well, so some people are adding replit to this loop

so that you can,

just have it do these live queries for you,

on the web. So because right now, it's sandboxed. It it's just based on, like, a a scrape of the Internet as of, like, 2021, supposedly.

But Right. It still thinks Twitter's a public company. I saw someone post that. It can it can write the code that will query an API,

and then so you're basically you have a chat user interface for querying any

open API on the Internet now.

Like, you can have your plain text interface

for for asking about, like,

you know, sports scores and then refining them. Again, you know, with this kinda hack of the repl adding replit in the loop. But I think that's the sort of thing we're

we're going towards. So we have, like, play basically, plain text user interface for almost every

API in the Internet. And then I somebody had a really good line. I I I'm I'm not I'm gonna butcher it, but it's like, it's not about,

hiring like a like a AI to be, like, replace one of your smart engineers. It's like it's like what if you had unlimited dumb people working for you?

It's like, like, what what would you be capable of if you had, like, you know, 1,000,000,000

junior engineers and you ask them to do things for you, and you conversationally would refine what that would be. It, obviously, is not gonna solve every problem,

but there's a lot of things I think could be really helpful with. It's it's really it's really fun to play with.

Unknown:

Yeah. I'm I mean, I I've been using, GitHub Copilot

since that came out, and that which is,

Unknown:

basically use JAD GPT 2?

Unknown:

Use the Is that different? Use the GPT 3, so, like, same

Unknown:

base type of It's the

old it's the old tech, though. Right? It's, like, the generation below?

Unknown:

Maybe. I'm not too sure, but it's just the same. They're both chap GPT. I've they're both GPT 3. I've heard some people call chat GPT, like, g p t 3.5, but they're basically the same conceptually.

Unknown:

Yeah. And that that, like, I think is

like, when I started using that, I was like, this sounds like a scam, but they gave me a free month, so I'll try it. And after, like, 2 days, I was like, okay. This is, like, the best thing that's ever happened to me. Like,

it's pretty insane. Like,

it probably wrote 10% of the vortex code base just because it would just auto fill functions for me. And, like, you know, it's it works.

You know? And, sometimes there's, like, bugs in the code, but, you know,

write good test and you can figure that out or just, you know, review it. Like, it's not like, like, I I love I never heard that line, but, you know, it's a 1000000 dumb people working for you. Like, it's extremely good at writing dumb things for you. Like, I I am terrible at Yeah. That's why I think it's gonna be great at engagement tweets.

Yeah. And, like, the you were saying about that yesterday. That's, like, I remember when you did that, I I really typed it in, like, 5

tweets about Ethereum, 5 tweets about Bitcoin. I was like, fuck. These are, like, perfect handheld tweets. Like,

these are this is gonna do so well.

And,

it's scary. Like, I was thinking, like, last, like, like, last night, like, you could literally have, like, our entire Twitch chat could just be, you know, bots now, and it'd be hard to tell.

And,

you know,

that's fine for, like, spam on Twitch chat, but, like,

you could, like, totally ruin, like, people's, like,

time and, like or, like and I saw, like, Stack Overflow is now banning,

chat GPT

Unknown:

response. But they have no way to enforce it. They're like, if we catch you, you're just done. Yeah. We're gonna sanction you. But it's,

Unknown:

I I agree. There's no way to enforce it. But it's like,

you know, this is gonna be a problem. And

I I think said on Twitter today, it's like it's gonna break the Internet because

we're gonna lose the ability to, like, be able to tell a human from, like, bot because, like, right now, like, you know, like, Twitter bots

Unknown:

are normally, like, even, like, you know Yeah. Like, the bots we have that just joined the chat, I I Yeah. I banned her, but it says, like, free Vert Girls Chat in all caps with a kissy face. Like, we know that's a bot. It's very obvious.

Unknown:

Yeah. But, like, if it was like I don't know.

Unknown:

That's pretty sexy.

Unknown:

The Twitter DM bots were like, how's your trade going? Like, you know, it's extremely obvious. Like, it's just like this random reply of, like, hot Asian girl versus, like, if they're in a way what Tim Pastor put in my comments? He, like, responded to my tweet complaining about the flood of engagement tweets we're gonna get. And it was you I would have thought he sent it to me.

