EB94 – Gavin Andresen: On The Blocksize And Bitcoin's Governance

Transcript

[Music] this episode of epicenter Bitcoin is brought to you by Ledger makers of The Ledger Nano Hardware wallet have peace of mind in knowing your private keys are protected by industry standard physical security go to ledger wallet.com to learn more and use the offer code eb09 to check out to get 10% off your first order and by Vol toro.com the gold to Bitcoin exchange trade gold to bitcoin instantly and securely starting at just 1 mgram go to volt.com to deposit some Bitcoin and start trading

[Music] today hi welcome to epicenter Bitcoin the show which talks about the technology projects and startups driving decentralization and the global cryptocurrency Revolution my name is Sebastian C and my name is bran crane so we're here today with uh someone who also needs no introduction like some of our other guests Gavin andreon so this is the first time we have him on the show but of course super exciting so Gavin for those who don't know or don't remember he was the guy who was involved very very early on in Bitcoin and he sort of took over from Satoshi the the lead development role and he had that for a long time and then he uh he pass on and well he was also the the chief

scientist of the Bitcoin foundation and uh he he passed on the lead developer role and and now he's at the the digital currency initiative at the MIT media lab so um G thanks so much for coming on well thanks for having me excited to be here yeah absolutely so I remember I met you once very briefly at the Bitcoin Foundation conference and I asked you how much does it cost in mind if you include a transaction because it increases the propagation risk and you said 11 cents that's probably i' would have to redo the calculations it's actually probably a lot lower today yeah um cuz technology has changed but uh but

yeah short interaction but it was interesting CU that yeah um so tell us a bit like you you're now at the MIT digital currency initiative what's what is that exactly well you really should do an interview with Brian Ford to get like the complete like vision for the digital currency initiative uh but basically MIT is interested in you digital currency and what does it mean um and is interested in having students and professors do research learn about it um and so they're trying to to you know figure out kind of how to how to make that happen uh and you know I have no experience

with this kind of academic interacting with industry kinds of uh initiatives but you know the MIT media lab has a ton of um experience with that you know they do it all the time and so the media lab is sponsoring the digital currency initiative and when the Bitcoin Foundation um kind of fell down last year um they volunteered to kind of pick up me Vladimir and Corey continue paying our salaries to work on the core technology um and you we're still kind of working out you know how much interaction will you know we actually have with the digital currency initiative um I mean

the digital currency initiative itself is still bootstrapping and growing and kind of getting its feet so I don't know we'll see what happens over the next year or so but so far it's been it's been great it's been a great place to great place to be so so do you actually work from MIT and are you also going to teach uh courses or seminars workshops things like that uh I don't work at MIT I'm actually two hours away from MIT in Western Massachusetts um and we'll see I mean it all depends on kind of you know what the professors are up to um as to I probably I won't there's no plan for me to teach a course I'm I'm busy writing code that's not really you know what I do um I don't you know I'm not a teacher I'm not a

professor but certainly going in for seminars um maybe doing guest lecture certainly something that I I think will probably happen uh over the next year or so and can we get an idea of how much uh money the digital currency initiative is putting into Bitcoin development um well they're paying three salaries um I don't you know again talk to Brian Ford about uh you know details of the digital currency initiative and I don't know if they want to make kind of you know the salaries that they're paying public um but it's a pretty good chunk of money so um yeah I think that's all about I can say about that yeah no I think I think it would be great to have

them on right because people have heard about it but at the same time it's sort of unclear exactly what it is going to do and it's definitely it's definitely great that you know University like MIT is is supporting that and putting their name and support and infrastructure behind it yeah absolutely but the I mean the only thing is that they're not on Reddit and they really need to be on there to take part in the global debate on bitcoin I you I think you do see sometimes see like Jeremy ruin on redddit uh anyway but yeah I mean it's certainly a busy time I I know it's a busy time you know getting kind of ready for a new semester uh so I'm not sure if Brian will be available soon

