CD95: Bitcoin Miner Incentives with Nick Hansen and BitMEX Research

Happy Bitcoin Monday, freaks.

It's your host, Odell, here for another Zillow Dispatch.

The show focused on actionable Bitcoin and Freedom Tech discussion.

As always, dispatch is funded by our audience. We have no ads or sponsors.

It is only possible thanks to the donations of

thousands of people around the world. So I wanna thank you all who continue to support the show.

The easiest way to support the show is by going to citadel dispatch.com/donate.

You can also donate to the show

via Pay NIMs via BIP 47 payment codes.

My Pay NIM is

Odell. Very easy to remember.

PayNIMs are supported right now by Samura Wallet, Sparrow Wallet, and Stack Wallet, and it looks like Blue Wallet is about to merge support for PayNIMS.

You can also support the show using podcasting 2 point o apps, such as Breezewallet,

Fountain FM,

and echoln.com,

podverse.fm.

There's many different podcasting 2.0 apps. They work like regular podcasting apps. You simply search to dispatch,

load up your wallet with Bitcoin,

choose how many sats per minute you think the show is worth, and then as you listen, those sats flow directly to my note.

Podcasting 2.0

also supports

something called boostograms.

Boostograms allow you to attach a

payment to a comment,

and the top 4 Boostagrams from the previous episode are read before every dispatch.

So with that said, from our last dispatch, we have Eric 99 with 50,000 sats saying stay humble, stack sats. Really great advice, Eric. We have at stimmy.

Matt, at minute at an hour 20 minutes, you described as satoshi's coins haven't moved. You're right. The plural does end in s.

Wrong. The plural of Bitcoin is Bitcoin.

We have atgarth with 20,000 sats. Give me a Shaka sign.

Cheers, Garth. And then we have ride or die freak, ape myth rendur with 14,444

sats saying, Matt, is one more bank collapse away from saying 200 k by conference day? I don't know what you're talking about, but I will say Caps O'Dell is hiding beneath the surface, and I have to remind myself to stay humble every morning

to keep him there. So with all that said, I wanna thank the freaks who continue to support the show. Also, I know

we've had we've been in a bear market. Looks like it might be over. I'm not gonna try and jinx anything.

We're definitely in a recession. Might be going into global economic collapse. So I understand if you can't spare any stats. You can also support the show by subscribing on your favorite platform. It's available on every single podcast app by searching Cielo Dispatch. Also available on pretty much every platform, YouTube, Bitcoin TV, Rumble, Twitch.

Subscribe to that. That really helps. Sharing with your friends and family. That really helps. Leaving reviews. That really helps. And last but not least,

dispatch is very unique in that we have a live

audience, an interactive live chat for the audience coming in through YouTube, our matrix chat, which you can find at sill dispatch.com/chat,

which is 2473

65 even when the show isn't going on, Twitch.

Anyone who participates in the audience,

I really do appreciate you. You guys make the show unique. I consider you host of show along,

with myself, and, I really do

appreciate it. So with all that said, I wanna introduce our wonderful guest.

Today, we'll talk about Bitcoin mining incentives.

We have mempools that are very full.

A lot of activity out there. Bitcoin mining incentives are more relevant than ever.

First,

we have Nick Hanson,

cofounder and CEO of Luxor. How's it going, Nick?

Great. It's a great day to be a miner, as with all days, but it does seem like recently it's been even better.

Love to hear it. Thanks for joining us, Nick. And we have,

Jonathan Beer, who's the man behind BitMEX Research and the author of The Block Size Wars. Great book. Pick it up if you haven't read it yet.

It's a bit of it's a piece of history that a lot of people that haven't been around might not be familiar with, but it's a very important topic.

So you should definitely read it if you if you weren't around during the 2017 block size wars. How's it going, Jonathan?

Very good. Thanks.

Nick, you asked me in the private chat how to,

participate

in,

the text chat, you can do that by going to dispatch.com/chat.

Or you can go to the YouTube feed if you wanna be on the YouTube chat. But as you can see, they're just coming. Or you you can just talk to them like I do and just respond to them. So with all that said,

I don't know where we should start.

I guess an, an an interesting icebreaker

is a topic that I'm very

passionate about, which is the fact that I think

we may never see mempools clear again. Do you guys think we will see mempools clear again?

Go ahead, Jonathan.

Yeah. I guess that's a difficult question. Yeah. They may not clear again. Of course, there is no the mempool. Everyone has their

own mempool. So depending on your local settings, your individual mempool

may clear. But, yeah, if you're running a very deep, large mempool,

you can you're gonna still have 1 sat per virtual byte

transactions that will probably end up clear,

I guess. But I don't know. Yeah. With the with the default with the default node right now, we're purging

anything below,

what, about 4 sat per v byte,

which is probably the first time that's happened may maybe since the last,

since the last run-in 21.

And certainly,

what we're seeing with, you know, with ordinals and inscriptions,

does seem like there is now a buyer of last resort for block space, which we've never really had before.

You know, we've started to see, like, there's a pretty firm floor,

well below,

or, sorry, pretty firm floor floor down there in, like, the low single digit

fee fee rates. So,

it we'll see. I mean, I guess there was a block just now that was a 2 sat per v byte.

But a lot of times when you see that now, especially with how deep the mempool is, those are pools that are, like, prioritizing different transactions for whatever reason. Like, you know, for example, there you know, a lot of a lot of pools mind their own payout,

their own payouts with 0 fee.

So

there will be, you know, sir certainly some low fee rate transactions getting through, but overall, it does seem like we've gotten to a pretty deep floor of,

or I said, I should say a pretty deep mempool.

Yeah. Of course. What one thing we really want is we do not want the mempool to clear. We want Bitcoin to be successful, so we want Bitcoin to be popular,

which means

we want people to submit transactions to the memory pool. We hope there's a lot of demand for that. And if it Yep. It never clears, that's, I think, a very good thing. If it does clear, then, yeah, in the future, we could run into all kinds of security problems. So we want a deep mempool, we want more demand, and it and it's a great thing if the mempool never clears, in my opinion.

Yeah. So, Jonathan, they're like, the one of the points about,

the the mempool clearing. If the mempool clears, that means there's no demand for block space. And right now, that maybe isn't too much of a problem because there's a subsidy of 6.25 BTC

per block. But, you know, in, 4 or 5 hattings, when the block reward is less than 1,

either Bitcoin price has to go up or there needs to be,

more, you know, more transaction fee volume to take up that to take up that delta to maintain the security of the chain.

That is, you know, obviously a point of contention as Bitcoin continues to,

continues to evolve. A lot of folks are concerned that maybe there won't be enough,

subsidy or or incentive for miners to continue continue,

doing the work that they do, and that could lead to a, you know, to a point where Bitcoin,

Bitcoin security model starts to become somewhat shaky. And that is, that's a concern that I share, but, you know, recently we've seen,

you know, a lot of,

yeah, a lot of fee activity. So

One thing, to keep in mind, well, first of all,

Freaks, I know

I know you guys are more sophisticated audience than a lot of other,

shows out there. But just a reminder, the way I like to visualize

mempools, first of all, as Jonathan Jonathan said, there's no one v mempool.

Every node has their own mempool, but the way I like to visualize mempools is almost aligned to enter a restaurant sorted by who's willing to pay the most.

When there is

a,

when there's a wait list to enter the Bitcoin block space,

people have to pay higher transaction fees, and the higher your transaction

fee, the more priority you are given,

by miners to be included in blocks.

When people usually interact with

mempools, they usually interact with Wizz's mempool when they're on mempool.space,

and that's what we have up on the screen right now. And to Nick's point about the block that came in with, 2 sets per byte, one of the cool features that mempool.space

has right now is if you click on an individual block,

it will give you

the expected

block layout that they expected based on

strictly highest fee paid transactions,

versus what

the reality was. What actual block got mined is a pretty cool,

way to kind of visualize, you know, what miners are including in blocks.

So,

guys, for,

thank you for answering that question for me. I mean, I just full disclosure. I have a little bit of pie on my face because when we had a fee pressure in 2021,

I got a little bit of fee FOMO, and I, you know, I said mempools were never gonna clear again, and then we had one sapper bite for a long time. So,

I was a little hesitant hesitant to make this call, but it does feel

like there is, you know, a mix of of buyer of last resort in terms of the d gen,

NFT people.

And then also

just, you know, more demand for Bitcoin block space than ever in terms of Bitcoin as a censorship resistant digital money that's an alternative to,

all this corrupt fiat shenanigans that we that we see right now. So it should be interesting to see play out.

I'm curious. So, Nick,

you're the the founder of Luxor.

Luxor was originally

not a mining pool. Right? The mining pool is only Yeah. Relatively recent. Like, the mining pool is what? Late 2021, like, October 2021 was when you guys launched the mining pool?

So we've been a rep so we've had a BTC mining pool for for quite a while.

We, you know, we started,

not even with mining at all. We were working in 2017 on something called what I called scia 3. So there's an all coin called scia coin, and, it's kinda like a decentralized storage. And we were like, let's see if we can build, like, an s 3 like API on top of this thing, kind of back when, you know,

and storage and,

made safe and all these things were kind of a thing.

It seemed pretty interesting, compelling.

And so we started building this, this, you know, we started building this API that's supposed to mimic s 3. But instead of being backed by servers inside of AWS, it was backed by this decentralized file storing network, which, of course, it didn't work, at the time, and it probably still doesn't work. I haven't really checked in on, where they're at in development now. But,

what it did give us was an ex exposure to an ecosystem that had a very centralized mine a very centralized mining ecosystem

and in that there was 98% of hash rate on a single mining pool, at that time for for Cyclone. So we started, like, digging into how do we build a mining pool for this? Because at that time, it seemed like they were certainly

most interested in, or that that that team was very interested in, like, figuring out how can we democratize mining for this ecosystem. And that was really where we cut our teeth with mining. And the reason that's somewhat interesting is that if Sai is not a fork of Bitcoin code,

almost every altcoin that exists out there today is, like, a code fork of Bitcoin. Even, like, you know, some of the ones you've heard of, Monero, Litecoin,

Dogecoin, those are all,

code forks of Bitcoin. And so there was a bunch of infrastructure for, like, getting those done, but they were,

like, getting mining pools built for those, but this was a completely different ecosystem. So you had to learn, like, from the ground up, like, how do you construct a block?

These things are all, you know,

applicable to, you know, whatever ecosystem you're in. And so we cut our teeth on, like, this new ecosystem.

Obviously, it's not really that important anymore, but, over time,

gathered a bunch of experience on how

to build blocks, propagate blocks quick quickly,

validate shares, all of the things that we need to do to build a Bitcoin mining ecosystem. Spent over the course of a couple of years, gathered that experience, and then, you know, decided to make the leap towards Bitcoin. Of course, making the leap from, you know, some, you know, small cap altcoin to Bitcoin was certainly a tough task because the stakes are much, much higher. You know, you're talking about, you know, hobby mining and and, and, you know, you know, doing something in the nights and weekends, not an enterprise grade mining pool like Luxor is today. So we had to spend a lot of time getting into

and learning about,

those that like, how, you know, the the construction of blocks, like, the mining infrastructure,

and then how do we apply that that knowledge to to Bitcoin. So, that was what we did. And then, yeah, pretty much have ever since just been very much focused on Bitcoin. That's why we built firmware. That's why we built Hashrate Index, which is, you know, one of the leading platforms or data platforms for learning about, mining incentives, mining economics,

basically, the the ASIC market, of course, the mining machine market,

and then, of course, you know, got into even, you know, financial derivatives on top of, on top of Bitcoin. So

that's, that's how we're where we're at, how we cut our teeth.

That's how I know you guys. I know you guys through your mining research. Just to backtrack real quick and not to get bogged down in this, but, Finch beat me to the punch.

Most

most Shitcoins are,

forks of Bitcoin's code, but you mentioned Monero. Monero is actually,

crypto note based.

It's a completely different, code base,

just for what it's worth. But that anyway, to go back to the point,

the fact of the matter is Lexar's mining research has been top notch. I've loved your research.

Also, Jonathan's work at BitMEX Research

has been amazing,

best in class for a long time. So I figured having the 2 of you on,

at the same time to have this

extremely interesting.

I'm curious, Jonathan.

When you look, obviously, it seems like

Luxor is kind of trying

to take the lead in terms of a mining pool that is focused on inscriptions

or that likes inscriptions, likes these ordinal inscriptions. And and for even listening to the conversation I had with Casey

a couple weeks back, you should definitely go and listen to that after this.