And then he admitted to underneath that it was

the chat GPT.

Unknown:

Yeah. I saw multiple of those. It's like, that's terrifying. And, like,

that's gonna

that's gonna, like you could, like if you just wanna, like, hey. I wanna fuck over.

Like, you know, I I I work at the Bitcoin company for, like, hey. Fuck Beat Refill. They're a competitor. If we just, like, sent a 1000000 chat chat GPT bots to, like, flood their customer support, we could probably easily do that. And, like, you know, you could fuck over a company that way.

And, you know, you could do the same, like, anywhere, and, like, that's kinda scary. Or or, like, or like content marketing. You know, it's super valuable for companies

Unknown:

to have, like, a,

like, a blog, and they they post, you know,

relevant

things on this blog. And a lot of it's about building,

engagement

and also their search rankings for various things.

So so I I just saw somebody like,

hey, chat GPT. Generate me some prompts for content marketing blog posts for a dog grooming business. And I was like, okay. Now write all of those for me. Okay. Now make me a website that, you know, serves the you know, like, it's a full it's a full stack spam engine.

I I this has already happened to me. I was I was trying to figure out something,

I don't know. What browning butter is not the right term. I was trying to figure out something in the kitchen. I was like, what how do you do this thing? Like, the recipe told me to do a thing, but I didn't know what that meant. So I was looking around. And I'm I'm on this article, and I'm trying to understand it, and it just it's just not clicking. It's like, am I the dumbest person? But it starts kinda repeating itself, you know, and it's kinda meandering, and it keeps on kind of recur and it, like, I oh, this is just AI generated text. Like, somebody made up a spam recipe site and just, you know, fed it some prompts, and it's coming up with, you know, SEO spam.

And, you know, I'm to a search engine, it I'm sure it looks pretty legit. And it took me as a as a full blown human a little while to figure out. I was just reading,

nonsense.

Unknown:

So, I mean,

I'm probably in

I don't know if paranoid is the right word, but I'm in, like, the

I think everyone in the I think both of you are

Unknown:

No. I'm with you.

Unknown:

We're in, like, the top tier of paranoid.

We're, like, in the top 2% of paranoid in the world or adversarial thinking, let's say.

And,

you know, this this all kind of comes together. Right? So we're seeing deepfakes getting

really impressive.

And at the same time, we have,

you know, this essentially what is a new kind of search interface. But it's not just giving you results. Right? It's also giving you kind of a human readout of results. Right? It's it's it's acting like it's it's trying to emulate what a human would say back to you. Now

the way it acts is basically what it's trained on. Right? It's it's they train it on all these different datasets.

They've been known to use public Twitter posts

to do it.

All these different places they can find data. Right?

So you mix that with this

surveillance

apparatus that we live in, like this Panopticon society where everything we do is recorded.

I've already come to terms with the fact that, you know, I have 2

I have 2 main shows. Right? I have rabbit hole recap and solo dispatch. There's thousands of hours of me talking about pretty much everything.

I'm gonna be one of the first people to be deepfaked.

You mix that with something like this GPT

stuff,

And

literally in,

you know, maybe in in the course of 10 minutes, you can probably do a podcast

where you can have me say whatever you want me to say,

and you just have to all it takes is, like, a couple sentences of a prompt.

There's a video of me on rabbit hole recap for a subset of those thousands of hours, so you might actually even be able to mix that in with a video deep fake. But you start going down this rabbit hole,

and we already kind of I think the safe assumption is just assume that everything on the Internet's fake

to begin with.

But you go down this rabbit hole, people have Alexas in their homes,

you know, recording every conversation they have. They have Amazon

cameras on their on their front doors

just aimed out on the street, recording everyone who walks by their house and walks up to their house.

They're sending their DNA to all these different,

to to these different companies that have terms and conditions that read, like, terms and conditions.

