ReardenCode Joins Pleb Underground To Discuss rBitcoin

0:00Hey everybody welcome back to the pleb underground I'm your host Phil happy Friday uh great way to kick off the weekend guys I've got a very special guest Brandon Black or Reardon code you may know him as that on Twitter he has been he's gonna he's actually a return guest he's been on the show before a long time ago but it was time for us right it was time for another get together a lot has happened lots of things have changed he's working on a new Bitcoin client implementation and I am obviously very curious is

1:02is Brandon is Brandon attempting to become Luke 2.0 I don't think so I've known him for a little while he doesn't give me the Luke 2.0 vibes I'm sorry Brandon I know you're seeing this in your off stage can't say anything yet I don't know man you know we're gonna find out okay anyways it should be a really fun show we're gonna talk about his implementation our Bitcoin we're also gonna talk about some some quantum stuff as well but as you guys know before we do any of that fun stuff we're gonna give a shout out to the pleb underground sponsor and we're gonna be back actually you know what no wait before we do the sponsor let's do the numbers okay how about that right because we're gonna get this stuff out of the way we don't want to bug Brandon with the numbers so Bitcoin Fiat Exchange 79454 new all-time high on the block height 965496 three sats per V bytes uh-oh everything is still affordable that's not good not good for the shit coiners uh total public lightning capacity 3753 BTC that's pretty good 79 Bitcoin now sorry seven yeah sorry 79

2:05iPhones per Bitcoin damn that'd be crazy if it was 79 Bitcoin per iPhone I'd be be stacking hard right now and 20% of a house per Bitcoin guys that's right those are the numbers here we go pleb underground is sponsored by our comedies imporium discover the pinnacle of precision engineering starting with their Bitcoin logo wall clock meticulously machined in main from a solid block of aerospace grade aluminum ensuring unparalleled durability and performance experience the difference of a product made without compromise head on over to our comedies imporium welcome to the show Brandon how you doing man I'm doing good man how are you very good very good Friday uh you know we're not we're not in the 40s for Bitcoin which I know is upsetting to some people who are expecting perfect numbers so they could double their stack uh but yeah I'm a happy

3:07camper right now man what's uh so yeah that's right Rorck it is interview time and Brandon the uh the chat does get lively so so be aware maybe some comments that that pop in we are going to take a look at those so last time you were here you were uh and I didn't know this I didn't know this when uh when Walton got you on the show you were you were working at Swan um you were your programmer so since how long ago how long ago did you leave Swan and let's start with that yeah I mean it wasn't too too much longer after the last time we talked I think I left Swan in January of 25 um and and you know life got in the way for a whole year there so I didn't really do very much Bitcoin oriented for about a year and then starting this year in January I started uh kind of my own consulting thing so I I consult on Bitcoin projects you can hire me if you want to review your protocol or uh it helped design something or anything kind of wallet related I'm available to do that on an

4:13ad hoc basis now I rather than having a full-time job which is why I can also work on my own projects very cool very cool a Bitcoin consultant so would you classify yourself as a Bitcoin expert? well the whole reason I started the thing is because I realized that to my surprise uh I knew a more about Bitcoin than a lot of people who were trying to build things in Bitcoin and so I don't know if that means I'm an expert but it means I can help me fair enough I appreciate that that that's a uh that that's a humble take I'll take that um all right so that look it's good man humility is very important I think it stops a lot of people from getting wrecked I think that'd be a loop 2.0 as you were saying exactly exactly so let's dive in like we're not going to talk about the window stuff right away we're going to inch over through it so look let's talk about the the arbit coin thing um I guess what would inspire you to to make another client so it's kind of a

5:13weird story I had some consulting clients and and for those consulting clients I had to start learning to vibe code I had to start learning to use AI tools to do stuff and I was frankly shocked if I keep abilities of these of these tools and my brain has been percolating on things I would do differently than Bitcoin core since I started working Bitcoin and even before frankly um I've always been frustrated with kind of the way Bitcoin core sinks and the way it caches data there's certain things it does that bug me and so as I learned to vibe code and I learned to use the AI tools I realized in like a flash one I almost came to me like a vision I can build my own node I I'm not restricted by like the amount of code I have to type to build a node now and so then I started working on it and and and so so I have to kind of thank my clients for exposing me to the capabilities of these AI tools and then it gives me the opportunity to build something I've wanted to build for almost 10 years very cool hold on a second I'm actually just going to pull up obviously I'm going

6:15to pull up the website right can't just our Bitcoin dot org 100 vibe coded so so one of the things one of the things with our Bitcoin is that I have refused absolutely refuse to write any code myself if I can't get the AI to do it it doesn't happen even this logo is vibe coded yes the logo is vibe coded and everything is AI generated all right so like I guess we'll start before we before we dive into the the real uh the meat and potatoes of this this client's what what what is this logo so I see the Bitcoin I see the crab or lump yeah okay yeah so our our Bitcoin is a rust implementation so rust they they famously use crustaceans as kind of their their spirit animals so I have a crab because it's in rust um it's ours so ours for rust and ours for Reardon and if anyone's familiar with the Reardon Metal story he's all about building rail bridges from his famous Reardon Metal and so we have a rail bridge and so ours for Reardon,