I'm curious like, Jonathan, when you when you look at at the new

hype around inscriptions, how do you think about that and how with with regard to how it changes minor incentives and how how minors are interacting with the network?

Well, I think the first thing to say about inscriptions is as Nick has said, is that it's boosted demand for block space and therefore,

boosted minor revenue. And that's

a very good and positive thing. And, I mean, that doesn't change minor incentives that much other than it just being

very positive. I think the only kind

of downside or potential negative is

that some of the inscriptions

are valid transactions, but non standard transactions.

So,

what that means is you can't broadcast

the inscription to the Bitcoin memory pool. I think that's if the the image you're inscribing

is over,

500 KB,

then it becomes a nonstandard transaction.

So in Bitcoin, you have this kind of thing where you've got policy rules

and standardness rules. So if it violates policy, the block is invalid.

And if it violates standardness, it just means your node won't relay it. If you send it directly to the miner,

the miner can still include that transaction in the block. So when Luxor made the 4 megabyte,

inscription,

that was,

a valid transaction, but nonstandard. So there are some kind of incentive issues there because it means that,

you don't broadcast it to the members. So it's not kind of this open

competition

with all miners trying to include it. And instead, the miner has to kind of communicate out of band with the person who's doing the inscription.

So, yeah, there's a potential kind of weakness there. And, I mean,

this has sparked some debate inside the kind of technical technical community kind of who are reevaluating

the whole concept of why do we have

standardness

and policy rules, why are they different.

Maybe this example illustrates we shouldn't really have that distinction,

to the extent we do and that

any valid transaction should also be a standard transaction. Don't know what your thoughts on that issue are, Nick.

Yeah. I mean, that was one of the very first questions that was raised around the,

you know, around the big block or the big wizard as people call it.

Was that, hey. What well, why why wouldn't my node have picked up that block? And as, you know, that transaction, then that's a really good point is that you're it's a valid transaction. It's just that it is a DOS,

denial of service attack vector. And so Bitcoin Core has introduced these policy filters so that way you're not picking up you know, you know, somebody could just spam 4 megabyte transactions.

Your

your mempool would get filled with them, and then regular transactions may be getting pushed out.

So that could be that could be you know, that's an example of,

why, like, policy and consensus don't actually add up.

You know, Bitcoin and a couple of other

I think Bitcoin is really the only one, but they don't have those policy filters. And so that's an interesting point,

because

those yep. Because those policy filters do make it a little bit weird,

in that you do have to do that out of band,

type of interaction to get your your transaction line.

I can see both sides. I I don't have a very strong opinion on whether policy show filters should be adjusted or if that should be, like, a person

making a very

specific and targeted,

change to their core node to accept those types of, those types of transactions.

It does

potentially introduce, like, some like like you said, some weird,

incentives where, you know, these types of payments have to be,

paid out out of band.

And and that's, you know, that is maybe not not not not ideal because then, you know, standard mempool mechanics don't don't apply.

I mean, to be clear, this is not the first instance

of out of band payments,

for transactions. In fact,

if we start to see mempools being more full is when exactly

when when we see this wait list start to expand is when we have historically seen out of band payments. I mean, I remember even as early as 2017,

we saw people who maybe paid too low of a fee. And then as a result, they wanted to get their transaction confirmed, and so then they would go directly to a mining pool,

and then pay a miner directly to get their to get their transaction included. Right?

Yeah. I think in 2017,

via BTC launched a service where you could go to their website, you give your transaction ID, and if it wasn't confirmed, you could pay them via credit cards and

it was, minor transaction.

So these outbound things have been going on for a while, but, yeah, they are a a potential kind of weakness for Bitcoin because, of course, it encourages

mining centralization because via BTC, because of their

their service where they took credit card payments could earn higher revenue,

then more miners could choose that pool, and then that could encourage centralization because they have that infrastructure in place. And the same thing with the out of band ordinal inscriptions.

If everyone who's using who's doing inscriptions always contact the same mining pool out of band, that mining pool could be more profitable, more successful,

and that's, centralization

risk. I don't think it's a major risk for Bitcoin, but, yeah, that's certainly

the the main disadvantage of these out of band,

structures.

Yeah. That's, that's effectively MEV. That was kind of what was happening on ETH when, you know, when ETH started going through, you know, when when DeFi started getting really

popular and folks were doing a lot of,

you know, degen stuff with these contracts, they were willing to pay, higher you know, willing to pay, you know, out of band fees or whatever and also needed, like, a private mempool.

And so that was, like, one of the re one of the things we started to see pop up in flash bots, if anybody's familiar with flash bots, was,

well, that was one of the one of the repercussions

or the consequences of this occurring.

As you mentioned, you know, those those are called transaction fee accelerators. It's not just VibeVTC that has them. There's a lot of those that have existed throughout the years, and, you know, they continue to, you know, they continue to operate maybe not as much as they used to because RBF is pretty prolific now. RBF being replaced by fees, so you can,

replace a transaction that exists in the mempool

and, bump the fee. So it's not as ubiquitous as it used to be, but it definitely was something that was out there and existed for a long time. So out of end payments are not new. They've definitely been around for a long time. And then there's also, like, a subset you I've heard of of of more, like,

more,

you know, folks that are trying to do, like, private transactions where they issue a transaction directly to a particular mining pool that don't wanna broadcast all over the network. That's another type of, you know, reason there would be out of band payments or or out of end transactions that are out of the next standard mempool flow.

But those are much more rare and not as common.

And I think that, I think the real the real,

like, the real delta here or the real piece that people are missing is the,

like, is that out of out of band payment for,

prioritizing different transactions as well as, you know, whether they're for acceleration or for,

you know, getting around the policy filters.

Yeah. I mean,

first of all, guys, I'm not your typical host. If you wanna respond to each other, feel free to respond directly to each other. Don't feel like I have to,

prompt a question every time.

I probably should have told you guys that before I started the show, but also I'm not a typical host, so I didn't tell you that ahead of time.

Yeah. So, I mean, I'm curious. Like,

let's let's dive into this a little bit because

you mentioned this term minor extracted value.

People that are, you know, in ethland already know what this term is,

but maybe people that are focused on Bitcoin only aren't really familiar with it. Jonathan, you wanna, like, just give us a little primer on on what MEV

refers to?

Yeah. Absolutely. I think MEV can be quite a big complex topic.

So the first time I thought about MEV was actually

way back in 2014, and I remember this because at the time,

you could bet on the, FIFA World Cup, which was in Brazil,

on the Bitcoin blockchain using counterparty.

So I remember I think the World Cup final was,

Argentina versus Germany.

And I I placed the bets

on Argentina, which meant I broadcast my transactions to the memory pool, which was

the counterparty interpreted as a bet on Argentina winning with

a certain odds. And then I'm thinking to myself, well, hang on a minute.

What happens if Germany score? Because they could score at any time the match was ongoing. So they could score within the 10 minutes

before the Bitcoin block was mined, before my bet had reached the blockchain.

If that happens,

what I would wanna do is, of course, withdraw my bets. But then I thought, well, hang on. What if the miner is who mines a block is also

watching a football game,

and they're also

looking at these transaction in the mempool?

They could match my bet because Germany's just scored the goal. Germany's more likely to win. If the miner is doing all of this, watching the football match, and looking at the mempool and having sophisticated

systems,

they could stop me canceling my bet,

and match against me, bet on Germany, and then they would, you know, be an advantage.

And, you know, NEV didn't exist at the time in 2014.

The terminology didn't anyway, but that's that's set kind of what it was.

And I think that's a good way of looking at it. It's like miners

analyzing the markets,

as well as mining

see interpreting the meaning of those transactions

if it's not a standard thing, if it's not someone paying from a to b, but they're doing something more complicated

like betting,

financial trading, swapping tokens.

Yeah. And and the miner can extract value. Now

the MEV has become very, prominent on Ethereum,

and almost all of that is arbitrage. It's occurring normally on Uniswap. So when people, like, kinda swap one token for another.

And

the

the block producers or miners were not only were they producing blocks, some of them were also analyzing Uniswap, looking at live kind of price feeds and all the exchanges,

seeing if a price moves, and then front running

someone who submitted an order.

And, of course,

the consequences of this are very negative. MEV is very bad for the user experience. It's great for the user experience.

For me betting on the football match, it meant, you know, I got inferior execution.

And for people trading on Uniswap,

they get worse execution

because

of MED.

So that's one of the kind of reasons that MED is a bad thing. And when you construct

your, you know, contracts or systems, you should try to reduce the opportunity

for people to conduct MEV such that your users have a better experience.

So while MEV in and of itself is bad,

you could also look at the high value of MEV on Ethereum, which I think is about almost $1,000,000,000 now. And that is a positive signal about the success of Ethereum

because Wait. That's a $1,000,000,000

in in what time period?

I think since since Ethereum's threat all of Ethereum's history,

I think. Got it.

So

you can look at that that number. Do you say, well, that's an indication that Ethereum is very successful. Right? Because

it shows that Ethereum is a very rich environment with a lot of trading activity. There's a lot of value, and therefore, there's a lot of MUD to extract.

So that's why at least some kind of confusions with people saying is MEV good or is MEV bad. I would argue that MEV in and of itself is a bad thing, but it's a positive indicator,

about the success of a platform.

So then what happened with MEV is,

MEV kind of quickly rose to prominence in Ethereum in kind

of November 2020,

kind of just after the so called summer of DeFi.

And as a result, this caused a big crisis in Ethereum because what was happening was

was that the

the the miners

were

searching the Ethereum blockchain for opportunities.

Once one miner found it,

broadcast that transaction,

The other one saw that, and they copied with a higher fee. So then there's this massive

kind of fee war with everyone trying to conduct MEV and increase the fees

to kind of in a kind of scorched at war against the other people doing MEV.

And this caused a big crisis in Ethereum where fees kind of massively spiked at the start of 2021.

And I think Nick mentioned Flashbots, and Flashbots

was the a solution to this or something that mitigated this problem. Because what Flashbots did was

create a kind of centralized MEV infrastructure

where

the searchers

could search for these MEV opportunities, submit it to the FlashBot server. So that and then they weren't then competing

against other

other

searches in in a fee war. So this kind of resolved that fee crisis.

So, yeah, that that's kind of what happened. And then the kind of negatives of that is people then argued, well, Flashbots itself is centralized. So whilst you've

solved or mitigated the problem,

you have, you know, caused centralization.

And then so there's often a kind of argument between Bitcoin and Ethereum, whereas Bitcoin is will point to the MEV and saying,

look at Ethereum. They've got a huge problem. They've got a huge MEV, and then their solution to this was centralization.

Whereas in defense of Ethereum, they would say,

well, you first of all, the fact that we have all this MEV is indication of success,

and we are kind of working our way through solving this problem with these highly complicated

systems,

like Flashbots. So we you know, the problems come and we're trying to solve it, whereas Bitcoin hasn't got reached this problem yet.

So, yeah, I don't know if if,

Nick, you wanna come in and talk about NEV? Because, yeah, it's quite a big tool.

I I love NEV. So now that you've invoked ETH, maybe I'm a little less nervous because I was, like, trying to dance around NEV and maybe my, like, altcoin history because I know the Citadel Hornets are gonna come after this stuff. But, now that you've invoked invoked ETH, like, we can talk about MEV pretty extensively. Like, MEV a is,

like you said, terrible for user experience, but phenomenal for miners.

And so it's like this weird competing,

sort of,

dynamic,

which is really weird,

but also super interesting and makes mining really, really interesting. Like, mining on ETH was crazy interesting because

of these mechanics. Like, you effectively had, like, high frequency traders, like folks that are able to build very sophisticated trading strategies coming in, issuing these transactions,

and and trying to, you know, affect you know, more or less trade against each other in real time. It's pretty crazy.

And and it was wild to see, like, that stuff play out in the mempool. It's like the most competitive thing you've seen. And then, you know, there were pools that were building either on flash bots, which,

you know, the ETH community somewhat rallied around is that's like, okay. That's the okay way to do MEB. Like, people are gonna do MEB. Like, let's figure out, like, a standardized protocol for this and, you know, try to push pools that direction. But there were pools that were building, like, their own MEV implementations,

and they were out competing flash bots in various ways, which was, you know, really interesting to see because now, like, okay. As a miner, I have to make a decision.

Do I go

mine the the more profitable pool, or do I go to another pool,

that's doing, like, what the foundation like, the foundation and, like, what they say is, like, the proper way to mine? Should I go mine there and make less and be, you know, like, virtue signal? Or do I go be purely profit driven and go mine at the pool that has a better NEV strategy?