but you should definitely have him on this show I'm sure uh he'd be happy to talk about the digital currency initiative and what the vision is and and what's likely to happen going forward yeah we'll do that so today the main thing we want to talk about well there sort of uh one topic which is of course the block size you know we've talked about this many times and it's it's an important topic and our listeners will be very familiar with it but that we don't want to make that the main focus here but we really want to talk a little bit about the larger The Meta question of how actually are decisions reached in the Bitcoin uh Community how is the code evolved what happens when there is a disagreements and arguments uh so really the issue of

of governance and uh I think yeah it's a very important issue to think about and it's certainly not going to get easier from here on so before we dive into that do you have a definition of governance um yeah I've I I got some flak on Reddit and Twitter for talking about governance in a very kind of broad way and when I say governance I don't mean government um I think we're governed by all sorts of things you know we're governed by our government we're governed by the laws of physics uh we're governed by social norms you know there're all these things that affect

our behavior and and what we do and that's what really I mean by governance is you know some some effect on you know what we do how we behave or how a technology behaves you know how a technology works that's all that all fits into my kind of general idea of of governance and when I think about governance issues I really think about all of those things I think about you know what's possible with the laws of physics you know what's possible with kind of social norms um what's possible with governments right I mean also you know I think if you really want to think about governance you have to think about kind of all those different levels um so you I think a lot of people

when they hear governance they immediately go to you know the a a a nation state governing and deciding you know what you can and can't do and that really isn't what I mean when I say governance yeah it's interesting that you define it like that cuz the thing that sort of came to my mind was like the the process by which decisions are reached right that that's but maybe that's a little bit uh a more narrow way of looking at it uh that is more narrow and and I mean that's a useful way of of thinking about it I think process is is always important um and I think if you have good processes good things happen um um but yeah I mean in general when I

say governance I really am thinking about everything I don't know if there's a better you know we could talk about process I think I I would probably say process if I'm specifically thinking about you know what is the process we use to come to decisions or what is the process we use to make changes to source code or or whatever you know I I don't really I do make a distinction between that and governance and in the recent debate around the block size that we've seen in the last few weeks and months what have you learned about the process that is in place whether it's formalized or not uh what have you learned about the process about Bitcoin governance as it exists

now I think the the the problem is different people have different ideas about what the Govern governance process ought to be or even what it is or even what it was um if you go back in history it it was really simple it was whatever Satoshi decided at the beginning you know that's really where we started we had one one one source code we had one pseudonym person uh who made all the decisions about you know what should Bitcoin be uh how should it evolve what should it do um you know that's where we started you know as soon as Satoshi kind of step back and and threw the project onto my shoulders um one of

the first things I did was to try to kind of decentralize that so that if I got hit by a bus you know it would be clear that the project would would go on um and so know that's why at this point there are you know five people who have Commit access to the GitHub Bitcoin Source tree um and there's this kind of consensus process for what changes are made to the code um and even you know what what you know what consensus level low-level changes to the Bitcoin rules um should happen um and and I don't know I think a lot of the the the strife and conflict comes with

you know we're going to we're starting to grow Beyond even you know even that to there are a lot more stakeholders there are a lot more people involved there are a lot more projects um and how do we evolve from you know it used to be just Satoshi making decisions to it was the small group making decisions to suddenly there's a much larger set of people who are interested in in in decisions and how they're made so you think that's the that's the challenge right now so we we have these five people that well you chose I guess right to have Commit access and of course you could say that well they just vote and it's a majority or or four out of five or or yeah I chose the first I

chose the first couple people and then kind of you know Among Us among like the three of us we chose a couple more people um and yeah I think I think that is the challenge going forward is is um I don't think any of us you know especially those five people like want to be the high priest of Bitcoin that's just not our role that's not what we want to do that's it's not what we want to be and so we need to find a new way of uh kind of a new way of of governing right a new way of coming to decisions and and and having everybody happy about the process and and the decisions that are

reached cool well well we're going to talk a lot more about what that could look like but before we do that you know when we emailed before the show you mentioned one thing that you wanted to stress and that is sort of confusing and a little bit unclear right so there that term forking right and of course forking is used for open source software when you know you take it you copy the code and then you know you have two instances you can make changes to one and all but forking is also used in Bitcoin right when all of a sudden uh there's several chains right that sort of run in parallel um so how do you think of of those two in the terms of of governance and and of those sort of decisions yeah