Like,

it can get it can get very, very dark very quickly. And by the time

we hit that, it's almost

it's almost too late. And then not to mention, I mean, Sam Altman's behind this chat GPT thing,

and he's of Y Combinator fame, but he's also the one of the main promoters of Worldcoin where he wanted us all to scan our eyeballs to

get his shitcoin airdrop.

So, like,

I I I think it's kind of unavoidable at this point. Like, someone's gonna build it. It's gonna happen. I'm not like

I don't wanna be the guy complaining about it,

but I'm the guy who's concerned about it.

Unknown:

Yeah. I I agree with that as well.

I

think it could be a useful tool,

but when it becomes too useful of a tool,

then that's that's maybe a concern.

I like, to not be as dark about the

the the

the world, like, the I guess dark as that was,

I from an engineering perspective,

to to kinda make it a little lighter, I'm kinda concerned about,

the reliance,

not just engineers, but everyone, their reliance on using,

AI to solve their problems

instead of, like, figuring it out themselves or having the context or intelligence to to do those things. So, I mean, sure. I just I, you know, Google search every engineering term that pops into my head when I'm trying to solve a coding problem. But,

you know, maybe maybe this is just a better version of that. But if it's too good of a tool,

that is that's kind of my concern.

Unknown:

I I don't think that really makes like, you could say the same thing about a calculator,

like, a 100 years ago and be like, oh, you should do the wrong And this is this is why I can't do math. This is why literally why I can't do math.

Yeah. But you don't need to. That's the thing. It's it's an evolution in technology kinda thing. Like, I know, like, it's gonna exist either way.

And, like, I I agree with, like, the data collection stuff is, like,

extremely bad.

But, like,

personally, like, you know, you can opt out of that the best you can. You know, don't have an Alexa, you know, coinjoin your funds,

you know, all that stuff. But, that we know. But

and, like, all the normies are giving up their data, so

it's gonna exist either way. And I think, like, it's a very powerful tool that's gonna benefit the world, like, in some ways good, some ways bad.

But normally, technology leads to a good thing. So,

Unknown:

yeah, there's definitely it'd be a huge learning. Curve. I mean, we haven't even we're not even halfway up over the hump of the learning curve as a species of how to use smartphones and not, you know,

Or, you know, whether to give a tablet to a kid or that you know? There are all all sorts of Do not give a tablet to a kid.

So, like, we're we're still

figuring that out.

So there's definitely gonna be a lot of figuring out here. My hope is, 1,

that

AI will be able to train AIs that will detect,

AI.

Unknown:

And AI

versus AI warfare?

Unknown:

Absolutely.

Yeah. It's it's gonna be that's our our our best hope.

Unknown:

It reminds me of you know the radar detectors? You guys have radar detectors in your car?

I think there's a very strong overlap between people who have radar detectors in their car and then people who care about digital privacy.

Unknown:

Like, like, police radar?

Unknown:

Yeah.

But but you do not. Okay. But radar detectors were, like, the first radar detectors were very basic, and they could detect

police using radar detectors.

And then and then police

had radar detector detectors.

And then radar detector companies came out with radar detector detectors that detected the detectors detecting the radar detectors.

Oh, my god. It was gonna be like that.

Unknown:

Yeah. Okay. So so there's that. There's there's also Well,

Unknown:

sec. I act I would I would be perfectly I would be way more fine with it if there is something where it's like, hey

hey, chat GPT. Did this come from you? Did this string attacks or did this code come from you? And every every pull request that Paul does that he does it, he just blindly copy and paste in code. I can ask gbt

Unknown:

GPT. Did it did did this code come from you? And if it did, I'm gonna make him rewrite it. You know how you could tell right now in our code base which parts were written by GPT? The The Russian code. The ones that are really well commented.

Sometimes I'll delete the comments just to to to

Unknown:

to You slip in a bucket too. Well no. I mean, that's the thing. I I think I think Matt had said it about it's using, like, human responses and stuff. Like, it is

intentionally putting in bugs as well

because humans also make bugs as well. So, like, it it gets things wrong

sometimes on purpose when it when it could, quote, unquote, know better.