7:20ours for rust and we have Reardon, Rust and Bitcoin on the logo dude it is it is all clicking that that is some fantastic uh some some fantastic sim uh what am I trying to say here like symbolic synergy uh but I can't put that on the fly that that that's a fill original yeah yeah see you just come up with a philosopher over there right this is what I do man that's what happens okay so look um okay who is this for like who is this client actually for like I know you said that you you you started to develop it because you were working with clients but sorry with you were consulting with with clients but who is this like is this for fill is this is this like a client where like I run this in a VM and I'm like super excited with all these extra features so I think the the target audience here is is frankly me but other people maybe are also like me um and so what I what I do right now is I run a Bitcoin core node and a electors uh kind of sidecar to serve my own wallet

8:21right that's how I verify my own Bitcoin is with Bitcoin core and electors and so I wanted something I've always been annoyed that I have to run both of these things they both have their own kind of costly indices their own separate processes they use kind of a lot of RAM um and I wanted to have kind of a single binary that would let me just serve my wallets straight out of it and so that's that's basically what our Bitcoin is it's it's a single binary that's kind of tries to limit it to resource usage and lets you serve your wallets out of a single thing so I can connect my cake wallet to it I can connect my um my bitbox client to it I can connect anything that speaks electrome to it sparrow all of these things can be connected to an orbit coin node over the electrome or explorer protocols um and I can serve my wallets that way with a single binary it has the as far as I know the smallest store size of any full archival node a full archival block sync in our Bitcoin uh without the electron indexes uh 690 gigabytes with the indexes 720 so we're below the one terabyte mark by a fair margin still with all the

9:22indexes whereas if you run Bitcoin Core plus tx index plus electors you're I think you're over a terabyte with those things you are um anyway so so I'm really trying to be resource conscious to serve my own nodes I run myself on a VM as it is those get expensive when you have a lot of storage needs um and that's what I was for is anybody who wants to serve their own wallets from a limited resource machine would you would you suggest that I test this thing out now or should I wait uh it depends if you I wouldn't use it in production yet I just yeah I just want to like let's say like I just want to install this right stick it in a VM mess around with it are we are we at that stage yet or no very close I have a release coming out probably next week it'll be the 0.6 release uh that's going to have like a lot of stability fixes a lot more test effort has gone into it I've I've done more benchmarking against different nodes so so I would say in the next week or two I'll have a release out that is ready for more people to experiment with uh on their own VMs

10:26as as of now it's a little sketchy it crashes pretty often that's okay yeah I just want to go back to you mentioned electors and it's just for the um my own edification but as well for the people in the audience that don't know when you when you say electors that that is the electrom server right like the core component of the electrom server is a third party server that serves the electron protocol so it's the rust electron server um electrom is the basis of the menpool electrom implementation and also the esplora electrom implementation so there's there's a whole ecosystem of different electrom servers out there uh I believe the first party one is electrom X or something and then there's full crumb and then there's electors there's a bunch of different electrom servers out there electors is the one that I happen to run the next my bitcoin node okay very very interesting so I'm curious though so is yours so is our bitcoin considered to be are you considering it to be an archival node it is and so one of the things that's different versus

11:28core there's no pruning mode it's always full archival okay so if you if you were pruning it uses more storage but if you were running a full archival bitcoin core node it uses far less storage than a full archival bitcoin core node I'm very curious how you were able to because I I mean like like I was just saying before so I run um I run electrom server on my you know my node in a box right and I recently had to upgrade because you know I had no choice I went over the one terabyte so I'm just I'm I'm curious did the eight did AI figure out where to create these efficiencies no AI in the first implementation that when I just told the AI to go build it it used much more storage and I had to then work down through the the details so so while I haven't written any of the code I am I am deeply involved in optimizing this and making it do what I want it to AI did not figure out these optimizations but but for example one of the things that that makes bitcoin blocks big is that every

12:34transaction input specifies the full 32 byte txid of its of what the transaction output it spends right so it's the txid plus output index in the transaction for every spent output well that means that we have that that same txid is specified many times in our blocks right because if let's say a transaction had a hundred outputs there's going to be a hundred later transactions that I'll specify that same txid to spend yet and so I I basically index all of those in far fewer bytes and so in the actual stored data we don't have the full 32 byte hashes ever-previous transaction we just have a pointer yeah you're pointing to the index yeah so that's an example of one of the things we do in arbitcoin to to make it more efficient in storage interesting very very interesting okay so so going back right like now you've you've made me feel better now I am somewhat confident that I can that I can run this at home so let me ask you is is this geared towards at all like a corporate customer like is this you know I'm guessing I what I'm trying to say is is this

13:40going to be for for for miners right is this going to be that type of or is it really just a client meant for a for a home user like who's I guess I'm asking what is who is the target audience for this client implementation and besides again besides you who built it for yourself but is this yeah sorry going yeah I'm not honestly sure when I when I first started building it I actually told the the AI we were not going to target miners at all um and then because in order to do serious testing you have to do a bunch of virtual mining right if the mine a bunch of regtest blocks in order to test the way block constructions work to make sure that your block connections and your consensus validation works in all scenarios um I basically was forced to bring in more of the mining functionality just to get the testing that I wanted so it as a result it can build block templates it can do that whether that's a thing people would want to use it for