The same thing is somewhat occurred has somewhat occurred in Bitcoin back in you know, when we mined the big block, we mined the big wizard, which they do call it the big wizard, by the way, doc. Dot Guru says, I dare say no one calls it the big wizard, but they do call it the big wizard.

And at least that's what the Taproot Wizards call it. I don't know who else you talk to, but,

that is, you know,

right after that, a lot of folks told like, that was when, the org has been affected.

To be clear, because

most of my audience is not familiar with the big wizard. This was a 4 meg

NFT inscription

of a the famous wizard, you know, the magic Internet money wizard of Bitcoin. And you guys mined it out of band, and the whole the whole block, the entire block was this single inscription.

Continue. Yes. Exactly. Okay. Thank you. Yeah. So apologies for,

the the assumption there. But, yes, that's what it exactly what happened and had to be out of band. And so, like, Udi Udi Wertheimer and Eric Wall came and were like, hey. We kinda wanna do this thing. This was back around in description number, like, 400 or so.

For context, we're at, like, 550, 550,000

now. Like, that's a you know, it's only been 4 weeks, but it's a long time ago.

It's

anyway, right after that happened is when, you know, this started this ordinal thing really started to kick off. And,

you know, Luke Dash junior put out his or disrespector patch, and we started getting pressure from folks. Like, I was on spaces talking about this stuff, like how interesting, you know, this is gonna make mining and how this is gonna change the dynamic, etcetera.

That was, you know,

that was starting you know, that was really, like, the kickoff.

And so we were on these spaces talking about this, and somebody was like, hey. You you know, you're a mining pool.

NFTs on Bitcoin are anti Bitcoin. You're attacking the network.

This is against the, you know, this is against the ethos of Bitcoin. You should be running the patch. And I said, okay. Let me let's let's, like,

let's

straw man this. So let's let's let's chat this through.

I'm out here talking about this. None of the other mining companies have really been at all public about what their strategy is with regard to orbitals.

Let's talk about this. Okay. I, you know, I virtue signal. I I put in the patch. I filter out inscriptions from my mempool.

Now my mining pool is less profitable than all the other mining pools, and that's because I'm I'm virtue signaling that that inscriptions are bad.

So

now all the miners that are on my pool, I'm doing them a disservice by not giving them the most profitable transactions to mine and thus maximizing their reward, which is what mining pools are here to do. Like, we're here to service our miners,

and that's, you know, what we were doing. And so if, you know, I I I put that in, and and the argument then was, well, you should do it because it's what's best for Bitcoin. And if you have a a lot you know, a short time preference or sorry, a long time preference,

then you should be, you know, you should be running this because it's what's best for Bitcoin.

And I said, well, what if if I turn this on now,

I won't get to long time preference because my pool will be dead because nobody will mine in my pool because nobody wants to make less. That's just the that's just the way miners operate.

So that's kind of where

this like, all all this, like, NEB stuff started to come to Bitcoin. And the reason we you know, Jonathan, the reason you chatted about MEB and why it was important to ETH is because we're starting to see that in Bitcoin, and it's really interesting. And it kind of started around the time of the big wizard,

which is around inscription 400, maybe 5 weeks ago.

And and so, yeah, this has been super like, really dynamic ever since.

Yeah. I I mean, I don't necessarily think these inscriptions or out of bound fees should necessarily be classified as NEV or at least a different kind of NEV than kind of the arbitraging.

Yeah. It has a lot of similar characteristics in that, you know, there's the the risk that individual miners could become more profitable

and gain market share and it could cause mining centralization. But I I I was thinking about it, and I don't think inscriptions, ordinals, or NFTs

are that conducive

to MEV compared to,

say, Uniswap or

or AVE or something? Because

that's you can't really front run an NFT because they're they're nonfungible.

So

if the if the market changes and, like, the price of the NFT suddenly will rally 20%,

you can't really front run that because you can't, like, buy 1 and then sell it back

to the bidder quickly, by front running and and ensure the miner because, you know, you're there's a kind of an individual market for each NFT. So

in that kind of the classic

or the more, like, kind of concise way of defining MEV,

I don't think it's that applicable to, inscriptions. Although, you know, it could in certain circumstances if you're doing

a kind of new mint of an inscription in certain conditions

that could lead to a small amount of MEV. But I I mean, I could be wrong, but I think it's very unlikely.

However successful inscriptions are, I don't think we gonna have kind of 100 of 1,000,000 of dollars of MEV on Bitcoin. I think just because

the dynamics or structure

NFT

markets are not as conducive to MEV as, say,

of

of Uniswap or something like that.

Yeah. No. That's certainly true. That's certainly true. We could start, you know, we could start to see maybe, like,

maybe a world where, like, PSBTs

become, you know and there's, like, a a mining pool that puts out, like, a centralized PSBT

marketplace, and then they're they're the ones that are, like, able to extract the value from those PSVTs

that then, you know, they create and craft. But, you know, that would be I I I don't know. That that's a pretty far fetched,

sort of scenario. So, yeah, in your like, in the definite like, what we what we know as MEV from, you know, our, like, from our ETH,

from the ETH land is definitely not what's happening in Bitcoin, but it is, you know, somewhat

you know, it it just all comes depth down to, like, mining economics. And, you know, we we need to maximize, you know, our mining reward for our miners.

And, and the way that you do that is by capturing the highest value transactions. And if you're trying to filter out the highest value, you know, some of some transactions because they don't fit into your ethos of what Bitcoin is, you're then, you know,

more or less virtue signaling that, you

know, and not actually,

how do I say, not actually maximizing profit. So

yeah. I I certainly I mean, I I definitely agree with you there that, you know, we should if if a transaction is valid, it doesn't matter what it is, what the ethos is. The sensible mining policy is to try to maximize

income.

And we should, you know, develop codes that is incentive compatible or consistent with miners trying to make the most profits because that's ultimately how the miners might behave,

in many scenarios anyway. So we might as well build build code that enables that. And that's the same kind of argument that was in favor of, replaced by fee. Right?

The argument was is that

replaced by fee

even though because, I mean, when replaced by fee was introduced, it was very controversial. I think this debate happens in around 2015.

There was this old policy called first seen safe. So the first transaction,

the miner saw by default was the one they included in the block.

Whereas if they saw a conflicting transaction, even if it had a higher fee, they would not put that in their block. But that's not incentive compatible because to maximize earnings,

you should switch to the conflicting transaction with the highest

fee rate that gives you the most profit. So that's one of the main reasons, you know, RBF has become

more popular. So it's a similar argument there that is that,

even if it damages the user experience, we have to accept that's the reality of how miners are gonna behave.

And therefore,

that's what we should we should allow miners or encourage miners to do whatever increases the

maximum revenue in the short term by constructing a block to maximize their earnings.

Yep. Yeah. Unfortunately, you know, miners are, you know, fortunately or unfortunately actually, I would say it is fortunately because that is the way the Bitcoin, you know, that is the way the Bitcoin consensus mechanism works is that miners are short term focused. They're focused on the very next block, which is, you know, 10 every 10 minutes.

They don't think they in they they don't have to think in that super long term. As soon as you become you know, as soon as you start thinking long term, you're more thinking with a holder strategy, which is different than a mining strategy. And those are 2 I don't wanna say different. They're definitely parallel, but they are you know, they're certainly not intersecting,

as much as, you know, say,

trying to think of a of a good analog here.

Like, oil producers,

when they produce, you know, they they produce oil, they immediately need to they need to capitalize and maximize the value of the oil that they receive right now because that's their you know, that's what they do is produce oil.

Now oil consumers, which would be, in this case, like, the I think maybe the holders, maybe this this analog is not gonna not gonna do very well. But,

the idea is that the producers need to be maximizing profit right now because you don't know how long that profit's going to exist.

You know, oil oil producers, maybe maybe a good maybe a better example would be, like, on a depletion curve, if you're, you know, if you're pulling out oil too quickly,

you know, it makes your, like, your long, you know, long term, your your your oil well depletes more quickly,

if you're pulling it out right now. But you have to really think about what is your what is your margin right now. Build a business on how things are based right now, and you can't really think about, well, you know, if I pull out all this oil super quick,

maybe my depletion curve is, you know, is accelerated or something into,

you know, 3 or 4 years down the road. Most most, you know, oil wells, you can't think that way. Same way with, with miners, you can't really think that way either. You have to think about what are the mining economics right now? How do I build my business and maximize my profit right now?

Yeah. I think I think you're right. And I think that's very interesting because I was making similar analogies back in 2015

with gold mining and oil oil producing

in the kind of 13 safe versus RBF debate.

Yeah. And I think

a a good way of thinking about it is that, you know, the oil price be volatile, the gold price can be volatile.

If the oil price goes down, the the profit margins that they all produces can be very slim. And same with the Bitcoin miners. Right? The if the Bitcoin price falls,

their profit margins can be very slim. It's a very competitive industry.

They may have debts. They may be struggling

to survive. They may be struggling to repay that debt, and they Yeah. Are going to need to maximize their short term

earnings. They don't have the luxury of thinking, well, this theoretically may damage Bitcoin in 10, 20 years time,

in some scenarios, and therefore, they will wanna construct the block to maximize their revenue.

And, yeah, I think it's very interesting to kind of compare it to kind of oil and gold and especially the super cycle that those commodities

can experience. Yeah.

And Bitcoin can have that same kind of long term cycle and then

which could mean that, you know, Bitcoin the Bitcoin mining industry is not always gonna be healthy and highly cash generative and all the companies have a strong balance sheets.

And, you know, we've been through in the past,

I don't know, a year or 6 months. You we've seen a lot of bankruptcies, a lot companies fail, a lot of companies

with a lot of debt, and I think it's ridiculous to ask them not to maximize the revenue in their blocks.

Yep.

I agree. Yeah. So, you you mentioned super cycles.

I realized also another another point here that's completely as an aside. I'm at a huge disadvantage to somebody with a British accent. Like, you just I just naturally sound dumber. But, but, you know, the going back to your super cycle point,

I do think that, like, Bitcoin has super cycles the same way that oil has super cycles. Really good analog to and they're they're they're different time frames. Right? Because Bitcoin's only,

12 or whatever. Would be for 14 years old,

where oil has been being produced for, you know, over a 100 years. Gold's been mined for 1000 of years. So those markets are much more established, and so their super cycles are probably quite a lot longer.

And also macro environment changes significantly. But we can see, you know, basically the same thing that happened to miners in 2022

happened to oil producers in the early 2010s.

You know, there was a huge run up in oil in, in oil price post, global global financial crisis, which caused 1,000,000,000, if not 1,000,000,000 of dollars in investment into new oil,

you know, into new oil production,

both domestic and abroad.

And over time, like, we saw that the price, you know, has basically can has just cratered ever since that point. And so a lot of oil producers have been going out of business. It's it's been a much slower,

a much slower process than Bitcoin, but we saw that, you know, that exact exact same super cycle play out in in oil,

you know, in early 20 tens and is now somewhat

it looked like it was gonna be somewhat revived maybe last year, but then, of course, global macro definitely shifted against them as well.

You know, I mean, the I guess the absolute bottom for oil was, was actually negative pricing, which hopefully I don't think we'll ever get to negative pricing in Bitcoin. But,

you know, it's it was a pretty interest you know, pretty interesting to see. You can definitely tie these 2 industries together in a lot of different ways.

I mean, I wanna jump back to minor extracted value real quick Yeah. Because it is pretty important topic. So, like, the way I look at it is

at its core, it's it's profit maximizing behavior,

of miners trying to to extract as much profit as possible. Now this is what we expect from Bitcoin miners historically. One of the cool parts about Bitcoin in general is the way the incentives are aligned.

You don't act you don't need people to act in a benevolent way. They act greedy,

and as a result of their greed and their profit maximization,

Bitcoin is more secure as a result.

Now when it comes to minor extracted value, though, the argument against it would be

and is

that we wanna keep this protocol as simple as possible so it's predictable and stable.

And minor extracted value is based on this premise of almost

an external reason.

It's it's a it's a it's a minor

using external data

that that will create profitability for them that is separate from the transaction fee that that people's men will

see, that the gossip player sees, and that it that is separate from, you know, the minor subsidy that everyone knows. This is an

the

the quite large NFT you had.

But recently, there was I I try and not pay attention to this stuff, but, ultimately,

I rely on Bitcoin on a day to day basis, and it's hard to ignore.