It's giving you a human response. So it's it's it's being trained on flawed information because we are flawed humans,

and it's got, you know, ramifications

Unknown:

for doing that. But it's more than that. Right? Like, I wanna

Unknown:

so I went on Obviously, no one here has, like, watched The Matrix or

Unknown:

or any of these. Yeah. Go watch The Matrix, freaks.

So I I

I tweeted out

out, our our institutions have degraded just degraded to such a point that it would probably be easier for us to relate to an AI generated politician than our existing

ones. And

someone responded

with,

let me see if I can find this response.

AI politicians would probably be less robotic.

No. That's not it.

Someone respond someone asked chat gpt.

Someone asked the AI. Oh, yeah. Raijendale,

he's been on the dispatch before.

He goes, write a short story about AI generated political speeches being targeted at individuals.

Because Wiz said, don't worry. Now everyone will have their own AI powered politician that writes custom rhetoric to extract votes for them. So then

Raijandale went down that route with Chat gpt,

and Chat gpt was, like, in the future, political campaigns relied on advanced AI algorithms to create personalized speeches for each individual voter. And there's, like, 3 different paragraphs that goes down how all these politicians were using it, blah blah blah.

And then the last two paragraphs go,

eventually, a group of activists banded together to expose the truth about the AI generated speeches. They revealed that the algorithms were not just predicting people's preferences, but were actively trying to change them. They exposed the dark side of the technology, and the public outcry was swift and fierce.

The politicians were forced to abandon the AI systems, and the public demanded more transparency and accountability from their leaders.

The era of AI generated political speeches was over, but the lessons learned would not soon be forgotten. Oh my god. That's that's AI psyop psyop.

Yeah. It's got built in propaganda

that's, like, telling you, like, oh, don't worry about it. Like, we're gonna generate the speeches, but, like, don't worry. The public's gonna stand up and stop it from happening. It it's it's a it's a reason to get,

Unknown:

it's a reason to get smarter

in some way.

I don't know if that's very helpful, but but, like, you you know how Gmail for a long time now has had this thing where you start typing in the actual Gmail UI? I I haven't used it much recently, thankfully, but it it I don't use Gmail? It's it's like it's like if it's

beyond just auto complete, like, that your phone suggests the next probable word, it suggests the next probable sentence.

And I use it kind of as a way of seeing, well, what would be the cliche,

boring thing

to say right now? And if you look at, you know, what a chat chat g p t response it is is it is kind of the most middle of the road

cliche,

boring thing to to say. I I posted in in the,

Citadel dispatch

chat that, you know, I asked it about,

you know, what are the potential pitfalls of building a lightning node for the web, or what are good reasons to build a lightning node for the web? And it said all the things that I've you know, a lot of the things I've been saying already, but they were kind of the most cliche, boring things that I've been saying.

So it kinda made me think, okay. Well,

why don't I think about this differently? And rather than just being what whatever the knee jerk first Reddit comment would say,

you know, what would a thoughtful

person who maybe spent a, you know, an hour,

with his eyes closed, thinking hard about something, you know, What could I come up with that's a little more original?

So I I kind of, I kind of view when AI, like, eclipses us on something that we thought was kind of a human activity, I think if it's kind of a it's a prompt to us of, like, hey. This is this is the stuff that we do that's not the most special.

Let's focus on the things that we are,

we can be better at.

So I I I hope that we, you know, that as we realize that the politicians

are targeting us with AI generated text, that we use it as a a a way to learn,

how manipulable we are and how, in some sense, simple we are, and we kinda have to level up a little bit.

Unknown:

Yeah. I mean, I think of it as, like, there's a lot of shit where you're just like, I need to do this, and, like, you you can think of it exactly in your head. Like, I'll do x and y and z. It's very simple thing. I just need to spend the hour doing it versus, like,

you know, in this is no thought process thing. You could watch TV while doing it versus, like, other things. You're, like, I need to, like, spend a week and learn this and figure it out and do it. Like,

AI will solve the first problem there. It's just like I need to spend time just, you know, doing the actual

monotonous work of figuring this of, like, you know, doing the thing I know how to do versus, like, the actual human work of, like,

learning stuff and, like, you know,

solving a problem

will be done by actually humans, like, you know, because we're always

we're all we'll always be, like, inherently smarter in some ways than AI because

AI can't, at least in this current form, like,

earn new things. It's just like, you know, it's just very good at Googling, basically.