14:45I don't know um the way the place where I could see clients wanting to use this like a you know maybe a corporate setting would be I think that we can serve more wallets like more electron clients in less resources than basically any other implementation out there you look at like I know block stream isplora in order to serve a heavy load of wallet clients they have a 900 gigabyte index on top of the regular blocks in order to be able to to serve clients rapidly they pre-index a bunch of wallets so it's a one key look up for for these large wallets um and I'm able to achieve something that's not quite as good performance but fairly close to his good performance without that heavy 900 gigabyte index isn't that like a bit of a tradeoff though if I if I understood correctly though the the 900 I mean to have a 900 gigabyte index I I'm assuming that some entity has to host this right so that's what block stream does on there for their isplora nodes they they host the bitcoin core blocks as well as the large 900 gigabyte

15:49isplora index and that lets them serve wallet clients extremely efficiently so you'll notice sometimes that if you query mempool that space for a large wallet it'll say too many GTXOs I'm not going to serve you that but a splora will serve you that because it has it pre-index and can serve those very large wallets well our bitcoin again can serve that without having the large index even those very large wallets you hear that block stream watch out I'm with you on this one man I'm I rather I rather count this client than a blockchain client but okay so I don't know if you can really see this or I have to expand it but this was actually something that you retweeted and I am interested in your thoughts this was from zender yeah I hope one day core and wallet will be two different repos that will not overlap and and I just I guess I just want to understand is he talking about like so that so that core just becomes like a a background engine almost like

16:56I this is going to be a terrible analogy but hopefully you'll you'll get it like it's kind of like .net frameworks right where .net frameworks is being used by the applications being built on top of it but the engine under it all is Microsoft's black box of .net frameworks yeah I mean I think there's there's I see where you're saying I think that if we were to go to that analogy we'd say that libbit coin kernel would be that .net framework but then there's like another layer of that where you you have I might even say that coin core becomes like the Google Play services it will use another bad analogy so it provides the service layer for the wallet you're going to build on top of it but no longer has the actual user-facing wallet functionality and that's something that frankly most other bitcoin node implementations intentionally omit a wallet functionality right so libbit coin has no wallet as far as I know sat d arbit coin I think btc d does have a wallets the one I can think of that does have a wallet like bitcoin core but but in general it's less common in

18:01other implementations to have a wallet than than not to so I think core has a very large maintenance task because they have a wallet and a consensus node in a single box and and I for for our bitcoin I kind of took a different approach where instead of having a consensus implementation and a and a full wallet I have a consensus implementation and a server that can serve wallets and I wouldn't use wise if over time core goes kind of in a similar direction where the core node itself is a server that serves a wallet and then the wallet is shipped separately for those who want it you actually answered my my next question because I was going to ask you does our bitcoin actually I guess host the the private keys or is it just talking to a wallet that you already use and that's exactly that okay so I think it also mitigates even a little bit of response you know responsibility on your end as well yeah well exactly so I don't have to have the same level of assurance yeah right it's more of like a coordination service okay hold on

19:05you get in questions here here we go good morning set stacker shots fine maybe problems okay don't worry we're going to get into the next we're going to we're going to start talking about his his possible Luke transition which I don't know what happened but this is why I'm bringing it up okay this one bringing it up because look you know knots was also started as an alternate implementation and look where we got okay now we've got a whole bunch of people to think crazy crazy thoughts okay we got a question here from ACP just got here I'm guessing our bitcoin is implemented in rust yes I can answer that one yeah I'll go a little deeper on it's all I can say is yes I did something that is controversial in our bitcoin which is that I did not use lipstick coin kernel or lipid coin consensus so I wrote the full consensus well I shouldn't say I I instructed Groc to write the full consensus script validation and block validation logic in pure rust I did not use an English screen library for consensus because I

20:06think one of the challenges of bitcoin core is the way that the consensus implementation is is spread in many files and I kind of deeply wired into all the C++ code so it's definitely not perfect in our bitcoin because you know I'm a relatively new user of clankers and it's a young project about two months old now but I think we have a better separation of the consensus logic from the rest of the data storage logic and everything then cordas and so I think it's valuable to have other full consensus limitations in other languages that are clear and and reviewable and so that we can do more comparison between different consensus implementations and so my clanker actually right now Groc is working on differential fuzz testing where it compares bitcoin core consensus to our bitcoin consensus to find any gaps between the two so I think that's a valuable aspect of our point is that it is a separate consensus implementation I got to say that this was kind of my first

21:09thought when I when you first mentioned our bitcoin I was like what effect isn't that a subreddit? yeah it is yeah no no you might I don't know you might you might you might end up doing a rebrand I'm not sure but I mean look I personally like it I think it's you know I think it's cool so going on this this topic of the the possible Luke 2.0 thing so I know that you wrote this client for you and also of course to help to help customers right possible customers but are there bigger aspirations? not directly I've actually talked to several folks since I started building our bitcoin about the benefits of having more different consensus implementations out there to potentially unlock future consensus changes you might think that having many clients many different consensus implementations would we get harder to change consensus but to the degree

22:12that we gain the ability to cross validate different consensus implementations we actually may be able to more easily change consensus similar to how lightning is able to change consensus with their bolts right now they do a thing where they they make sure that there's at least two matching implementations that all that both meet the spec in both interoperate before they consider making a consensus change with this coin we don't have that right now because we don't have separate and cleanly separate consensus implementations the test against but as we're as the ecosystem grows as hopefully our bitcoin there's another employee shop that called Hornet node there's one called sat D there's one call libbit coin I've never heard of who works on the Hornet node this guy Toby he's a he's a he's a pretty deep engineer he did an interview with Brink recently if you want to check it out about Hornet node and the Hornet consensus specification okay cool yes I want you to have us out there problem right like this is kind of the this is what you're talking about