it's really important for people to to separate in their head Bitcoin the protocol you know Bitcoin the system that we're all using to transact and the Bitcoin core open- source software project that lives on GitHub and has a bunch of people contributing code to they really aren't the same thing um you know I call the I call Bitcoin core the reference implementation and that and I've called it that for for years I mean that implies that there will be other implementations of the Bitcoin protocol um and when we think about governance I think we have to think about the governance of kind of how will the protocol evolve as separately from you know how will Bitcoin core the reference

implementation code how will that evolve and how will that be governed right I think there are two separate governance processes um and I think it's I think most people because we started with this you know this one source code defin the protocol and was all anybody was ever running I think in in a lot of people's heads they don't make that separation um but I think it really is important to to think about the protocol separately than you know this one set of source code that has been admittedly the source code everybody's running um but I've been saying for quite a while that that I want to get to a point where you know

there are there are multiple uh robust implementations uh of the protocol and I think I think we will get there I think it might be a little bit Rocky but um you know that's a really important distinction to make as we talk about uh governance processes and consensus and you know all of these things now do you think that in terms of software design it was a wise decision now in retrospect to have the protocol sort of embedded in the reference client rather than the protocol be defined somewhere else and have several clients for instance like what ethereum is doing with the release of the the ethereum client the frontier launch they they have multiple clients um what are your thoughts on that I

think if Bitcoin had like lots of money to start like ethereum did uh it would have made sense to try to like start with multiple implementations uh it's having multiple implementations is great because you find kind of places where things are not well defined and as you know as Bitcoin has developed alternate implementations um like for example conformal has a implementation of Bitcoin written in go which if I had infinite time I would spend time playing with and I'm actually hoping to to spend some time working with it and and trying it out um and they've found you know they've as they reimplemented things you know they kind of refer back to the C++ reference

implementation and they find little edge cases that that like weren't tested um and submitted tests to say you know this is the way the system behaves um you know we ran across this as we're reimplementing it and you know let's you know make sure that that we're aware of all these hairy L edge cases so it's definitely a great thing to do um Satoshi could not have done that right uh Satoshi didn't have millions of dollars to hire lots of developers to do four different implementations at the same time do you think it would be wise today at the point where we are to separate uh and and really have clear

protocol documentation do you think that would enable more people to come in and start developing clients I mean the the problem with documentation is it always lags behind code so um you know in in in early in my professional career I worked on a 3D graphics specification um and I saw you know how hard it is to to get all of the details right in a specification and even if you get the the details right in what you write down in a specification like actually getting implementations to interoperate or produce the same results given a

specification is is really hard um so I'm I'm a much bigger fan of uh like test cases for example um and actually if you look the work I've done on bitcoin core you know one of the first things I did was implement the test net so we have a place to to test things out and alternative implementations have used the testnet as a place to to test things out you know a bunch of block early blocks in the test net have all sorts of weird transactions in them so that if you are writing an alternative implementation you know syncing with the test Nets actually a pretty good first pass of have I implemented the Bitcoin protocol correctly and so that kind of thing you know that kind of like actual code you can run to find find out if you're correct or not I'm a much bigger

fan of that kind of approach rather than just writing words on paper that are subject to interpretation and uh and I know people will you know read the same sentence and come across with completely different idea about you know what does that really mean you know how should I put that into code yeah and what's also important to notice regarding ethereum because we we just did the episode with italic where we actually asked about that too and he did say that it was sort of a mistake that they try to do so many implementations at the same time and you know they sort of abandoned that and felt was resources wasted kind of and and CU right now they didn't launch with several implementation right that's interesting um so if anything he said it

was a it was it was a mistake to do that or at least to have that many maybe they would have had two or something instead of four yeah yeah yeah I think two or three yeah four is a lot uh two is probably enough that you would find you know almost all of the little nasty edge cases that will you know that you wish weren't in there right there a bunch of bunch of like things in Bitcoin that are kind of there by accident that we really wish weren't there um we just got rid of one of them with um bip 66 and that soft Fork that we just did was to get rid of a a nasty little Edge case um it was actually a consensus bug right it could have uh if people had known about it uh