There was some kind

of NFT sale or something where, like, people needed to get into a certain block. Right? And if you if you weren't aware of that NFT sale,

I think it was done by the the Bored Apes guys, like Yuga Labs, but I'm not sure because

I wasn't exactly paying attention to it. But if you that you had to get into a specific block. Am I correct? And, like, as a result, there was

some people I think you were alleging on Twitter that there were some shenanigans around this block and, like, there was

can we go into that? Yeah. Yeah. Yeah. So so just for, just so we're all on the same page, you YUGA's this wasn't related to YUGA. This was DGODS. DGODS is the number 1 at NFT on Solana.

They had a bunch of n f t like, NFTs from Solana that they burned. I don't know how that works, but they burned them, meaning, I guess, they can no longer be spent. And so the war was that they're gonna bring them back to life on Bitcoin, which they did

in a in a what we call a sequential mint. That was another bespoke block that Luxor produced, basically taking all of those inscriptions, putting them into a single block so that when they were mined, they all had sequential mint IDs, and it was very, like, aesthetic for the up for the art people.

So that's what happened. And then what what what that's and that happened long ago. That happened, like, about long ago, and borderland is, like, 2 weeks ago. But, feels like 3 years ago.

The

what what what you're talking about is the auction process. So they,

Friday afternoon,

or late evening,

they had a there was a a block. They said at block height x,

we're going to open the auction. And so x plus 1, you have to get your transaction into x plus 1, or it'll be the first 500 transactions

that,

that complete the auction.

And, you know, the the the thesis was that most likely there will be you know, the entire sale will happen within 1 block, which was the case. And so that one block ended up having, you know, 3 and a half Bitcoin worth of added fees to it as a result of this kind of war,

or race to get in. And so that's kind of the precedent. Jonathan, anything I missed there you wanna add?

No. I've I mean yeah. I wasn't that much aware of that. But, yeah, I think the

these are all kind of, yeah, an any of you type it's difficult to know whether that like, meet the definition of MEV or not, but they're MEV type

activities and or outbound transactions or a race to kind of mint your NFT within a certain,

time horizon. But I I wasn't following the that particular

thing probably to the same degree you were.

Yeah. So that was All I know is if you go if you go on mempool.space,

there's one block that has an insane amount of fees in it, and then the blocks on both sides of it do not.

Right. Which is in in in his

where before we had these inscriptions or whatnot, you would never see that. That it makes no logical sense if you're not paying attention to the external data that is this there's an NFT inscription or whatever going on. Mint.

Auction. Auction. Yeah. Auction. Auction. Like, you know, I it could've been, you know, I could've been selling off 500

Charizard cards that I ship you in the mail unrelated to anything, you know, with relate related to the ordinal or the inscription that had been done long prior. They were sitting in a cold wallet.

The the what was actually happening was the auction process. So you needed to be,

one of the first 500

transactions

to complete.

And so that's why folks were racing to get their, their transaction in there. So that's that's what that was the mechanic that caused such a run. Okay. So I see that's yeah. I would say that that is,

yeah,

a bad auction process if you're saying it's one of the first 500 transactions on chain because you're encouraging

MEV, which is

as which is bad for users, but pretty advantages block producers, tools, and miners. So if you're designing the process, you wanna benefit users. So, yeah, I I would say that that's not good to choose.

So the, it wasn't actually so there's a little bit of a little bit of a, a nuance there, which is and so this is, I mean, this is pretty common in, in in NFT land. I I didn't I didn't really know this, but, like, the way they do Mints over there is is it they want I think they kinda want hype and the chaos. They need, like, you know, what what would you call it? Like, you need an an impetus to act, and so there needs to be, like, a compressed time like this to happen.

Yes. Yeah. Exactly. And so that is, like, the reason you put, like, such a a strict, like, timeline on things,

is it's very, very purposeful.

And so the mechanic here actually one one new one of the mechanic here was that the it was actually based on mempool time.

So it was when your transaction was seen in the mempool first. And, of course, that was,

you know, there there's there can be, like, a second or 2 of of latency around the world. But in general, like, it didn't really matter what fee you put,

and we you know, the the team tried to portray that message or or convey that message,

that you don't you're not gonna need to put, you know, a 1,000,000 sat per v byte.

Probably only gonna need about a 100, maybe a 150 to be included. But really what what it broke boiled down to was people don't understand how TXVs work in Bitcoin.

In ETH, you would you know, the higher your fee, the more priority you get in the block, and so you end up getting ordered higher. And that results in, you know, that results in you maybe getting the MIM where somebody else doesn't. In this case, really all that mattered was the time that your, your transaction showed up in the mempool. Okay. So that's pretty much what's

that?

Isn't that gameable?

Probably I mean, all of that. Mempool are we going by? Or they're going by d god's mempool? They're going by their own mempool?

The the node. Yep. The d god's node, which, of course, they did I don't think they published, like, the the node ID or anything like that. So it wasn't, like, people could, like, peer up with it.

So that that process seems fine then. Right? Because that's not that's not necessarily causing an EV. Just broadcast your transaction, and whenever they receive it first, they'll give you an award.

Yeah. Yeah. Certainly. You're not, you're not

it wasn't it it was clear to anybody that was paying attention that this was gonna be a large fee event because

for whatever reason, like, the message of not needing to have, you know, a 1,000,000 SAP per vbyte didn't resonate because there were multiple transactions that had, like, a 100,000

plus SAP per v byte in them and paid, you know, 1,000 of dollars of fees.

Yeah. But, you know, it's it's yeah. Like, looking here,

yeah, there's a couple that paid, like, over $1,000 in fee for a for a Wow. For, yeah. Here's one that was 10,000,000,

10,000,000 fee rate. Oh, I'm sorry. 40,000. But it means 1,000 SNAP per v byte.

I mean, but that might not it just depends on on how people wanna set up these mints. Right? Like, we could see ones that are based on

on confirmations

or not based on are based on mempool time, I guess, if someone's mempool, who knows who's mempool.

In the future, we could see all sorts of different

schemes here.

So, I mean, I don't really wanna get hung up on, like,

the semantics of what is AV and what isn't ME. Yeah.

If said Jonathan's book, The Block Size War,

one of the main the the two main reasons the block size war was fought,

at least from the winning side perspective,

was to prevent node centralization and mining centralization. And if you increase bandwidth requirements

by increasing block size,

you end up with more node centralization and you end up with more mining centralization because the blocks are larger on the node size,

bandwidth costs go up and accessibility,

goes down. So it's it's harder for people to run nodes. It's more expensive for people to run nodes. And on the mining side,

there's more latency in terms of block propagation,

which helps

centralized miners and results in more mining

or 2 pools that are in Kahoots or whatever, get more and more, or or 2 pools that are in Kahoots or whatever, get more and more,

share of the hash rate. Now when I look at something like this,

I see a lot of concerns in terms of

incentivizing

minor centralization

because regardless of how this mint is set up, if there is one block that has way more fees than all the blocks around it,

all of a sudden and it there were some allegations that

shenanigans happened here and that there were malicious minors. It looks like after the fact there wasn't, but that doesn't mean there won't be in the future.

It it enters all of a sudden, it adds this stronger incentive for selfish mining where miners are withholding blocks and releasing the blocks all at the same time. It incentivizes

reorgs, which which encourage minor centralization.

I'm curious, like, what you guys think here if it seems to me that it it is obvious that

this kind of situation

will lead to more centralization

among miners and mining pools.

Well, yeah. I I think MEV

can be very centralized

had a very big centralization pressure, and that's very negative. And you said you said it on ETH. Right? Because you had to be basically a trading shop and you had to have, like, 100 of 1,000,000 of dollars of capital to exploit the arbitrage opportunities.

And that was a, like, a major problem. I don't think

we're anywhere near like that on Bitcoin. And I I don't necessarily think,

like, an auction for, like, an inscription mint or whatever,

should be classified as MEV or has any of those kind of incentive problems.

I mean, it it could just be they're bidding loads of fees,

and that isn't a unique problem. Just yeah. Yes. We have the real risk if there's high fees, but that's no different to any other kind of high fee environment. Right?

If every block had, like, 200 Bitcoin in fees,

whatever was causing that, there's there's an incentive to

to Rio could try to get those fees. It it doesn't necessarily make a difference whether that's

whatever that is. If it's, you know, people in emerging markets spending £1,000,000 across the world,

or if it's people bidding for

ordinals or whatever these ordinal auctions were. I think it can it has the same kind of dynamics. So I wouldn't classify that as an MED related problem. It's just a normal problem of of high fees, I

guess. Yeah. And so, you know, you you make a really good point. So one one thing I wanna talk about just before we get there is, like, when you talk about the block size so I don't know if anybody knows this, but, ESV is, like, 7 terabytes. Their blockchain is, like, 7 terabytes because it's been just filled with the, like, random shit.

So a,

like, I don't even it's like you wanna go figure out how to get a 7 terabyte plus drive to sync it, BSB node. Like, come on. That's, like, insane.

And their mining their mining is in insanely centralized there. There's 2 pools.

One of them is trying to be, like, the, like, good BSP pool that mines all of their,

you know, mines their super huge blocks, which are, like, multiple gigabytes.

But that pool is massively unprofitable compared to this competing pool, which just mines empty blocks because they don't wanna take all of the they don't wanna take all of the time to, like, construct that block template, which has 2 gigabytes worth of random shit, transactions in it. So this is, like, on the extreme end of what could be happening in Bitcoin if we allowed big blocks to go through. So I'm very happy that we have this very concise, very precious block space that is not, you know I think they they I think their block size is 2 gigabytes. So imagine 2 gigabytes being added to your node every, every 10 minutes. I think it's 4 gigabyte 4 gigabyte block size limit on BSP. Four gigabytes. Okay.

2 to 4 I've tried to sync a node, so I I know very well about the there is at one point, they had, like, 24 gigabyte blocks in a row.

It's crazy. Yep.

It's crazy. And so, anyway, you see, like, what happens to decentralization there. There's 2 there's there's only 2 pools, which have, like, very little hash rate, and one of them is, like, they call it attacking the network because it's, it's basically

mining empty blocks, and the other one is, like, mining all of the transactions, but it's massively unprofitable because it can't keep up. So those are the things that you have to look forward to if you increase block size.

So let's not do that.

Back to the point that you were making about NED. I don't really think yeah. You're right. This isn't in, like, the traditional sense NED. It's not on chain activity that's been trying to be extracted

via strategies,

just like you mentioned. It's not really

trying to, like, reconstruct the mempool or or trying to construct the mempool in such a way that your transactions,

are are you're able to, you know, sandwich folks or things like that. So, like,

in in

in ETH, the transactions are executed in order. So each of, like, the top transaction gets executed,

that sort of thing. So if you wanted to sandwich somebody's order as a mining pool, you could see their order and then put your transactions on either side of them to either add or remove liquidity, let their transaction occur, and then add or remove liquidity at the end of it. So that's, like, classic MEV, which is what Jonathan's mentioning. What we're seeing what I'm seeing here is just, like, dynamism in the the mempool,

which certainly is something we need to sort out because this is coming. Like, in a couple of havings, what happened on Friday

is gonna become the normal where

transaction fees across multiple blocks will vary by 100 of percent. You know, the the transaction fee you know, let's imagine,

you know, let's imagine 4 havings into the future where you're trying you know, Bitcoin subsidy is maybe less, you know, is less than 1 one Bitcoin,

and we're seeing, you know, 4 or 5 BTC worth of of fee,

The you know, in a per block.

Maybe we have a long block that's an hour long, and the mempool gets saturated with super high transaction fee, and then the next block is very low.

The dynamism between those two blocks profitability from a mining pool perspective is gonna be incredibly high. Meaning, the the the long block will have a lot of transaction fee and will be very, very valuable, and the block that comes right after it's very slow, will be very low value.

And so in that way, we need to figure out, like, does the consensus mechanism work in that type of environment? And I think that, you know, Jonathan and I were it sounds like we're just about to get to the point where we start talking about what happened on Friday and how that, breaks down for,

for Bitcoin in the future. I think this is a snapshot of what we get to look forward to in, you know, say,

3 or 4 havings.

Yeah. Exactly. I I'd agree to you. The problem of let the big high fee on blocks is is what we're gonna face when when the block reward comes down. And I think

I I wouldn't it's it's not necessarily ordinals or NFTs creating that problem.

It's it's just a problem we're gonna face. And I think there's variety of things we can do to mitigate that.

The main thing is, of course, we want Bitcoin to be more successful. We want a vibrant usage of the blockchain,

all different kinds of uses.

Ordinals,

transactions,

ecommerce,

savings, like derivatives.

And if if there's all these multiple uses, then that will kind of smooth out,

fee fee revenue more. And another important thing, which is, you know, was when we added lock time into Bitcoin

several years ago because you add a

a lock time to your transaction, so it's only valid, you know, from the current block height forward. And that, therefore, if if you use lock time, that's one of the big kind of mitigations against this this risk because it means you can't kind of re they can't reorg you back into the past and still snipe your feet.