23:12like I never heard of that one and there's another one you just mentioned that I that I never heard of thank you sat D like how long's that been what do they do yeah I think I mean sat D is exploring different data storage models as I kind of think they're their main thing they're less focused on the the consensus details and more on the data storage Hornet also has a unique custom UTXO database so one difference between different node inflations whether we have a UTXO database or not our bitcoin does not Hornet does core does libbit coin does not there's just a lot of stuff happening in the in the node space I learned about some of this after I started our bitcoin just to be clear like this is what it is right it sends you down the rabbit hole okay so I'm just going to ask you flat out do you think you need to save bitcoin no no no we're trying to open up are you not going to get coin it's all he's doing possibly he's not the savior of bitcoin okay so

24:14we're just we're just clearing that up okay sorry continue yeah well I think I many people go through a phase where they feel like they have to save bitcoin and and I did do when I when I kind of first started working on trying to push covenant changes forward for bitcoin I kind of felt like I was I was a guy you can help save bitcoin and then the longer the longer if you don't go crazy with it as longer you're here you realize that bitcoin doesn't need saving it just needs building and and so building more things opens up possibilities as opposed to saving anything dude you should you should tweet that out that's a banger here's like that that is a fricking banger like that's what it is bitcoin doesn't need to be saved it just needs builders this is exactly it's all right psycho train yeah he is he is a smart guy okay this is why I wanted him to come on the show plane and say the magic words that I cannot and if I do say them they make no sense whereas he says them they make sense because he knows what he's talking about okay so so look um

25:19did you think that I I don't want to do too much stuff on the blip one 10 stuff but I just want to at least ask you this did you think that it was going to blow up as big as it did did you think that it was going to go down the road that it did and did you think that we were going to end up with this you know this this total shit fork and the mess that we're in right now while we're not in the mess because we're just in bitcoin then then whatever the hell they're doing over there so yes largely it was it was very clear from early in that debate that they were they were going to eventually propose consensus changes they were going to eventually have to hard fork because or they'd have to give up and of course people don't give up so since people don't give up they had to go down this road I was surprised at how much support it got I was surprised that so many people fell into the kind of the rhetoric trap of mechanics sultry tones um but they did and that part really kind of sucks like that that bothers me I spent my time at conferences last year just

26:22having one one conversations with people that were sucked into the mechanic creditor rhetoric and at helping them kind of balance their perspective and and trying to help people not get sucked into any further because I think some people are are losing material money by by you know renting hash rate and selling their bitcoin for for a shit coin and and I hate to see that I totally I totally agree um ACP is just commented saying Reardonness politely saying that he's unawcifying the network by decoupling from course look that that's a very interesting question I've always I've thought about this and I am very curious as to what your definition of ossification really is I I believe that many of us have we've all drawn lines in the sand like we all have these lines in the sand where we believe this is ossification but then if I speak to you we don't have consensus on what ossification is and I'm sure that ACP who who made the comment he's going to have a

27:26slightly different definition too so just curious what this what does bitcoin ossification look like to you I mean I would say ossification would be that we stop changing consensus rules because we're afraid or unable as opposed to because it's it's good enough we don't have a good reason to change it and so you know right now it is we are not in a place where it's good enough and we don't need to change it we actually have maybe not the right change yet but we have many proposals out there for ways that we know for sure would improve self-custody would improve security would improve bitcoin in many ways even post-quantum as an example you want to talk about that later we have lots of things we can do to make bitcoin better and so we shouldn't stop improving it because there's concrete clear specific things we can do to make it better so ossification to me is that we stop changing it accidentally because we're afraid or unable and I agree with ACP that more

28:27clients out there that clearly specify the consensus rules helps us to keep making changes that improve bitcoin interesting yeah because I'm never really sure because for me it's like okay so I the hard stop for me is is the consensus rules right like you know the hard cap and everything like that like it's like it's to me these are like the foundational pieces however I do understand people wanting to develop right right you want to continue to work on bitcoin you quote-unquote want to continue to build however there's other people who have of course drawn their line in the sand even further out right and it's it's not just consensus it's more it's also ossification also means policy right which is kind of what we saw with you know with the blip-on-tens stuff so I well policy doesn't really exist is the problem and that's like the whole thing they made

29:27the mistake is though it's a real thing right but it's just not it's an accident of protecting nodes from DDoS and and kind of confusion of the history of bitcoin you know because like Satoshi was actually confused about how bitcoin worked he had that first-seen safe idea that that a transaction in the mempools already safe even though it's not in the block even though Satoshi himself had written the consensus rules to solve the consensus problem of coordinating what's valid and what's not he had this idea that it was kind of pre-valid and that that just doesn't exist that whole idea the pre-valid doesn't exist only blocks exist and so that's an important thing it's true it's and it's so crazy that that people were able to get confused and again I I do believe that it has to do with the emotional the emotional tugging right it's it has nothing to do with actually understanding how bitcoin works it's just like well that I don't like this and I feel a certain way about this you know so yeah um okay so speaking of which right

30:27because okay obviously you know bitcoin is captured right big bitcoin is here right we're all we're all freaking screwed and and bitcoin is dead and on that note on that note the next battle front quantum right okay so look you had very strong opinions on quantum already several months ago I remember seeing all already you were like debunking a whole bunch of stuff so I'm just going to ask you we're going to start with this question is quantum actually a threat to bitcoin let's say in the next six months to two years based on what based on the general understanding of where quantum is today the next is what you're definitely not definitely not okay so next question based on based on that okay um so how do I word this um so how to figure out how to word this so it's not a threat do you think

