TFTC: A Bitcoin Podcast - #241: Mining pools, Braidpool, resistance to attacks, adversarial thinking, and dev funding with Bob McElrath

Episode Date: March 26, 2021

Join Marty as he sits down with Bob McElrath to discuss: - NY BitDevs - Mining pools - P2Pool - Braidpool - Making decentralized pools viable - How they may fail - Adversarial thinking - Bob's Bitcoin journey - much more Follow Bob on Twitter Bitcoin Munich Meetup Video Braidpool IRC Shoutout to this week's sponsors. Cash App. Start #stackingsats today. Use the promo code: "stackingsats" to receive $10 and contribute $10 to OWLS Lacrosse when you download the app. Lend at Hodl Hodl is a new non-custodial bitcoin backed lending platform, that allows peer to peer lending and borrowing between users, globally, anonymously & on your own terms. Create you offers & set your own terms on lend.hodlhodl.com Compass Mining. Compass wants to democratize bitcoins hashrate and give everyone the ability to strengthen the network while mining bitcoin profitably. Thanks to Compass, it’s never been easier to mine bitcoin. Check them out at compassmining.io Braiins. Braiins are leaders in the mining industry; running Slush Pool, creating firmware to make miners more profitable (Braiins OS+), and building out Stratum V2 to help further decentralize mining and make Bitcoin more robust. Check out their mining profitability calculator.

What's up, freaks? It's your Uncle Marty here to introduce this episode with Bob McElrath. But before we begin, I just want to give a heartfelt apology to all my Texan freaks who may have listened to RHR. That was posted last night, March 25th, 2021. Officially 1952, Moscow time. I fudged up. I fudged up. I got my Texas geography wrong. I'm terribly wrong, embarrassingly wrong. That's on me. In the moment, I had a brain fart,

got Dallas and Houston mixed up with each other. I know that's blasphemous. But I'm going to try hard to right my wrong and to earn the love from my Texan freaks back. Yeah, I'm sorry. I woke up feeling embarrassed, frankly. Truly embarrassed. I already can't pronounce words too good.

Now I'm just proving that my geography is not up to snuff. I'm going to try to get better. For you freaks, you deserve it. You deserve better. Now we'll try to give you better. this episode was brought to you by our good friends at the cash app cash or how many stack sets and sets receive sets sell sets if you so please we're saying sat sat sat sat sat sat sat sat because that's

the standard is 100 million sets in one bitcoin you can stack whole sets using the cash app you can also get cash sats back when you use your cash card boost anywhere visa's accepted and you have your uh bitcoin boost initiated you get sats back when you go purchase things somewhere it's pretty dope cash app can even be your bank account they have offering account numbers rally numbers so you get your paychecks direct deposited into the app the whole shebang go figure it out go download the cash app use the code stacking sats that's s-t-a-c-k-i-n-g-s-a-t-s when you download the app you're going to get ten dollars and ten dollars going to go to our good friends at owls lacrosse that's owls lacrosse owls lacrosse this rip is also brought to you by

your good friends at hodl hodl hodl hodl is helping you if you're a bitcoiner in need of some liquidity and you don't want to sell your bitcoin they have their lend product that is available to u.s citizens all right it's an incredible product that's allowing you to keep your bitcoins okay landed huddle huddle is a new non-custodial bitcoin back lending platform that allows peer-to-peer lending individual users meet on the platform globally anonymously on their own terms you put your bitcoin up as collateral in a multi-sig escrow account you hold one key you're getting some liquidity by borrowing you don't need to entrust someone with your funds right you have the collateral locked in escrow and never be rehypothecate you hold one key you

can always track that collateral you control key to it if you have some stable coins and you want to get some yield on those you want to get some return landed huddle huddle also gives you the ability to put those up to be lent out so that you can get returns on that you're saying as high is 25% in some cases, APR. So go create your offers and set your own terms on Lend.HodlHodl.com. That's Lend.HodlHodl.com. This group is also brought to you by good friends at Brains. Excuse me. Brains doing incredible things. Okay. They've been around for a while. You may know them because they operate Slush Pool. Slush Pool is pretty popular. It's been around first ever bitcoin mining pool okay they're also creating brains os os plus this is

firmware for miners you download the brain brains os firmware and it makes your miner stack more sats i'm just going to put it like that simply is that we've had jan and pavel the founders of brains on the podcast before uh we've had edward evenson on the podcast as well go check out those episodes uh brains has been incredibly bitcoin focused throughout their whole existence uh they released the first free and open source firmware for amp minor s9s so that miners would have an alternative to bitmain firmware this was big back in the day uh when amp lead was around they discovered that there was a backdoor in amp minor firmware that could give bitmain the ability to remotely power off asics and they provided miners running s9s with the ability to

to avoid that situation 2018 brain team verified that s9s could run about 13 more efficiently with overt asic boost they made that available to their customers as well let's go check out everything brains os basically auto tunes firmware to give miners more control over their asics and help them stack more sats like i said currently support s9s s17 t17 generation and the dev team is working to support what's miners now when what's minor and they're working on it soon tm they're also working on the open source initiative which we talk about in this episode particularly which is stratum v2 along with matt corallo and some square crypto devs also published a lot of content to educate bitcoiners about some of the lesser known aspects of the mining industry their