And that's one of the main kind of defenses against this problem. So, yeah, that this is one of the big challenge Bitcoin's facing. Will all these incentives work in

a few years' time

after all these halvings? I don't I don't see inscriptions as contribution to the problem. I think I see it more as part of the solution because it's just creating more demands.

Yeah. And I Yeah. And I I'm not,

I I say problem. I'm not sure it's a problem.

It's a it's a thing that exists. And if it is a problem, let's expose it now. If it's not,

you know, that's that's great. But, like, I would rather expose these types of, you know, these problems, potential issues for us to, you know, to cover now than, you know, in the in the future when Bitcoin is potentially a reserve currency or, you know, is being used, you know, in in a way like that we all expect and hope, for it for it too. So I I would much rather get to the bottom of this now.

I have another point about mempool saturation, but I don't remember where I was headed with that. So go ahead, Jonathan. Anything else on,

on this?

I will yes. I think you're mentioning go on. Go ahead.

Go I mean, do so do you guys is this a minor centralization risk?

I think

in in general, people paying higher fees for inscriptions

and having these auctions and all these spikes and fees, I would say is not the minor centralization risk. I think the out of bound stuff

is a minor centralization risk.

So centralization risk, and it also breaks the consensus mechanism. Meaning, if you're getting paid outside of the mech like, if you're getting paid outside of the standard consensus,

that could potentially become a problem if it becomes ubiquitous.

And and the reason for that is because you're no long you're not incentivized

by

the the consensus

mechanism that is built into Bitcoin. You're incentivized by some external factor, which may have competing priorities with what the consensus mechanism of Bitcoin is.

And so that's why I think out of it like, out of band payments in Ubiquiti could potentially become a problem.

And and I'm not I'm not I'm actually not as concerned. Like, I'm not sure any of these things in particular are

inherently centralizing.

My my concern is the is the,

would be the security model

being viable,

in a more dynamic fee environment.

What it what is the what is the I think that is,

I mean, we can we can cross that bridge when we get there,

But I personally am not concerned at all about the transaction fee market being viable for security. I think it's mostly FUD level by Bitcoiners.

We've seen Bitcoin purchasing power increase over time. We've seen adoption increase tremendously over time.

We've seen, although the Bitcoin,

reward mining subsidy has decreased in Bitcoin terms,

it has increased in purchasing power

throughout its entire existence.

So I am personally not concerned about that, and we can save that conversation for dispatch in 10 years.

But I'm like, so let's just drill down on this out of band payment centralization risk. What is the concern there in terms of

well, how does that incentivize miner centralization?

Well, you have to have an infrastructure in place in order to receive the outbound payments. Right? So you may be it could be, you know, you're you're a mining pool operator and you're very well connected and you're friends with all the people doing order inscriptions.

Or you could do what via BTC does and have a website accepting credit cards. So then if you're if you're a smaller mining pool,

you don't have the infrastructure. You have to build that infrastructure out and that's a cost. And also these other things like paying for outbound fees,

the the the market for that, you know, that there may be economies are scared and everyone just always goes to VIBTC to pay over. Right. Luxor Pool is very well known inside the inscription community, so they always just reach out to Luxor and ask them to Yeah. Put that.

And and that's that's all centralization.

Yep. 100%. And you're you're totally right. And the the reason

so we we don't we did the out of band payment one time,

realized this is probably not a good idea. Let's we we put everything else in as TXV. So if you go look at any of the other you know, we did,

we did a a big art piece called War Bonds and something called Skrilla, who's an old,

counterparty artist,

that did a did a video. These are, you know, these are inscriptions

that we we put the fee in the in the transaction itself. So it it does remove that portion, but I'm doing that because I love Bitcoin and wanna see it, you know, prosper. But if there becomes a reason why

you wouldn't wanna do that, set those fees at 0 and take the payment out of band, you know, somebody that doesn't have,

you know, somebody that doesn't have the, I guess, how do I say the,

like, the the I get I don't know how else to say it, but, like, it doesn't doesn't love Bitcoin as much as I do would would go in profit max maximize for those those advantages, maybe for the centralization or or whatever. Whatever those advantages are, they somebody's going to take advantage of it. If I've always said this, if the protocol allows it, somebody's gonna take advantage of it and allows you to make more money. So

Yeah. So I think we as

Bitcoin just needs to be vigilant. And if we see any of this kind of out of band stuff increasing in value, we could say, oh, that's a security risk and we, you know, try to,

you know, do something if we can to stop that activity. And it's very similar. There's another problem on Bitcoin, which is very similar, which is,

merge mining. So merge mining is very popular. So the mining pools you can see in the Coinbase transaction, they have a extra upturn output, which are other protocols such as,

RSK is a big one, also Stacks.

So these mining pools earn other revenue on other protocols that often call themselves side chains, for example.

And that's extra infrastructure the mining pool needs in order to mine it because they've also got to run stacks, they've got to run RSK,

they've got to run 4 or 5 things. So that's

another problem or potential problem that could cause mining centralization

that is, you know, has similar similar characteristics

to the outbound fee problem.

Yep. And if we ever see that become like, right now, fortunate fortunately or unfortunately for them, I guess, fortunate for us, that the the revenue for that is, like, very minimal. But if you look at you know, we can look outside of our ecosystem into Litecoin and Dogecoin, and we see that the merged coin, Dogecoin, is actually the more profitable coin, which I don't know how that I don't know what the implications of that on their security model is, but it definitely can't be good. I wouldn't want a global reserve currency built with that potential,

unrest in, you know, in it.

So let's just to go back to outer out of band payments, there's this another aspect of out of band payments. So people tend to think of,

especially,

NoCoiners,

or Bitcoin deniers.

They tend to group mining pools as these single entities. Right? They're like,

oh, there's only 5 mining pools in the world that control this much hash rate. Like, Bitcoin's controlled by these 5 people. But in reality, mining pools are comprised of many individual miners that are all contributing hash to a mining pool. Yes. There's one mining pool operator.

They are operating the node.

There's a proposal stratum v 2 that changes some aspects of that, but the point is is

mining pools are made up of many individual miners that are essentially profit sharing and combining their hash rate. Now if you take an out of band payment,

because,

that is not a transparent market like,

typical mempool gossip where you see all the transaction fees and what's included,

That payment needs to be settled in a different way,

and there's not necessarily verification by the individual minors. I'm curious, Nick,

at Luxor, like, how are you guys approaching that? Because

I if as as someone who who does do mining,

I would be concerned that as an individual miner that maybe you are not telling me the truth about how much your out of band payment is, and as a result, I'm not getting my fair share as the as the minor.

Yeah. So well, one one thing so Luxor operates at PPS, meaning we pay based on the expected value of your hash rate. It's effectively purchase purchasing the hash rate from your mining machine at a market rate discounted by your fee. Whatever your fee amount is, we discount the, the amount we pay you by that amount.

The concern would be with, with a, like, a PPLNS pool where the transactions that are being included in your block are distributed equally amongst everyone,

that would be a place where you would be you you would be either

subject to, I guess, greater reward if your mining pool is very efficient at getting inscriptions into the mining pool, and taking those, those out of end payments and then distributing them effectively.

Or you could be on the other side where, you know, know, in the case of, like in in the case of the big wizard,

for example, where we took 0 fee, and filled up the entire block,

and then we didn't if we didn't we did distribute that. But if we didn't, then you would be out all of the transaction fee, or the mining pool would be out all of the transaction fee that was in there, if they didn't distribute it. So that's the concern and and it really breaks down to the the different payment methodologies.

So I do think that But it goes further than that. Right? Like, there's no way to verify how much you were paid for that.

That is true. Yeah. There's no there's no cryptographic way to verify how much we were paid. You have to trust your mining cooperator. But again, running PPLNS, you really have to trust your mining cooperator entirely anyway.

That would be the, you know, that would be kind of the,

sorry. I got a little distracted. But, yeah, that would be the,

there's no like I said, there's no cryptographic way to prove, like, how much you got paid for that, and you would have to trust the mining pool. But in a PPLNS pool, you have to trust that mining pool anyway,

because they could be injecting shares. There's there's a lot of things because there isn't there is no way to you can

so I know that Flush does their proof of hash rate, which is great. It does help prove how much hash rate they had at any given in time. But the the reason they don't do it for all shares is because, like, the amount of share data we consume as a mining pool is insane. Like, we're taking billions of billions of data points a day, which becomes, you know, multiple terabytes of data. It's very, very difficult to prove all of that hash rate.

Stratum v 2, maybe it improves this, but but it won't. It's it's not the solution to every problem, especially in Bitcoin mining.

Jonathan, any thoughts on, like, how a mine like, if a miner were taking out of end payments, how would they prove that they're taking,

you know, that they're taking that route amount, a,

honestly, and then also paying it out?

Yes. Very difficult. Of course, they could receive the payment

on chain via Bitcoin and publish that, but then, you know, you can't prove the absence of any other payment. Right? So,

yeah, it's very difficult. I think it's impossible to prove the absence of a payment. So

Right. They could have got paid separately.

Yeah. Exactly. They could have just, you know, got a brown envelope or whatever under the under

secretly, and you you can't prove that that didn't happen.

So,

yeah, that's one of the reasons why these outbound payments are bad because they're yeah. This is nontransparent.

And, yeah, it's an opportunity for the mining pool operator potentially to, yeah, take money from the miners.

Yeah. But p I mean, PPMS pools could do that already by injecting shippers, but

So can we just quickly go through what are the top,

so so you keep saying PPLNS. That's a different payout method

that mining Yeah. Mining pools have different payout methods. Right? Can we just go real quick into the the different popular ones

and which pools use which?

There's 3 that are ubiquitous now. There are a lot of different payment methodologies that exist.

But the 3 that exist now, FPPS is by far the most popular and common. That's what Foundry uses that will that's what Luxor uses. That's what, you know, Binance, and I'm pretty sure Antpool. But, basically, all the big pools do FPPS. And what that means is you look back over the previous day and see what the average fee rate was and use that as your Coinbase reward in the calculation you use for the PPS calculation. So,

the PPS calculation is just the Coinbase subsidy divided by the network difficulty times the number of shares that you you have submitted.

And in the Coinbase subsidy,

it's Coinbase subsidy plus fee. And so the the the plus amount there, the fee is basically the average fee over all the previous blocks. PPS plus is very similar. So you get paid the base rate. So right now, it'd be 6.25

divided by the network difficulty. And then the plus is they do a PPLNS round for whatever blocks that they found.

I think that, you know, that

that way does introduce a little bit of trust because you have to trust that the mining pool is doing that properly, that calculus.

But that's I think that's how f two pool does it and

maybe via BTC. I'm not exactly sure which other pools do PPS plus, but that's another common one.

The one I'm talking about PPLNS,

you can think of it almost as like a prop it it stands for a pay per last n shares.

Oh, and I guess p p FPPS means full paper share, and PPS just means paper share.

And so PPLNS,

pay per last n shares is a proportional

reward. So, you know, if between the 3 of us, you know, we've submitted you know, I submitted, you know, a 1000 shares total. I submitted 300 of them.

Jonathan Jonathan submitted 500 of them, and Odell submitted 200 of them. We would get paid out. You know, I'd get, you know, half, you know, and then we would or I would get

a 30%. Jonathan would get 50, and, Odell, you'd get 20%.

That's the way PPLNS works. The reason they call it PPLNS is because they want to,

reduce what's called pool hopping,

so you take a much longer look back window. So you have to look over multiple rounds,

and that's how PPLNS works. The doing a full prop is is called proportional, which no mining pools really do now. I think maybe some of the private pools like Marathon and Terra Pool do prop. But prop is basically,

for a particular round, you know, n to n plus 1,

you proportionally pay out how many shares were were paid.

Proportionally pay out based on how many shares were submitted. So those are the differences. And you can see how with PPLNS and prop, you really have to trust that the mine and even plus to some PPS plus to some regard. You have to trust that the mining pool is accurately reporting how many shares were submitted. So, you know, if, if say there were a 1,000 shares you know, the pool reports that there were a 1,000 shares submitted, but there were actually only 800, they're effectively able to, I don't they're effectively able to to steal 20% of the reward for that block. I don't wanna use the term steal because it is pretty negative, and I I would never accuse,

any of the PPMS pools of doing that, but that's just the mechanic that they would use to do so.

Jonathan, any any thoughts on that? Like, do you think any of that is, inaccurate or needs clarification?

No. That sounds fine.