31:35that bitcoin is going to require a hard fork in order to solve the eventual quantum issue that we don't really actually know is going to happen but there's a lot of fun around it and I know that people's projects their their financial livelihoods are are writing on it happening what are your thoughts yeah so not a hard fork for sure we can absolutely solve any quantum issues with a soft fork there actually a lot of soft fork proposals at various stages of readiness that that would improve the quantum situation and and and some are kind of more step-wise some are kind of all at once but there's lots and lots of work out there um I would say check out recent issues of the uptech newsletter I help write to the change in consensus portion of that and in there we talk a lot about the work in quantum readiness so so there's that and the other thing that I wanted to say here um you mentioned like ossification around consensus versus ossification or other parts of the software

32:36yes um as part of talking about quantum I want to also draw like a broader distinction where there's there's actually different levels of consensus rules with it right we have and actually a hornet node that I mentioned kind of draws this line pretty clearly we have consensus rules that are kind of the absolute the the invariance that nobody's going to change and the only person talking about it is Peter Todd because he's Satoshi what uh and what which is the things like the supply no of course not um uh the supply schedule the the the the the the coordination around the difficulty adjustment things like this are things that are never going to change but then there's like script validation rules those are a whole different level of consensus and if we add some additional script validation rules we're not we're not breaking the foundation of bitcoin the way we would be if we changed the difficulty adjustment so so we need to not say like consensus as a whole we need to say well there's certain rules that are absolutely already ossified and then there are

33:40other rules that are in consensus but that we're probably going to change whether it's for quantum or whether it's for covenants or whether it's for um the the consensus cleanup bit 54 there are lots of smaller detailed rules in the consensus rules that we are going to change and I say that as a as a definite we are definitely going to change them so hold on a second you just mentioned the the consensus cleanup bit 54 is that the same thing that rusty was working on or that great restoration restoration versus cleanup yeah yeah I have no idea and to me yeah are they the same thing totally so I got that wrong okay so what's the consensus cleanup what's bit 54 if you don't mind me asking if you can you know give like a because I I kid you not I have been confused uh understandable there's a lot of acronyms that a lot of stuff out there yeah so I for me every time I hear the consensus cleanup I'm like that's rusty he's doing the work he's not doing it's not the idea

34:42it's not gonna but all right so it's a bit 54 stems out of work by Matt Carolo from many many years ago of identifying some non-critical but weird kind of wrong things in bichotin's consensus rules so like for example bitcoin consensus rules specify how the times of different blocks must relate to each other with this rule like it has to be no more than two hours in the past then it can't be more than so far in the future from one block to the next and but that rule actually allows for some really weird stuff if miners were to coordinate for example they could about 10x the block production rate okay they could actually like one 10th exit with the existing consensus rules and 54 cleans that up by tightening the rules around the the block times so it's like a very small change as of right now all blocks are already within the the 54 rules for the the time specification

35:44there's no reason to mess with it but if there ever were to be some weird thing where miners are coordinating to make more blocks we would want to have a consensus rule that protects us from that so that's one part of bit 54 another one is that right now legacy scripts meaning scripts of like the p2pk and yeah era can be constructed in a way that they take many many minutes like half an hour plus to validate on a modern computer and it's a valid block but it would take half an hour to validate one block on a modern computer and so Portland has has dubbed this DOS blocks and those are valid blocks but there's no reason for a miner to mine them right because the miner would slow its own block propagation and delay its next block if it did mine it so it's not an immediate threat to the network but it's still something that could mess with people's nodes and so we want to make a consensus change in 54 to restrict the ability to dos nodes with these weird blocks of legacy scripts and then there's two more changes in there that are somewhat less important but the point is it's it's cleaning up these weird edges of the consensus rules so that

36:49the network is more resilient over time is that actually is bit 54 actively being worked on yeah it's active in inquisition several miners are already enforcing the rules of it at hawk and and basically oh sorry since I said that wrong and merged directed me last time I need to be more precise several miners are already fully following the rules of it oh my god he is currently enforcing it I should say that precisely I always wonder about the the Brock the Brock the block propagation because it to me it's it's very reminiscent of inherent latency in a network right like there's something that that people like if you don't work in networking or IT you don't you're not really aware that like there is inherent latency in like essentially when you're just trying let's say for example you're sending a packet from North America over to to Asia or something like that right like you can essentially go to like the Verizon website and see all the inherent latency all around the world so to me it seems as though we right or you guys as you're as you're

37:55developing you have to still counteract that latency so maybe the latency would actually amplify this issue and make it worse because now you're dealing with a miner that has to do more workers or sorry it takes longer for them to actually propagate the block and then on top of that you're dealing with inherent latency in the network yeah exactly and Bitcoin has actually an excellent story around that latency right with compact block relay but well connected nodes can often receive and validate a block without any full round trips they just get the information from their peer and they validate it without any round trips back to that peer so that so a Bitcoin is actually very this is one of the things Bitcoin core has worked on over the years what's up it's less chatty yeah it's less chatty in the in the good case so just like things that happen in Bitcoin and why like relay policy doesn't exist so last summer we had subset summer I don't