content is credible their profitability tools were incredible if you're a miner go check them out if you're a bitcoin or want to learn about mining go check them out brains b-r-a-double-i-n-s.com b-r-a-i-i-n-s.com last but not least this episode is brought to you by good friends at compass mining all right compass mining is trying to make it easy for you freaks to go from zero to mining bitcoin uh as simply as possible they they take away all the hard stuff for you they help you acquire a miner they find a hosting facility with competitive energy costs and they they set up that relationship they allow you to go on the site you buy a miner you pick a hosting facility and they get that miner and they plug it in at that hosting facility and they start

streaming you sats from the miner compass mining.io c-o-m-p-a-s-s-m-i-n-i-n-g.io make it easy making it easy they want to help distribute mining to more individuals you can go from zero to mining at a competitive energy cost you don't have to plug it in at your house. Know your wife with how loud it is. Know your parents with how loud it is. Know your in-laws with how loud it is. Heat up the house. Drive up the electricity bill. Compass Mining is here to help you avoid all those hairy situations. I hope you guys enjoy this episode. To all my techs and freaks, I will make it up to you. Only time will prove this, but I promise I'm going to make a concerted effort to make up for my terrible mistake. I love all y'all. Enjoy.

themselves to devalue their currency bitcoin wins in the world of fiat currencies bitcoin is the victor i mean that's part of the bull case for bitcoin if you're not paying attention you probably should be probably should be what is up freaks it's your boy marty bent here on a thursday afternoon fighting through a cold but very excited for the conversation i'm about to have it's a very important one we're getting back to basics here i wrote about it in the newsletter. This morning, I came across a YouTube video with like 120 views, and it was a conversation about mining pool protocols and how to make sure they're sufficiently decentralized to be resilient against state attacks. The conversation revolved around Stratum V2, its potential shortcomings,

and in another type of pool that may want to focus on pool protocol uh braid pool which is sort of an iteration on p2p pool if i'm uh remembering correctly sitting down with bob mckellarath bob welcome to the show thank you very much thank you um for all you've done for bitcoin throughout the years uh pivotal on my journey because i believe if i understand correctly you helped start the the new york bit devs meet up the white paper series specifically correct uh yeah i didn't start it i mean justin was the guy he's the brains behind that uh yeah i actually rebooted it there was an iteration of the white paper wednesday before i got there but yeah i went to new york in 2015 i believe it was and so ran the white paper wednesday for a couple

years there which is quite fun um just kind of on the back of my previous career in theoretical physics where we would do what we call journal clubs in academia where we just do a close reading on paper that's important yeah well thank you for revamping it because that was right around the time that i moved to new york and that was the first uh the first bitcoin meetups i went to those white paper wednesday really awesome awesome um in person union square ventures was was the venue back then correct yeah absolutely yeah um so thank you for that i have a funny story it's been told on the podcast quite a few times i was in the back of the room nervous having uh only ever seen the people in the room on twitter and i pointed to the the the clock

crossed the way across union square and said to peter todd hey how prescient is that we're we're across the uh away from a deck clock and we're here talking about bitcoin he was like that's a time clock sir i was like oh wow uh but the uh no yeah i mean the the reason i brought you on already described, I came across this discussion you were having with Eric Voskuhl, the Bitcoin Munich meetup YouTube channel about mining pool protocols and how we should be designing them to make sure they're sufficiently resilient against state attacks. And you were mentioning something that you started working on in 2015, which is Braidpool. And I'd love to just dive into the whole topic like why we should be focused on the mining pool protocol level and the solutions

that exist today and how we can make it more resilient via something like braid pool yeah absolutely i mean um so for those that are kind of not aware how of how mining works you know you basically you try and mine a block and your miner doesn't hit the the difficulty target for bitcoin it sends this incomplete work you know they're in there there is a proof of work but it's not enough work to qualify for a Bitcoin payout, you send this to a centralized entity called pool. The pool puts this in their database, they do accounting on it,

and they count how many of these that you produce. And then if anybody within the pool produces a block, they then take those funds and distribute them proportionally to all the people who contributed, what's sometimes called weak blocks. In other words, the shares that you contribute. So that's what pools are, that's the way pools work.

And kind of the risk factor of pools is that there are only a handful of them. They are easy to censor. You know, if a government, you know, wanted to censor them, there are only a couple of entities to go to and say, hey, you have to, you know, censor this address or you have to not process transactions from this entity or this minor is in a jurisdiction we don't like, you're not allowed to pay him. That would be very easy for government to do, right? and you know that's that kind of goes against the the whole narrative of bitcoin but the reasons pools exist are kind of an economic necessity and uh it all comes down to two concepts um eric likes to call it proximity premium and the other is variance so let me um interrupt me at any point if you want to clarify or discuss further but i'll just ramble on here keep ramming okay so

So variance is the concept that, you know, every day there are 144 blocks produced, right? The error on your profit in that 144 blocks is basically the square roots. The number of blocks that are produced in a day is 144 plus or minus 12. And that is simply statistics, right? That is Poisson error on the number of blocks in a day. If I'm a smaller miner, in other words, I'm not earning all the blocks. I'm earning, let's say, 10% of the blocks or 1% of the blocks, right? If I'm earning 1% of the blocks on average, you know, I might produce a block one day, and I won't produce a block the next day. And, you know, in practice, most miners are much smaller than 1%. So you might produce a block a month. And the variance on your profit on a yearly basis can be quite high, right?

