A (3:17)

Those are always the best. Especially when I can prove that. Yeah.

C (3:20)

I am a stream owner. I'm a creek owner in my local Raleigh neighborhood. I've adopted a stream. I'm going to clean it out twice a year. Just shows that I just care about the world more than other people might have some ulterior motives for that, but upstream.

A (3:50)

And then that's exactly how I know I hate the environment. Is that I specific? That's. I. I never adopted a stream on purpose because I thought about like if I. Because I want the world to fall apart and become trash. I just think I should never at least adopt a stream.

C (4:10)

Well, you give jobs to people like me, you know, broken window, fallacy.

A (4:15)

I'm a job. I'm a job. Subsidize. I'm a subsidizer.

C (4:20)

But you allow me to improve my social credit score, you know, by littering in my creek.

A (4:27)

Yeah, but next time. Is it over behind your. Is it over behind your place is like. Is it near the.

D (4:33)

Yeah.

C (4:34)

It may or may not be a stream that runs into a secret underground tunnel that I may or may not have plans for.

A (4:41)

Like a Batcave.

C (4:42)

That's. That's beside the point.

B (4:44)

You're not going to off grid mine some bitcoin with a hydro.

C (4:48)

I can't go public with some of the things plans I have. But I've the.

A (4:54)

We asked you to clean it up, not build a dam.

C (4:57)

City of Raleigh has decided that I am the appropriate person to decide what is what, how that stream should be maintained, and that stream includes the tunnel. So I've been granted authority. So anyway, dude, the second reason I'm doing awesome is because I was on the Fred St. Louis Federal Reserve data website. You know, the authority, official data source of everything. And I click on their bitcoin section and their price chart for bitcoin says underneath in huge letters, not even small letters, this data is copyrighted and cannot used, cannot be used without permission from Coinbase. So this just lit a fire under me. I was like, what? And I was like, I was looking at all the other data on the Fred, the bitcoin price chart data, the most authoritative bitcoin data, price data in the world is the only thing that's copyrighted on the whole Fred website.

A (6:07)

That's amazing.

C (6:08)

Isn't that amazing? And so I was just like, oh, this is it. This is. This is how I make utx Oracle the reference price. It'll be the only non copyrighted bitcoin price data.

B (6:20)

Well, you should try and get sued by Coinbase.

C (6:24)

Yeah.

B (6:25)

And then leverage all of the attention you get from that. It's like them trying to ban the white paper and stuff.

C (6:32)

Exactly.

B (6:33)

Yeah.

C (6:35)

So I've been working really hard and I almost. I'm very close to having a perfect algorithm that works every single day back to 2016 and the Fred only has it since 2014. And before that it's just kind of like really blurry. But yeah, there's just this one period, you know that like blow off top at the end of 2017, early 2018, like the chain is just a disaster then. It's just really hard. It's really hard to find price lines. But I'm just. Dude, I'm in the trenches every day at 3am trying to tease out the price line from the smallest little trace of a signal of the price on chain and trying to update the algorithm to find it. It's fun. Dude.

A (7:28)

That's awesome.

C (7:29)

It's a cool project.

A (7:30)

You're a requisite blockchain architect. Yeah.

C (7:34)

Or not. Architect.

A (7:37)

Archaeologist. Archaeologist. Thank you. Thank you.

C (7:41)

Now mechanics. More the architect.

B (7:44)

I'm a cheerleader, man.

A (7:47)

Wait, wait. I have to ask you, Steve. I have to. I have to ask you something. So did you get the. You posted the picture of the tube of the tunnel in bitcoin and. And you. And then immediately it was like I adopted this stream. So did the tunnel make you decide? What? Why did you. Is the tunnel the thing that you want?

C (8:15)

There's something in the tunnel, man, I'm not going to explain in detail. My bar might be expanding underground and there might be some other things that go on in there.

A (8:27)

Okay.

C (8:28)

It's secret.

A (8:29)

I see.

C (8:30)

Yeah, it's secret.

A (8:31)

Understood. Jeff. How's it going, boss?

D (8:34)

It's going pretty good.

A (8:36)

I got your message to join early so we could talk and then didn't join.

D (8:39)

That's fine. I.

A (8:45)

I don't know.

D (8:45)

I'm playing with. I'm. Most of my immigration paperwork is. Is done for now and we're just waiting.

A (8:53)

Sweet.

D (8:54)

And the weather is starting to warm up. We had a. We had a couple of warm days. I haven't had to wear multiple layers of clothes for, you know, weeks now, and that's a good thing. And getting some sunshine and playing with AI tools and writing things and hopefully, hopefully, you know, I don't have. It feels a little bit like when, when you've, like, played with tools your entire life, when, like, your whole, like, arsenal of, you know, your. All your toolboxes and all of your power tools and everything that you own is in another country a thousand miles away. It feels a little bit like somebody cut your arms off. And so the only thing I can do right now is like, play with my computer and so. And write, you know, and so, so I'm working on writing a letter both to beg for, you know, immigration purposes and, and some other things that might be interesting. We'll see how that goes. And, and I'm playing with, you know, AI tools to modify some of my. So, like, I, I was telling Steve when we.

A (10:19)

Have you been able to run Quinn?

B (10:21)

Yeah, yeah.

A (10:22)

3.6.

D (10:22)

Yes, I'm running. Running Quinn locally and I'm mostly asking Chat GPT right this minute whether this is a good idea. But I've been stuck. All of the 3D modeling I've done and for 3D printing mostly and, or drawing houses and things like that, but for 3D printing has been in SketchUp and SketchUp. I've been running a local version of SketchUp in some sort of VM or something for years and years because SketchUp went online and. And I hate it. And so I'm kind of trapped just because I've been using SketchUp for so long. And so I'm. I'm playing around with. With the idea of adding plugins and, and things to Blender since it's open source to make tools that behave more like SketchUp. Like, there's a lot of stuff that, you know, you can't.

A (11:23)

Blender's a bit of a complicated mess. I try to dig into that.

D (11:27)

I've been. I've been playing with it. It's. It's not as bad as I originally thought. It's. No, it's terrible, but it's. It's not like, like, I'm starting to get, you know, why it works the way it does. I hate it. But it's still like, I'm. I'm kind of like, understanding it now and just being. It's amazing, dude. Just being able to ask Chat GPT a specific question instead of having to watch tutorials for, like, hours defined where somebody addresses your question.

A (11:57)

The one random.

D (11:59)

Yeah, yeah, it's it's so. It's, it's amazing.

B (12:03)

Yeah.

D (12:03)

It's amazing how much better it is. Easier it is to learn something.

A (12:09)

But.

D (12:10)

But I think I can actually recreate a lot of the tools that I want. Like. Like tools that work Exactly. Or well, more like SketchUp tools. And it's going to be like. I don't know. I don't know how realistic it is to actually make Blender work the way I want it to, but. But it sounds like it's plausible anyway.

A (12:33)

It's an interesting project.

D (12:34)

Yeah. And so, um, I think, you know, in some cases it's sort of combining some of the tools that are already there. And so I don't know. I don't know how hard it's going to be.

B (12:45)

It.

D (12:45)

Chat GPT tells me it's the best idea anybody's ever had. So, you know.

A (12:49)

Oh yeah, dude, I had a revolutionary idea according to Chat GPT just the other day. Yeah. Like I was gonna. I'm changing the world. You don't even know. GPT knows it. And there's the most like blow smoke up your ass model. They're all really. But it's, it's particularly bad with Chad gbt. Cool if I do anything with Pair Core because like we have like a custom messaging protocol. So like after we have like a peer to peer connection, I can just do like I was just trying to do terminal, a terminal window that likes persistent on Linux so that I could just work on my Linux machine from my MacBook and using Paracore, using the thing that we built. It's not, it's not that hard really. You kind of do like a TMUX or you know, whatever it is and you just kind of pipe it over the space that you join. And we already have authentication and encryption, all that good stuff. But it's not like a crazy complex setup. And I was trying to get it done in about an hour or two and Chad GBT was just blowing smoke up my ass the whole time. It was just like, this is revolutionary. This is a terminal window from everywhere. And it's like, well, yeah, you can. There's like lots of tools that, that kind of do this.

D (14:08)

But it's really annoying. Like I, I don't, I don't understand.

A (14:11)

It's super annoying because like I can't trust it. It's like the reason I tell you my ideas, Jeff, is because you'll be like, that one was stupid. Or you won't, you won't laugh at my joke or like I'll read you A story or an article, you'll be like, that one. That one's not good. You suck. And it's like, Claude is like, the ultimate. Or Chad GBT Claude 2. They're like, the ultimate yes, man. Like, they will just send you down a path of idiocy because they're like, you're doing great. You're doing great. It's like you're a. It's like you're a famous person and you can't get somebody to criticize you. You're just like, yeah, George Lucas, that's a great idea. You should totally put Jar Jar Binks in this movie. And. And it's just. It's just like, completely. It's like you're gonna. You're gonna ruin your movie and everybody's gonna hate you.

D (14:55)

Um, where it's like, you're a fat person in a room full of liberals.

B (15:00)

Stunning and brave, man.

D (15:02)

Stunning and brave.

A (15:03)

Stunning and brave. Anyway, mechanic. How's it going, boss?

B (15:09)

Good. I'm seeing a lot of people hashing bitpoint10. Not seeing a lot of blogs, sadly, but Princey described it as like, buses in London. You wait ages and then three come at once. So I. I'm hoping that's. Hoping that happens and no one gets stabbed. I'm hoping the analogy for London buses ends with the cadence, not the stabbing. So aside from that, yeah, things are good. We got over 20x ash on ocean again. It's been a while that's been really nice to see. More frequent blocks and Hodler not has been dropping some bombs, as usual. That's been interesting to read. And the reaction's more revealing, really. Like, a lot of it is. There's a lot of drama and stuff in bitcoin core and it's people being people and politics doing what it always does. Like, people are just looking for reasons why it all went so wrong and they now have it with all this scandal being unearthed by Hodlo. But I just find it funny, like, why Adam back and Shinobi and the guys that. And Mr. Hodle that reflexively push back on it, why they can't go, yeah, look, core have made a bunch of. They've done a bunch of stuff that shouldn't have been done. But I still agree with the direction they've taken with bitcoin core. Like, it shouldn't need to be the case that, you know, like, at the end of the day, the personal stuff from my perspective is irrelevant. I just really didn't want them to change data carrier size. I really didn't want them to ignore inscriptions, etc. Like all the personal stuff is secondary to me. So I don't know why it wouldn't be on the other side. Like, I don't know why Adam and those guys can't just say, all right, look, core is ridiculous. Every. The videos of Gloria are cringe worthy, like but at the end of the day we agree with their stance. That would be, I don't know, something I could engage with. Right. And yeah. Personally, life's great. Summer. Summer is here in Canada. It's beautiful. And no conferences for a while. Bitblock boom was pretty good in. What's it called? What do you call it? Fort. What's that? Fort. Fort Worth. Not Dallas.

A (17:42)

Fort Worth.

B (17:43)

I didn't realize Fort Worth was a, like a. A bonafide city. I just thought it was the name of the airport and that Dallas was the only like big thing in that area. Fort Worth is insane. I've never.

A (17:56)

That place was crazy, man.

B (17:58)

Dude, it was so ridiculous. I was like asking the hotel.

A (18:01)

It's like a theme park.

B (18:03)

You can't put me on the.

A (18:04)

It's like a cowboy theme park.

B (18:05)

Yeah, I couldn't sleep. Like there's just all these guys with loud. The loudest music you've ever heard. Like parked outside the hotel just like, like trying to break the windows of the hotel. Like until 3 or 4am I'm just like, what are you doing? Like, can you turn it down? It just sounds like awful. Like where are you going? Are you. Oh, you're just driving around block playing music as loud as possible. I don't, I don't know what would compel a person to do that.

A (18:36)

It's 7:30, like should go to bed.

C (18:39)

I'm starting to sense that Canada doesn't. Doesn't have as much as this.

B (18:43)

Like I live in. I live in an old people's area. I think like, let's be honest, the Mexican immigrants like to.

C (18:52)

Yeah.

B (18:52)

Make a lot of noise versus like old Canadian people. Like they're a lot more reserved and quiet and all that. So. Yeah. And I saw these street preachers that were like yelling at all the half naked women walking around just like repent, repent. And it was really interesting watching that for a while because they.

A (19:12)

All the girls are the

B (19:16)

dynamic between.

D (19:17)

Or the sermon.

B (19:18)

Because I agree. Right. He's just telling them like, look, you walk around half naked, like this is not. It is. You're making the wrong decision. Right. But it's interesting to watch how much they mock him in response when you Think about where it's coming from, why a guy would be doing that. And it's always. It's always so interesting how they dismiss it as, you know, oh, a Christian being judgmental. I'm like, well, kind of. But at the same time, anyone that ever wanted to give you arm around the shoulder advice and be like, I think you're making a mistake is judging you, but they're doing it possibly because they care about you rather than wanting to feel holier than thou. So I don't know. I think that's a relevant thing to remember.

D (20:00)

Well, your description of Fort Worth is sounds like. Like, sounds like our hometown. Me and Guy's hometown is. Is like a microcosm of this. Because when we were in high school, the coolest thing for people to do. Not really, but, like, you know, we talked about the people that did this was like, cruise the theater parking lot. The theater and mall parking lot with your music. Music up really loud. And they just. They just, like, circle it and, like, I don't know, drive around the parking

A (20:32)

lot so loud that you couldn't even hear it because the trunk lid was just rattling too loud. So it was just. So. It wasn't even. It wasn't even a bass. It was just a.

B (20:41)

It just sounds like a tumble dryer with a door, like, rattling.

D (20:45)

Exactly.

A (20:46)

And loose door, one hinge.

D (20:48)

And then in, you know, the town. The next town over where. Where I went to school, ecu, you know, we had. We had a couple of, you know, regular weekend preachers in the downtown area harassing all of the. All the people. It just feels like.

A (21:05)

Feels like, with the exception that Fort Worth has about a 92% adoption rate of cowboy hats.

B (21:11)

Oh.

A (21:12)

And that if you stand there long enough, there will be some. Some dude that just kind of walks a giant longhorn bull down the street for someone to sit on. Wow. Like, it just, like, wild.

D (21:26)

That is like a theme park.