Unknown:

Well, what And a a definition

what humans are now anyway? Like, what A definition of wisdom is, like, is like being able to it's not just being able to provide good answers. It's being able to provide good questions, which is kinda what you're learning to do when you're having a conversation

with with my best friend, Chad G.

Unknown:

So, like like, I would say, like like like, Paul, I'll give you example of, like, asking the lightning privacy stuff. Like, I did not give anything in our article

or, you know,

Hippo, whatever you wanna call it, thing. Like, we get very monotonous, like, you know, use Monero or something, like, you know, default answers that Google will give you that's not a real solution.

Versus, like, if you wanna do, like, the actual, like, hard work of, like, figuring stuff out, you need to, like, actually, you know, find a human and think about it.

Unknown:

But did you see the answer about building a lightning wallet in a browser? It's, like, a pretty decent answer.

Unknown:

It it it was a pretty decent answer. I was impressed by that, but the the true test was all it had to do was copy and paste my lightning article from a year ago since it would have been included in its,

in its metrics,

and it didn't do that. So not impressed.

Unknown:

Maybe you could ask it for some SEO tips.

Unknown:

I'm I'm number 1 on Google. You know that. Right?

Unknown:

For lightning privacy. It's true. It's true.

Unknown:

I mean, I think, like, if you

Unknown:

I don't know. I,

it should be interesting.

I mean, I I know the AI will never, you know, fuck up people's names, will never curse on air.

Like, I think Citadel Dispatch is

is pretty safe just because it's it's incomplete. Right? It's not a polished show. It's not a

Unknown:

Raw humans.

Unknown:

But, like, if you compare if you compare

it's unedited,

it's live,

like, it there's there's a difference there. But if you compare it to something like

I mean,

people already have a negative view of the New York Times.

Like, in the current state of

this

work in progress,

closed source

app,

it could probably already produce a better newspaper

than The New York Times

if no one knew that it was AI generated.

Right?

Unknown:

Certainly better than the verge.

Unknown:

Well, you'd still you'd have to to to supply it with

you'd have to supply it with the,

the the news event, like like a 2 sentence summary of what just happened,

and then,

like, give this a,

a liberal elite spin,

and then you'd be good. Then you have a newspaper.

Unknown:

It is hard, though. Like,

because, like, it's probably it's like, if you were to, like, make a newspaper based off just the AI,

like, reading the Internet, like, it

would largely just, like, agree with everything that the normies say. So it'd be like, oh, yeah. Bitcoin is blowing the oceans. This is a fact, and it would report on that every day about Bitcoin because that's what the norm is. What it's trained on. Right? Yeah. Like, the dark stuff, like okay. So so this is a show about

Unknown:

dispatch is a show focused on open source software,

individual empowerment,

self sovereignty.

Like,

this is a closed source stack controlled by a monetized company

with questionable motives. Like, no one knows what the motives are. Right? They can feed the how much they wanna feed it. Yeah. The dark shit happens when you start feeding

non public

normie stuff.

Like, what it like, if you had

and this tweet didn't bang yesterday because it wasn't written by AI. It was just written by my drunk self.

But, like, if you had

if you had,

an AI trained on on

BlueJeck's,

like, private DM history

Unknown:

or

The Twitter files. Arnold Schwarzenegger's

Unknown:

kitchen conversations from his Alexa.

Like, things start that's when it gets really

that's when you start mixing in,

you know, the data surveillance machine in a closed black box type of operation,

then things get really dark really quickly.

Unknown:

Right. Or even,

you know, declassified

government, you know, records or even classified ones or something. Or I mean, I think there's even

an art I think there's a whole movement of or some organization or something that's trying to stop,

AI from running on code bases with with, you know, more restrictive licenses on them,

because that's certainly something like a mat like, you feed in a bunch of GP 3

code bases,

and and all of a sudden, like, okay. What? They can just monetize

my AI data?