you can't really stand in one year for your profit to be 50% lower one year and then 50% higher in another year it's very difficult to do planning it's very difficult to you know procure your miners and know what your budget is and things like that I did an analysis for I gave a talk at the magic crypto conference in 2019 in New York and I did a little bit little analysis asking the question if I want my annual variance of my profit to be less than one percent how large does my pool need to be and on bitcoin the answer is about 20 each pool needs to be 20 in order for my annual variance on my profit to be one percent uh your monthly variance is higher it's like three and a half percent or something like that but your annual variance is down to one percent and that

you know gives me confidence on uh you know my budget and my mining operation now um that if you look at the pools exist today you might notice all of them are around 20 percent and i hypothesize this is kind of a natural equilibrium point for the size of pools miners choose pools they don't want them to be too big if the pool gets above you know 33 percent we're starting to get into um 51 attack territory definitely not above 51 right but they don't want them to be too small either so the net effect is that you know if each pool is 20 you have five pools globally right five entities to attack to essentially shut down bitcoin um so how do we solve this problem right um we we have to trust these entities um pools they are often in strange

jurisdictions lots of them are in china slash pools in the czech republic your mining operation might not be in the same legal jurisdiction and they owe you money there's counterparty risk here there's contracts most pools don't even have any contracts you just send them your your shares and hope they pay you. But as the industry grows and becomes more professional, these questions of legal matters are going to become more important. If my pool doesn't pay me, what is my legal recourse? And in practice, people have basically YOLOed it because switching pools is relatively easy. You can switch pools in a relatively short notice and only miss a couple hours or a couple days worth of blocks. But long term, what we really want is we want to have a decentralized

pool. We want something that integrates directly with Bitcoin itself. If the block rate on Bitcoin was a lot, lot faster, we wouldn't have had this problem in the first place. But for various reasons, that's not possible. So what do we do with this? Years ago, there was a mining pool called P2Pool. So P2Pool was itself a blockchain. It had a 30 second block time. And its purpose was to mine Bitcoin blocks. So with a 30 second block time, that's 20 times slower or 20 times faster than Bitcoin. So on average, one out of every 20 blocks on P2Pool was actually a Bitcoin block. And then P2Pool used its blockchain to account for those other 19 blocks on average in that time interval. And from that one block that actually was mined on Bitcoin, pay those other 20

people. And if you go back and look at the blockchain explorer and some of these older blocks, you'll see some Coinbases that have between 20 and 100 addresses in them. These are P2 pool blocks. The Coinbase has like 20 addresses or 100 addresses because it's paying all the other miners. So that's, that was a great idea. It didn't work for a number of different reasons. But it was fully decentralized. It is a blockchain. And it's frankly a great idea. Why didn't it work? Well, as you decrease the time interval between blocks, you increase the probability that two miners accidentally produce a block at the same time. And you increase what's called the orphan rate this is what eric bosco calls the proximity premium um if you are close

to another miner uh in space and in time um you can produce blocks very quickly after each other because i hear about his blocks faster because i'm close to him in space and in time um but on a global scale uh there's a you know minimum time that it takes to propagate block to or propagate get any message to everybody on earth um and you know using uh the analysis i did in 2015 this is about six seconds using the bitcoin network to get a block to everybody um today it's probably closer to around a half a second to a second so there's a window of time between a half a second and a second where two miners could produce blocks at the same time this sets an absolute lower limit on how fast you can produce these blocks and the idea here is that you've got a a blockchain that's

producing blocks very fast because each one of those blocks is a share as you'd account for in a mining pool but we're doing our accounting on a blockchain with blockchain rules and that everybody's enforcing these blockchain rules everybody's enforcing that everybody gets paid and by using a decentralized network like this there's no one to attack if you want to take it down right it's also more fair in that you don't have any counterparty risk you don't have a contract with somebody you don't have to worry about what jurisdiction this pool is in It would be a lot better in a lot of different ways if we had something like this that worked. So I did some work starting in 2015 through about 2019, looking at a different structure for a blockchain called a direct acyclic graph.

And if you're interested, I gave three talks about this total, I believe, under the title Braiding the Blockchain, if you just Google that. and really what I contributed to the conversation is how to organize a blockchain such that every block can have more than one parent and the purpose of doing this is to allow the block time to be very fast which corresponds to allowing a large number of miners to contribute shares and get paid so that's my contribution to it you know at this point I really would like to pass the torch on this subject I've worked on it for some time I made my contribution and for various other reasons in my life i don't have the time to write the whole thing myself but there's enough people in the world who are interested in this and need it but i'm hoping other people pick it up

um so there's a fellow uh junglee on twitter who is now um hell-bent to write a spec for this get a prototype implementation and that's where we stand today okay so i want to dive more into like the specifics of this this is a very interesting concepts and i didn't realize pita pool was its own blockchain yeah it is and so like instead of getting instead of producing hashes and attributing hash to this blockchain instead of getting a reward and a token you're getting shares in the payout of the bitcoin block reward so i guess the token would be that's right a claim on future shares that's right so there there isn't necessarily a native token on these blockchains with the separate transaction system. So P2Pool did not

have a native token, right? So it wasn't a shit point, which is one of the things that made it great. But instead, it simply did accounting for shares. And its only native token was Bitcoin. One of the things I have proposed for a future potential break pool would be to actually have a native token. And that native token would be proportional to the number of estimated hashes contributed in a share and this enables a number of things it enables different miners to mine at different rates without overwhelming the system by sending too many blocks it allows derivatives because this this enables hash rate derivatives because one token is essentially the hash rate the other token is bitcoin so i can do you know d bitcoin d hash rate and create futures and