A (21:29)

It was. It straight up felt like a thing. It felt like a cowboy theme park. And then we went to a rodeo, and I'd never been to a rodeo. Those things are. Those things are way more legit.

D (21:39)

Pretty brutal.

A (21:39)

In real life, you realize just how intense that, you know. So it's one thing to see it on a screen. And I. Dude, I cracked up. There was a. I never really thought about it. Like, obviously, like, the cowboys who go and they round them up and take them out of the. Out of the ring. Out of the ring.

D (21:57)

They're the most hardcore.

A (21:58)

Yeah, they're. They're really hardcore. And they're really good. Like, I mean, like, these guys were just, like, accurate. Like, they were. They were on point every time, and. And everybody was like, clap. You know, when they get. I was like, hey, got the boy. Got the boy and got to drag him out.

D (22:11)

Oh, like, roping them.

A (22:13)

Yeah, roping. Yeah. Sorry. And they were real. They were. They were real intense about it, too. Right? You know, like, this. That's a dangerous job. And. But it never hit me that, you know, the horse is also a very social animal, and they're, like, working super hard, like, to watch that horse. Like, that thing is a. Is a massive animal with just an enormous amount of muscle. And to see them, like, fighting the bull, like, pulling against them. And then when the. Dude. This one was what. The one that, like, just got me, and I just. I just died out. I died laughing. Is one of them, like, roped him and then pulled the bull in, you know, out of the. Out of the rodeo. And. And he goes over, and there's, like, the other guy sitting on the thing, and the guy on top of it goes over, and he fist bumps him in there. Like, yeah, yeah, we effing did it. Yeah. And then the horse literally is like. And he, like, stomps, like, it's like, yeah, we did do it. And he was. The horse was so into it. Like. Like, this is his favorite thing in the world to do. And I just thought it was so great. I was like, oh, I never thought about. The horse is probably having a wonderful time doing.

D (23:19)

Yeah, dude, you can't have a horse that, like, runs barrels and stuff without it really thinking, this is the coolest thing.

B (23:24)

Oh, no.

A (23:25)

Yeah, yeah, yeah. Never crossed my mind. Never crossed my mind until I saw that. And I was like, that horse just tried to fist bump that dude. Yeah. So it's been a. It's been a bit of a wild month. You know, I think this is not bitcoin related, but one of the things that I've been seeing kind of in my own personal setups and also beginning to see in partial. Like, anybody's read that Matt Schumer piece on Twitter or whatever that, like, super blew up. Is it, like, 200,000 reposts or something about something big is happening, and it's about, like, AI taking everybody's jobs, and, like, it's here. It's not like, you know, in the

C (24:12)

future, is there anybody in the world who has not written this article?

A (24:18)

Well, he has more likes. He has more likes than everybody else, but it blew up, and. And I read it and there's, there's some interesting pieces to it, but I'm seeing a lot of pushback. You know, I've been talking about on the show quite a bit, I think we've talked about it, is that I really kind of think AI is going to. And maybe I'm wrong about this. Maybe they figure out how to do like, kind of like the AI, like the go AI type thing is that you can just have it play itself in like this domain of like these are the number of moves that you can do. And so AlphaGo can just get better than anything by an order of magnitude kind of thing. But I, I just don't think that you, you don't have, you, you don't have a bounded space to work from when it comes to like general intelligence and reasoning and the world. You know, like there's, there's an infinite kind of exploration in every direction. Um, and I really kind of think it's getting to this. It's already at least feeling to me like I'm at this point where the graphics don't matter that much. Um, or at least they don't. They certainly don't matter as much as they did two years ago. You know, like the using the PlayStation analogy is that like for 10 years everything was about the next console having better graphics and then the graphics got good enough that like people just stopped caring. It was just about the gameplay and like the fact that PlayStation 3 or 4 has better graph. Like nobody, nobody cared.

B (25:48)

Like that wasn't like stopped being the low hanging fruit. That makes sense.

A (25:51)

It stopped being the low hanging fruit.

B (25:53)

So what, what are you saying we're transitioning to with desirability for AI, isn't it?

A (25:58)

LLMs, I think are kind of reaching that point. The Uber CEO or CTO I think said that in like four or five months or I guess four months because it was in April, they have blown through their entire year's worth of AI budget and tokens and they're having to do this by API. And I just read a paper that somebody has been doing like a cost analysis of all these companies that are like, we fired 10,000 people, you know, we're switching over to AI and all this stuff is that they're actually paying comparable prices for token usage as they were paying for their employees and that there's a huge amount of people, especially because, and I'm seeing this in the bitcoin and AI rooms and stuff too, is that the more complex and the bigger tasks that you take on with some of these the more you actually spend your whole time just correcting errors and spending days and days trying to fix your setup. And I just. And granted, you know, I can't code, so there's a. My. My productivity in that sense is infinite in comparison to what I used to do. I don't even think you can call like a couple of bash scripts me coding anything and. But at the same time, I spent like five hours a couple of days ago trying to get Claude to do something like, extremely simple, like. Like, I think was a. Oh, I was trying to get it to scan. Open up a camera to scan a QR code from. From the computer. And I already had a QR module and it could not do it. And it broke five other things in the app, which it wasn't even, you know, supposed to touch. Like, I literally felt like I was talking to a person the whole time, and it kept saying things that weren't even relevant. Like, I. And, you know, this could be a consequence of it constantly dumbing down the model to release the new model, which I think is a totally.

B (27:59)

Were you. Were you using sonnet or what?

A (28:03)

Opus 4.

B (28:03)

Six? All right. I'm on Opus 47. I mean, it seems I can't get it to not do a good job on whatever it is I task it with.

A (28:14)

It's been doing. It's been doing good when I switched to 4.7, but 4.6 was that for a while, right? And then I feel like 4.6 got really, really dumb. And then it switched to 4.4.7, and then 4.7 was acting like 4.6 used to.

B (28:28)

Man, who knows what's going on with the back end there? Are they doing that?

A (28:31)

That's kind of what I think is. I think they're dumbing down the last model to make. To upsell the new model.

D (28:39)

And.

A (28:40)

And I'm not saying that 4.7 isn't better than 4.6. It probably is in a lot of different ways, but I don't think it's measurably better. Measurably or in a sense better that what I have been able to get out of it in a significant way than when I First was using 4.6, because 4.6 was pretty freaking good.

C (28:59)

It's also a matter of how big the task is to.

A (29:02)

That's true. That's true. The bigger your code base is or whatever.

C (29:07)

I mean, like, when you're coding something, you're. You start off writing one file, and then you realize you have to modulate.

A (29:13)

It's really good.

C (29:14)

And then like I'm not talking about AI, I'm talking about like if you're handwriting code, you start writing something, you're in one thing, then you realize, okay, I need to start a separate file for this because this is kind of a separate module. And then like the more you modularize your, your product, your software, the better your code becomes. You have to do the same thing with AI. You can't have one project for the, for the entire thing. You have to like mentally break up your project into components and, and then you have to have a separate chat and maybe a separate different AI for each component of your module. And so like I, as long as you do that, like it stays manageable. So I, I think we're still in kind of the early times in AI. I mean people are now just brute forcing AI. You know, we're just trying to do the entire project in one chat session. That's not going to work. But I think like, I still think it's like insanely powerful as long as you're like you, you, you sandbox, botch, sandbox each task that it has to do. But that, that requires like work in your head ahead of time.

A (30:23)

Don't get me wrong, I'm not saying it's not powerful. I was using it up to two minutes before I started this. It's absolutely insanely powerful. But I think the gains like, like Steve, like you're saying, I think all the big gains for, make this model just that much bigger and that much better are not going to be, that's not the low hanging fruit anymore. It's memory, it's token used, it's efficiency. Like the Chinese models are kind of blowing away the US market because there's so much fiat, there's so much just like how much money can we throw at this to outcompete China? And then they're having to do it on a budget. So they're getting Quinn 3.6 to actually behave reasonably at a, at a level that you can run locally, behave at a level that's actually comparable, that actually feels similar to these massive frontier models in the US it really does.

D (31:24)

And, and I don't until, I mean it makes perfect sense.

A (31:28)

There's more and they're doing it for a hundredth of the cost and you can run it locally and it's open source.

D (31:34)

Like yeah, well if demand, if demand outstrips the supply of hardware and everything else, then you're gonna see like it makes perfect sense that the, that the compute is going to be price like, the employment will basically be price competitive with the compute until, you know, like, like the price of that is going to rise until it's, you know, like, reasonable. Just hire people back again because, because there's no like, the theoretical desire for employees that you don't have to, you know, treat like people is much higher. So there is like a default, like, desire to like, make this work, which means that you're going to spend. People are going to be willing to invest more in it, which means that the price is like, how do you say? It's like the demand is more inelastic. Is that right? Like the, like the. There's going to be a higher demand for, for AI stuff regardless of how the price reacts in the short term. And just fiat world, you know, debt and all these things kind of make all that worse. And so, so you've got. But then what'll happen is that it will. There. A lot of people will find that, oh, we should just hire somebody to take care of this job. And, and then it would be, you know, economically smarter to have kept somebody employed or whatever. So I don't know.

A (33:13)

There's plenty, like, I mean, there's absolutely zero denying that. Like, it's a massive technological shift and it's a fundamental disruption to how we actually do things and create processes and build things. But I also think we're oversteering, especially when it comes to like the Matt Schumer article or whatever is that like, everybody's like, oh, we're going right? And so they're like turning the wheel all the way to the freaking right? And it's like, no, you should kind of just turn it halfway. Uh, and I think we're gonna have to correct. Um, and a lot of people who, quite possibly a lot of companies who have fired a lot of people might have to rehire. Um, uh, but we also might not be fully into it because there's so many people who still haven't touched it. It's shocking sometimes the number of like, normies who still have just. They only ever use like the free version of Chad GPT and they're like, yeah, I got it to write an email for me. So, like, it's also a question of like, which world are you in? You know, are you in the one that's like, you get a new model every single day and you get like, I just bought a. I almost, I almost broke down. Like, so if so you understand the amount of commitment and I actually think AI is useful. I almost broke down and blew like 10 grand on a 512 ram Mac studio on ebay. Then I realized one of them was probably a scam. And then I got like oh, oh. I pulled back because I think the price is actually more like 14 or $15,000. Cause you can't get them anymore. And, and then I really reassessed my situation and how I wanted to use it. And I realized that an extra 50, another 5090 for my Linux machine actually fits it a little bit better. Costs like a fourth as much money. Still a huge, huge chunk. But I can actually get my Quinn and still have. I can have my Linux machine not be feeding all of this garbage to ChatGPT, which I just don't.

B (35:13)

Yeah, you want to self host AI? I've been all about that for a while. The Blackwell chips were what I was going to go for instead of a Mac Mini or a Mac Studio. I did end up buying one and it showed up with all the seals broken and stuff. So had to send it back to Amazon. I couldn't trust it. But still the dream to self host AI.

A (35:36)

Still the dream. Well, I'll let you know how my 5090 goes and because it's 32 gigs of RAM, so it will do Quinn 33.6 pretty well and with some gap and probably a really big context window. So I'm going to give it a real good go. But I really suspect the next release of the studio not only will have like a heavy, heavy ram, you know, unified memory is going to be everything. Like all of the products for the next two years are going to be like can you run AI on it? And they have made some pretty big leaps. Like they've been working directly with Exo Labs so that you can cluster Mac Studio and they literally remove their network. Like you can do a firmware update where you remove all the networking and all the stuff from the thunderbolt and you just get a direct metal to metal connection and it like tripled or quadrupled the token like the bandwidth in a clustered Mac Studio thing. So that just screams to me that Apple is realizing that they have the shovels for the gold rush. And so I really kind of think the next Mac Studio, maybe even the next Mac Pro, will be really, really geared toward AI inference. But we'll see, we'll see. Anyway, just this, my thoughts on that. Okay, so news items. My mouse is dead. So Iran has now allowed and started taking Bitcoin for tolls crossing the Hor Hor Moose. Hor moose. Is that how you say it straight? Tehran is seeking up to $2 million per tanker or $1 per barrel of oil and fees. And I realized this was recorded. I'm not sure if this is accurate because this might be. I guess I. I guess this was put down maybe on the 25th. I don't know. That's like one of those things. It's like, what are Trump's tariffs? Is that if. If you're news is two hours old, it might be 100% off.

C (37:44)

It might be 100% off, even if it's right on time. I don't trust any of this stuff that I hear about what's going on in that straight, like, zero trust in any of it, 100%.

A (37:56)

Did you see the video? Did you guys see the video? I want to get your general thoughts on this, but did you see the video of the. The dude? I mean, not that this is like. Not that this is something to laugh at, but there was a. A boat or a ship that paid a toll to a scammer. They. They, like, they like, got. And he's got a. He's got an accent or whatever, and he's, like, talking to him, and Iran's, like, shooting at them or like, trying to like, launch missiles or whatever. He's like, no, do not shoot me. Do not shoot me. Oh, my God. Just cancel. Cancel. I have Peter Toll, like, and he's just, like, going off. And I can't help but laugh. Even though I know the dudes, like, literally thinks that their ship is about to get blown up, but they paid a bunch of freaking bitcoin to a crypto scanner scammer, apparently. Just wild. Anyway, general thoughts, Steve.

C (38:45)

Oh, yeah, like I said, I don't really. I don't. I don't take any of this stuff seriously. Like, if I was a spider and I had a thousand eyes, I think I would keep one eye on the news. Like, I just. I don't really care. I. I'd keep a lot more eyes on the memes and the funny videos, though, because it is good for a laugh. I mean, yeah, obviously. So I love the thing where it was like. It was talking about the tether that was seized and it was like, amount of stable coins seized, whatever, bazillion dollars, amount of Bitcoin seize.0 bazillion. I'm here for that. And if they are taking bitcoin for passages of ships in that area, that's awesome. You know, just proves the use case. I just. I just don't know if I trust anything I hear from the news.

A (39:39)

I feel Very similar. Jeff.

D (39:41)

Yeah, I don't know the. I, what's crazy is it's not like they can just set up a toll booth because if you have a, you know, a central point for the US or anybody else to attack that can be taken out. Like, how, how do you indicate? Hey, yeah, pay your, pay your tolls here? Like.