they could have forked the the network they could have caused machines running 32-bit operating systems to disagree with machines running 64-bit operating systems uh which would have been a mess so yeah it's a kind of thing that it takes a long time to find um and you know it just it it takes time and experience and you know having another implementation does help find those sorts of things it doesn't guarantee that you'll find those sorts of things um but it certainly helps so there so you just to point out there are other uh there are Forks of Bitcoin so there's another C++ client called Bitcoin XD which we've talked about and we'll talk a bit more later I believe there's a JavaScript one as well I think

yeah I think bit pay I don't know if bit pay's implementation is is fully validating yet or not um but yeah I mean certainly we're seeing other people tackling doing other implementations of the protocol in in other languages yeah and of course some companies have their own proprietary implementation although yeah well certainly there's Toshi at uh at coinbase which is a Ru Ruby implementation uh that they use quite a lot um so I mean this is this is happening right we are getting alternative implementations of the protocol and in my view that's that's healthy that's a good thing so one another thing I wanted to point out with regards to what Brian mentioned regarding forking the code or forking the blockchain is

that the way way I I see this if you Fork the blockchain it implicitly probably means that you forked a client and made some changes to the protocol however you can Fork a client and not necessarily Fork the blockchain because it can be it can be compatible with the reference implementation that's right yes yes um so yeah when we say forking the blockchain we mean changing the consensus rules of Bitcoin in a way that's not compatible with the way it used to be um and so you know that that is on the kind of consensus side

um and that almost always means you've you've changed some code so you've maybe forked the code although we've certainly had problems with with forks with Bitcoin core where you know there was no fork in the source code it's just kind of old versions of Bitcoin disagreed with newer versions or like with bip 66 the same code running on a 32-bit operating system might disagree with it running on the 64-bit operating system and that that could have caused a blockchain fork it didn't in this case uh Peter V found the problem and fixed it uh before it became a problem um but so we so you can have blockchain

Forks without you know software Forks um you certainly have soft sofware Forks without um blockchain Forks you can have a fork of the Bitcoin software that's completely compatible uh with the reference implementation and all the other implementations let's take a short break so I can take you to Paris I dropped into lamei Bitcoin the house of Bitcoin in the heart of silicon saner home to many startups including Ledger and I spoke with nikoda Baka Ledger CTO about the level of security in their devices oh Ledger is based on the smart card chip that exactly the same SM chip that you have in banking chips and in passwort in passports today so that's something that has been validated for the industry for more than 20 years

today's a favorite second factor is the phone um when using a phone you are transmitting an encrypted version of your transac to the phone to the screen that can be displayed to you and validated to you so we leverage in two insecure devices your desktop and your phone to create a secure validation with Ledger we believe that Hardware wallet are very primitive today they are just use to confirm transactions and to display some validation and that's it uh in the future as Bitcoin protocol move towards smart contracts to have a lot more use cases we believe that those devices will evolve into something that I would call consensus devices which basically will be able to confirm for specific Services specific rules how

easy would it be to hack a ledger it wouldn't be very easy to be honest I think best would be for you to try physical attacks first like glitching like side Channel attack if you want to really attack it you will need to get a bit more equipped with like a focus y beam one station or a scanning electronic a scanning electron microscope so that's going to cost you probably a few Millions uh and as well you are going to be able to deap the chip so that's going to be a bit risky for you as well uh but you are free to try of course and let us know if you manage to do anything Ledger is building an infrastructure which will provide the best level of security for the Bitcoin industry you two can benefit from this technology and get an affordable secure

setup for storing your Bitcoins with The Ledger Nano so go to Ledger wallet.com and use the offer code eb09 to get 10% off your first order that offer code is valid until September 30th of 2015 we'd like to thank Ledger for their continued support of epicenter Bitcoin now you know you mentioned in the beginning there was satosi he was making the decisions and it was a you know it was a small world back then uh and of course there was no payment processes there governance governments weren't interested in it there but today it's a very different world right so we have miners there's startups there's the users there's the Bitcoin holders there's the developers there's Merchants then there's Regulators there are all

kinds of different people who care about what Bon does and where it goes so who of those entities should have a say in how Bitcoin evolves and where it goes and and how should that be you know how how much weight should each of those entities or groups of entities have um I think everybody should have a say um it's hard to balance you know letting everybody speak and not just being drowned out by the noise um so you know if you look on Reddit every anybody can speak there and talk about you know whatever issue about