options out of that which is exactly a financial instrument that miners really need so would this be like a is this like technically a sidechain or is it completely separate like i'm trying to no it's not a sidechain um it is it is a merged mind blockchain right so if you're familiar with like uh dogecoin and litecoin so dogecoin litecoin are merged mine so the same mining hardware can mine both at the same time uh so this is similar in concept the same mining hardware mines both the blockchain for p2 pool or for brady pool and bitcoin at the same time if so the block header that it's mining is always actually a bitcoin block if it doesn't satisfy the difficulty target for bitcoin but does satisfy the difficulty target for the braid it can become

a node in the braid right but not a bitcoin block and then sometimes those blocks are actually bitcoin blocks um and there are rules about what has to be in that bitcoin block to pay everybody involved on the on the break okay and those rules are enforced by the blockchain rules not the bitcoin blockchain rules and practically the way it appeared in p2 pool was ruled about what had to be in coinbase in other words your coinbase had to contain um all of the address from all the other people who submitted shares in proportion with the right calculation of of satoshis they were owed right it's a simple way to think about it that's what yeah that's actually another question i had like illuminating the old coin bases with many addresses like how

like these miners in the this decentralized pool are sending shares they just attach an address to every share or every block okay that's right all right which is what you do the mining pool anyway yeah yeah all right and so i guess again to dive deeper into the conversation that you and eric were having with the trade-offs with this right like you already described some of it with the proximity like i guess the big knock or question that needs to be answered is latency um and profitability for smaller miners versus using a decentralized pool versus larger miners using a more centralized option? Yeah, so that's the problem I think I've solved.

There are plenty of unsolved problems. But using the directed acyclic graph, allowing miners to essentially choose their own difficulty so that rather than producing five blocks within a time period, they produce one block. What you can essentially do is set the braid up such that everybody targets the same variance rather than the same block rate, right?

And if you look at any pool, they will scale you down. So if you're producing shares too fast, they will double your difficulty and double again, double again until you're producing shares at a rate that doesn't overwhelm your systems. So you can do the same thing with the grade. And what I've showed is how you essentially add those together. So two different blocks with two different target difficulties

can be added in a reasonable way using a directed-discipline graph. So that should be solved, and I'm happy to contribute to something you want to carry this forward. plenty of other unsolved things um like you know what is the native token how does that work does it have a transaction system um do we do atomic swaps so that people can essentially instantly sell their shares for bitcoin uh through market makers on exchanges um there's there's lots of

interesting problems itself fascinating and again going back to the way pools are set up now and the way stratum v2 if adopted widely would work you said in the talk of the bitcoin munich meetup the stratum v2 improves in a lot of things just like the payout mechanism is where everything's hung up yeah yeah i mean stratum v2 fixes everything except for the fact that someone owes you money right and you don't know if they're doing their accounting properly you don't know if you can sue them if they fail their responsibilities um you know you don't really have any recourse uh and they are a target for potential attack okay and so like what if you could open source the the payout mechanism right

and they cannot have it that's the goal i mean there has been open source um pool software in the past uh but you know you never know what's being run behind the scenes um at a centralized pool right by using a blockchain those rules of who and how how you paid are in the code itself you're running it everybody else is running it um and if they violate those rules you know your node is going to reject their their beads in the braid if you will um because they didn't follow rules it's the same as bitcoin right it's the same way we enforce rules of bitcoin yeah yeah um you know those rule enforcements are completely decentralized right there's nobody that's holding your money uh the distribution of funds is actually coming from the blockchain itself

there's no entity who has keys and you have to trust them to do something with those keys in order to be paid yeah that said there are designs um uh chris belcher has uh done some work kind of um jumping off of my grades work um where he has uh what he calls um uh what are they called these kind of federated nodes who actually handle the payout which is another potential solution um anyway it's worth looking at his work as well uh if you're interested in this topic but i would i would prefer to have all the payouts happen on chain so you can use the blockchain to enforce them yeah and i guess the biggest hold up there is like if you you have an individual miner within a decentralized pool that constructs the block like how can you guarantee that they pay

out to the other miners contributing to that decentralized pool because he has to prove it with every block he submits okay so uh let's say i submit a block okay um so i'm mining a block that didn't hit the bitcoin difficulty target i did hit a lower target required by the pool now not only do i have to submit a bitcoin block header that contains this lower proof of work target i also have to submit proof that within that block are payouts as required by the pool so in the case of p2 pool what that means is not only do i reveal the header i also reveal the coin base in the transaction merkle tree and that has to contain the addresses and direct payouts as dictated by the by the p2 pool blockchain so if i don't submit all of those things it's not

considered a valid share okay and my my share is rejected and i won't get paid for that share that makes a lot of sense and so like if this is successful if like a successful and working grade pool implementation comes to market we can end up with like one decentralized mining pool essentially correct or would yeah would there be that's the goal i mean there's really no like 51 attack risk with this by having one pool um you know everybody could be on this pool safely uh without creating any risk unlike with centralized pools where you do have this if any one of them gains um 51 it would be bad yeah are you optimistic that this can come to market

i i am optimistic you know i haven't been paying much attention to it in a while um the bitcoin munich guys asked me and eric to talk about it so we did um and you know i think this has sparked a bit of interest um and so i'm happy to see that interest um including from you know potential people who might fund this so i'm beginning to think in the back of my mind that i might be able to put some grants together or something like that um if that's something that interests you either from a funding perspective or from a receiving funds perspective to build this um if you talk to me personally i would love to pull that together because i don't think I can do this myself but I would love to facilitate other people creating it yeah so I