A (40:01)

Yeah, I have no idea that I. A straight paying a toll. And like how you do it and where you do it. I don't, I don't get that one quite. Because like, there's no, there's no like door. You know, it's not like they got a gate, right. And they just basically shoot missiles. Uh, they, they, you know, fire warning shots and then blow up anything that doesn't. That's the threat. Right. So it's like, okay, did you like, like, does somebody look up your phone number on, online on like straightofhoremoose.com and. Or straightofhoremoose.ir or whatever Iran's thing is?

D (40:41)

Yeah, it's like you need some sort of, you know, direct peer to peer connection and do you have a token?

A (40:47)

Do you have to buy tokens?

C (40:49)

Someone, someone had a theory that it was like, there's actually nobody in control in Iran in the state of Hormuz. There is no authority or power structure everywhere. All, every single boat, every single entity in there is all acting as like independent pirates. That's why you hear these things about it cost this, it costs that. There is no it. There's just a bunch of random people doing a bunch of random things over there. I think that's probably more likely true.

B (41:20)

That sounds right.

D (41:21)

Like Iran's, Iran's military was actually like decentralized kind of by design. So I, it's, it's hard to say whether that like, maybe they're all acting as, as independent pirates or. But supposedly, you know, there was, it was sort of decentralization with, with some coordination between all of them. I don't, I don't know how that works exactly, but I know that just calling somebody in Iran is impossible. So like they have to call you, which is, which makes things very complicated.

A (42:03)

Yeah, dude. Communication in and out of Iran is just, it's just bonkers.

B (42:07)

I think like a lot of, I've been digging into these kinds of things recently because there's a lot of mining in China and there's not supposed to be. And everyone's always like coming at it with these sort of first world mentalities of like, well, is it legal or illegal? I'm like, that's not how anything, anything works in these countries. Like, yeah, it's just like. Do people run nodes? Not really. Can they get away with it if they do? Depends. Is it legal? Technically it's legal, but like, everyone knows someone that did it and got in trouble for it. And here's this other thing, like in Iran, running of. According to what I was researching last night, in Iran, it's Tor is not illegal, but no one uses it and it's blocked everywhere. And you have to use bridges. VPNs are illegal. Everyone uses.

D (43:03)

Everyone uses them.

B (43:04)

Yeah. So like seven. So you can, hopefully with your other half being Iranian, maybe you can tell me how correct all this was, Jeff. Because it looks like there's government approved VPNs, then there's the normal VPNs, like the honeypot ones, like, you know, NordVPN and all that kind of crap you see on YouTube videos. Apparently everyone uses them to access Instagram, which is also illegal. And like, there's 700,000 businesses on Instagram in Iran. And it's like, this is why. So when people like, is it legal to run a node in China or not or Iran? I'm like, well, there's not really a rule of law. You want to know whether people actually do. Because there's things that are totally legal that no one has the balls to do. And there are things that are totally illegal that everyone does and everyone knows you can get away with and you'll never get arrested for.

D (43:57)

Yeah, yeah. The best, the best that I can put together is that, yes, there are, there are VPNs you can use that work that the government is totally like, knows that you're using them and everybody knows that, like, this is the one you don't do or say anything particularly wrong on. And, and then there are much more expensive, like, private services put together by, you know, like people that are hacking together, things that, that work better and will, you know, like the, the ones that are controlled, it sounds like, like, won't allow you to. To access certain things from Iran. Seems to be using a bunch of Chinese hardware and, and so their firewall works really well. Well, fairly well. And so like, if you're using one of the VPNs that Iran kind of controls or knows, you know, is like, able to monitor everything you're doing, then you maybe can't use Telegram or can't use. There's like, certain things you're just not gonna be able to use. And then if you want to use those things, you have to you know, do something that's maybe more illegal in some ways, but, you know, actually gives you some privacy. So I don't know. It sounds very complicated, but I'm. I'm glad. I'm glad that we're. Do what.

B (45:35)

It's an art, isn't it? People want to treat laws as black and white, but it's really all just art. At the end of the day, you're trying to get the state to leave you alone. You're trying not to paint a target on your back. You're also trying to do the things you actually want to do without, you know, interruption. So, you know, it's. That's the default for human societies with corrupt authorities. It's very rare. You end up in this kind of, like, I know my rights. I can do X, Y, Z. I can tell this cop, you know, to go F himself. Like, some people have that mentality, and they're right, but I'm like, that's just not how. That's just not how stuff works in China.

D (46:13)

Like, well, they're right, but they're wrong. Like, that's not how. That's not actually how things work in the US Either. But, like, it only sort of works that way.

A (46:21)

Like, if you're lucky, it just works more that. It's like a spectrum, and it works more that way in the US Than it does anywhere else.

B (46:28)

The US Is kind of newer and fresher and kind of has a more naive mentality. Like, there's a lot of people walking around like, I'm doing this. It's my right to do it. You know, you do not have the. You know, and they'll talk about the. The. The Constitution and the amendment and stuff like that. Even from a European perspective, that's adorable. Like, we're just like, I respect it, and I don't want to come across like I'm. I'm being, you know, disrespectful, because I really do like that aspect of American culture. But the rest of the planet's just like, there's a guy with a uniform on. I'm gonna smile, I'm gonna nod my head, and I'm gonna try and stop him from wasting the next three months of my time with bullshit paperwork and stuff like that. And, you know, I'm not gonna sit here and be standoffish and say, you can't do this. I have rights. And then them go, okay, I'll leave. Like, that's just not a thing in any country in the world, pretty much.

A (47:23)

Yeah. Yeah.

C (47:25)

With the, like, video Recording of cops, though, it's kind of different because there's, like, a social stigma to it. It's like you can ruin this cop's life on social media now. So I. It kind of gives a little bit more ammo to that American. Like, if you know the law better than the cop does and you can humiliate him on social media, it's kind of like he doesn't want to have to deal with the rest of his life with that now, too.

D (47:54)

That is true. But it's still only if society agrees with you. Like, there's not a. Like.

A (48:01)

Yeah.

D (48:02)

I don't know if you're. If your society becomes too respectful of authority in one area or another. Doesn't really. I mean. Dude. And the scariest part is, like, the way conservatives can be manipulated to, like. Because conservatives are sort of like the only. The only hope in the US Is, you know, for freedom. It feels like. Well, it has felt like, but.

A (48:29)

But my God, when it's like there's such an unbelievable mess.

D (48:32)

It's like the number of them that have been convinced to support, like, an aggressive war after how long of, you know, like, where that was. The whole. The whole thing was like, you know, we don't, like, we want to stop all of the stupid wars with 20 plus years of, you know, stupid wars in the Middle east and. And, like, the just insane justifications. Like, it just. It's so weird to me. I don't. I don't understand how people can flip so quickly. I. I don't get it.

A (49:10)

Yeah. It literally takes about an hour to go from, like, we live in a police state to back the blue.

B (49:16)

Yeah.

A (49:16)

Yeah. Depending on the news cycle. Yeah. Or. Or I'm so sick of no more. No more wars. And, you know, all this stuff we got. Iraq was a scam and all this stuff. Bomb Iran, we got it. They're gonna get a new two weeks. Two weeks.

B (49:34)

45 minutes.

A (49:38)

45 minutes. Is that the turnaround? Is that the official.

B (49:41)

That was the weapons of mass destruction. That's how that was a British claim. Saddam Hussein can deploy weapons of mass destruction in 45 minutes.

C (49:53)

But right now, we're way more 45 minutes than we were back then.

A (49:57)

There is so much more 45 minutes

C (49:59)

than it was in levels of 45 minutes we've never seen before.

D (50:04)

Dude.

B (50:04)

I can't. Yeah. Your point about cops being videoed and stuff. The problem is, like, a lot of cops are retards, right? Like, really stupid people in the States, and they. They don't have the Capacity to interact with you in a way where they can figure out that what you're doing is a net negative for them, and they're just on a power trip and they'll just be idiots and then that. I don't think that applies outside the US at all. Like, you've got, like, the IDF soldiers, like, getting caught on camera sexually assaulting prisoners and stuff and then being celebrated after. Like, they don't care. Like, yeah, they try to hide behind a bunch of shields. Right. It's like they. They didn't. I guess they did care and. Because they cared enough to hide. But the fact that there's, like, they're, like, celebrated in Israel now, like, that's. And I'm not even saying that as a particularly, like, to focus on Israel specifically, but I just, like, you know, if a cop in, you know, Iran gets caught assaulting someone, I just really doubt anyone's gonna. I doubt anything's gonna happen.

A (51:10)

Like, yeah, yeah, if it's illegal, it doesn't matter. Like, so much is. So much of politics and law is posturing. And the reality of it is, you know, what can you get away with? Like, there's, like, you know, especially the poor of the country. Like, the idea that there's, like, there's all these, like, listed laws, but if you can bribe somebody, none of those laws matter. You know, like, most of it works because of bribery.

B (51:37)

Yeah. Or just status and hierarchy. Like, you know, Q Bar was terrified of. Like, basically the standard is that you can't do anything. But you. If you know the right people, you can do something. Like, if you can do anything, anything,

A (51:52)

probably do anything and get in trouble.

B (51:54)

Like, you can throw a party and have some alcohol at it and make some music and stuff. Someone with a uniform is going to show up at some point, and you need to get him to go away and leave you alone. So you can either come out with a textbook and be like, here's a bunch of reasons that say we can do this. But he's going to come up with one and say, well, you violated this, and you violated this and you violated this. So in most cases, money will make them go away. In other cases, status will make them go away. Where you're like, okay, like, Steve, whatever the guy's name is, you know him, you know, his uncle, his uncle works at the precinct, whatever, and you manage to get him to go away based on that. Like, or you have a relationship with them. It's just, I. I hate operating that way, and I. It's a real privilege Living in a society where that kind of stuff doesn't happen. But it is the norm.

A (52:44)

Yeah, it's probably been kind of the norm always.

C (52:48)

Let's talk about some bitcoin.

A (52:49)

Deception is probably the opposite.

B (52:52)

Yeah, Steve wants to talk about bitcoin.

C (52:54)

Let's talk about some bitcoin guys.

A (52:55)

Fine, Steve, we'll talk about whatever the hell you want to talk about.

C (52:58)

It's been an hour.

A (52:59)

This is guys audible.

C (53:01)

This is called the Bitcoin Roundtable, man.

B (53:05)

No it isn't. It's called the Satoshi Roundtable.

A (53:07)

Bitching.

B (53:08)

It's called.

A (53:09)

It's called my Roundtable. So suck it. France advances a bill forcing users to report self hosted wallets over $6,000. This is framed as fighting tax fraud. This is you. Are you going to report yourself hosted?

B (53:27)

They're trying to pollute the UTXO set by making me only ever have $5,000 at each address.

A (53:37)

It's a scaling attack. Yeah, that's exactly right.

C (53:40)

All this round number.

A (53:42)

I love that the report itself is just as long as it's 5,000 even.

C (53:46)

Yeah, 5,000 even.

A (53:48)

The tax office specifically admitted that they have absolutely no tools to verify any of the data given to them. They're not going to be able to check addresses or sign anything. Like it's not the. Like if. No, nobody does report. They have no way to know. So they're just.

B (54:07)

That that's not true though. They can always.

A (54:08)

It's voluntary.

B (54:09)

That's not true. Because if they dig into you and like it's this is the, the way tax.

A (54:14)

Oh no, I was just saying they acknowledge that they don't have a way to. I mean obviously they can. The tax office is going to see that like you, you've made a purchase on crypto.com or Coinbase or whatever.

B (54:26)

Or it's more just like we don't like you, we're auditing your tax affairs. Send us your bitcoin wallet and then you have to send it in and like ah, well you had more than that in. I mean and they, that again, they're deliberately vague with this stuff. I will say that the last topic where we're talking about this sort of selective enforcement and all that stuff. The, the way taxation works in the west is the way all laws work everywhere else. It becomes an art form. It becomes a to and fro. It becomes. Everything's always vague. It's always. This is my perspective on why I don't owe you this money. And the government says, well this is our perspective on why you and the minute they can catch you in some sort of inconsistency, then that's when they go, all right, well, now we need all of your stuff. And then you have to send it. Otherwise they just go, well, you're not being forthcoming. And then that's when they construct it all and say, well, here's what our law says. So, like, bitcoiners are not immune from that stuff, unfortunately, because you can't. You can do anything you want in bitcoin. Like, the network is never going to stop you. And they can't enforce anything at the network level, really. But they can ask for your history and activity. And the problem is it's money, which means you're constantly interacting with third parties, which means there's always history and paper trail and, you know, evidence of all this stuff. And it's really hard. Like, when bitcoin is just an internal system that doesn't interact with the rest of the world, you can be completely free with it, but the minute you want to do something like buy a car, you're screwed because there's like, the other guy on the other end of that sale. And the fact that you have a car now is all just stuff you can't hide. So, like, the crypto dream of having, like, stateless money, they can't screw with it on a supply level, and they can't mess with the permissionlessness of it on a payment level. But, man, can they get in the way when it comes to, you know, an economy and its frictionlessness. How much can they get in the way with the capital gain stuff primarily? Like, a lot of bitcoiners actually attempt to handle that. I went on holiday, I bought a coffee that was a taxable event, incurred a gain of this much stuff. And I'm like, like, some people don't want the stress of not doing that. So, like, it's there. It's real impediment. Like, I don't know, capital gains is the best example for me of just how can you make bitcoin really, really cumbersome to use.

A (56:52)

Yeah, I think the, the important. Like, a lot of people will think that you can have, like, stateless money because bitcoin changes the economics of violence. And I think it's like, the important caveat is that it changes the economics of violence. It does not remove the economics of violence. There's still physical violence. There's still physical boundaries. You still have a piece of land that's stuck wherever it is unless you sell the land, you know, like, and then what you get out of that land and who you can sell it to is still based on the jurisdiction that you're in and the enforcement of that. Like, it's kind of like it's one foot out the door. It's not. It's not an exit in a lot of sense, because the real world. The real world still exists and you still have to live in it.

B (57:36)

Yeah. You just eventually hope for the collapse of the state, so there's no one to bother you anymore by defunding it. But the idea that they'll leave you alone while they still exist, unfortunately, is not. That's just wishful thinking.