Bitcoin they care about you know it's hard to it's hard to it's hard to to to kind of listen find a signal above all the noise um um I don't think it's impossible and and I think you know I think it is fairly clear kind of what the consensus on Reddit is about the block size issue um but you know it's it's it's hard to find more than kind of kind of a you know a high level you know we think something should be done about this issue when it gets down in the details of you know exactly what should be done I think then it gets even even harder um you know who should have the influence um I mean well the the really comes down

to you know who has the um you know who's what code are people running and you know how influential are the people who are running the code so for example exchanges are incredibly influential at this stage of bitcoin's life right exchanges are are the place where a lot of trading is done right now there's a lot of speculation exchanges are where that speculation happens so what the exchanges want to do matters a lot um miners are also very influential um so kind of what the and and more than miners the mining pools right that that pull together miners uh are very influential um so I think they have a a

big influence and a big voice in in what happens um ordinary users um or Bitcoin holders uh it's harder for them to have a huge influence um you know I think if you're not a a a if you're not a technical developer you know you can have some fantastic idea about you know how we could improve Bitcoin and if it's like a technical idea you need to convince a developer that it's a great idea and that's going to be hard because developers are busy for example so I think you know the people who have the really the the biggest voices right now are the developers um the exchanges and kind of the mining pools those are kind of the

three I think biggest players um I like to think that you know as a developer I try to listen to all the stakeholders I try to listen to you know kind what long-term Bitcoin holders want what people who are actually transaction transacting with Bitcoin want what the companies like bit pay who are supporting Merchants want um and try to balance all of those things when I think about kind of what should the you know what what am I personally going to be working on uh as as I write code as I submit poll requests as I decide what pull requests I'm going to review um you know I try to channel uh you know kind of everybody

um to to to prioritize what I work on and I think all the other other developers do that too we probably have different opinions on kind of what's important to work on uh next and through what channels do you uh get all of those opinions I mean there's it seems like you'd have to be everywhere to get everyone's opinion it's hard um you know for the kind of for the big exchanges um Merchants you know I I I'm on the uh technical Advisory Board of a few companies so I hear there kind of what their concerns are you know if they're concerned about transaction malleability

or they concerned about instant uh zero trans confirmation transactions if they're concerned about the block size uh so I get that from there uh I think the best we can do right now for like ordinary people transacting with Bitcoin are channels like Bitcoin talk and Reddit um and again it's really hard to kind of hear above the trolling and the noise I think at the Amsterdam conference last year as part of my state of Bitcoin speech I said I want a better way of uh of communicating you know I want some way for the best ideas to rise to the top and for you know a better way of measuring consensus um we'll see going

forward what happens I don't know uh maybe prediction markets will be a good way of of G consensus I know that's an idea that that people have um I think that would be fantastic to you see kind of wisdom of crowds condensed in that way where people actually have some value at stake um but I think it's it's still an open problem as to you know how do you get these diverse stakeholders to come together come to consensus um to communicate um it it's it's not an easy problem I think I I don't think it's been completely solved kind of anywhere where you have

the distributed way of of making decisions but do you think that discussion about our governance how decisions are reached is is that important now do you think things the way they are right now are are broken or do you think they work well enough I think it's important to have the discussion um I don't think it's going to like make or break Bitcoin coin if the discussion goes nowhere uh because when it comes right down to it you know the governance is whatever code people are willing to run so you know it's it's the the the blocks and transactions that get mined and then the blocks and

transactions that get accepted by whatever Merchant or person you're paying um or exchange you're you know transferring into kind of you know it's the code that people are running that defines the consensus and that will continue to be true uh no matter what happens kind of at a higher level of how to and and I think Eli Dorado uh wrote a what I thought was a great blog post on how should Bitcoin be governed and he talked about this about how you know the the Nakamoto consensus algorithm this consensus for coming to you know what is the blockchain and what are the valid transactions doesn't care about other consensus processes right we could slap