mean it's important and you're speaking to somebody who's been really championing stratum b2 disclaimer brains is sponsored the podcast but as you should yeah no as you should right I mean stratum b2 is a step in the right direction it doesn't solve all the problems but you know you got to start somewhere and everybody should be using stratum b2 we'll stop right there's no excuse for using strategy one anymore but there are remaining problems that we would also like to solve yeah and i think just for any newcomer who may be listening to this podcast and isn't i mean we've already talked about it a bit but like how important is it that we ensure that this layer is sufficiently decentralized in the future like right now everybody's playing nice

and um we haven't seen government's pressure pools to censor but it could happen in the future and what about jurisdictional arbitrage that's something i wrote in the newsletters today like it would have pools are sufficiently domiciled across the globe and one government says hey uh we're going to force you to censor some transactions and another government says hey we're not going to force you to do that so point your hash rate at a pool in our our borders so can collect those tax revenues from the mining pool well it's happening right i mean um we already have uh the ofact the office of foreign asset control has you know listed bitcoin addresses that are you're not allowed to do business with basically now um by and large i don't think miners

are censoring those addresses but if any service you deposit from an address like that they they have to take action they're legally required to take action now does that mean miners have to do that. Not necessarily. Now, that said, there are a couple of companies who have announced they're going to make pools that are more restrictive in various ways. One of them, I believe, said that they're going to make KYC of the miners required. And the discussion is often had about whether those addresses that OFAC put out are going to be put into blacklists by miners. And the fact of matter is, it doesn't matter. There's always going to be somebody out there that's willing to mine those addresses. And it doesn't matter. In fact, if you're interested in law enforcement,

you're interested, and you know, for clarity's sake, these addresses are truly despicable people or people involved in, you know, various kinds of crimes, and you know, it would be nice if they could get caught. Well, the way you can get them caught is if they actually deposit those funds at a service provider in a jurisdiction where they have committed the crime, and then those funds can be seized, right? If we don't allow the transactions to be processed at the mining layer, they can't move the funds. They can't move them to a service provider, and they can't be seized. So from a law enforcement perspective, it actually makes sense for miners to continue processing those transactions, because that's what allows those funds to move to places that

law enforcement could seize them very good point it's actually one i haven't heard before but it makes a lot of sense i guess the whole blacklist thing is like we don't want anybody profiting from from the movement of dirty funds that's disgusting yeah i mean blacklisting you know if if blacking excuse me if blacklisting was 100 effective what you effectively done is destroy the coins you destroyed the money right um that's not necessarily the best the best option you know watching the movement of funds uh involved in crime is itself evidence right um and being able to see what they're doing with it uh you know it's a way to gather information now we should all be using z cash we should all be using uh you know things that have more privacy uh but that's not the

world we live in with bitcoin today um we'll get there but slowly um but you know it frankly doesn't make sense for miners to be censoring i these pools as i understand are focusing mainly on KYC-ing miners so that other pool participants are confident that somebody who's on the OFAC list is not mining alongside them.

I don't think it matters that much, frankly. Go ahead, do it. It makes businesses comfortable. It makes businesses comfortable. It grows the industry. That's good for all of us. Agreed. Agreed. There's always

another rabbit hole to fall down in the Bitcoin in the bitcoin space the mining pool layer is one that many individuals don't think about um too often it's one um we're starting to think about more great american mining again like as we expand here in the u.s and you see publicly traded mining companies posturing like you like you mentioned with kyc individual miners that's something we want to stay as far away from as possible yeah me too but the fact of the matter is you know when you get one end of the mining economics is a publicly listed company and the other end is an auditing firm and the third end is a uh you know custody provider like i previously worked for fidelity they're all going to ask this

question like we have to make sure that we are not facilitating crime right that's their legal obligation and they have to do it right uh you know the the fiat governments of the world have co-opted financial institutions into uh financial law enforcement now i don't think it should be that way but that's the way it is and that's uh what they have to do so again um do the things that make them comfortable but at the same time building something like uh grade pool if we build something like grade pool it has to be demonstrably better in essentially every way right so on the financial side of things you remove counterparty risk that's a big benefit to these uh these big players who are getting into mining now um it removes the possibility

of government attack because there's no one to shut down right uh if you introduce native coin um you have the possibility of creating hashrate derivatives which is demonstrably better for everyone um these hashrate derivatives are a new market new liquidity new volume right um we need to glue all those things together right we can't end up in a situation that p2 pool had where i mean p2 pool failed um because it wasn't better uh it was worse and making something that's worse but more decentralized is not going to work it has to be better and one of its side benefits has to be that it's more decentralized agreed well hopefully people start thinking about this more because it does feel like the

pressure is going to be heating up pretty significantly as bitcoin crosses the one trillion dollar market cap like level it seems like there are i think there are six companies have now filed for etfs in the u.s this year the most recent announced yesterday was fidelity my former employer, filed for an ETF on the back of Fidelity Digital Assets, which I helped put together. Very proud of that. Congrats, guys, if any of you are listening. But I also worked for one of the previous ETF sponsors, which is Solidex, who later partnered with VanEck. So I was there when the ETF got thrown down twice. And this is going to be a point of contention for all six of these ETFs, right? You know, the SEC in approving an ETF is going to ask very hard questions about

ensuring that money laundering isn't going on, or that the ETF provider is doing it. So, as you said, pressure's on from many different angles. If we get an ETF, that's a pretty substantial source of pressure. And privately, you know, there are people getting involved in investing in mining who have more restrictions on their ability to invest. You know, we're moving away from hedge funds and private money invested in Bitcoin to public institutions, public funds investing in Bitcoin. And they have a much higher auditing and reporting and due diligence requirements than private funds.