Okay. Cool.

The,

and so and so when I mention yeah. I guess when I mentioned, the reason that, you know, when when an FPPS if if an FPPS pool were to take out of band payments, it wouldn't really affect the the payment scheme because the p p the FPPS rate wouldn't really be modified,

much. Now if everybody was doing out of band payments, the FPPS

calculation would be completely broken.

Miners would be getting paid out much, much less than the reward that the pools were getting. So in that way, you would have to do PPS plus,

but that also would introduce Right.

Another another layer of trust.

Not yet.

It's it gets really interesting if if a lot of of if the majority of pools are are taking out of band payments.

Then then our whole vision on on what the actual fee market is is completely wrong. We have no real idea of what's going on.

Yeah. Exactly. And yeah. Great, Contillion, Jim. Thank you for adding, just talking about v 2. Unfortunately, v 2 does not solve this problem.

Well, so so v 2 has a couple different

proposal changes. Right? Right now, we're using Strava v 1.

V

2, encrypts the connection between the miner and the mining pool, so you reduce man in the middle risk. And then Yep. There's some small efficiency gains in terms of

I'm not sure where the efficiency gains are found. And then on top of that, the big thing that everyone likes to talk about is block instruction, which right now the mining pool operator,

is choosing which transactions are included in the block. But in stratum v 2 is the idea of distributed block construction where the individual miners are running their own nodes rather than the only the mining pool operator running a node Yep. And they're choosing

which transactions are included in a block.

Right. They get to propose new transactions that get added to the block, but I

I don't think that I I'd have to go check. I I haven't checked the spec in a minute, so do correct me if I'm wrong, somebody, maybe people in the chat. I don't think that the miner can can reject the transaction that exists in the block that otherwise

they wouldn't

meaning that they wouldn't mind unless a transact or if a transaction's included that they don't want to be included.

Well, I think point about

Yeah. I think the the I that hasn't been implemented yet, but the idea is that that will be implemented in Strass and p 2 eventually, and then miners will be able to

select the transactions themselves. But, yeah, I don't think they can fully do that yet if they're doing p 2. That's development work that's ongoing.

Right. And I think that part of the spec is still being scoped out.

Moving back to a couple of the points that you you already mentioned, Odell. So I wanna clarify, you know, I I am

I am a big fan of stratum b 2. I think that it's an important and necessary upgrade to a very legacy,

to a very legacy mining protocol.

The problem that I have now is that

the problem that I have now with with, with stratum b 2 is that it doesn't it's not gonna solve the problem that people want it to solve, which is job negotiation.

So back to a couple of your points, Odell.

The encrypted net the encrypted connection, super important. That is a phenomenal improvement. It does reduce censorship,

at the ISP and networking

level. So it means it's it's it's harder to tell whether your traffic is mining traffic. Like, for example, in China right now, if you wanna mine in China at a pool that's not based in China, you have to use, some sort of, like, encrypted proxy

that masks what your like, what the data you're

transferring is,

more than just an SSL, connection. So, like, SSL has been around for a long, long time,

and, you know, a lot of miners,

some miners use it, a lot of miners don't. Most miners just use plain text. They don't really care about encrypting their data, but d two will have that by default. So it's a really big improvement from a censorship perspective. The point that you were making about improving the efficiency of the, of the protocol

is, that they're gonna be able to encode the protocol in binary.

Right now, the protocol is JSON based, which is a text based human readable protocol, which

the stratum protocol does not need to be human readable. So it's, like, much, much more verbose than it needs to be. And they're claiming that they can compress it by around 30:30 to 50 percent, I think, is the amount they were saying,

if you're if you're using a

binary protocol.

I I do know that you're able to you're able to compress share submission by, like, 90%,

which is a huge, huge improvement. So I think that's something,

awesome. So

the,

the point that everybody wants to talk about with stratum b too though is job negotiation.

And I don't know if job negotiation is going to give you the,

you know, the the,

what do you call it, the silver bullet to all of these, like, transaction selection problems.

Because, unfortunately, at the end of the day, the mining pools

still operate within a jurisdiction. And I'm my biggest concern is, is is regulatory capture if, you know, if if governments decide to try to,

push mining pools to let's just say there's some transaction that governments don't want to get mined by pools within their country. I don't think Stratham b 2 gets you around that problem.

But yeah. But if if Stratham b 2 is, like,

does everything that it was promised to do, which was, you know, individual miners can choose their own transactions, in theory, that should

significantly mitigate the outbound fee problem because,

you wouldn't be submitting it to the

you know, you could pay a pool out of bands, but then the individual miner might not include that transaction. And therefore, the out of band fees didn't work, and

that would significantly misget the problem. But, yeah, that I don't think that's been implemented yet.

Yeah. No. I I think Pavlenecks is, is one of the

one of the important people to listen to here, and he's chatting in the in the,

in the message or in the chat here. He says there is no negotiation

in the reference implementation.

The pool can accept the proposal or close a connection in which the in which case the miner falls back. And what he means by falls back is it falls back to another pool. So what that means ultimately is that the pool your primary pool ultimately won't take your transaction if for whatever reason it says that it's not allowed. Conversely, Jonathan, what you're proposing is or sorry. It also doesn't solve the out of band payment. But,

conversely, what you're saying is that if the transaction ID is included and the miner decides to reject that transaction ID, let's just say there's some out of band transaction that shouldn't be in the mempool,

then eventually,

that eventually, you you know, that that miner will decide, okay. I need to switch to a different mining pool that isn't including that transaction and forcing me to mine on the transaction that I don't want to.

But it will ultimately still come down to the mining pool. And if, you know, for whatever reason, say,

let's use a, you know, big publicly traded Bitcoin miner here in the US that wants to use

a US regulated mining pool because of regulatory capture,

there isn't really a way for them to get around, that, you know, let's just, you know, for whatever reason they decide they don't wanna mine this transaction, they won't have fallbacks,

because they need to be mining to US pools. And there's really no way to get around that. Yeah. Of course. If if both the mining pool and the miner together wanna censor this transaction, then, of course, Strass and BT is not gonna sell that for them. Well, if a mining pool wants to send through a transaction if a mining pool wants to send through a transaction and the miner does not,

then the miner has one choice, which is to switch to a different pool.

And d two does have that, that implemented that you can it has more rapid switching capability. Right now, getting switched over to another pool and ramping up that that connection,

it does introduce some late like, some

some lag or some degradation in performance.

But the stratum v 2 implementation should remove that, so you can fall back to another pool. And I think the point that I'm making is that a miner that's based here in the US or or wherever inside some regulatory environment where their pools is also incorporated,

they won't have the option to fall back. So that that transaction will ultimately end up getting censored regardless.

Sorry. I'm a little bit distracted because, we have good friend, Pavel, next to the chat, and he's, the lead maintainer of

the open source project that is Stratham v 2. Yeah. So I was trying to get him to join us. I sent him an invite link. It looks like he will not be joining us, but

I will hopefully get him on dispatch,

after they ship in 2 weeks,

and we can discuss it at length then. But, yeah, stratum b 2 in general is this idea of trying to improve the protocol that miners,

individual miners are

interacting with their pool operators because,

I mean, from where we're at, we've been using

for, I believe, like, a decade now.

Yep.

It's, it was it was written when we were mining with we were written with we were mining with GPUs.

Yeah.

So, I mean, guys, this has been a fantastic conversation.

I personally

could care less,

or couldn't care couldn't care I couldn't care less about NFTs.

I think we're about to hit a global economic

collapse.

Eventually, I'm not gonna say I'm not a Balaji. I'm not gonna say it's gonna be 90 days. You know, it could be in however many days. It's probably gonna be after conference day. I'm not making any predictions

about conference day.

But

and we need a money that is that is independent of of corporations

and governments. I think Bitcoin is that money. I think Bitcoin is gonna probably price out a lot of things that are that are bullshit and nonsense and only only important thing or is that people value will be included. Some of those things might be NFTs. Some of those things might not be. But I do think that,

our mempools,

how blocks are being constructed

are very important. I think minor centralization risks are very important. That's why I I thought and I continue to think that this conversation has immense value to, the dispatch audience and and to myself because I'm trying to understand it more.

Yeah. I have Jonathan,

just messaged me who's on here. Let's talk about before we wrap up,

let's talk about this one block

situation that happened.

And can can we get to get some context to start first about what went on there?

Yes. So,

this was on Saturday, and I think,

1:40 AM.

So there were 2 blocks that were mined at about the same time. So,

Foundry Pool USA minor block with the time stamp 1:41

and 10 seconds.

And then 7 seconds later,

via BTC found a block,

of the same height.

So this is just a you know, this is obviously why you have proof of work mining to you know, it's a timing problem. So presumably via BTC didn't receive that foundry block,

in that 7 seconds

and therefore, you know, mind a conflicting block, and then there's a race to see,

which chain was extended first. And this is a very normal thing that happens,

relatively

often in Bitcoin. So I'm looking here at the data from my notes. So,

even though the the time stamps have a second 7 second gap,

I saw the foundry,

pool block first

at 1:41

and 40 seconds.

And then 35 seconds later, I saw the via BTC block.

And the VIA BTC block is the one that eventually won. So that is the one that had the blocks built on top of it.

So that means my local node experience to real. The way I kind of think of it is that, you know, this is a stale block situation.

A stale block may or may not result in a real from the perspective of your node because I saw the losing block first.

That meant that my node did experience a real.

And, yeah, I think it was something

For those in the back, can you say it again?

I would say a reorg is when your local node,

you know, has a chain tip,

and then that that tip loses, and then you go backwards and then forwards again on a a different chain.

So you just stay on the Whether it's one block Yeah. Whether it's one block or 2 or 3, it's a reorg. Yes. Exactly. So I think the what happened now

was a reorg from the perspective of my node. However, if I if my node saw the via BTC block first, my node would have never experienced a real. Right? It would have just carried on moving forward. It wouldn't have cared about that Foundry USA

Yep.

So the reason and I guess the reason I became you know, I got a little vocal about this is because there's

in no circumstance

that Vibe BTC didn't see that block in that 40 seconds.

There is no circumstance. They're all peer together. We all share peers. Like, we peer together. There is no chance that they didn't see that.

But then what what what yeah. Because I think there was a tweet alleging that it was, like, kind of foul play from BYOBTC and selfish mining.

And I

I don't think that's right because,

the the nodes the the tweet is saying that, well,

look, my log show update change it happens,

and it's got the 3 blocks. The block before, the block where there was 2 blocks, and then the next block all happened the same time, which was 5 minutes later at 1:46

AM

UTC.

Was that I think that's completely normal. That's how the. Right? That that node first saw I think that was your node. Right? It first saw the foundry Yeah. Yeah. Mhmm. The foundry block.

It regarded the So I think everybody's notes saw the foundry block. The note the foundry note, the foundry block first because it was, like, 40 seconds previous. So I think everybody reorg. But, anyway, I know that the tweet you're referencing was my tweet, and it's been, a point of contention.

And, and a lot of people have been very mean to me about it, but I stand by it. K.

Because I don't think it was But but let's yeah. Continue. Foul play or indication of selfish mining because right there was a 6 a 5 minute gap. Right? So first this happens

and then the VIA BTC block was built 5 minutes later.

So if if it was But, Jonathan, can you define selfish mining real quick for the audience? Okay. So

the selfish mining attack, which I think was first discussed way back in, like, 2010,

maybe by, I think, Dan Larimer was the first person to talk about this. It's basically,

you find a block, but you don't publish it.

Instead, you you just keep your block secret and keep trying to build on top of it. And then if another block is published,

you then broadcast your block.

And, hopefully, what you're trying to achieve is you build, like, a secret chain of maybe 2 or 3 blocks. And then if someone else has published publishes their block, you publish your whole chain,

and you're the winner,

because you've got a longer chain. And this

supposedly,

you know, if you exclude

kind of other considerations, the game theory is if you have 33% of the hash rates

in the network, you can be more profitable by selfish mining,

because you you just make more money by kind of

excluding other people from

from getting the block reward. So that's the idea of selling share with the network.

It's a very negative thing. With the competing miners. Exactly. You keep blocks held secret, and that's a very bad thing for Bitcoin because it just means there less information is published, more stuff's happening in secret. And again, it just causes mining centralization. So I don't think that did happen. And I think if it did happen that or it was happening, that would be very negative.

But I I don't think there's any evidence in this case that it that it happens.

Yeah. Because yeah. I know, Nick, you wanna comment on this as well?