38:56know if people remember this event on a Bitcoin X we had subset summer and subset summer was was this thing where even though Bitcoin core still defaulted to requiring one set per V byte for relay a bunch of sub-sat per V byte transactions were relayed and confirmed and this messed up block relay because Bitcoin core nodes using the default configuration didn't have all the transactions so instead of just getting the block immediately evaluating and confirming it they had to go back and say where are these transactions I don't already have because all those subset transactions weren't on the default Bitcoin core nodes yeah Bitcoin core was kind of forced to change their defaults to accept these subset transactions into their mempool so that we could get block propagation back up to speed I didn't actually know that those the subset's transactions were we're happening the only reason I found out was because I was posting like look at how cheap these fees are one set per V byte and people like no fill it's even cheaper than cheaper and I go take a look at mempool.space and I'm

39:56like what the hell point one I mean that's insane yeah yeah and so again this is this is demonstration that relay policy doesn't really exist only blocks exist and relay policy is forced to follow what gets into blocks to maintain good block propagation I want to go back to I actually want to go back to the the arbitcoin stuff because there was a question I wanted to ask that I forgot to ask you and I have to do with the differences between core and arbitcoin I wanted to ask you like you know what is arbitcoin actually supposed to do that Bitcoin core doesn't do because it seems like you guys are both doing the same thing but you're doing it in a slightly different way or the what's under the hood is different yeah they're really it's supposed to be a compatible consensus implementation you know and I worked hard with rock to build it that way that it speaks the same language on the wire as bitcoin core the v2 transport that is

40:57in bitcoin core we speak the same language we support most of the same features in terms of the appeared appear it's a it's a fully compatible bitcoin node it tries to do all those things and the the point really is the under the hood stuff so that it can combine in a single again in a single binary in a single data store the ability to efficiently serve many wallet clients and also be a full consensus bitcoin implementation without having to get a bolt multiple things together that's that's really the core or the central difference between the two is just that that integration of a multi-client wallet service and a core consensus implementation into one binary with with your arbitcoin have you added anything that had to do with with quantum quantum resistance or anything like that with it are you even looking at that now I made a hardline for myself that I'm not messing with any kind of consensus proposals until bitcoin other wise done so so in arbitcoin reaches my definition of done like when I get to a one point to release then I would potentially start looking at taking some of the consensus changes that are in let's say

42:02bitcoin acquisition and making an acquisition compatible version of arbitcoin but until until everything else kind of on my main punch list is done and I get to kind of a one point of release I'm going to leave consensus alone completely when when do you think the other one point over leases it's so hard to win you know because this is because this is a this is a total just whimsy project if you will a flight of fancy you know I think it's useful I think it's good to have it out there but but how much time I spend on it depends on how I feel that day so but I can say one point oh could happen in the first half of next year perhaps okay this is very interesting I like that I like that vibes vibes capital huh look at that I think that's uh that that's how the vibes today they're good we're releasing um exactly is is saying is it fair to say that rear-end and attempting to build a consensus spec that future clients will one day build off of

43:03to some degree yes so so the person is really doing that specific thing is Toby over at Hornet node um but I I wouldn't be surprised if there ends up being some collaboration there over time he and I do talk on X from time to time um I love his work I think he he likes what I'm doing as well um and we've each tackled two different parts of the kind of the storage problems in bitcoin I focused on the block storage portion he focused on the UTXO storage because I don't have a UTXO set in arbitcoin um but I think again there's there's kind of a synergy there uh I had my clanker this morning making sure that every one of Toby's English written consensus rules has matching tests in arbitcoin um I wanted to I wanted to make sure that we're in consensus between our two nodes at least based on his specifications I just want to go back because now I just realized what you just said and you just said there's no UTXO set so now I'm relating this back to an earlier

44:06comment you made when we first started is this why the node um does an initial block download faster is sort of so there's no set but you are validating against an index correct and the index is local or is it okay sorry continue so there's there's many ways to skin this particular cat not to be luke about it or anything jokes jokes okay so I'll compare four different bitcoin implementations out there core libbitcoin hornet and arbitcoin so core has a somewhat naive UTXO set and that UTXO set is a significant portion of the cost of cores initial block download um libbitcoin does away with the UTXO set and has a totally asynchronous validation for initial

45:10block download where it basically streams in the blocks in any order puts them into its storage and then later checks whether they all connect to each other correctly so that's libbitcoin I mean that's it just shifts the work time yeah we still have to make it fast so so libbitcoin right now is and the the two fastest ways to sync a bitcoin node are libbitcoin and uh hornet node and so hornet then has a UTXO set but it has an extremely efficient UTXO set much more efficient than cores UTXO set and it can also sync the blocks basically as fast as you can download them using UTXO set so libbitcoin and hornet are the fastest two ways to do it core and arbitcoin are not as fast however they have tradeoffs so so core has a a very simple off-the-shelf database for the UTXOs which lets it kind of reduce its code around managing UTXOs although hornets UTXO set is only a 3,000 line database it's still a custom database for UTXOs

46:15and then arbitcoin has a custom data store so it doesn't store roblox it stores this database format essentially of the blocks and that lets it have a much smaller data store but it syncs it about the same speed as core interesting so there's all these different tradeoff spaces out there I am not as fast as libbitcoin that wasn't a goal of mine uh my goal was more around the resource usage in once you've synced I just want to go back here to what you just said about core's UTXO set being the validating slower is is it because of all the features that core attempts to to incorporate into this one package right because they they have the wallet function no okay not to be a question but no no the UTXO set is more just um there are some tradeoffs in the way core is built in using an off-the-shelf database for the UTXO set it's not the ideal structure for bitcoin data and you can see that when you look at hornet node which uses a custom database for the UTXO set it can be dramatically faster because it takes the structure of bitcoin UTXOs