So, yeah, the pressure is on. What if we get, if we use the First Amendment, can the First Amendment come in here and protect Bitcoin, particularly like miners and ETFs? It's just a messaging protocol. in speech no no no you don't think so why not um oh i'm not a lawyer don't ask it's a complex question that's something i push for it's like right it's just the messaging protocol at the end of the day like how can you how can we don't how can we don't go after the telecoms when two terrorists uh basically collude to to attack

something over their networks i'm not the right person to ask this question of not my expertise but i will say that um the courts have ruled over and over again uh and refused to to dip into campaign finance reform and the argument is usually that money is a form of speech and your ability to contribute to a political candidate is a form of free speech um so you know maybe there's a path there i don't know uh again not my my specialty but i think uh it it doesn't matter you know if this thing is facilitating crime it's what they feel um without knowledge by the way um you know the rates of crime are much smaller than rates of crime within the fiat industry but um you know if it does facilitate crime then they're going

to do everything they can no no legal hyperbole is going to stand up to an actual threat right yeah i feel you there it's the fire in a crowded theater argument right um yeah you have free speech but you know at the point that your free speech is harming people uh that you know your right ends yeah the battles are just beginning freaks it's uh like yeah that's the other thing like who like who's willing to stand up like or should we even depend on people being willing to stand up or should we make it so robust that you don't even have to worry? Well, what I want to do with Braidpool is not force people to have to stand up because no one wants to put their life on the line, right? No one wants to go out on a limb and say, I think decentralization is so

important that I'm going to put my time, my capital, my life, my risk of prosecution up for this goal i want to get it as a side benefit by providing something else which is demonstrably economically beneficial right uh you know if your if your primary goal is you know decentralization and freedom um it's a hard sell you know most people are short-term thinkers uh they're looking at their bottom line they're looking at their profits it's a lot easier to sell improved profits than it is to sell uh high and mighty goals of you know resisting the government yeah right we are short-term thinkers but i guess that segues into an interesting question which i always wonder like what got you into bitcoin originally being a physicist was oh

um well so uh yeah prior to physics um i have a whole other career in theoretical physics sorry part of bitcoin i have a whole other career in physics um so yeah i got my phd uh university of wisconsin and i held postdocs at um uc davis uh cern and switzerland and then university of heidelberg in germany and um so this was i graduated from grad school in 2003 i was at cern in 2009 when the bitcoin white paper came out and you know i've always been a software geek uh you know long before i went into bitcoin i was writing a lot of software i science project in 10th grade or specifically 10,000 lines of C++ code in like 1993. So you know I'm self-taught I used to write a lot of software and then when I once I got into physics

basically you know every single one of my projects revolved around software and involved a lot of software one way or another but the output was always some graphs you know in a paper rather than a software product but you know going into software after physics was always kind of a natural thing and um i was also kind of always a crypto geek you know i had a pgp key from very very early days was always very frustrated that there was nobody i could send email to you know i think i had a couple friends that i you know exchanged pgp keys with back in the day and so i always kept an eye on on this kind of technology um and so when bitcoin came out it caught my attention like right away i have friends from when i was at cern who swear that i was

talking about it in 2009 i think i may not have discovered it until a little bit later than that i think i i think the first slash dot post is when i discovered it but it may have been earlier i don't know uh in early days it didn't pay that much attention but another thing that was going on was that um you know i was living in switzerland and germany and i had accounts in three different countries i had you know three different currencies in my uh my bank balances um and i basically started fooling around the foreign exchange market because i'm like well do i keep my money in swiss francs or in euros or dollars i don't know so i started fooling around and um taught myself a lot about finance uh in in those years and you know when i uh when the writing on the

wall became clear that i wasn't going to get a professor job in physics uh and i started looking for a job you know doing something in software was obvious and doing something with bitcoin was obvious um i did do a bit of mining way back then so i mined with my gpu back in like 2013 i think i think i mined 0.1 bitcoin and at the time the um the price of bitcoin was 30 dollars so i you know for a week turned my computer into a noisy space heater uh and earned three dollars and i said screw that screw that i'm not doing that um i should have kept doing it not um but instead I went to work for some companies in the space. I got a software job working on Bitcoin in 2014.

And so that has led to several other jobs in the space. I don't regret it for a second. It's been a wild, interesting ride and a very important one, I think. Yeah. I mean, I'm thankful that you found Bitcoin and got into it because I think you're a great thinker in the space. I think this conversation around Braidpool proves that we need adversarial thinkers who are thinking about the long term. I know the price. Yeah, we do.

That's one of my, I don't know, long-term frustrations is that I have this kind of thesis that we're going to end up with a discipline, which I sometimes call cryptographic engineering. And, you know, Bitcoin really is a form of cryptographic engineering. You know, Satoshi glued together a data structure, which is a blockchain, which is really a linked list, with some signatures, you know, which is cryptography, with some hashing for the mining. They took all these cryptographic building blocks and glued them together. And this really, really requires, as you said, an adversarial thinker. You've got to think about all the ways this goes wrong. And most software engineers are not adversarial thinkers. most people write software to get it to the point that it works and they don't evaluate the edge

cases where it doesn't work and you know in in this industry you have to evaluate how it won't work first because that'll be the you know that'll be the barn door through which the truck comes and destroys the whole thing um and uh you know finding that adversarial thinking is is rare you know i think all of the core devs have it but you don't find it among the general software engineering population i would like to see that developed kind of discipline that's taught in college i don't know if it's really possible to teach adversarial thinking um you know i kind of found it on my own you know in the physics world uh when you propose an experiment you have to then go figure out like what all the different ways are that uh it might not be measuring the