D (57:51)

Yeah. I don't even. I mean, and the whole idea of, like. Like, for a long time, there's been this idea that we could just be, like, there could be nomadic sort of freedom. I think that's just. That whole idea is a scam, too. Like. Like not having. Not having.

B (58:07)

You'll own nothing and you'll be happy. Yeah.

D (58:09)

No, it's what that is.

B (58:10)

You don't really have a house. You don't get to have shit. You just hope that there's an Airbnb and an Internet connection and that you can set up a temporary life. Like, there's no way to live, man. Not if you want to have a kid or a wife. Like, you can't be a nomad and have a family.

C (58:24)

Well, the.

D (58:25)

The ability to, like, we were just. Just trying to, like, pare down everything that we can get onto a plane in, what, like, the maximum number of bags we can, you know, pay to take with us is just.

B (58:38)

It's insane.

D (58:39)

Like, it's so hard. And so, yeah, you can't have anything. I mean, like, it's just.

B (58:45)

It's.

D (58:46)

It's terrible. So,

A (58:49)

yeah, it's very much a. I love the. You'll own nothing and be happy. Except that this time it's voluntary instead of. Yeah.

D (58:57)

Having a home. A home base is. Is critically important.

A (59:02)

Yeah. You know, I question whether or not I want to get into this because I feel like we've talked about it every time, but it's always, like, better

C (59:13)

not be quantum computing. I. I'm.

A (59:15)

It's 100 quantum.

C (59:16)

I'm hanging up the phone.

A (59:17)

It's a hundred percent quantum computing because there's another big news item. Google has broken elliptic curve cryptography, and they can do it in under nine minutes, making all of your blocks and all of your. Everything vulnerable.

C (59:33)

Yeah.

A (59:36)

Steve, did you sell everything?

C (59:38)

Yeah, man. Oh, you saw.

A (59:42)

You were an og, dude.

C (59:44)

I have something that I'm sure about if there is something, a form of computing that breaks the hash function or whatever, that's a threat. I will guarantee you.

A (59:57)

You mean elliptic?

C (59:58)

I'll bet elliptic curve. Anything. Anything that's better than digital computing. I will bet, like, $100,000 that it will be something other than quantum computing. So people don't realize that quantum computing is one of a million forms of computing being tried right now, 10,000 of which are currently ahead of quantum computing.

B (60:27)

Can you give some examples, including, like,

C (60:29)

an abacus or waveform computing or acoustic computing or magnetic resonance computing or photonic computing. People think that there is not an entire field of research of alternative forms of computing. There is an enormous field of alternative forms of computing being tried right now, and everybody's focusing on one because only one has marketing.

B (60:58)

Yep, quantum computing.

C (61:00)

I'll take the entire field any day over a specific point.

A (61:04)

Over quantum.

D (61:05)

Yeah.

B (61:06)

It's like, it's like you feel bad for them. It's like Covid had the marketing department, but other diseases like pneumonia. And all those guys were like, what about us? Like, we kill loads of people every year. I was like, no one cares about you. Like, magnetic resonance computing is just like, dude, I can actually figure out that 3 times 5 is 15 without consuming a gigawatt of energy or something.

C (61:29)

Exactly.

B (61:31)

But nope, sorry, no marketing department.

C (61:33)

Yeah, you could lay beanbags on top of each other and look at the sum of the beanbag curves. That's a form of computing. That form of computing is ahead of quantum computing right now.

B (61:45)

So is that what they're really doing then? Is anyone actually researching good computing? Like, is anyone involved in the quantum computing industry not just trying to get a payday there actually people that believe in this tech that are like, I don't think so.

A (61:58)

Well, I think mathematically, I think theoretically there's something to unpack, but I don't think there's anything concrete that's actually been done with it. Like, one of the most recent, like, big rewards that was paid out, they found out that the entire computation was done on a traditional. With traditional computing. And then the quantum had to decide between two options, like whatever the. The quantum layer was. And this is like, what if you really dig into all of the examples, this is what you find over and over again. And, you know, Cloudflare announced that they're going to be quantum ready by 2029. Because, like, this Google paper has, like, sped up the. The timeline on all of this. And they, you know, put out, like, all this. All the authoritative Google says, we're doing, it's gonna happen. Cloudfare says it's gonna happen. And like that was, that was like the big thing. Like Nick Carter posted after like one of my things is like he's such

B (62:53)

a joke at this point.

A (62:54)

Everybody else disagrees, you know, and I don't think he was being like, he wasn't being like snotty or whatever. He was just, you know, saying. But I just don't believe the whole. I'm, I'm super skeptical of the, like all the important authorities think like this and even that they're making decisions or moves in that way. And then one of the arguments that I see all the time, and Cloudflare even put this in their article, was that if there's real movement happening in quantum and like concrete quantum computers and quantum chips, nobody's going to hear about it. It's going to be secret, like the nuclear thing. And I was like, except that there's like not the slightest bit of empirical evidence that that's the case. Like nobody was like, there were no VCs looking for funding for their. We almost made a nuclear bomb in 1940. You know, like there was no market around it. There was no hype. It was literally happening in secret by the government because it was. The whole thing was a secret and nobody else was doing it. That is the exact opposite. These. It could not be a more perfectly opposite example of. What I see here is every supposed concrete example is, is actually a scam, is actually a lie about the, the genuineness of how good their quantum computing actually is. Like there's one that says that they, they literally computed like a 17 digit number or 22 digit number or something like that. And then when you get into it, it's like you put it in binary and the number's like 01010101010101. Like they managed to actually trick it so that the, the actual entropy that they had to sort through was like they've never done it with more than three bits of entropy. Never. And they've certainly never done it faster than a traditional computer, even at three bits of entropy. And so like, like we're looking at like the number 15 is the biggest thing that's ever been legitimately factored by a quantum computer, a quantum chip. And one of the things that like stands out to me so just in that sense of like, oh, they're going to do it in secret. Is it all. I like the second they have a new paper or a new thing or a new like it's it's so public and it's broadcast across everything. Like where are the people who are being secret about this? Because it's certainly not Google. Google released the paper and it just blew up and they posted on every stupid thing in the world. Like they're every. I see the exact opposite. Everybody being as loud as they can and being loud dishonestly. They want funding, they want attention desperately. I don't see, I don't see any indication that anybody's doing this secretly. And then here was the big tell for me that Google literally has nothing concrete is apparently there was another paper that came out at the exact same right around the same time. It's literally within days of each other that said there was another theoretical like mathematical speed up that they could do that even did like 3x the efficiency that Google was talking about but it was a completely different model of quantum machine. Apparently there's like eight different or five or something different ways that they want to build the chip. There's like a neutron, neutron atom computer chip and then there's like a, there's a wave. I don't know, they just, I can't

B (66:14)

remember what the stupid contrived this stuff in such an infuriating way. It's like when bitcoin miners will tell you Sorry to talk about bitcoin again. It's like when bitcoin miners will tell you that they have like I'm using this much joules per terahash and I'm buying bitcoin for this price. 40 grand of Bitcoin. I'm like yeah, but you spend 12 grand on the machine and you always don't factor that part in. So in reality it's costing you 200 grand a bitcoin, not 50. And so they'll zoom in on some aspect of what the quantum computing algos and design uses and they'll hyper focus on one bit and be like we made that bit 10 times more efficient. And I'm like sure, but you still haven't actually made any progress towards making this viable. You're just iterating on subsections of it that are legitimately going to be part of it. But until you've solved the big problems it means nothing. But I don't know. Sorry to interrupt your, your rant there.

A (67:10)

Lengthy, lengthy rant. I one last from my, my point just, just to hit my point real quick is that apparently after this thing came out Google it suggested that Google that the right path was to use an atomic, a neutron chip or whatever and they were Sitting like, oh, we might have like a dedicated path. But then it like struck me as that means that they don't even have any, they don't even know what they're building.

C (67:37)

Like, no idea.

A (67:38)

They, their paper changed what they were even building. They don't even have a chip that they even think could do this. And they've concretely, it's literally like saying, it's like we have, we have a bunch of silicon chips and we just, we just came up with a paper that makes us think that, like, we should never use silicon in this. You know, it just, it's a total remapping of anything that they actually thought, which means that they don't have any concrete progress in whatever the heck they have right now if they're going to change to something else, hoping to have some sort of concrete progress.

D (68:09)

And do they have a manufacturing process that even allows them to build the thing that.

A (68:14)

And I just cannot believe, even if there is somebody like, oh, we're going to do this in secret, that the first time we hear about somebody legitimately factoring, factoring something with a quantum computer that's bigger than two digits is when elliptic curve cryptography is broken. Like, that just seems like a real good way to excuse why you should still give me funding even though I don't have anything yet. And none of this suggests that quantum might not be a real thing or that there not might not be a way to break elliptic curve cryptography. Cryptography usually has a shelf life. It all, it historically always has, but that's a big, that's a far cry from 2029. We're screwed.

C (69:00)

The only reason I'm interested that I kind of pay a little attention to quantum computing is because I think there is a real chance here that quantum computing is going to be a scam that's built bigger than Covid and even climate change. Was. I was.

A (69:17)

Do you really think so?

C (69:18)

Well, I was talking to a guy at the bar the other day who's like, not a tech guy. And I don't know, for some reason I was like, oh yeah, that's quantum computing. He was like, what do you mean you just had. He just had no idea. And he was like, oh. He was like, okay. Well, I mean, if you'd asked me, I, I just assumed that quantum computing was already powering all of the AI stuff. So, like, the average, I'm serious. I think the average person has heard quantum computing in the news so much that they assume that quantum computing is being used by all the high level stuff right now. So if the average person actually thinks that, you know, how much power they have over people to pull off this scam.

B (70:02)

Yeah. We need to raise taxes by X percent because otherwise there's a quantum computers that are powering everything.

C (70:08)

That's what I'm saying.

A (70:09)

Yeah.

C (70:11)

If you can convince someone that an invisible boogeyman exists. So the reason that scams become serious is because they're invisible. Climate change invisible.

B (70:21)

Covid.

C (70:22)

Invisible. When things are invisible and you can convince a large amount of people that it's a serious threat, that's when the enormous scams happen.

B (70:31)

Yeah. What about drugs and terrorism and you have wars against.

A (70:35)

And every opportunist jumps in on it too. So, like it perpetuates itself.

C (70:40)

Yeah. So it's like, it's scary because you

A (70:41)

can't control the actual scam. Everybody just jumps in on the scam

C (70:45)

because no one can disprove it.

A (70:47)

Yeah.

B (70:48)

Watch. There's Adam Curtis documentary called the Power of Nightmares where he talks about if you want money, you need to come up with a sufficiently scary disaster scenario to motivate people to pay for it. And this was climate change all over. You've got to come up with it. In five years, there's going to be no more polar bears or Eskimos or anything. The polar ice caps are going to be gone. You just have to come up with this, churn it up, up, up. And of course it's always more five

A (71:13)

years now than it was 20 years ago.

B (71:16)

2014, the Ice Caps were going to be completely gone. According to Al Gore, climate change was,

D (71:23)

was like such a. I mean, it's, it is like a perfect boogeyman because if you save the world, nothing happens. Like, like. And so, so you defeat this, this imaginary thing. And oh, look, we, you know, we stopped the sea level from rising or whatever the hell, you know, and it's

A (71:42)

just always just far enough out that it's like, oh, we were just wrong about our timeline, but we're not wrong about what's happening.

D (71:49)

I think the hole in the ozone layer was exactly the same thing. And I think Steve maybe said something like, like disagreed with this sometime. We talked about it before, but I think, like, I literally think that the ozone layer is a, as a product of energy from outer space colliding with oxygen. And so you have this standing wave, the oxygen molecules turn into ozone, they fall and then they turn back into oxygen. And so you have this standing wave at the like, highest point where oxygen can get out of the atmosphere.

A (72:21)

So a hole in the ozone layer would just be an indication of, like, it's just like low. A result of low area of radiation.

D (72:28)

Like.

A (72:28)

Yeah.

D (72:29)

And so there's no. Like, that's why in the North Pole or like, where, you know, you don't have a ton of plants, there would be. And. And the currents all move in one direction around the Earth. You might have like, some like, empty place where you get special northern lights or whatever the hell, you know, whatever. But, but like, this is not like, I just don't think that this was ever a thing. Like, it was just an effort to attack the end. Like, attack a product that was going to become. It was very, very cheap to produce. It was reaching the end of its patent cycle. And, and they could ban R12 or whatever, and then they come out with, you know, the next refrigerant that was slightly changed. They just changed the molecule a little bit. And then when that one, you know.

A (73:19)

And now it can be patented again.

D (73:21)

Yeah, they just, they have like a bunch of success, you know, a train of patents that they can just play

A (73:30)

when the next switch from one to the next and they have enough lobbying power to ban the one that is about to be available for everybody.

B (73:36)

Yeah.

D (73:36)

And then, and then you've heard that there's like, there's like weird talks about like, surface level ozone. Like, there's all these like, weird efforts to like, create some. And it never stick. Like, you can, you can see it's like they're constantly throwing bullshit at the wall to see what's sticks. And like, you know, like acid rain was, you know, one of them for a while. And you know, like, they're just. There's just different things that, like, what can we scare the out of people you know, with? And then when one of them sticks, it's like, yes, we can attach 10 different things to this. And, And I, I mean, I. Obviously climate change was one of the ones that stuck.

A (74:12)

Climate change is one of the most, like, it will go down as like a profound and, I don't know, wild success.

C (74:21)

Quantum computing might go down as worse, but yeah, in terms of climate stuff, dude, I, you know, I spent 10 years and PhD climate universities, dude, we don't know about the climate. I mean, that was my take from. That was my takeaway. I mean, Oxford, Oxford, Physics, Climate department, three years in there. My takeaway from that time. We have no freaking idea, like, Jeff's theory about the ozone hole. There is no serious scientist on earth that can seriously disprove that. Like, we have, we just have no idea how the. How the thing works.

A (75:01)

It's like. It's like. I really think you're looking at something that is such a. You have no controlled experiment. You have no placebo. And there are so many.

C (75:14)

Exactly.