47:19and maps that precisely into how they're stored so that the lookups are very very fast Toby talked about in this talk he did with brink that the the lookup UTXO requires one page read and the importance of that is is just that um when the UTXO set no longer fits in RAM you're gonna have to go to the disk to get some UTXO inputs and in core you might have to look several places on the disk in hornet it's always one place any UTXO comes from one place on the disk the whole thing and that's just an example of where the optimized data storage can make the UTXO lookups faster and that means then that hornet does not need to have a separate coins cache you know that db cache flag we all set on core yep hornet doesn't have that and it's still fast and that that's always been like a bugger boob of cores is that that db cache need is because the database isn't very efficient on disk and that means you have this like double RAM usage you have the RAM usage that relate to the data on disk and the RAM usage for the coin cache fighting each other

48:19but it was if i'm not mistaken though right bitcoin core was originally written for windows right like as a windows client so that that to me right there explains all the inefficiency right there it's like they may as well have used not net frameworks you know like they may as well have died know it's a bad analogy but sorry that's the only type of thing i ever troubleshot that made me want to throw myself out of a window yeah i think it was i think it's not that it's written in windows that the problem although it relates because you know maybe because libbit older windows because libbit coins also first written for windows it's one of the two fastest clients out there oh shit well that blew that out of the water but it but libbit coin is written by by vosko who's like a deep export on the windows memory management and ios system and so he was able to optimize the crap out of it by understanding how the OS is going to interact with that data i worked with i worked with engineers that used to have to try to figure out how microsoft sink frameworks actually worked moving data around so if you've ever done like database synchronization and stuff like that

49:21you know that this thing is like a fucking black box yeah and what you're explaining to me about this guy it's like it sounds to me like this guy is needed but that's cool knows his windows very very yeah yeah no he he seems to i watched a i watched a talk with him about uh tail emissions and how he just completely destroyed it um and i was like look i didn't understand what he said for the first 30 minutes um all i understood was the last three minutes of this is why it doesn't work you're not doing it yeah yeah yeah we're not doing it and this is why it makes no sense and i'm like that made sense um okay going back to uh going back to the uh we're talking about the utxosat the blockchain so going back to arbitcoin though so if something goes wrong right in the initial sink with with arbitcoin can can an end user are we at the stage where an end user can can try to fix this themselves or or no and i'm pointing to myself as an end user but i'm not the typical end user because i am technical in my background this technical so i don't know

50:23somebody you know maybe slightly less technical than me i something like that nothing you can really fix yourself what i would say is that it's it's built intentionally so that when it screws up you can kind of resume you can kill the process kill dash nine and and start it up again and it should keep going from where it was so so from that perspective like yes if you see it screw up and you just stop the process and start it again it should fix itself all right we're gonna hold you to it when i try to break i used to do qa so among other things oh god i'm in trouble yeah no no don't worry you're not i don't have any of my automation tools or anything it's you know i mean i can afford the licensing for that stuff in my own you know so don't worry about it um okay i guess now a little bit more like light you know light kind of questions here because we're we're gonna you know we're coming close to uh to wrapping up has anybody um have any not that you have to

51:24mention names but has anybody come forward to help fund our bitcoin what has been the feedback if you can share yeah i mean i've had a couple of kind of small conversations about it um for now i've gotten a few small donations and i had some small conversations about it but but as of now it's just it's just me um my hope is that because it's a specifically useful client for serving individual nodes then it i mean i'm gonna use a term i kind of hate but it fits in with the show here i hope that it's only useful to plebs and that and that's some degree i don't need a ton of funding to keep working on it but if there's some pleb donations that come in to help support me it will keep me motivated to work on it very cool i'll have to see what we can do i have to see what we can do here uh that's uh that's good stuff i look i um you know on a different notes i i've always appreciated i've always appreciated your takes not necessarily right it's not about

52:27necessarily agreeing or not i've always found that when when you present something you do your best to be objective about it you do your best to actually talk about the trade-offs right and i think that that is i think that that's very important so i do appreciate you working on this i this is the other the other issue that i have with these implementations all the all the different implementations i still haven't figured out if multiple implementations for bitcoin is a good thing or a bad thing like and i know i'm just making it sound really general and stupid um but it's like i don't i just don't understand like in my mind it seems as though this is an open system it's an open system it's voluntary wouldn't it be good to have like multiple implementations to figure out all the ways that this could break so i do find it kind of strange that there's some people that have been in bitcoin for a very long time and they tell you that other implementations are shitcoins and i'm like how yeah it's a it's a it's a nuanced topic right there it is because we

53:33don't have a written spec for bitcoin consensus there's this risk that we end up with accidental chain splits and so then you actually could have like an arbitcoin node could be a shitcoin literally because there's an accidental chain split um so we do have to be very careful in how we build these things and and and that's something that i'm trying to be as very conscious of um James O'Baron proposed to me for arbitcoin that we take the actual same functional tests that bitcoin core runs and we run them against arbitcoin and so i've been working on that i've got 60 something of the core functional test scripts the ones that core validates itself against also running against arbitcoin because i want to make sure as much as i can possibly do it that arbitcoin will stay in consensus with bitcoin core now we may eventually move forward in a way where we make backwards compatible consensus changes and maybe those are coordinated on arbitcoin libbitcoin and hornet and core follows later core might not be part of the main consensus building in the future i don't know what the future holds but i think having a vibrant ecosystem of consensus implementations that that try to work together