thing you think it's measuring and it's very adversarial and then you go give a talk at conference and people say oh that won't work you know it's a very adversarial environment so i kind of academically grew up in such an adversarial environment so switching to bitcoin was was kind of natural to me but that adversarial thinking is is really not most people unfortunately i would like to see that developed into a proper discipline well could it be distilled even further to simply critical thinking right no it's more than that um it's expecting the unexpected it's uh you know trying to figure out how you know it's murphy's law right that that which can go wrong will go wrong don't you know prioritize figuring out what might go wrong it you know you're you're doomed

everybody has a plan so you get punched in the face i think mike mike tyson really distilled that yeah very exactly very succinctly creating something that works in the space is not enough you have to create something that doesn't not work there's a completely other world yeah well if you're listening out there in your developer adversarial thinking it's important and if you're a bitcoiner in general who's in it and very excited about the gains these gains can be diminished if bitcoin can be diminished and bitcoin can be diminished if we don't have enough adversarial thinkers so yeah how uh i mean you mentioned that you'd be interested if anybody's out out there listening who'd be interested in uh contributing to to taking the torch and running

with braid pool where can they find you how should they reach out is there anything i think there's like a free note irc channel correct uh yeah there is an irc channel on free note um you're welcome to join it's a good place for discussion um you can reach out to me personally on twitter I've been thinking, you know, since this conversation got started by the Munich Bitcoin guys, I've been thinking more and more that I should probably orchestrate a way to get this developed and in particular, a way to get it paid for. Because I've spoken to a number of people, including some very large miners, who are very interested in this. And I have a few ideas for how to essentially fund its development. And that's been one of my challenges is like, okay, so I could dump a year or two of effort into this. without getting paid and i just couldn't afford to do it right um and nobody can afford to work

free and that's kind of the tragedy of of open source and blockchain software is that uh you know satoshi put something out there as a public good the fact that his coins made billions was you know is kind of an accident he didn't know he put you know years of effort into this before launching it he couldn't he couldn't be sure that it was going to actually make money um so yeah uh you Most of the core developers are often struggling for funds. And so those of you who have the capital, please look into finding ways to contribute to them because they are working for the greater good, as it were. And this is a similar case.

This is really a blockchain on par with Bitcoin. Similar amount of effort to build and similar amount of people need to eat. People need to be funded. So I'm saying this here for the first time. i would like to orchestrate funding these people um i will try and organize a series of grants um i would like to collate people i won't take any money for this i won't get paid i don't want to get paid um but i will help orchestrate uh getting money to the people who are working on it and evaluating those people who are working on it and deciding who

really are the best contributors um to bring this forward awesome how big of a technical lift is it think uh it's big yeah you know some of the some of the building blocks are already there so in the munich podcast or meetup we were talking between myself eric and uh junglee uh about using lid bitcoin so eric bosco has written the bitcoin which is a bitcoin clone um and uh it has the property that it can run multiple blockchains at the same time you know the idea of like recompiling the code to run Litecoin when it's only like a one-line change is kind of silly.

And so he put it all as config flags and he could run more than one at the same time, which is something that's required for Braidpool. You have to track not only Bitcoin, but also the Braid. And P2Pool does the same thing. It was a piece of Python software, but it had to connect to a Bitcoin node

so it could track it. So I think the two of them are very interested in this and would like to use LitBitcoin. I'm not opposed to that. You might connect it to a Bitcoin node instead tracking that one she do have track two and i think eric has done some very very smart things as far as uh organizing live bitcoin when you re-implement a piece of software um and refactor it you know the bitcoin's written from scratch you don't have to make all the same mistakes

and the fact the matter is that uh bitcoin core code base is kludgy and ugly and old and it's it's grown organically um and you know historically what happens is people rewrite these things you know we aren't still using netscape got rewritten into chrome opera and microsoft edge right each time that that happened um it got a little better and mistakes of the past were able to be left behind uh it's a big effort to rewrite something like that but uh in this case it's already been done the change would have to be made to the bitcoin to track a directly secret graph is not that large probably talking between 100 and a thousand lines of code um and everything else is basically kind of standard bitcoin transactions signatures all that kind of stuff that's already

there so um i hope it's not that big of a lift but i would estimate at least a year i mean in bitcoin terms that doesn't seem like that long of a time right yeah maybe you say that now that now that becomes what's old well like would this need would this need any changes at the protocol level i know you're mentioning lib bitcoin i know with core you need for shroud and v2 if they want to do the job negotiation part they have to get something changed there it's not consensus code but it would need to merge at the core level when this be any different would be for it to not um require any consensus changes right uh i don't think there's any anything that actually requires consensus changes which is actually kind of

interesting from a prototyping perspective because um i've spoken i've spoken to developers from like monero and litecoin are interested in the idea too um you know we could make braid pool for litecoin and that would be kind of a great test bed for it for pouring it on bitcoin you know the difference between litecoin and bitcoin is is minimal it's just the hash function really and the name a couple numbers change um so not requiring those changes in the parent blockchain is important and i don't think it's hard yeah all right well freaks let's get on it let's make sure we're thinking adversarially and making sure bitcoin is robust as possible right because right now what 93 of hash rate is dominated by pools in china is it that high oh boy i think