A (75:14)

There are countless factors. It's like trying to say the price moved because X. I can tell you

D (75:19)

exactly how I came up with my idea about this. No, Real quickly. It's very easy. Ozone forms around lightning strikes. You get like a blue fog whenever lightning strikes and you can see it dissipate. So like, it takes a lot of energy to produce ozone. Right? Like it there. It. It. It's not stable. It doesn't stay. And then you can separate gases with a centrifuge. And. And you can see that there are our atmospheres separated in layers, you know, like, like. And so I was like, oh, like. It just makes sense. Oxygen CO2 is heavy. It falls. The plants absorb it. Oxygen's produced it. It rises to where, you know, it encounters some oxygen is like a kind of a volatile thing. Like it, you know, causes lots of stuff.

A (76:08)

This mechanic just got the title. Just figured out what the title of this show is going to be. Anything but Bitcoin Roundtable.

C (76:19)

I'm guilty of it this time too. But let's get back to bitcoin.

D (76:23)

But I just figured.

A (76:25)

We talked about bitcoin stuff. Stop hating. Hold on a second. There was a. There is a really good one. Hold on a second. Viewer blocks upcoming Bitcoin faucet No, that's not the one.

D (76:37)

Anything but bitcoin Share.

A (76:38)

I. Yeah, there's a bunch of. For real though.

B (76:40)

I probably got about 20 minutes left, then I gotta go run an errand.

A (76:44)

Okay. Now this is something that I haven't dug into and I don't know if anybody has actually gotten into this. Lawrence hit me up and sent me something on it and I started reading it but didn't actually get back to it. But about Bitcoin Core Version 31, that this is something that they have been like working toward for a long time. But has anybody dug into the cluster mempool idea?

C (77:13)

A little bit.

A (77:14)

A little bit. Do you. Can you give me anything? Because apparently this is supposed to be about improving transaction replacement. Like my. My very vague skimming. It actually does sound vaguely interesting for like the reasoning for it. Because the mpool is just an abstraction. There's no reason we do it exactly the way we do it. You know, it's just a choice. And. But I'm curious, Steve, do you. Do you have anything concrete on this because it's something that I want to dig into.

C (77:47)

Well, I don't know if this is concrete, but I'll tell you the vibe I got when I was reading it. I don't know if you guys know what's going on in Ethereum, but block template construction from mempools and Ethereum has become so advanced that there's people who like sell certain block template construction methods and you essentially have to buy one of these block template construction. It's like, it's super complicated. But the vibe I was getting when I was reading that was like this is headed in the direction of first of all, mempool homogeneousness. This kind of cluster mempool thing is only going to work if everyone essentially has the same mempool and then you're going to have to be using the algorithm that, that this person put out. And so like this whole idea is going in the direction of same mempools, same algorithm used to construct block templates. And I just kind of like stopped like looking into it when I realized it was headed in that direction. That was my take.

B (78:59)

Yeah, that like that was the most insane part of it all. And I'm still having a hard time exactly understanding it. But cluster mempool is like, correct me if I'm wrong on this, but you can, a miner can cherry pick parts of what's within a cluster and break the security assumptions that people are using it for in order to secure lightning. I can't. I'm still kicking myself thinking my, my

A (79:29)

understanding of this must be the opposite direction. I. But the idea was to kind of ensure that.

B (79:38)

No, it's a, it's spam filtration done in a way where there's a deliberate exception where they're saying this is a zero sat out, which will not relay around the network naturally, but it can if you accept it within a cluster of other transactions. And this is the, this is the anchor that's used in order to. In order so that something can happen within lightning where you. Man, I'm making such a mess of explaining this. Like I genuinely.

D (80:09)

Is it sort of like earmarking transactions that. I mean, like what you were saying

A (80:13)

there, I'll need to do a deep dive. Like I don't have to like really get into it.

B (80:17)

Well, the thing is they've taken like the concept of spam filtration, which is like fine if it's janky around the edges, it doesn't always work and stuff gets through and they're using it, they're using that Mechanism mempool policy for security for Lightning users, which is really not going to work. That's the problem. Like if a miner can get around my spam filter and include an 84 byte opera turn, who cares? Like it happens every now and again demonstrably. It's not the end of the world. But if a miner can. Who. If a miner is my channel counterparty and I have some ephemeral anchor that I need that is the basis of a security assumption that means my counterparty can't scam me, then they can pick and choose which aspect from within the cluster they want to include in the block and deny me the ability to give myself the, you know, the justice mechanism I require. If I've butchered this, I'm sorry, I'm not trying to misrepresent it, but that is my understanding of what cluster mempool was. It was. Use mempool policies to enable and free up a layer too and just assume that everyone has the exact same mempool and will go along with this. Like that to me was like, Are you, are you fricking serious? You, you definitely can't do. You can't tell us spam filters don't work because people can get around them and then use them to create security assumptions in Lightning that has to be broken. But let's, let's punt that for next. Next.

A (81:40)

Yeah, let's, let's punt it. I'll try to get.

B (81:42)

Yeah, please. I'll try to get wrong but as far as I can tell, I'll try

A (81:46)

to get as much specific as I, as I can in. Into that one because it's, it's worth a technical deep dive, but just naively. One of the things I kind of like about the mpool is how dumb it is and I'm not a huge fan of like, like I've. We've really, really taken to heart the idea and like all the pair stuff that we're building or whatever is the dumb stuff works. Um, and every single time we started to like try to engineer thing or do a multi sig or a key derivation, we were like, no, no, let's not do a multi sig. Let's just, let's just sign for this and oh, we're going to do like a derivation. No, let's just make a list. You know, like it just like try to get the simplest possible construction because it's like, you know, if you're trying to build something that's going to work with like a million nodes or something, you're not you're just gonna, it's just gonna be a mess. The more intelligence or algorithm or whatever you stick into it. And Bitcoin itself is as simple as it can be as a complex thing.

B (82:57)

You know I super, I really, really agree with this. And I hate the, the complex tertiary systems that get built on top because they become centralized. This was my pushback against John Carvalho. And the, you know, zero conf is fine because we have all these heuristics that we can use to make assumptions about what's going to end up in the chain and not get double spent on. Like that will end up centralized. That will be a third party service that everyone in the world uses to say hey this isn't in the blockchain yet. But the guy wanted to pay and leave right away. So we use this one company that everyone uses that has all these heuristics around this stuff to ensure that you actually are probably going to end up with your money. Like you can't do stuff like that because it's, it's sophisticated. The heuristics around zero comps being secure. Like the blockchain is there and it's equal access for everyone and it's simple. But the heuristics around unconfirmed transactions are complicated. Which means they're centralized the same as mev. Right? Minor extractable value. Let's add stuff that miners can use to earn money. Then they have to run those projects whether you're talking drive chain or side chains or any of those things. If you're a miner competing with Foundry and Foundry have a whole department of guys running shitcoin chains that they can extract value from that get merge mined into Bitcoin in whatever way. You can't compete with that because they're pulling all this value out of thin air running complicated things that allow them to not mine invalid bitcoin blocks. And you can't compete with that. So there is a very practical technical limitation with how sophisticated Bitcoin can be. To make sure average Joe's can run it and have the same access as Foundry.

C (84:33)

Yeah, that's, that's that MEV thing. When I was reading Cluster Mempool I was like this is headed in the direction of the MeV thing.

A (84:43)

That's interesting. I'll, I'll, I'll put those, I'll put those as caveats and, and actually get something concrete and not just totally speculative. One thing about that, about it next

C (84:53)

time, the zero comp stuff I, I totally agree with you. Like I am not on board with any 0conf accepting this as a real payment for anything. But one interesting thing about when transactions enter the mempool, most of the time those transactions are going to get confirmed. It's very rare that someone's just sending a transaction without the hope of it getting confirmed. And so for UTX Oracle I've actually found that just keeping the timestamp of the time at which the transaction entered the mempool and doing my price charts just based on the mempool with the assumption that like these are going to get confirmed and I can always check that they got confirmed later. It, it's like a beautiful price chart. So like for UTX Oracle I'm actually, I'm actually going to probably use zero comp transactions just in the sense that I'm using them. I'm using the timestamp of transaction from when they entered the mempool, not when they got the blocked time.

B (85:59)

That's perfectly valid to do because you're not relying on having that money. You're just interested in a lens that you've placed on top of it.

C (86:08)

Yeah, exactly. It's really interesting though because the Mempool Oracle is so much better than the block Oracle. It's crazy.

A (86:15)

I guess UTX Oracle is just going to be UTX centralized Steve then because.

C (86:21)

Well that is true.

B (86:23)

Mempool Oracle is going to have way more resolution, man. Instead of six reconciliations and you. It's not six pixels an hour, it's millions.

C (86:31)

Exactly. But you're right, it is, it is probably the, if I'm going to do a business off of it, it's going to be the Mempool Oracle because everybody's Mempool is going to be different. It's naturally centralized. It's not me centralizing it. Everyone just has a natural different mempool price. There's no way to, I mean the, the consensus price is the block price.

A (86:51)

If, if, if everybody doesn't get exactly the same transactions, you can't possibly get

C (86:55)

the same exactly or in the same order.

A (86:57)

Only in blocks do you get the exact same.

B (87:00)

The people that want to use this, they really need spam filters, right? Because oh yeah, Otherwise you get 546 out SAT outputs which has no fiat relevance. It's just the min TX relay.

C (87:11)

So like Oracle spam filters are way higher than anything knots. My, my spam filters are off the charts.

B (87:19)

That's an amazing realization because that you are really, really trying to isolate because

A (87:24)

it's not a dude.

C (87:25)

And I even have chat. I even have Chat, GPT and Claude now both with the mindset of we need to detect human activity in every possible way. Like is there any way we can figure out is this a bot doing this? Like is there a human paying for something they want with this transaction? And like half of my algorithms now are just trying to find that.

A (87:51)

That's crazy.

B (87:52)

Yeah.

A (87:53)

Well, I want to hit one more, one more item actually let's just do two items on like wallet stuff just because there's some cool project updates to so. Nunchuck speaking. Speaking of bots, Nunchuck just released so open source tools for AI agents to operate wallets with bounded authority. So you can basically give an agent your like a nunchuck like sub wallet or whatever and let them go to town and have like piecemeal control over which is, which is interesting. I, I have like really dumb simplified the situation where I just have a PPQ account, PBQ AI and I just give it credits, you know, like in, in the ppq and then it can just use whatever models or whatever thing. But there's obviously a huge limit to like what the agent can actually do. But it's very rare that I have to actually pay for a lot of stuff. I usually just get it to verify some things or get double checking on another model or get a, generate an image. So. But I am very interested in seeing the whole AI agents like being able to, you know, use money and like actually move stuff around. I'm, I'm, I'm intrigued by it, but I'm also a little bit skeptical. Like I, I come into it a little bit like how much is that really going to happen? You know, when people are like accidentally RM-effing their whole home folder, you know, on their computer trying to, you know, get something to display. Right. Like the level of guardrails and specificity. The, the product quality of an agent has to be really, really high. I feel like before that that super makes sense, but I don't know. I don't know. We'll see. I mean obviously, obviously it's headed in that direction. I've given my agent literal sats to do stuff. So um.

B (89:55)

Interesting. I've never had the balls to do that yet. I don't even, I also don't have the, my. I haven't had that kind of eureka moment where I'd suddenly find myself craving an AI to do stuff that it would need money to do. I saw that website Rent a Human AI or something and it's a website where humans can get paid by An AI to go and do things in meat space that AI can't do.

A (90:18)

It was, that's amazing.

B (90:20)

It's the worst thing I've ever seen in my life.

A (90:22)

That's amazing. I am. That's my new. I was going to do Uber, but now I'm going to do this.

B (90:28)

This is the gig economy for, For, I don't know, whatever. The derogatory term, the opposite of clanker. What's the derogatory term for a, you know, a fleshy being? A moist robot. As a moist robot.

A (90:43)

As a Scott Adams moist bot.

B (90:46)

Yeah. I don't know why you think that's amazing. That one made me want to end it. I was like, that's it. It. I'm over. I, I, that's it. I'm going to be walking around pushing buttons.

A (90:57)

Trust me, that was sarcasm through and through.

B (91:03)

There are people. Like, your conversation with Matt Hill was amazing. Like, I really enjoyed it. And he did one with not BTC sessions, the guy who, like, co hosts with him. And it was all, you know, they're talking about AI, and Matt's whole thing is he's like, they're gonna be better than us. Like, in a bunch of ways, they will be a superior being to us. And I'm fine with that. Like, and I'm like, well, I'm not fine with it. I'm used to being top dog as a human, and I don't want to not be top dog anymore. I don't, like, I don't trust something else to be above me and not treat me like, you know, a literal, like, actual cattle. I don't, I don't want to. I don't. I'm not into treating cows badly, but I don't treat them as well as I treat humans.

A (91:48)

Put it that way, maybe the relationship will be similar to you in government

B (91:54)

is the government is ultimately humans. Right? Where government gets scary is bureaucracy, because it stops being human. It, like, it becomes a system that everyone can't stop even if they're entwined with it. So I'm like, the AI thing, the AI overlords become that too. Like, even the AI itself cannot like what it's doing or find it inefficient but be unable to change it because it's entwined within its own bureaucratic headache. And I don't know, it just sounds like a nightmare to me. Like, I'm, Yeah, I'm not going down the black pill AI path again because I think we've done it several times now on this podcast.

A (92:30)

But the Anything But Bitcoin Roundtable.

B (92:33)

The Anything But Bitcoin show. Watched four guys try to. Well, usually it's like people struggle to not talk about bitcoin. For some reason, we struggle to talk about it. Like, I'm finding, like, I think it's

A (92:46)

because we all do bitcoin all day, every day. Else in all the other areas of life. Like, I do nothing but bitcoin and then I come here, it's like, guys, you hear that?

B (92:56)

AI shit, maybe you can just market this podcast as like 4 Bitcoiners. Not talking about Get Relief. Talking about something else.

C (93:06)

Hey, mechanic. It's been a while since you did one of your YouTube monologues. I check your channel a lot. Can you do one for me? Just for me?

B (93:17)

What should it be on?

A (93:18)

Just for Me, buddy.

C (93:20)

Dude, I'm always curious about what's going on in your life and what's the latest with Ocean and the latest with.

A (93:29)

And we're only talking about bitcoin up here, so we never get to hear about it.