Yeah. So so the the the,

how do I say this? The consequence of of selfish mining was spy mining, and spy mining is where you are your mining pool listens to other pools, and they broadcast,

a new job that appears to, indicate a new height. You can update your internal state with that job, and start and start processing it and basically

send out an empty job template. So a lot of times when you see a, an empty block, there was one actually earlier today, a lot of what you're seeing is either spy mining or or a a a variation of selfish mining. Selfish mining doesn't have to be just one block. It can be multiple blocks. And Didn't we used to call that something different? Not,

because it was happening a lot. Spy mining. It was is there another term that it's referred to as?

So, you know, the the consequence of spy mining is empty blocks,

but we also see empty blocks as a result of the covert segwit, which is what was happening back in 2017 when Antpool was mining

empty blocks. Jihan Woo has their has this famous tweet about, you know, fork your mother,

all of that. That was, that was due to covert segwit, meaning you could mine in that in that in that case, you could mine,

that they they figured out a software improvement that required you to to mine empty blocks, and it reduced the energy consumption of your miner by about 18 to 20 percent. And so they decided, well, mining fees are only, like, 1 to 2%. I can reduce my my cost by by 18%.

Just gonna mine empty blocks. And so that's what we saw empty blocks at that time for.

What you're and and I think is that what you're talking about?

Yeah. But they were it was, I remember, like, Adam Back, like, talking about, like, it was, like, header first mining or is is that am I, like, making that up? Or is Well, SPV yeah. It's it's mining. You you build on the next block before you download and verify the block, just so that you're you're faster.

Exactly. And that's that's a that's a variation, I would say. Maybe a variations of selfish mining or or you can call it SPV optimization where you get the header, from some other from some other pool, or maybe you're listening to their stratum connection and getting the note getting the updates from the stratum,

the stratum server, which will have the most up to date information. So if you'd notice a chain tip update from the stratum server, you can immediately build a block internally with that information and then start,

start your miners mining on it. And that's called spy mining. But, like, what Jonathan's talking about is SPV mining. Yeah. I think SPV mining and spy mining are very, yes, similar concepts. Right? So spying is when you're you're connecting, you know, via stratum or you're connecting to the pool of your rival pool

to get information about this block header faster.

And then

the the other thing is you're is starting to mine on next block without fully validating the previous block. Yeah. And that's, it's a you know, it's only a second or 2 improvement,

but a second or 2 is very big in mining when, you know, miners are talking about, you know, single dips in, in in,

you know, in revenue Yeah. In revenue changes across pools.

Yeah. So in in this in this scenario on on Saturday morning,

you know, when you well, basically, all all major nodes experienced this one block real when they had the foundry pool

block

at 141. And then

6 or 5 minutes later at 146,

there was another block that built on top of the VIA VTC block. And you showed your logs,

which had, you know, update chain tips. So what the what the log show is when you jump from

the shorter chain to the longer chain, you

go back one block to, you know, block height 781-276.

Then you move forward one block to the via VTC block

781-277,

and then you move again

forward 1 block to 781-278,

and that's all in basically one instance. So your log showed the same second Right. These 3 new blocks. That doesn't mean those 3 new blocks were received at the

that same time. It just means you reconsidered a new tip and moved to a new tip at that time, which is moving back 1 and then moving forward. In your, in your note, when you're saying that you received the block at a particular time, what are are you looking at a a log, or are you looking at,

something in the in the block data itself?

Okay. So, yeah, what I'm talking about is this website, fault monitor, which,

yeah, is a bit different than a normal note. So I think this

that first scene time is when we,

write to our database

the

details of that block. And that's so so that's basically that's not really a Bitcoin

core output. That's just something from Fort Monitor's own website that produces that time stamp.

Is there is there more more data available?

Like, I was actually I was chatting with Peter Todd and some other folks about this. Like, it's it's hard to get the exact time that a a block was, like you said, without making some changes or or whatever to to get the exact time that your node received a block.

We were not Remember the the data. Well, exactly. Because remember,

the

when everyone regarded that foundry pool block as the tip, and then the VIBTC

block came in, The default behavior of Bitcoin core is not to download that because that's not the chain tip. So the default behavior is you only wanna download a full block of the most work valid chain. Otherwise, it's like a DOS attack. Someone could submit all kinds of large blocks

that aren't, you know, the the most proof of work and span the notes. So I think most notes wouldn't have downloaded that via BTC block at all

until it was built on and became part of the longest chain, and then they downloaded it 5 minutes later at 1:46

AM. So that's my understanding of how a normal node

would operate.

Would have operated. Yeah.

The reason it's so suspect

is because of how close in proximity it was to this event occurring,

the meaning the the d gods block. The d d gods block. So we're talking about block. Let's say we're talking about block n.

The d gods block, the block that was gonna include all this transaction fee was block n plus 2. And so for it to happen at this exact time is incredibly suspect given how I don't wanna say rare given that these things happen maybe once a month, but for it to happen at that exact time in a 30 day period is certainly suspect.

Yeah. But the the the 5 minute gap between 141 and 146

is potentially an indication that

nothing until what happened. It was just

the the you know, they they may may have been trying to I'm sure they were trying to spy mine or whatever in the first few seconds, but then they only actually found the next block 5 minutes later.

Is the,

is are you are you able to get, like, the the data around the blocks around the other blocks themselves? Like the,

it doesn't look like that's available here, but it would be super interesting to kinda see the rest of them. We found the block with all the fees in it.

And pool.

That was another block, but I think that that that incident was was a was a a wiley.

So, yeah, I mean, there could be something untoward here, but I don't think

the the the time gaps in this particular incident look like it is a normal

one block reorg to me, I guess. Right.

But, but, yeah, you never know. They could What block so much we could do. What is what was the block, the one that had all the fees in it? What block number was that?

Shoot. I just had it up here. It was

781279.

So if you go to mempool.space,

you'll see 781279

had, like, 3.5 BTC,

worth of reward in it,

which is And that was 2 blocks after the reorg, basically.

Yep.

Exactly. And to be clear, we see these one block, like, stale blocks,

orphan blocks, whatever you wanna call it. We see them pretty common.

Like, what Nick said, like, once a month.

Yeah. Like, once a month every every few weeks. And this is why Yeah. This is why, like, the common

this is partially why the common,

recommendation is to wait 6 blocks on a confirmation,

is to be abundantly clear that you're not gonna see any selfish mining. You're not gonna see any reorgs that happen before you release your goods or services that you're selling your Bitcoin for. Yep. Settling the transaction. Yep.

Yep. And also, on this particular

incident on block height 781277,

fault monitor did not detect any conflicting transactions.

Whereas, there was another real Yep. The the next day where there was one one small,

conflicting transaction Yep. Of, like, the conflict was of 0 value, that there was a small

Yeah. The the conflict I I think the conflict, it wasn't really, this in this case, it wasn't really, an attempt to get a, you know, get, you know, double spend or something through. It was just to maximize,

minor value or minor reward, I should say.

But regardless regardless if this was in a deliberate process

by and I think I think it's safe to assume that Antpool and VIA BTC are, like, part of the same cartel. They're, like, the Chinese miners that all hang out together.

If if regardless if that was the case in this situation,

I think if we start to see more of these,

you know, one block

NFT sales or inscription sales where the fees are way higher than other blocks,

Like, there's gonna be shenanigans. Like, miners are gonna fuck around with that,

because that's their incentive to do so. Right? Like, they're gonna wanna try and get those fees.

Yep.

Well, yeah. That could happen, and I really hope not, because I think that's that's very bad for Bitcoin. Yep.

If we have these kind of if if they're willing to not build on the longest chain just to try and snipe some more fees, and we need to try to mitigate against that risk as best as possible. So how do we mitigate? Do you have any thoughts on how to mitigate that, Jonathan? Yeah. Well,

one way is is the lock time thing. Right? So if you Yeah. I think Bitcoin caught us up by default. If whenever and you use wallets that set a lock time on your transaction such that your transaction is only valid at the current block height or higher,

then it can't then the the miners are incentivized. Never they can't reorder transaction back in the past. Let's keep those fees. So I think one good way to Right. Get that is is lock time.

For your that's for your own security, for your own payments. Right?

Well, it's it's kind of well, it's also not necessarily just for it's also for everyone else. Right? If I'm sending someone, like, a a big transaction with a big fee, and I put a a lock time on it with the current block height, then

the the miners that won't incentivize the miners do a one block real, then that one block real could damage the security of other people as well. Right? So you could argue using lock time is making you a good citizen

and rich helping reduce the incentive to real back.

But it what it I mean, it would whoever gets the fee, like, just because the the end just because the lock time is set, the the miner that ends up getting the fee for that, that transaction still wins the fee.

So there is like, it doesn't really I don't think it changes the incentive to reorg to capture more fees. Well, it just means they can't reorg back in the past. They can only do it from this point forwards.

So it's any mitigation. It's not doesn't really solve the problem that well, but it stops them going back into the past to to real you you to, like, combine your fees with even more fees from the past. Yeah. Understood. Exactly. So, like, you know, in in in the case, you know, there's a, you know, big mint tomorrow night,

there's a big big mint tomorrow night. Like, mining pools will know that that's coming. You know, say there's gonna be 10 BTC worth of fee in there.

They're all gonna be very incentivized to selfish mind at n minus 1, wait for that them to build that block, build n, and then release as soon as they found n.

So so that you'll think that that might have been what VIBTC were trying in this instance?

Oh, yeah. 100%. And that's what I and and I was gonna say, like, the way that I would have executed this, like, my pool is too small. You know, we're maybe 5 or 6%. You need, you know, you need a all you need is a a little bit bigger pool than that, maybe in the 15 to 20% range, and then a little bit of luck. You need to get a little bit lucky. And I think what if I were executing this within that in that case, you can build a statistical model for what your expected value from doing something like this would be and determine how long should I be holding on to this block in my memory like, in memory before I distribute it to the network,

to decide, like, what is my, you know, what is my profit? What is my expected value here? That expected value goes down over time. And the longer you hold on to a block without

broadcasting it, your expected value for that block goes down. You know, if you release it immediately, your expected value is a 100%, but there's a different curve based on how much hash rate you have, which is why 51% attacks can occur.

In this case, I would have attempted I would I would have attempted to mine n minus 1 or maybe even n minus 2.

Hold on to that block for a moment, try to get n or n minus 1, and then and then, you know, basically build my own internal chain that's 1 or 2 blocks long, and then release that as soon as I, as soon as I've mined the target block. That would be the way you would attempt to build to do something like this. And it see and it was just so coincidental for it to happen at that exact height with all of the hype that was going on. It's crazy.

Okay. So you and but do you think their their attempt was coming they were collaborating with AMPU and that their attempt was successful or did or their attempt failed or I think they I think they I think they missed, I think they missed by one block. I think they got success I think they were successful because they were able to get the target block, but the the fork occurred one block too early.

The the the block that they were attempting was one to one block too early.

But yeah. But they could have been collaborated with Ample, and Ample did did get that high c block. Right? Ultimately ultimately did win it. Yes. I do think they ultimately won it. Okay.

But yeah. I mean, if that did happen, that's, you know, pretty bad, and we should definitely try and look into this and try to tell me if it did, I guess. But I'm not sure I mean, it's unavoidable. The other It's just a free market at work. Like, how can you

you have to wait for more confirmations.

Like, that is

Yeah. Exactly.

That's it. I mean, that's why we have the 6 confirmation,

target or maybe, you know, some places have 10, but,

that that's that's the reason for it.

We've always been a big fan of confirmations.

Not a huge fan of anything that relies on, you know, low confirmation numbers. There's 0 comp. I'm not a huge fan of 0 comp at all.

I'm a big fan I'm a big fan of, full RBF.

You know, I I I turned that on for our pools. It unfortunately got turned off because of a misconfiguration. But, yeah, turning it back on,

big fan of a full RDF just because I I think that we need to rely on confirmations for the security of the network. I guess, I think, you know, the a bit like sustained high fees. At 141.

Alright. Continue, Jonathan. Go.

Yeah. If if at 1:41 AM,

VIBTC

saw the foundry

block first. I mean, they really should have just

started this and they didn't have a a big enough secret chain. They should have just started the selfish mining attack again

after that block at 141

rather than

Right. Carrying on something. And and

and they may have. We don't know that. You don't know that.

Because you you would only you would only know it if they release the block if they had mined the blocks to release them.

Yeah. But they they

they built they built on an alternate. They, you know, the chain forked and VibeTC

produced another block at 7812 77 whereas if they were

yeah. I, I don't know. I mean, you could if they were selfish mining, they could've they could've just said and then all they were trying to do is capture that big fee that was in T Block's time. They could've tried to do it on top of started salvage mining again on top of that foundry block, and that might have been a Right. Policy. And I think that I think that if I had to guess, it was a coding mistake that somebody was trying to do the block that you're talking about, and they were off by 1.