whatever consensus process we want on top and It ultimately comes down to you know what is the code that people are running um that is the consensus at at at at at base and that is what governs kind of what Bitcoin transactions are accepted which is really uh you know the whole purpose of of the Bitcoin Network yeah I I thought that was a good article but there's there's one thing that I don't really agree with there and and and he says that we can arrive to this consensus more easily by improving communication between groups and stakeholders and ensuring that the debate is grounded and realistic assumptions about the world so the first part about uh improving communication between groups I think that that's quite

a challenge because not all the groups I mean now we're a relatively small community all the stakeholders that are pretty active are communicating but as Bitcoin grows uh you're not going to necessarily have all stakeholders on one channel of communication where they can communicate and a lot of them overlap and and and are separated by geographical uh uh barriers right so I think that's complicated and that and the other part ensuring that the debate is grounded in realistic assumptions of the world I mean we can see now how realistic that is right with most of the bitcoiners in for instance North America thinking that remittances are going to

save uh Africa or the developing world I think that that just goes to show how these realistic views are not always very founded um I think that the what's important to perhaps realize is that the stakeholders that we have now are not necessarily the ones for whom this technology can benefit in the future so if we look at the internet for instance uh perhaps very Sim similarly in the beginning we had a small amount of actors involved and then as I mean now everyone's online and if there were some fundamental changes made to the internet

it would affect a lot of people and I sort of see Bitcoin going in the same direction where I'm worried about the future stakeholders the People For Whom the technology is going to be really important in the future that perhaps aren't even born yet and I don't know if the interest of coinbase or today's miners or today's startups are necessarily those of those future stakeholders I think that might be right I mean I'm I tend to be pretty pragmatic so I tend not to try to plan you know 10 years ahead I I I would much rather plan 10 weeks ahead um just because I think if you if

you if you try to plan too far ahead you never get anything done you just you'll ignore if if you if you ignore short-term problems because you're trying to solve some long-term problem you'll never get to the long term right you won't get past those short-term problems uh to actually get to the long term um so I mean yeah I kind of it's fun to think about future stakeholders but I it's it's hard for me to imagine that I know who those future stakeholders are going to be I just don't know how Bitcoin is going to evolve I don't know if it's going to be you know if it's going to take over Africa and instead of having all these

you know little currencies that are that are uh that hyperinflate every once in a while that you know everybody in Africa will be using Bitcoin as kind of the panafrican currency I don't know that's a possible future and you know if I knew that that was going to be the future then there probably technical decisions we'd make that that might be different um I don't know you know maybe Bitcoin will become the you know the the replacement for um some financial transaction that's inefficient right now that I don't know about because I know almost nothing about you know stock and bond and derivative trading transactions but I know that there are people

experimenting with that on the Bitcoin blockchain um if I thought that that was going to be big that was going to be the win you know maybe we' make technical decisions differently now to kind of satisfy those future stakeholders um I'm actually not that interested in those future stakeholders but you know who am I to decide you know how this technology should evolve so I don't know I I have trouble uh you know I have I've I want to find solutions that that that you know no matter how Bitcoin evolves in the future uh well first I want to make sure we get there um and second I want to I want I want the technology to be like

the internet to be as kind of General as possible so that it satisfies as many stakeholders uh potential future stakeholders as Pro as possible absolutely and I I agree with that and I think that that is fundamentally what needs the sort of the the fundamental core principle is that moving forward whatever governance or whatever decisions get made that that is always what's sort of a you know front of mind is making it possible for anyone to innovate permissionless Innovation um transactions at low cost Etc that's just my opinion no I agree I mean that's the part of Bitcoin that I

find really interesting is that for for really the The First Time online we do have permissionless Innovation when it comes to money and and transacting and we didn't have that before you had to you had to go through a a gatekeeper um and I I think fantastic things are going to happen um just exactly because of that today's magic word is Fork f o r k head over to let's talk bitcoin.com to sign in enter the magic word and claim your part of the listening [Music] reward I would like to come back to what

you said regarding you know the governance of of the code and and the blockchain is sort of two separate issues cuz I don't know if I agree with you on that one cuz it it seems like you know most people they just default to running whatever code comes from that reference implementation so I think that that sort of I mean yes it is true in in the extreme case people could switch the code and and stuff like that but I do think there is a sort of an enormous an enormous weight that the reference implementation has I mean I think well I look at the history of the internet right I was back I I was around