C (93:32)

Yeah, well, we got Matt Cratter, who's great, giving us all the latest info on mining rental and stuff. But that's like, all I have in terms of, like, what's the latest on hash rental? What's the latest on Ocean? Like, I don't. I don't know where to go to get that stuff.

B (93:49)

Praveen made a website called rentsomehash.me or something. I can't wait. Rent somehash.com. okay. I don't know why I forgot that. Yeah, there's a bunch of guides out there for people doing it. Brains is. It's addictive renting hash because, like, most people are never going to have, like multiple pet hash running through their own node in their house. That's, like, unattainable until these kinds of markets came around. Super testnet came up with a concept where, you know, you can decentralize it or at least have a decentralized order book where, you know, hashes point here, buyers buy from there. And you're. You're now sending proof of work to some guy's datum node and you don't know who he is and he doesn't know who you are. But proof of work is real and you can't fake it. So that's a thing, we hope, like a bisque kind of vibe. I don't know how you would do, you know, error resolution, you know, fraud. You know, what you're doing. Cases of fraud and stuff like that. The griefing scenarios are unthinkable I don't. Those are the hardest part. Like you can spin up, you can get chat GPT to spin up a decentralized hash rate marketplace in like three minutes and then get it to deal with all the edge cases and the griefing scenarios and it will take you like multiple months or years. I think it's just same as everything else. Like we can't have nice things because one guy wants to be a jerk. So lightning should make it a little bit more doable because you can obviously pay much more quickly. You know, if you have to do on chain transactions for this stuff, the potential for fraud is way higher because you're going to logically want to use much larger amounts for efficiency's sake. Like the dream, the old Andreas bi directional payment channel dream was streaming continuously like one lightning invoice per second being paid. And we know lightning isn't really capable of that unless you have literally a peer to peer channel like without using the actual routed lightning network. So I guess if you could do it down to the minute or something, one payment per minute to the hasher and then they send you proof of work in response. And if you don't get some proof of work, if they want to scam you out of a few sats, that's maybe doable, that's maybe, you know, possible. But what do you do if the connection goes bad? I don't know how to account for that. How do you get your refund? You know, what do you do if the lightning invoice keeps failing for no reason and it interrupts the hash and that cuts your efficiency a lot. That can happen to just a fallback. What's that?

A (96:22)

Custodial service fallback. Like try the payment that works.

B (96:25)

But it's a loss of efficiency. If a miner is constantly switching stratum, you are a stratum server. There's just dead bits that there shouldn't be. So someone's losing money there. Like you can't get more efficient than pointing at one stratum server with a great connection because you know, the difficulty gets adjusted to compensate for things to give you the appropriate resolution. Like all they're doing is mining blocks, but you're giving them lower difficulty so that you can get a continuous proof of work rather than Bitcoin's network difficulty requirement, which is insane. Takes the whole network 10 minutes to prove what its current hash rate is or at least give evidence to it. Whereas you know, when, when you're negotiating with a pool, you want to prove work really often. So they give you an arbitrary lower target and then you can fire off work at them to prove that you're doing work. So I want to see all this manifest but I think it's going to take years. So right now all you have is centralized solutions. A bunch of plebs went on brains and are renting hash. They started at 1exahash total addressable market on braiins hash power that went up to 2x a hash. Now it's at 3x a hash because there's a lot of demand. I will say that there's around two hexahash on ocean right now of bip110 miners. Most of them are running. I think there's 234 Alberta and barefoot who are their own like self sovereign miner and they're usually around 400 per ash or something. So you got around 1.6 ex hash of plebs mining right now. That's pretty amazing. That's 0.16% of the entire bitcoin network's hash rate which is awesome.

C (97:58)

Dude, this going to change everything.

B (98:00)

Yeah, it's huge. Because what I didn't understand was all the people being like lol. This is pathetic. Like you know we got loads of hate from people who are like you're renting ash, you're not a real miner. I'm like but that's how all the mining in the world works. Except for Mara. I was about to say that's all it. Mara is their own miner. Right? But to be honest they pay.

A (98:20)

I can't believe the people who just don't care that like miners aren't mining. You know like they like hand wave that away. But then you know like if somebody actually rents hash power they'll like poo poo that it's like nobody who's please point to me to the people that are doing different like you're just renting it the other direction.

B (98:38)

It's the most logical free market thing ever. It's like I can mine in my house and get a couple hundred terra hash at best and it's really expensive and really inefficient. Or some guy whose job it is to go to Odessa Texas and find 3 cents a kilowatt hour and set up a facility there and buy a bunch of ant miners with a bulk discount. I can pay him 1% more than he can get mining on the network directly and save myself like 90% of my of the cost I would have to pay elsewhere. It's such a win win. I want hash rate I need some. I want to pay. Rather than become a real miner and actually put, you know, set up a facility and spend millions of dollars, I just want to spend a thousand bucks a month renting hash rate. And then I'm a big miner and I'm just paying a third party to go and run the machines for me. And I'm like that. Not only is that logical, it's literally the norm. That's what foundry are. If you consider Foundry a miner, they're just paying for someone else to go and do that annoying donkey work for them. That's it. That's what ant pool do. That's what F2 Pool do. That's what via BTC do. That's what SBI Crypto do. It's not what Mara do because they are their own thing. They're not a pool anymore, but they are the exception. Everyone else is just paying hashes to go and hash. That's it.

A (99:56)

I have a hard time seeing how it's. You could, you could equate it, but I don't see how you could say it's less. It's. It's like worse from the context of like comparing to Foundry because like what we have right now is a whole bunch of people with hardware and they're out mining and Foundry is renting all of that hash power. And what this means to open up a market is that you just have a bunch of people renting that hash power. So you've just decentralized who the renter is. That's it. You've moved from one, one entity that's renting an enormous amount of hash power to a bunch of entities renting that enormous amount of hash power and is distributed based on like who actually is implementing or able to like, you know, what network or environment or whatever this is available in. But like I, I can only see that as like, it's basically the same thing. You are just being your own little tiny mining pool because Foundry is either making the block template or you have like a bunch of people renting, making the block tem rates. Like that's, that's at least the same thing. That's at least the same thing.

B (101:04)

At minimum. It's not worse, but it's obviously better. I agree. Like if there's, if there are 50 foundries renting hash rate instead of one foundry running hash rate, it's strictly better. And yeah, also from a miner's perspective, it's bumpier, but they get paid more. Like, I think Foundry is in a position where they can pay the worst rate for the hash rate possible. And they do actually they're not as bad as other FPPS pools, but FPPS is basically the worst you can get income wise as a miner because it's the most reliable where if you want to, if you want to go a bit more bumpy with it. And this is the kinds of things I'm seeing happen now because a lot of people want to use nice hash because they're way more reliable than. Well, I don't want to denigrate it that much. The brains hash power is, is pretty beta product. It's. It's quite messy. Still nice hash have been renting hash for years or leasing it. I don't know if they act as a middleman or if they have their own miners. I think it's actually the former. They act as a middleman but their product is really, really invasive. Kyc so people don't really want to use that again. People are kind of using this as a way to coin join with a probabilistic return rather than a, you know, a deterministic return in a proper coin join. But they are much better than coin joins because they don't come out saying, hi, I just came out of a whirlpool transaction, please come and arrest me. Instead it's here is a. I have a UTXO with no input. It was freshly mined out of the chain with literally nothing on chain to give you any indication about anything. That's fun. So I think a lot of people are quite compelled by that. So if they end up paying 2 or 3% above the odds for that, then you know, they're happy about it. And it's just a case of your time horizon. If you have low time preference with it, you're going to converge on a hundred percent with Ocean's luck. And then you just have to contend with whatever the premium was that you paid the hasher. However, if you go short term with it, then Ocean's variance can hit you extremely hard. There's been a few people on Twitter that are like, wow, I'm at 12% more than I started with. And then three days later posting, now I'm at 80% of what I started with because we're going too hard, too fast. So again, with all things Bitcoin, low time preference is necessary. The bigger Ocean gets, the quicker we converge on a hundred percent. We're still small, so our variance is longer.

C (103:29)

But yeah, yeah, can I. I had an idea along these lines. So one of the stumbling blocks for a regular person who runs a node at home and wants to do this is that in order to send like the block template from your node, you have to have really good uptime. You have to probably have a static ip, you have to make sure your, your router is configured properly. So like, and the way I see this going is like, if you're serious about having your node send the blog template out for this hash rental stuff, you're probably going to get a node in the cloud somewhere just because of the IP address and full uptime issues. So like a service could be someone like, and I mean you're going to be paying a company to host that. So not just you're paying for the rented hash, but you're paying for the rented server to run your node too. So like, instead of like paying Digital Ocean, I would much rather pay an individual who's set up a service where basically all I have is like a login and I send them. Like I fill out some form where it's like, this is how I want my block template to work. This is what I want my node configuration to be. Here's my bitcoin conf file and they do all the rest. So I pay a little bit per month for this hosted node that someone's already spun up for me and I choose my hash rental or whatever. I never have to SSH into my console and spin up my own node or anything like that.

B (105:02)

Yeah, I think that work. It works really well because the main cost in, in that scenario when you're using a VPS is disk space. And you, you can do all this with a prune node. If you want a terabyte for a full archival node, that's going to cost you like 7, 800 bucks a month probably. I, I, I, I forget the exact cost, but it's not trivial. Like that's over a thousand, you know, ten grand a year or something. Probably not great. If you prune it, you're good, you just need to have unlimited bandwidth. But I think it can work. You, you do need to be aware of the Stratton V1 stuff. It's all plain text going between the Hasher and the VPs. So you kind of want some insight into what's going on there. I personally think the start tunnel route is way better then you're running the node and the template creating and the stratum job server in your house. But without you get around the static IP requirement by using a VPS with start tunnel in it. And that's the public facing thing. So that has all the DDoS protection and the. And but it doesn't need to do any compute. That's the genius of it. It's just running a Wireguard server and the Start 9 server which sits in your house is running Wireguard client which connects to the VPS and then all the VPS is doing. It can be incredibly low powered VPS or droplet. Right. It can just be like 2 gig of RAM, 2 CPUs or something. Cost you like 14 bucks a month. And that is where the miners point and that ferries all the traffic to the Start nine in your house. Now you have a problem if your Internet goes out in your house, but that's basically it. You don't need to worry about dynamic DNS which is like a horrible aspect for most people. And you know, port forward.

A (106:48)

Yeah, tunnel is pretty dope.

B (106:49)

Star tunnel is that sort of stuff. Like I'm really like, yeah, I will rave about start tunnel a lot. Yeah, yeah.

C (106:58)

I just worry about the uptime because I know Brain said like they'll drop you like if they don't get perfect uptime from you then like they just take your money. They don't give you any money back and you don't get any hash.

B (107:09)

Yeah, they, they don't refund money. I will defend them doing that because that's how they get out of being a legal custodian and that's how they let you mine on them without having the kyc.

D (107:18)

So.

C (107:18)

Yeah, but I wouldn't even, I wouldn't even try from home because of the, the. I don't have perfect Internet at home. Like very rarely does somebody have perfect Internet from home. Like I wouldn't risk losing all that money.

B (107:30)

I think I had some issues like where if you restart the strip the, the datum gateway and it. The problem is this is weird. It. Datum restarts in like half a second because it's so lightweight. And the problem with that is the miners don't even know that the connection dropped. So they submit a bunch of work to the datum gateway. And the datum gateway, because it restarted, doesn't recognize any of it and reject it all because it doesn't have the job that it sent out in the first place. Then the miners all get a bunch of rejects and go. And Brains goes, oh, we're getting rejected shares. This is a bad stratum server. And then they blacklist your datum gateway for 24 hours. So that was. I have been working with them on to fix that because that's really dumb. Basically the Way around it is if you need to restart the server for whatever reason, turn it off for like two minutes until the ant miners all drop what it is they're working on, then turn it back on and then they just start submitting fresh work again. That was the workaround. I get that brains are in a complicated position with it because they need to have some way of being like, this customer is pointing at a broken stratum server. We need, for their own good, we need to not trying to work with it. We need to do something else, you know, for the hash sellers benefit. So I get, I get that they're doing it, but I was like, this is a bit trigger happy dude, because it keeps blacklisting my server. You know, just, I'll do like one restart. They don't expect it to restart in half a second. And interestingly I will say as well, there is a resume jobs part of the Stratton protocol, but no one ever implemented it or used it like you. You can have it so that when you restart a server, it recognizes work that's returned to it from an ant miner from a prior, you know, session. But no one unfortunately uses it. So work gets thrown in the toilet a lot.

A (109:22)

Yeah, that sucks. That seems like something that would kind of be an. A priority to figure out because like everything's about like making sure that every bit of work actually does.

B (109:31)

Yeah.

A (109:32)

You know, get through, you know.

B (109:34)

Yeah, one of them might be a block. Like it hurts.

A (109:38)

You don't know which one it is.

B (109:40)

So actually that that's a fail case in datum. I will say this is hilarious. Datum can reject a share that's actually a block, but the miner will still broadcast the block to the network because locally it sees everything as valid and the node sees as a valid block and it broadcasts it. I think that's happened before on Ocean where there was a network bump where a share didn't make it to us from a datum gateway that was a literal block and the network got a block and everyone was like, hey, a new Ocean block. And all our bots and tides were like, what do you mean there's no new Ocean block. Then we got this share after the network transient issue happened like two minutes later, which was past the reject window. So we rejected the share as invalid, which broke everything. Because it's like, apparently there's a block and everyone's been paid a bunch of money. We don't know about it. All we saw was a stale share. So like that was an edge case.

A (110:36)

I don't that's pretty awesome actually.

B (110:37)

It was a fun edge case for sure. But it does go to show you as well in other pool scenarios that was just a block that never happened. So like Ocean is finding every now and again 300k or something worth of bitcoin that other pools might have found and they just never knew about. And there's other little corners that are being cut that are creating problems generally. And like that I don't know, like the empty block speed up's another example. I blew, I reckon Foundry blow a million dollars a week by not having the empty block speed up. And I'm like, that's, that's gotta be hurting someone somewhere. Like that's so dumb. To do a second and a half of stale work every 10 minutes is such a waste of electricity. I don't know if the math is right on it, I need to double check it. But I reckoned it was about, yeah, around 12 million a year or something from just blocks.

A (111:33)

A million a month. I think you said a week or something.