i think so yeah i think it's that high but that what you were saying earlier like it would take like that's one thing in grand american mining like we we have prepared like we need to switch pools our head engineer austin just thinks we can do it within 10 minutes he's pretty certain like we need to switch like yeah it's perfect you know in lieu of bitcoin or sorry braid pool and you know before something that exists you know miners need to be able to switch um quickly and that's you know in many ways it's kind of just a script you have to write on the back end it'll do it uh but there's authentication and you know pool authenticating issues and things like this my experience has been to the miners that i've talked to that they are actually pretty sticky

with their pool they don't like switching pools um now i think this has changed somewhat you know Well, back in the day, we used to use what was called PPLNS, which is pay for last end shares. And this was essentially what it did is it averaged over the last end blocks and then paid people for the average of that contribution during that time period. Today, most schools use PPS, pay for share. So you submit a share of work and you essentially immediately get paid. Now, this introduces counterparty risk because you don't know if the pool has the funds, and that's a very important question. But it's more immediate, and in particular, it makes pool switching a lot easier, right? Because if I stop contributing hash rate, switch to another one, it doesn't affect my bottom line.

Whereas in the past with PPL&S, one of the reasons we used PPL&S was to prevent people from pool hopping. because uh on some of the smaller coins that got 51 percent attacked um it was done via pool hopping uh there were you know things like nice hash right so people take a look at all the coins out there and say which one's the most profitable mine right now and they would switch very quickly to another pool or to another coin to mine it and um in so doing you could you know increase the catastrophe on one coin by a factor of 10 in seconds you know all of a sudden you have a 51 percent attack on that coin accidentally just because people were following a profit um but anyway this largely today this largely doesn't exist people are using pps and that means switching

pools is easy so for the miners out there be prepared to do it yeah be prepared and hopefully if break pool comes to market we won't have to worry about any of this yeah it's what we should be striving for so thank you for again starting this work in 2015 and then shout out to the the bitcoin munich meetup for for forcing a conversation about it recently and hopefully this podcast pushes the conversation a bit forward and we can make sure bitcoin is sufficiently resilient to these government attacks because they will be coming they will be coming governments are not going to take uh you know uh removal of their fiat hedge money and the the powers that gives them flying down right they're going to fight tooth and nail yeah so yeah anybody who wants

to contribute to this uh you know please reach out to me bob mcgillrath on twitter again there is a braid pool irc channel irc.freenode.net i've spoken with junglee we'll be setting up a mailing list and possibly other communication channels soon um and if you are interested in having something that's created and are willing to put some some money where your mouth is to fund it please let me know also and I will try to connect you with some great developers who want to build it and get it done awesome I guess one last question before we leave just to be adversarial about this particular topic like what's the knock on this what's the knock what do you mean like why why could this be a bad thing um if at all why could this be a bad thing uh

um oh i can give several things but they're all fairly deep long conversations so like in terms of eric's uh latency um or proximity premium it can't be completely solved because you still have to post into bitcoin which does have a proximity premium um you can certainly make it a lot better but you can't completely solve it um another one that i like to think about sometimes maybe a benefit maybe a detraction is that um there's this concept out there of a feather fork and the idea is that uh the underlying blockchain like grade pool could become the main blockchain over time with the legacy blockchain

essentially becoming uh you know if if nobody's using everybody's running grade pool then grade pool becomes bitcoin now this has been proposed under the under the name feather fork as a way to radically alter the rules of bitcoin right it's a way to do a hard fork that without doing hard fork um you know it's kind of an opt-in mechanism for a hard fork you know grade pool will work if you know only 10 of miners are using it it doesn't really care but if it ever got to like 100 of miners using it then well what is the purpose of the the original blockchain if 100 of everybody's using the new one you've effectively executed a hard fork now that could be good or bad depending on your perspective um that's a whole nother can't worms oh my gosh exactly i

told you right worms i guess the number one question that opens up there is like do the full node validators on the the original bitcoin network like would they ever accept like would they ever go transact on that other blockchain well no but they and they wouldn't know about right again they don't have to know about very cool existing it looks like just another pool producing block to them right but you know if all the people i need to transact with are running breadpool instead and breadpool can carry bitcoin value in special transactions um in addition to its native sharecoin value uh i could trans i could transact bitcoin that way right it's it's similar to a peg side chain like liquid networks or the risk networks

except that because it's an underlying layer that produces bitcoin blocks as a side effect it can actually take over the parent blockchain that's fascinating that i mean i've heard the the argument like liquid lightning could like reduce the protocol layer of bitcoin to something like rye stones where nothing moves and everything happens on those layers sounds similar but a bit different and is that a big problem um i guess that's a discussion i would i have to know if it's a big problem i'm a big fan of lightning um a fan of liquid too to a lesser extent uh but you know i think that satoshi had it right the bitcoin's a settlement layer and the real value is going to be transacted on like like it yeah i agree i'm a big fan of lightning network i've

been using a lot and actually people are going to be streaming me sats as they listen to this podcast awesome it's a crazy world we're living in awesome well bob i know you're a busy man i think i appreciate you taking some time to sit down with me and answer my stupid questions today but i think it's an important conversation at all appreciate the opportunity and i hope uh hope we can get this off the ground i've been trying and failing to get it off the ground for a long time so i hope this time around uh people will reach out to me and we can really get this built yeah i know um i mean this bitcoin munich youtube conversation got thrown in our great american mining chat so i know our engineers are thinking about it and i think we'd be willing to help out

these efforts as well so we should have a side chat about that let's have a side chat awesome we'll um we'll coordinate that side chat all fair but i hope you enjoy the rest of your day uh and thank you again have a good one peace and love freaks