Like, if it's if it's closed source licenses or,

you know, restrictive licenses, like, that AI shouldn't be fed those those, code repositories. Right? But I'm sure there's, like, GitHub terms and services that are out there that just, like, give it

kit where they can do that no matter what license is on it, or they at least feel justified in feeding that your code bases. And, you know, for for, I think l unsploit

was on,

a self hosted

Git instance, Gitia,

and we were doing that with Mutiny web for a little bit. We were running in a lot of CI problems, and it was just easier when we're moving so fast to just get back on GitHub for a little bit.

But I wanna move back off of it

explicitly because I don't want I don't want my damn code bases training AI.

And, sure, like, like, you know, if it's a public repository, they can always do that, but I feel like they would be less inclined. It would be harder for them to,

like, you know, someone like GitHub to scan,

code bases outside of its own platform than code bases that it has unborn. Yeah. They'll do the low lift thing first. Right.

Unknown:

Yeah.

So did you see what Spy Hunter posted

in the chat?

Unknown:

I can't see it.

Unknown:

You have to, like, pull it

up separately. He did

describe the dialogue between a group of people, Tony, Matt, Ben, Paul, on the Citadel dispatch podcast discussing how AI will impact the future of humanity.

And, I mean, obviously, it doesn't know which one of us will think which things, but it kinda just sums up the conversation we just had.

Unknown:

Yeah. So Tony,

in this case says, I think artificial intelligence has the potential to revolutionize

the way we live and work. We can see a future where machines interact with humans

in ways that are more effective and beneficial than ever before.

Unknown:

And, Matt, do you wanna say how you go about it? I go, absolutely. I think AI could be a huge asset to humanity, allowing us to solve problems faster and more accurately than before. It could also help us create new forms of entertainment like virtual reality experiences.

Ben, are you gonna read your part? Are we doing a table reading?

Unknown:

Yeah. I'm not so sure. I think AI could be used for nefarious purposes as well, like creating advanced weapons systems or manipulating public opinion?

Unknown:

I think it's important to remember that AI is just a tool, and it's up to us to decide how it will be used.

We need to ensure that we use it responsibly and ethically in order to ensure the best outcome for humanity.

Unknown:

Yeah. We're all fucked.

Unknown:

We're so fucked.

Unknown:

You know you know the move to would be so so we what we really need in AI is it to be training more responsively, but you could you could imagine

you you get chat g p t. So you record the first half of a podcast episode,

and then you just give it, like, 5 topics and say, like, hey. Riff on these for the next half. Finish it. Just finish this episode for me.

I got somewhere to be.

Unknown:

Fuck.

That's gonna be out of job.

Unknown:

No. I mean, I'm fine.

Unknown:

Make sure to stream your sats to chat chat GPT.

Unknown:

This is right.

No. But, like, the Peter McCormicks of the world are I mean, he's just gonna use like, why why even record the podcast? You might as well just

feed it information and then just have it go for it. Right? It'd probably do a way better job.

Unknown:

Pretty yeah. I don't know.

Unknown:

This is why it's important to I like,

everything is fake except for Bitcoin.

Unknown:

So Do do you sign the, like, the hashes of your episodes before you release them?

Unknown:

No. But it's just live. It's live unedited.

I mean, it's gonna take a while before

they can do live unedited interacting with the live chat.

But you could if somebody said, hey. Matt said this on a podcast episode, you'd be able to prove it. Well, so you wanna hear something that I was thinking about today? So so like, every morning for the last

50 days, I've

woken up. I've enjoyed my morning a little bit. I've had a coffee, and then about a half an hour later, I could sign on to Twitter and I tweet out, stay on good morning, stay humble, stacks, hats. Right?

And I type that myself.

It's not scheduled.

I don't know if that really matters. I'm obvious I'm not doing it for engagement. It's literally I'm I'm mostly doing it for myself, but everyone else might as well

hear my morning

motivational or whatever, my morning,

mantra.