54:38is is a overall good thing even though it is also a risky thing so we need to be mindful of those challenges i would say yeah i appreciate that and to that point i also think um more implementations i think would create an environment of i'd like to believe honest actors right good i i hate to use the the terms good and bad actors because an individual can be a good or bad actor depending on the situation but what i'm trying to say is that by having more implementations i believe that it keeps all actors or uh does a better job of keeping all actors honest what are your what are your thoughts on that absolutely that makes sense if there were if there were to be an attack you know the core maintainer ship is a relatively centralized group and if that group were to be compromised in some way i don't think they are to be clear i think they're they're they're good actors to put it in your terms at the moment um but if they were to be compromised there's there's two ways

55:40to kind of work through that one would be to fork the bitcoin core code the other would be to use an existing implementation and kind of re re regroup let's say it as a bitcoin network around another implementation as like the new core um both of those are valid and i just think both should be out there that we should have the option of forking the bitcoin core code or we should have the option of of a kind of regrouping around something else yeah yeah i agree i um you know i've always understood that that that competition is healthy right and from what i've seen it it helps to keep all the players honest um so and it drives innovation too like there they've been a about the right between between me and some of the core guys about the size of the tx index database and and they made a big improvement would actually drove me to make mine even smaller and and so there's there's this fun fun and and and i think positive competition between node and limitations you know libbit coin drives them to be faster and i drive them to be smaller and

56:44it's all good yeah i think you're absolutely right and i think that that's that's very important i know that in you know in the field that i was in right oftentimes uh you would see other products coming out with these like super slick ux innovations and you know what i mean now all of a sudden you got to step it up now i know with bitcoin it's right it's different it's not the same thing these are not the same type of polish products however right it does right like let's face it you know if you have another implementation that downloads right that does the initial block download 30 40% faster right as an engineer you're going to want to take a look at that and be like well what are they doing and can i implement some flavor of this in some way you know like can't maybe i can add this improvement to what we're doing and then all of a sudden maybe that happens and it forces another the other project to be like oh they just did something differently than we thought even though we did a good you know a good performance improvement well now this gets the wheels

57:46rolling right the wheels turning again um you're being asked you're being asked for a price prediction ACP you know what i'm going to give some serious hope you i'm after you give yours um can reared and give an end of year price prediction and don't be bearish all right so in other words he's asking for your prediction but he's telling you the prediction to me i want your numbers man but better not be bad but let me put a caveat on this which is that i don't bet often but i lost a bet this year when bitcoin did not get over 200 000 before the end of may this year so i it was a it was a friendly bet but i did lose a bet this year i i owe someone a flight to to meet up for coffee or something um so with that said um my my price prediction for end of year and i i i'm going to say that there's going to be a dip in December so i'll have a peak higher than this but we're going to end the year at 120 you know what i'll take that i'll take that okay i drew lines yesterday and i put out a

58:56post um and my prediction is not for the end of year but it is for july 1st 2028 the bitcoin price will hit 9,612,200,311 and 10 cents look i'm following the power law thing i just just i called wise i drew a line and that's where we end up that day so guys don't get mad at me so there you have it okay you can you can take rearedens 120k by end of the year and then in about a year and a half from that point you can get behind me okay there you go dude to always freaking awesome to have you on the show man i really appreciate what you're doing uh i love i love that you're doing our bitcoin and um i genuinely uh when i listen to people like you talk and explain what you're doing in bitcoin again it gives me hope the same way of like talking to like a gen zir about the future i'm like yeah you know what we're gonna be okay these these kids

1:00:01they get the right yeah they're all right you know so it's the same thing i speak to you you know i speak to merch i'm like you know i'm like we're gonna be okay these guys are paying attention i appreciate this um okay so we're gonna what we're gonna wrap up where can everybody find where can everybody find you find your stuff how can they bug you how can they donate guys these links are gonna be in the show notes yes to check out arbitcoin.org it's a 100% clinker driven site 100% clinker different different project um this is intentional i want to see the capabilities of these these new ai models uh that's part of the point along with the actual bitcoin benefits so check out that arbitcoin.org um you can find me on x reareden code also on other platforms the same name reareden code uh and hit me up with questions ideas i do want feature requests for arbitcoin you can open issues it's uh uh github.com slash reareden code slash arbitcoin that's also linked from the website and yeah come either help out suggest donate whatever you can i

1:01:02i'd love to have more people helping with arbitcoin. Awesome awesome guys links gonna be in the show notes psycho trainer saying thank you for your time reareden i feel less dumb than before this started so see small wins great compliment thank you right small wins we'll take it we will take it all right guys um so look everybody thank you so much for for joining us for chiming in on the chat right that's good stuff i hope everybody has a great weekend Brandon if you could stick around for two seconds after i end the show guys you know the deal back monday 1210 yeah i know it's labor day but i still think i'm gonna do a show anyways right because i because i love you guys and i'm sure there's gonna be tons of shit to complain about uh so yeah monday 1210 pm eastern hope to see you guys and of course next Tuesday we're doing it guys it's gonna be bitcoin tonight everybody have a great weekend make the memories do all the good stuff and catch you all next week thanks thanks for your take it easy