Yeah. That would be my guess. Many such cases.

I mean, there's, off by 1 error in the Coinbase Genesis transact the Genesis transaction.

So it's a difficult deal. I error everywhere. It was a seg

it was a segway 2 x joke. The

Oh. I wonder how much of it,

I wonder if the calculation

came into play at all about them essentially. They were competing with Foundry,

which is the largest mining pool. Like, if they were

selfish mining against, like, a slush pool or a smaller pool, if

they would have been more aggressive with it?

I mean, potentially, you could hold on to I mean,

in that case I mean, you're it doesn't matter who who's your, like, trying to to,

I guess, trying to orphan,

be or who whose block you're trying to reorg,

because you're competing at the n plus one block at that point. You're competing against all pools again.

So I don't know if it would matter who in particular mined the block. Well, it matters because Foundry

Foundry has the strongest incentive to just say fuck you and continue mining on top of their block because it's their block. That is true. Yep. That's true. And they probably And Foundry has the most hash.

And Foundry, I I don't have the I I wish I knew how to do the math more effectively, but they probably could have like, if they had done it, they probably could have reforped,

and got their, you know, got their reward back.

And I imagine that as mining pools get more sophisticated, this sort of thing will become more ubiquitous where, you know, it makes sense for you to, like, effectively defend your block, as, you know especially as a pool that has, you know, what what do they have? Like, 45%

40% Hash rate?

They have 31.6

according to Oh.

The 1 week chart on mempool.space.

Okay. Well, that's good. I thought it was much higher, but that's good that they're,

yeah, you're right. 31.6. So that's good.

So what I was gonna say earlier, I mean, in terms of mitigation,

a sustained high fee market,

like, if there's if there's constant high demand for Bitcoin block

space Mhmm. Could reduce

the frequency of stuff like this, I think. Right? Because there wouldn't be so much variance from block to block

in terms of how much Yeah. Exactly.

If there's always a big mempool and there's always a lot of fees, then, you know, why do you wanna reopen and go back in the past when there's it's a rich, healthy environment with a lot of fees available. So If we had empty mempools here at this time when this happened, it could have been even more chaotic back and forth in terms of

who fighting over that block. Because they might have been able to do the math where it's like, we could wait.

We can do this. We can keep this up for 4 blocks and not really lose that much opportunity cost. Yep.

Yeah. It makes sense to, to like you said, you know, from an expected value perspective, it make it could make sense. There are times where it would make sense.

And volatility is gonna become you know, mining mining is already an EB game where you're you're effectively, you know, capturing expected value of every hash.

You know, this will become an even more dynamic. You know, the expected value of block like, right now, the expected value of n and n plus 1 and n plus 2, generally about the same. But in the future,

imagine, you know, a very long block with which accumulates a lot of fee in the mempool,

or in its mempool.

That block has a much higher reward at n than n plus 1,

in in theory.

And and there may become a time where mining pools start to really optimize for for this, and it it it I don't wanna say it will break

the consensus mechanism, but it will certainly require confirmations. You're gonna wanna make sure you have, you know, 6 or 7 confirmations for, for any high value transactions because this could potentially be a a place where, you know, more forks and more reorgs become, where forks and reorgs become a bit more commonplace.

Well, is it

I mean, correct me if I'm wrong, but I think it's Eric Vascule that talks about this a lot in terms of

in that type of situation,

you have a strong incentive to pay a high fee

for your for important transactions because you basically want

the security that that provides in terms of reducing

reorg risk.

Yeah. I mean, his his argument is that, you know, the censorship resistance is only

proportional to to the fee you pay, and those fees you're paying are for censorship resistance

Right. Which I'm sure in the long term is is absolutely right. The censorship resistance,

is not is not coming for nothing. You you have to pay.

Right. This is a bit of a chicken and egg. That's why I don't think

I really think this idea that there's that transaction fees can cover

just it just it's not

it's not based in actual reality of what we're gonna see, but, I mean, we'll see what happens when

when we get there. Well, yeah, I mean, the the critics of Bitcoin tend to argue it both ways. Right? They argue, look, Bitcoin's stupid. It's so expensive.

That's not gonna work. The fees are so high. And then they also argue,

it's not sustainable how you're gonna,

you know, pay for mining incentives. So, obviously, both those criticisms can't be right.

But at the same time, there is a lot of uncertainty. Can we really get the balance right?

And that the fee market is healthy enough to incentivize mining good also.

It's reasonable enough that people can use it for transactions and it's not too too expensive.

So

Yeah. I think,

the point,

I think both points are valid. So, Adel, I think what you're saying is that, you know, the concern being that, you know, Bitcoin,

the the security model may may not work long term because of the low like, because of not enough transaction fee. The con the what I'm kinda mentioning is that in the future when transaction fees are much more dynamic or not just transaction fees, when the block subsidy is much less predictable based on,

fee fee volume or just just standard block times. You know, the fee volume could be exactly the same, but The block reward. The block is

block yeah. Sorry. The full block reward, not just the subsidy. The full block reward. Thank you for clarifying that. Because it'll be mostly transaction fees at that point. So it'll change from a block to block Yep. Point of

view than a situation where where the block subsidy is the majority of the revenue for the miner.

Right. Yeah. Exactly. And as a percentage, like I said, we'll probably see maybe not this having, but the next having for sure. We'll see or hopefully see.

Fingers crossed. I mean, hope isn't a strategy, but it does seem like we're headed that way where,

the transaction fee will be more than a 100%

of the total block reward.

You guys have anything to add on this on this topic?

I shouldn't say over a 100% of the block reward. It will be more than a 100% of the block subsidy making up a much larger portion of the block reward. It can't the block reward I think that was a 100%.

Yeah. Yeah. Hopefully, that yeah. Hopefully, that was clear, but, I wanted to clarify that because it was The block reward in the block reward is the total amount the miner earns, and it's it's it's it's split up between the block subsidy, which is what splits in half every every 4 years and the transaction fees within the block.

Correct. Yep.

And so the statement I made that a 100% of the block reward more than a 100% of the block reward comes from

fee is inaccurate because a 100% is it's always gonna be a 100%. And what I was saying was that it will be more than the transaction fee will be more than a 100% of the block subsidy, making up, say, more than 50% of the total block reward for any given block. And as that number becomes greater,

the dynamism

between 2 particular blocks, say, n and n plus 1,

can become much, much greater.

And that could

start to introduce

more of these shenanigans as you call them.

It's gonna be interesting.

Yeah.

But, I mean, also, yeah, if the we just gotta make sure the mining industry is also more decentralized. Right? Because the more decentralized it is, then

the more chance the salvage money attacks or realtor tax have been failing. So we wanna Exactly.

That's, I guess, the the biggest tool we have is to try to make the mining industry as decentralized as possible.

That is 100% true. And I hope that I hope that we can get there. It does seem like we're seeing some centralizing effect here, especially in the US as,

you know, the the,

a, the rule of law here is very strong regardless of what, you know, what the media says and what, you know, everybody thinks. The US

the US' property rights and rule of law, very, very strong in comparison to other parts of the world. That's why lots of, you know, big investment is being made in the US, and that's why most of the biggest miners in the world operate in the US is because they know that their investment is going to be protected and honored by the US government,

where that and that is not always the case in most in in a lot of countries.

Yeah. I mean, we saw that firsthand with China.

But then the concern is with America is America is the land of financial regulations and KYC, and

we don't want too much cash in America at the same time.

Yeah. I was, you know, I was oh, I always thought that that you the US would be able to compete,

at, you know, one for 1 with, with China. And I was actually disappointed when they banned Bitcoin mining because China had a very like, the most robust mining ecosystem.

And I wanted I wanted, you know,

US miners to be able to go and compete and win on merit, not because of some swipe of the pen.

But alas, here we are, and, you know, we lost

a really large portion of, mining cash rate. So

Well, my bull case my bull case, at least personally, for the distribution of hash is ultimately

the mining game is a game of reducing

energy input cost,

particularly when you start to see ASICs,

the the life cycles of ASICs last longer and longer before they get obsolete.

And as a result,

miners that are able to utilize waste heat,

as a way to reduce,

their input costs,

will have an advantage, and it's easier to do that on smaller scale. I think there's a bit of a diseconomies

of scale once we start hitting that threshold,

and I think we're treading in that direction. So, like, I expect my bull case for Bitcoin

is, you know,

boilers around the world in apartment buildings and hotels and whatnot are all have ASIC miners in them.

And as a result,

these large warehouse miners will have a

a smaller percentage of the hash,

but we will see.

Yeah. I think that's exactly right. If you look at the history of ASICs, right,

Earlier on, there was huge, like, percentage improvements in efficiency of, like, 10 x improvement.

Yep.

And I think since Bitcoin launched, ASIC efficiency has probably improved by about a1000000x.

Whereas now, the new generations are only, like, twin 20, 15% improvements on the previous generation,

which means that these ASICs will be around a lot longer. And the mining industry is gonna be much less about a big race to,

you know, get a better relationship with the foundry and the AC manufacturer and deploy CapEx and buy all the machines to us and everyone else. It will be more about

finding the best energy assets, the lowest cost energy assets, and as you say, these alternative uses for heat.

And hopefully,

that results in a more distributed mining industry than this other dynamic. But you don't know for sure, but there's it's definitely a potential there

that that kind of where the cheapest, best places to deploy them from an energy perspective

will result in a more decentralized mining industry. Certainly, probably will be more decentralized than,

say, Ethereum's,

proof of stake where there's no it's nothing to do with the energy infrastructure across the world. So Yeah. Right. There's a chance there that that could hopefully, you know, result in more decentralization.

But it's as yet unproven. It's just what we believe could happen, I guess.

My my bull case for mining decentralization is more around, the the sparsity of energy.

Energy more or less is rather distributed around the world pretty pretty equally. It just happens to be that we can't capture it in all places equally, and,

the regulatory environment of all places is not equivalent.

The US happens to be, you know, kind of a utopia in that regard. There's a lot of energy here,

and the rule of law and and strong property rights are pretty are also very strong here. You know? I I do imagine that, you know, places like the Mideast start to become, you know, pretty big producers of hash here over the next decade.

Yeah. Put differently, you can only get so much cheaper energy in any given location

before you start driving up the cost of energy. Yeah.

Way way more succinct than what I said, but yes.

Awesome, gents. This has been a fantastic conversation. Before we wrap it, I'd like to end with final thoughts.

Nick, final thoughts.

I've never been more bullish on Bitcoin mining,

regardless of what we think about NFTs. I'm not gonna lie. I a lot of times I miss the NFT.

A lot of times I miss kind of how, like, NFTs work. It's not really my purview, but, I'm very bullish on what they bring to the fee market in Bitcoin, and I'm very excited for it. And I've never been more bullish on Bitcoin mining.

I want Bitcoin mining to continue to prosper here in the US, but also elsewhere.

And so, yeah, I'm I'm very bullish, bullish on Bitcoin mining. I'm super excited

and love building out the these these types of systems and wanna continue to do so. And, very appreciative of Jonathan and the hard work that they do on on BitMEX research. So, kick it over to him to let him give his final word as well. Thanks, Nick. Jonathan, final thoughts.

Yeah. Thanks very much for this. Thanks, Nick.

And, yeah, then the final thought is, yes, any I think it's May 2024 when the

subsidy halves, and then 4 years after that, it's gonna be 1.5625

Bitcoin, and that's,

quite a low number. And we really wanna, you know, hope by then we have

a very, you know,

strong, rich fee market with all kinds of activity with ordinals and inscriptions.

And the more diverse that is, the more robust that that fee market's gonna be. And, hopefully, the higher fees that's gonna be in a deeper mental there's gonna be, which is gonna keep, you know, mining incentive strong. So anything like inscriptions, which adds diversity,

richness, and bidding to that

to to the fee market, I think, is a is a positive thing and and it should be encouraged.

Awesome. Thanks, Jonathan.

I wanna thank both of our wonderful guests, for joining us today. Thank you, guys.

I wanna thank,

all the

the audience who joined us in the live audience and live chat. I appreciate you.

Thanks to all of the the freaks out there who support the show. The easiest way is at dispatch.com/donate

or through your favorite podcasting 2.0 app. If you weren't around in 2017,

consider picking up the block size war.

Jonathan's book is fantastic.

It really it's a really important,

time in Bitcoin's history, and and we should never forget the lessons that were learned during that period.

I love you all. Stay humble, Stacks hats. Thanks, guys.