So I was thinking, like, oh, like, I could do

a hash of the message

with the block height and the hash of the block.

But, like, you could just totally do that.

Like, that could be faked too. There's no way to prove

there's no way anyone can prove any tweet

was not a scheduled

GPT tweet.

Period.

Right?

Unknown:

I I think latest block, like, if it really was the latest block.

Unknown:

But why don't you just tell GPT,

like, every morning, send out a tweet, take the most recent block hash and block height, put in the tweet, and press send tweet. I see. You can make a bot to do that super easily.

Unknown:

Yeah.

I mean, Ben, you've run a bot that anchors people's opinions into the blockchain.

Unknown:

Exactly.

Yeah.

People yeah. I I know there I know there is a bot that uses my bot. So

Unknown:

So my point is

short of actually being in the room with you guys

and, you know, you're, like, groggy, you're waking up, and I, like, turn and look at you, and I go, good morning. Stay on with Stack Sats. There's no way to prove anything

going forward

except for Bitcoin ownership.

Unknown:

You you can still prove things in the past

if you let's say you wanted to review it at a later date.

I know, like, the Peter Todd open time stamps thing is something like that. But the problem with that is that that GPT just did that at the time and then

Yeah. I mean, you know, maybe if it's signed with your own PGP keys, like, are you gonna give a bot access to your gbg gbgp

Unknown:

g keys? I mean, you might. Right? So Maybe the bot fucking generates them for me. Maybe that's the solution to people finally using GPG is

that they they just ask the fucking chat app to generate keys for them.

Unknown:

Right. But I would I would assume you would want to I mean, you know, for for people like yourself or, you know, for hardcore people, you would assume that they're not gonna give they're not gonna let a bot generate their keys, and they're not gonna give a lot of access to their keys. You could you could, like, code your own bot self hosted that pings the open chat API

Unknown:

to get responses and then does the signing locally.

Unknown:

Yeah. Like,

Unknown:

after turn by, I don't like, Twitter doesn't have my private keys to my lightning node. It does it itself.

Unknown:

I mean, the my point is, like, we live in a post

Unknown:

truth world now. Right? Truth. Yeah. For a while. Truth is out there.

Unknown:

It's the entire life, really.

Nothing is real. This is all an illusion.

You you can't verify anything that's not directly in front of you.

It is all fake anyways.

Except And that's the and that's the episode.

Unknown:

My dear boys. Well,

I'm glad we got to talk about AI for a little bit. You can't trust

Unknown:

Tony when he says something like that because it's self refuting.

Unknown:

When are you boys coming back to Nashville?

Unknown:

I wanted to come this week. It's just Mexico and this hackathon thing. Oh, yeah. I'm jealous of back. You wanna show Mexico real quick

Unknown:

to the Yeah. Yeah. Absolutely.

Unknown:

Nifty is hosting a a banging BTC plus plus conference in Mexico City.

I I believe tickets are probably still available. It's all privacy focused, not specifically on chain privacy focused, so it's not too late to get tickets.

The hotels and everything are super cheap.

So definitely sign up. I'm talking Ben aped in, so he's gonna be talking about Fortex.

I'm gonna be talking about, like, some,

ways to use, like, on on get on chain privacy through lightning.

But, yeah, it should be should be a banger. It's,

Friday. Yeah. You got the dates pulled up. December 9th, 10th, and 11th.

Should be a lot of fun. It's BTC

Unknown:

plus plus dotdevs, and you have to spell out plus plus.

So plus is how you spell plus. BTC plus plus dot dev.

If you care about Bitcoin and privacy, consider going.

It's gonna be really great.

I will unfortunately not be there.

But I have massive FOMO, and Nifty is amazing.

Unknown:

Yeah. I think that this speaker list is hella stacked. If you wanna scroll down, like, I I can't remember now who the top like, Adam Gibson.

Yeah. Tony.

All these guys are really good.

Unknown:

Nothing much

Unknown:

Yeah.

Is a big one.

Unknown:

Thing, Peter Todd, Sergei. Yeah. Dusty.