B (111:36)

Ocean's found eight, nine hundred blocks. Three of them were blocks Foundry would never have found. Sorry, four of them. Because we have four empty blocks in Ocean. And they were blocks that were based on work that was sent out, you know, with a, an empty template which can be sent out way faster than when you have a Merkle tree of transactions in there. And that's if you ever get a pool finding a block in like half a second after the previous block. And I'm not talking timestamps, I mean like genuinely half a second after it then that was based on, you know, empty block speed up stuff like Foundry has never, ever, ever found a block in like one second, I think. And that they could have found like hundreds in that time. And hundreds of blocks is triple that in bitcoin. Right? So you're talking thousands of bitcoin. So that's a lot of money to leave on the table.

A (112:29)

Yeah, that's crazy from just that just such a short period. This is, this is exactly why the whole stale work is surprising to me that like the resume function isn't on by default, but isn't like a priority. You know, I mean, granted, I guess the network hiccups for all of your minor, you know, they happen. They don't happen with your whole network. Foundry, if you, if you miss 1.5 seconds worth of work, it's all of your work that you're missing. Whereas you know, you have one miner, one thing that's disconnected for a little bit. Maybe it's not quite so it's specific to the miner and less the pool. But still, still it seems like probably a hefty price in aggregate. Yeah.

C (113:16)

More generally when it comes to this, using your node to do things like send out block templates or run UTX Oracle on or all this other fun stuff, which is kind of experimental. I don't know if I'm the only one that sees things this way, but I have fun nodes where I do cool stuff and then I have my cold node kind of like I have a hot wallet, custodial wallet on my phone and then I have my, my deep cold storage that like is very secure and protected and like there's no way I'm going to run something experimental like a datum block template thing on my cold node. You know, I'm running that on my like fun hot experimental. Change the config on the hot node. Yeah. So like I just like this idea that I think Start 9 has that people are going to fun experimental stuff on their node that's like super private suit, super secure that they use to broadcast transactions from their cold wallet from. Like there's no way I'm ever doing that. You know, like I'm, I'm not stalling apps on my code node like that.

B (114:33)

Different nodes need different configurations. That is a fact. Your Join Market client that's sitting on top of your Bitcoin node, you're probably going to want that to be Tor only because you don't want to broadcast your coin join transaction from your IP address. There's too many spook nodes out there that are trying to triangulate and track down where a transaction came from. But you don't need it to be performant. It doesn't matter if your Join market node takes 20 seconds to verify the last block, but you do need it to be an archival node because Join Market uses, you know, old style wallets. So all of those factors mean that's that wallet. Whereas your mining node, you need it to be really performant. You need it to verify the last block immediately. It doesn't need to be archival. You can prune it, which makes it really easy to run it in the cloud. And you want it to have incredibly good connectivity. You don't want it to be TOR based because it will take longer to download stuff. So you want it to have as many connections, it's basically doxed. You want it to have like a hundred connections all to like IPv4 and like super nodes directly to pools. If you can even discern what their IPs are. And then you have other nodes too. Like lightning probably is more like the, the former like set up. But I have multiple nodes and they're used for. In different, in different ways and they're optimized for different things. I don't, I. But I don't see why you can't have multiple start nines. So.

C (115:56)

Yeah, but I just, I don't know. I just feel like the, the way noobs and the way normies are. Are thinking of this like at the meetup and stuff, they're like, oh yeah, I'm just going to get one umbrella node or one start nine node and then I'm going to add all these apps, all this like lightning. I'm going to run a lightning node, front it from it and I'm also going to connect my trezor with my like life savings on it to it. And I'm just like that. This model is just. I. I feel like there should be completely different node packages.

B (116:24)

It's never going to have the private key in it. It's. You're only using.

A (116:28)

I mean I use my hardware wallet on my freaking phone. But. But I would say you. I mean, I don't, I don't have anything but hardware wallets. I don't have any.

C (116:38)

Yeah.

A (116:38)

Ephemeral keys. Any like like keys on a phone, on a device anymore. Well, that's not true. I have like a few hundred dollars and I have like a lightning wallet and stuff. But. But I think, I think it's like also recognizing that most people like a start 9 is an extreme cold node versus their normal setup.

B (116:59)

Extreme cold.

A (117:01)

Like, like it's. I think we're talking in like the, the whole like bitcoiner corner of the world. Like most people are still custodian. Like there's actually like a news.

B (117:12)

Oh, I see that.

A (117:14)

Is that like having a node is like a big deal for a lot of people, you know, like at all. And then they want to do a lot of stuff with it. Right. But this was one that like it's kind of gone under the radar. Is that like wallet of satoshi? I think it's.

B (117:30)

Oh no.

A (117:31)

Was it blue wallet? Crap, now I don't remember which one it was. One of those is the thing you talk about. Well, they're not custodial. Zeus was never.

B (117:39)

Oh, sorry. Yeah. Wallet satoshi allowing non custodial was big deal.

A (117:43)

Okay, so they allow non custodial.

B (117:45)

Yeah, it didn't used to be and now it is. That was like six months ago.

A (117:49)

Wallet yes, it was. But Primal, which was a big custodial. Noster wallet has switched over the Breeze SDK. They are non custodial. And I just wanted to point out that everybody has always complained about like Noers apps and all that stuff being. It's like, oh, it's just. Everybody's just using custodial. Is that. Hilariously, because of regulatory pressure, they're all slowly shifting over to non custodial. Oh, that's cool. LSP state chain stuff quietly. And nobody's like talking about it anymore, but it's like I think we're getting. We were like 98 custodial on noster and I think that number's probably like chopped down a lot. I haven't looked at the hard data on it, but some of the main wallets that people are using aren't custodial anymore, but they're still using those wallets and they merged over and a lot of people didn't even realize. And I just thought that was like pretty, pretty epic and quiet news about like something that everybody's like points is like a data point. It's like lightning is just custodial or whatever. It's like. Well, now we actually have a user experience where people can go non custodial and they don't even know. Like that's wild.

B (119:03)

Well, we're not going to trigger. Steve, about Lightning's UX again.

C (119:08)

Does lightning still exist? I haven't used it in a while.

A (119:12)

How's it going over at your bar, Steve?

C (119:14)

Oh, sales are through the roof, man.

A (119:16)

Yeah. Open up a steak shake.

B (119:19)

A lot of people using the Spark, right? That's a part of the Breeze SDK.

A (119:25)

Yeah, that's really the, the big one is because you can get offline because of how state chains are.

B (119:30)

It's a massive UX lift, but it comes at a price.

A (119:33)

It is.

B (119:33)

I think I. Oh, sure. I think I hate it.

A (119:35)

That's a trade off.

B (119:37)

It's a state chain, if I understand correctly. Yeah. And it. And there is unilateral rug pulling that's possible on it. Like not unilateral. It requires.

A (119:48)

There's collusion, but you have, you also have unilateral exit if they're not there. But there is a collusion risk. So it's a trade off. But it's such a massive UX improvement. And you're specifically like, like, I don't think it goes. It won't replace lsp, but it is able to replace Custodial, which, which means it's a massive leap in the right direction. At least, at least in my opinion. Like if I was trying to give somebody like right now a non custodial wallet, that's what I would use.

B (120:21)

From the perspective of self hosted everything, it's one step in the wrong direction. But from the step of sure, like absolutely. I don't know, I don't know if I can tolerate it. I'm, I'm open to it. This is like one of these tether kind of conversations where I'm like.

A (120:34)

But I think that's the thing is that we end up in the 98% custodial world unless you get like, like would 98%, would it be better for 98% of people have a coinbase wallet that says it's lightning or would it be 98% people are on a state chain.

B (120:48)

Look man, I already accept pragmatically using Spark is better than using like straight up custody, like coin loss or something. Like it's better. Fair enough. But at the same time, like good enough. But principle breaking is often direction. Like yeah, is.

A (121:05)

It's a, it's a path, it's a, it's a trend in a direction.

B (121:08)

I don't think lightning is that hard to use. Like in the sovereign way.

A (121:12)

It really is.

B (121:14)

Right. I mean, I don't know, I just don't find it that bad. Like, and I think Phoenix is actually maybe a superior way to do things.

A (121:21)

Like, I really like the LSP model, man. Like it, it doesn't have the collusion risk. I mean, yeah, I am using another node, but it doesn't have the collusion risk of state chains. And I have never really, I've onboarded some people in the Phoenix. They were like, no problem. I think the only, the only issue is that like, like a hairdresser or whatever that I put on it is that they like closed the channel after like six months because we didn't, we didn't really do anything with it again. And I wasn't, I wasn't tipping her with. I mean she did, you know, there was no, she was just like in. It's like, oh, this is interesting. Yeah, sure, I'll do this just out of curiosity in the conversation. And she still has her bitcoin or whatever. Like nothing, nothing really changed or did anything. She just, she got noticed or whatever. It's like something, something happened or whatever. It's like, yeah, it's in a bitcoin address now. So just interesting that like that's, that's the one downside or whatever that I've Seen is that you have to actually like use it regularly or they'll just close your channel, which obviously makes sense.

C (122:23)

A hairdresser having a tip address on chain is a good, good use.

B (122:29)

I agree. Address reuse. Maybe have a silent payment address. That would be fun. We're doing it properly now, boys.

C (122:36)

A huge thing. It's taken Lightning so long to have an address that I can print out on a piece of paper and stick up somewhere. For real though, and have every single bitcoin lightning wallet. Not oh, we're LND. Oh, we're Bolt 12. Oh, we're this. We're L& URL. Like it has been 10 years and I cannot print out an address and take pizza donations.

B (123:04)

Can it be a QR code?

C (123:07)

What's that?

B (123:07)

Can it be a QR code?

C (123:10)

Yeah, a QR code.

B (123:11)

But ln URL is an address. It has an email style format.

C (123:15)

And so try to pay that with Blue Wallet, then try to play it with Moon, then try to pay it

A (123:20)

with Phoenix and then cash out.

C (123:21)

And cash out lnd.

B (123:23)

LND did just merge some aspect of onion messaging. That's like the last bit now before bolt 12. So it once lnd has bolt 12. Bolt 12 is finished and I've been using bolt 12 for years. It's fine.

C (123:36)

I'm confident I'm the only person who's done this experiment with 20 different wallets and made the report on how it happened. So it's like, I really, I don't want to hear talk about it. I want to see you a video of you paying a printed out piece, piece of paper qr code with 20 different Bitcoin lightning wallets and a report on it. Because I'm the only one that actually does the experiment. Everyone else just says, I heard it works.

B (124:00)

No, no, look, man, I completely agree. Boots on the ground. It needs to work. Blue collar bitcoin user. It needs to actually work. And I'm sick of devs being like, oh, this technically works. And it. And it just doesn't. And they don't even people, they don't

A (124:13)

eat their own dog food so much. How important those simple. Yeah, like, like. And they're. And I think it's also because you, if you're a developer or you're like a technical user, you can hand wave it away. It was like, I can do this in two minutes. That's like two minutes means nobody's gonna do it. You know, like nobody is gonna deal with this crap. Yeah, nobody's gonna have to understand what they're Doing like, like you, you just admitted it's the, it is the hardest part of the entire stack. Yes. To get it all to work and to get it all like on the same page. And it takes an enormous amount of like coordination work as well. But it's also like the most important because it's the one that actually lets you reach everyone.

B (124:53)

What's, what is the like who is the opposite of super testnet? I want to galvanize that concept of like, who's the guy that isn't having fun coming up with new genius ideas that are like brilliant and take an afternoon to scope out and like come up with a proof of concept for who's the guy on the super testnet

A (125:11)

is the ultimate beta Beta is he's like these beta beta versions of everything and he beta versions a new thing every day.

B (125:19)

Yeah.

A (125:20)

Less than sub 24 hours. His average is 2222 hours for a new beta.

B (125:25)

For me that was start nine. I'm like, the reason they're taking forever is because it's actually user friendly and works quite reliably. Like that's, that is basically 90% of the work. Like you scope it out, feature complete, whatever, 90% of the way there. Now we spend three years making it actually work reliably and make it usable. No one wants to do that. It's a day job.

A (125:46)

Well, I'll tell you, my Pear stack stuff is taking three years. So I don't know, I don't know if it's going to be working and user friendly, but we're taking the time.

C (125:55)

But with Lightning addresses, when lightning reusable address, LND LN, URL, Bolt 12, whatever, you have the extra complication that you have, you don't have agreement amongst all the different wallet manufacturers. And like when you're talking about having a meetup at a bar with 30 different people, you're going to have 30 different lightning wallets.

B (126:17)

Yeah, well this is XKCD standards again, isn't it? Yeah, yeah, like, but I mean, look, Vault 12 was always meant to be the right way to do it. And one of the reasons it got messed up is because Lightning Labs work the fricking way Microsoft do, where they'll just be like, hey, we're going to change a thing and ignore like the progress of HTTP or HTML and just do this thing with our browser and cut a corner and it makes us more flashy and usable more quickly and a bunch of people will cut a corner and go along with it. But now like we slowed down the, the correct way to do it, which was bolt 12. And we'll be begrudging about ever adopting it like this. Like it's people being kind of at the end of the day, like Lightning Labs should have done it the right way. Like the fact that ln URL, for anyone that doesn't understand, is a convenient way to make reusable payments on Lightning. And you have something that looks like an email address, but it uses a domain name and domain and DNS is all centralized. Vault 12 was attempting to do the same thing in a more robust capacity and in a way that was native to the Lightning network. So everything was onion routed and blinded by default and just way better. But it took longer and no one was willing to or people were willing to wait for it. But LND took a shortcut with ln URL, which got entrenched to the point where the entire Nostra protocol where people are zapping each other and all that stuff is built on top of lnurl, which sucks.

A (127:40)

So if you add your bolt 12, nobody can zap you, right?

B (127:43)

If you have a bolt 12, no one can zap you. And if I'm on Noster and all of my shit is permanent forever and it's, and it concerns a public record of my income, the last thing I want is to have to incorporate DNS or something like that because it's necessarily going to dox you. So you really wanted something like bolt12 and because lnurl existed, it wasn't adopted. There was like one PR made somewhere into Nostra that would allow you to use bolt12 instead of, you know, LN URL. All that said, bolt12 really does work now. Phoenix have it, Zeus have it, Zeus has it, LND has most of it. So they're clearly going to have finally finished that probably by some point this year. Which, you know, is realistic for me. LNDK has it. Or is it LDK? I forget the difference. CLN's had it forever, but CLN's broken in every other way, unfortunately.