All good people.

Unknown:

I love.

He's a real one.

Unknown:

I think the, the February,

Nashville meetup should be pretty fire. I'm gonna try to come in for that. I think it's all about open source.

Unknown:

Doxed.

Unknown:

Doxed.

Unknown:

Yeah. I mean, you should definitely come for February. February, our topic is false, so it'd be great.

Unknown:

Maybe I'll have to come. I was talking to some Bitcoiners recently. We're like, we traveled so much this year, and we're, like, enumerating the conferences. And I was like, well, there's only, like, 3 or 4 conferences.

And we realized it was just we went to Nashville, like, 10 times this year.

Unknown:

But

this is fun. I love Nashville.

Unknown:

The the Nashville Austin channel is strong. Very

complimentary cities. I've been to Austin many times,

this year. Well, anyway, boys, this has been a pleasure.

Before we wrap up, let's let's do some final thoughts.

Paul, final thoughts.

Unknown:

Chad Chad, GPT is great.

Don't be afraid.

Just embrace just embrace the future.

Unknown:

Thank you, Paul.

Ben, final thoughts.

Unknown:

Use the link privately, read our thing,

use mutiny,

stay humble.

I don't know.

Unknown:

Thank you, Ben. Tony, final thoughts.

Unknown:

Final thoughts. Don't use mutiny yet.

Don't listen to Ben.

But we are we we do wanna make it really good and and a lot better. So, you know, definitely reach out if you're

interested in helping or even getting user feedback and stuff. But, besides that,

read the lightningprivacy.com,

you know, book and and,

help contribute to that as well.

But besides that, thanks thanks, Matt, for having us on.

Unknown:

Awesome.

I appreciate you boys.

This conversation was great. I mean, I just really missed you guys, so it was just fun having a conversation.

And we obviously broadcast it out to the world as well and to all the AIs out there.

I look forward to kicking with you guys sometime soon.

Huge shout out to our freaks who joined us in the live chat and to the freaks who continue to support the show, whether that's through podcasting 2 point o or going to the geyser.fundpage@satsperminute.com.

Like I said before,

we broke geyser.fund.

You can still contribute directly there.

You just can't

view the leaderboard, but the leaderboard should be up there

midweek, something like that, they said.

There'll probably be another dispatch this week. I got a bunch of ranchers coming into Nashville. We'll We're talking about the beef initiative. We're talking about self sovereignty, securing the food supply.

A lot of fun stuff. A lot of good people. If you're in the Nashville area, come on Wednesday. Oh, tomorrow, we have, I think, our 3rd or 4th bid devs.

We've been averaging around 50 people there,

including a bunch of kids from Vanderbilt and Belmont, which is awesome.

So definitely come to BitDevs if you're in the area tomorrow,

and then the meetup is on Wednesday.

It'll be a live rip of rabbit hole recap.

After this week, probably going on a holiday break,

then starting dispatch up again,

in January with a mining dispatch in person because we have a great mining

meetup,

happening

that we expect over 200 people for in Nashville. You can find information on that at bitcoinpark.co,

which redirects you to meetup.com,

for now until we build the website.

But, yeah, just I wanna thank the freaks for supporting the show, keeping us on the top of the fountain charts is really

amazing. I appreciate you all.

And, if you subscribe on if you use YouTube. I'm like so

this is, like, the problem when I have to, like, battle with influencers and stuff. Like, I don't even think you should use YouTube. But if you use YouTube, subscribe to the show. You'll get notified. If you use Twitch, which is owned by Amazon, subscribe to the show. You'll get notified.

Otherwise, subscribe in your favorite podcast app, ideally an open source app.

You know, I really like antennapod.

It's completely false.

Share with your friends and family over Christmas. Enjoy your Christmas. I think we're gonna have another podcast before Christmas, but if not, enjoy your Christmas.

Get a Christmas tree.

They're great. They smell good.

Don't get, like, a plastic tree. Those are fiat. Don't ask the AI what kind of tree to get. Just do it.

Enjoy your friends and family.

Stay humble and. Love you all. Thanks, guys.