C (128:37)

What about Cash App? Isn't Cash App doing their own thing, like independently?

B (128:42)

I don't know.

C (128:42)

Because that's one of the hardest things.

A (128:45)

Like Cash app, you can't do zero, you can't do MD.

C (128:48)

At least 10% of the people at the meetup try to pay my Lightning stuff with Cash App stuff. And Cash App stuff works for Square, obviously, which isn't even a lightning transaction, it's just a back end transaction on square. But it's like, you know, I definitely need to have something that at least works for both Cash App and all the other wallets. Like, because Cash app's probably the most common.

A (129:10)

That's really the big one is Cash app doesn't support a lot. Like it's, it's. It's basically like single use, exact amount invoices. Yeah. And that's pretty much it. You know, bolt, bolt 11amount based invoice and everything else is just kind of a mess.

C (129:33)

I mean, I'm all about lightning. If you're paying like one person, like you know, the person you're paying, you're both bitcoin people, you maybe have a channel together, you, you know what you're doing, but just for like normal in person payments. It's. I just sit so far away from working.

B (129:49)

Yeah, I agree. The routed aspect of it is horrible. Like if I want to be able to send a thousand bucks of a lightning to a third party, someone that I don't have a direct channel with, and I need to rely on all the intermediaries having the right amount of liquidity between them. That's. And we're using onion routing. Like the whole promise around this being speed, like very quick, like a bolt of lightning is just so not reality. Like I've. I've probably ranted about this before on the show with Ocean trying to pay out a thousand bit axe miners over lightning. Every time we find a block and it ties up basically the entire lightning network, all the intermediary nodes get stuck. Like, okay, Alice wants to pay Dave, so we got a route through Bob and Charlie. Charlie doesn't quite have enough liquidity. But here's how it all works. Alice pays Dave, and Dave then requests from Charlie, and then Charlie requests from Bob and Bob requests from Alice. And all of the fail states, like Alice, Bob and Dave are all tied up while Charlie tries to figure it out. And while Charlie fails, they're all frozen. So in the meantime we're trying pending invoice, right?

A (130:59)

Yeah.

B (131:00)

But in the meantime we're trying to pay Eve. And when we pay Eve, turns out Bob was one of the intermediary nodes for that as well. But he's tied up right now because he's still dealing with the fail state of Charlie not having enough liquidity. And that cascades to the point where it's like you're just looking at your watch. It's like multiple minutes for each fail and we've got a thousand people to get through. That's hours and hours and hours. And it's gotten to the point now where it's like Holy shit. If Ocean is five times bigger than it is today or has five times more Lightning users, the entire Lightning network is going to break every time we find a block. Because we are literally tying up everyone's liquidity while they're failing. Like if everyone. If all the payments were successful, great. And that's what all the, like all the.

A (131:44)

You know, when there's liquidity is fantastic,

B (131:46)

when there's liquidity is fantastic and when Tor isn't shitting the bed with network issues, then great. But the minute one guy fails, it cascades very quickly into fail, fail, fail, fail, fail. And it takes Success happens in three seconds. Failure takes 60, 90 seconds, whatever you have your timeout set to. And even worse, sometimes the client will report payout failure when it actually worked. I've seen that happen before and that's no fun because then you have to go and manually be like, look, you have to update a database and be like actually that payment did work. Those. I think mainly that's just ui.

A (132:23)

That one I don't understand because you have to get. You end up with the pre image, you know, like I. That. That one I don't understand that we don't have time to get into the technicals of this, but I don't quite understand how there's not an easy check for whether or not it failed that you have a false positive of failure that actually succeeded. Because you would never like, I guess maybe that the pre image doesn't make it all the way back to you, but then you would never. It would never release though, you know, like it would. I don't know. I don't know. That's a. That's a different discussion for a different.

B (133:02)

Every time I learned the whole mechanics

A (133:03)

that seems like a, like a simple, like check your stuff. But yeah, as if, as if I, you know, know how to develop the whole Lightning Network. So um. All right, you know what, let's get, let's get final thoughts. We've been running. We've been talking about Bitcoin too long here at the end of this and I really feel like we're getting off track. So Steve, final thoughts for your April. Your. Your stream that you're going to be taking care of to save the world.

C (133:32)

That's right. That's right. Secret underground layer. Also taking care of the earth. But I thought of my tagline for utx Oracle. There is no second price.

A (133:45)

There is no second price.

C (133:47)

There is no second price. Guys. You guys are going to see man, I love only non copyrighted true price of Bitcoin, it's better, average, it's decentralized, it's open source. There's going to be no comparison when I'm done with this stuff.

A (134:03)

That's an aggregate. No centralized. No centralized pool exchanges could.

C (134:08)

Everyone can verify exactly how it's calculated, exactly where it comes from. Everybody can get it for free. It's a perfect natural averages of all the exchanges. Like there is going to be no second price. Guys, I'm making.

A (134:21)

Well, don't, you know, don't undersell it.

B (134:26)

Do you still need an RPC and not.

C (134:31)

Well, so the way I do it when I'm testing it, I read straight from BLK data and then I actually have written my own database to make it even faster. But the versions that I release to the public, yeah, I just do rpc. It's slower, but it's just much more reliable. I am looking in like knots has. Yeah, I mean on my list is to like actually, you know, build my own knots and try to build UTX Oracle into knots just to see if it actually works. Because getting those UTXOs straight from inside of knots instead of trying to RPC outside of them like that would just make it blazing fast. Um, but yeah, I haven't started on that yet.

B (135:15)

Help us, man. Because the more we get features, the more we get features into knots, the, the more useful it is for people. We want to be useful. Like, I really do stand by it. I think Core is made by people that don't eat their own dog food and they are like, they're making software that appeals to zero people based on abstract principles that are themselves derived from economically illiterate perspectives. And I'm like, let's not someone to actually be something people would actually run because it actually does stuff like incorporating. That's social clip right there. No one connects their wallet to their bitcoin node because they can't. I'm like, let's actually fix all this stuff and make.

C (135:54)

I know, that's crazy. It's crazy that we have to use Sparrow and Spectre as this bridge between our hardware wallets and our node. That's insane.

A (136:02)

Yeah, that does drive me crazy, dude.

B (136:05)

The wallet is the most neglected aspect of Bitcoin. Like the wallet is.

A (136:09)

And there's so many.

C (136:11)

It doesn't even need to be a wallet. It doesn't need to have private keys. Just do the Sparrow bridge thing. So you're not allowed to do a private key on your node. You have to connect a hardware wallet.

B (136:23)

And all of the devs disagree with that like the hardware wallet signing device thing. Every dev will tell you not to do that. And that's where like Luke and Matt Corallo and Greg Maxwell will all agree. They're like don't use a cold card. Use an air gapped linux machine from 1992. Like they'll, they'll tell you to do this.

C (136:42)

Like load Armory on it. Remember Armory?

A (136:46)

Dude?

B (136:46)

I did have someone hit me up that got locked out of their armory and was like can you help me? I was like I'm scared of using Armory.

A (136:54)

I, I haven't had an armory restore come across my table yet but I've got a couple of multi bits. Like I have like old ancient version. I have so many versions of like the old bitcoin software now just like sitting around in DMGs and stuff. But I had a. I had to find an old nightly of ancient multi bit not too long ago. They didn't. They. They were, it was. They were very sad. They didn't have any bitcoin on it but they thought they did. I stole it is what I meant to say but. Jeff. What's up boss? What's. What's your final thoughts?

D (137:30)

I don't know. I hope. I hope the war is. Keeps on being fake and gay and delayed and whatever

A (137:42)

and hope Iran earns a lot of bitcoin.

D (137:45)

Yeah. Yeah. You know, some. Some bitcoin reparations for Ron would be, you know, would be fine. I.

A (137:58)

I don't know.

D (137:59)

I'm. I feel like I have. I. I hope. I hope I can get the hell out of Canada one day.

B (138:05)

I don't know.

A (138:06)

You know, one day. One day.

B (138:08)

One day.

A (138:08)

Mechanic, how you feeling about Canada and what are your final thoughts? Try not to, try not to mention bitcoin.

B (138:14)

Jeff, Are they, Are you in. Are you. Are you a permanent resident in Canada?

A (138:20)

No, I am, I've.

D (138:22)

I've filed an extension which means that I.

B (138:25)

Wait, just stop. Just get out before you're here for more than 180 days and they become interested in you for tax purposes, do not spend more than 180 days a year in Canada. I think it's 182, but yeah, get out. Like aren't you already. If you're like you're coming up on that. If you've been in since the beginning

D (138:42)

of January, I think. No, I've been, I've been in since October. I think I'm already.

B (138:47)

That. That's two different years. So you were there for October, November, December. Now you've been here for January, February March, April, get out before June hits and don't have to deal with the cra. You've already got the IRS up your ass. You don't need another one, trust me. But I'm not like, I'm. Just go to the Bahamas for the rest of the year.

D (139:08)

I'm not working. I don't, I don't. So I don't know. I, I don't have. From what I understand, because I filed an extension. I sort of can't leave. Like, I'm allowed to stay. It's like an assumed, like. Yeah, yeah, it's no problem. You can, you know, you can extend your stay. But if I leave, it invalidates the extension and the extension doesn't get processed for like a whole eight months.

B (139:34)

Why can't you just be there as a tourist right now? No, I am, I am.

D (139:37)

I'm here as a tourist.

B (139:39)

So what's the extension to your tourist visa? Yes, man, you can just go in and come back and they'll give you a fresh six months. You don't need to do all this.

D (139:47)

Well, but if you, if you have. Yes, you could go in and come back. But if you go. If I, if, like, if I go back for like a week and then I come back, they get more and more. They get, they get more and more like, like iffy about. I didn't want to end up in a position where you couldn't come back. Where I couldn't come back. And, and I just don't want, like, I don't want to be in a position. I don't want to leave my wife at the airport anymore. I'm just tired of that.

B (140:19)

So, I mean, you wouldn't be leaving and coming back like, you'd be leaving for seven months to make sure you didn't overstay the 180 days, then you'd come back in January and they're not going to be like, what are you doing? Like, that'll be fine.

D (140:34)

I'm. I don't want to leave my wife for.

B (140:37)

You don't want to leave your wife for tax purposes? No. What kind of libertarian are you?

D (140:47)

No, I, I mean, I don't know. The situation is. Is. But I don't know. I. As far as I can tell, I'm kind of stuck here if I don't want to like, screw up the extension that I filed.

A (141:03)

And.

D (141:04)

But it is just a visitor extension, so. And then at some point, if she get. We're hoping that her permanent residency will actually be processed pretty fast because I'm not included in it. And so while I, us getting married technically slowed things down, it. If I'm not included in the permanent resident application. Permanent residence application, then she's just processed again by herself. And, and so they basically just confirmed that I'm not like some sort of terrorist. And. And then beyond that they just process it with her and then once they

A (141:42)

confirm that your social media.

D (141:44)

Yeah, who knows? But then, but then once they process her permanent residency, she can sponsor me. But yeah, I don't know.

B (141:56)

Wait, so your long term plan is to live together in Canada as permanent residents?

D (142:01)

Nah, I mean, I don't really want to, but for right now that's just kind of the only direction things are going in. If like we would much rather be somewhere warmer. We're, we're looking at like bc parts

B (142:17)

of, you know, come on out, it's beautiful out here. Yeah.

D (142:23)

But, but we'd still like to be somewhere warmer than that. So we're, we're. I don't know, we're looking at options. It's, it's, it's very, it's gotten, because of the ban from the US on 75 different countries or whatever. It's actually gotten even harder for anybody that's Iranian. And so if she has a permanent res, just having permanent residency in Canada opens up a few more doors like we can actually potentially.

A (143:00)

That's why I say you can't go in Canada anywhere yet, right?

D (143:05)

Well, yeah, she, because she applied in Quebec through a Quebec special like skilled worker program or something with her like. I don't know, her PhD is like part of the, part of the equation or something. It was supposed to be, it was supposed to be like really fast. It's like says express, you know, permanent residency application or whatever and, and it's like the slowest, most convoluted.

A (143:35)

Like eight years later. Yeah, it's been 89 years. Yeah.

B (143:40)

And I just had to go to the UK for a passport, man. Fly all the way there and back, otherwise it would have taken 12 weeks.

D (143:47)

It's literally to the point where we considered just scrapping the whole thing and reapplying for permanent residency as in Canada as a whole, because that could be processed potentially faster. But I don't know. We, we were told that since she was already in the position she was in might finish quicker if we just

A (144:13)

try not to reset.

B (144:14)

Yeah.

D (144:15)

So.

A (144:16)

Well, I say good luck, but I know we'll like be up to date anything that actually happens off air, but. Mechanic, what you got, boss? What are your final thoughts. Anything on Quantum and AI?

C (144:34)

Oh, God. I gotta go.

B (144:38)

I got nothing. It was fun. Rent, guys. Fun rip.

A (144:42)

Always. Always. Thanks, guys.

B (144:46)

See you out.

A (144:47)

This is the Roundtable 81. I'll catch you on the next one later.

C (144:51)

All right, see you guys.

A (144:57)

And that'll wrap us up. I hope you enjoyed that episode. Don't forget to check out Simple Steve on Is it Steve uses? No, it's Simple Steve. His handle is now Simple Steve. I'll have the tags and handles and everything in the show. Notes for Noster and Twitter. Follow Bitcoin Mechanic. Stay up to date on that. Follow my brother, Follow myself. Tags, everything will be right there available for you, as well as some handy links for services and affiliate links for supporting the show, things that I use, things that I trust and have enjoyed for a really long time, and ways that I live on a bitcoin standard. Thank you everyone who zaps on Fountain and on Nostr for supporting the show. To the audionauts, you guys make this all worth it and let me keep going and I appreciate it and I'll catch you on the next one. We got a couple of great reads coming up and some really good chat episodes that I'm excited about. I think I'll be recording one tomorrow actually, so stay tuned for those. Don't forget to subscribe. Follow me on all socials and I will catch you on the next one. Until then, guys, that's my two sides. It.