back when NCSA Mosaic was the web browser that you used if you wanted to you know browse the internet um right this was the original web browser written by Mark andreon when he was at the national Center for supercomputing applications um and then he founded ncape and everybody was using Netscape as the browser that everybody was using until Microsoft invented Internet Explorer which actually was a fork of the Mosaic web browser initially if I recall correctly and you know and then suddenly you had an explosion of of web browsers and you know way back when it was in CSA Mosaic you know that was incredibly influential on web standards

um but then web standards once there was more than one browser we went through a really nasty uh kind of transition period where you know you had to write your website one version for Netscape and one version for Internet Explorer and that sucked and everybody knew it sucked I was a web developer during that period oh it was terrible wasn't it oh oh man that was bad I I was learning to code web development at that period and then the 2000s just yeah killed me yeah no it was bad you know I was doing some web development back then too uh down with ie5 you know it's it's terrible um and and then you know kind of people got together and now we're in a in a better

place where I think there is a better model for kind of evolving HTML as you go forward it's still not perfect it's still messy it's still you know the browser manufacturers still try to be proprietary and uh still try to you know get market share by playing in the standard space um and I think Bitcoin will going through a similar Evolution right I I think we will evolve past just there being one reference implementation that everybody uses I think it's it's natural for things to specialize so uh you know the code that Satoshi wrote was a minor it was a wallet it was a full node it you know it did everything in

one package and I think he had to do it that way because you wouldn't want to download six different pieces of code uh to be an early adopter because you did need a wallet and you probably did want to mine and you probably you know you definitely wanted to be part of the network um but as Bitcoin gets bigger kind of we're seeing the pieces specialize so you're seeing the Bitcoin reference implementation is not the wallet that most people use anymore most people are using a a lightweight wallet on their phone I have three or four white light lightweight Wallets on my phone I have you know Hardware devices that are Bitcoin wallets um and that kind of specialization is is good and it's natural and we're going to see that at

kind of all layers of the Bitcoin stack so you know I think we will see an implementation of the code that's targeted just at miners or mining pool operators that you know has super fast validation make sure that blocks propagate insanely quickly um you know they just kind of focus on kind of what do Miners and Mining pool operators need in terms of software um and I don't think that'll be the reference implement I don't think the I think the reference impation will continue to be kind of trying to do everything for everybody um and so I think you we will see this Natural Evolution of you know it won't just be the one piece of code there will

be you know lots of different pieces of code implementing the the Bitcoin protocol it's time for a word from our sponsors volt.com the go to Bitcoin exchange now we all know there's been no shortage of Bitcoin exchange scams and hacks in these recent years and that's why when Phillip and Josh the two brothers behind voltro decided to start that Exchange in the wake of the mount gaau disaster they wanted to do things differently so they're really pushing the bar forward and innovating in terms of security transparency and auditability to ensure that customer funds are safe secure auditable and so that there's no there's none of this Mount gax you know stuff going on it's

all on the up and up so if you've been listening to the broadcast of course you know volor and perhaps you like volor and you like what they do well something new is happening something really exciting is happening and that's volore is doing an equity crowdfunding campaign through back to the future of samim Dixon who of course you know as well by this point uh so if you're interested you now have the chance of actually owning some equity in a startup which is sort of a new thing equity crowdfunding and not just a startup but a great Bitcoin startup a great startup in this space and you can even invest with Bitcoin so if you're interested check it out that's on bank to the Future so bnktothefuture.com and of course we'll put the link in the show notes and we

would like to thank Vol for their support of epson Bitcoin and hope they're going to have a fantastic crowdfunding campaign so it some people have called for to have like a more formalized governance model where you know it's sort of there's a little bit more sort of structure to that but it sounds like from sort of the way you talk about it it really doesn't make any sense to talk about that on a network level right because on the network it's just what whatever client people choose and whatever that client rules are in the majority you know that that's just the sort of the facto standard for the network it is I mean you know you always

want to encourage interoperability right you want all of these pieces of software to be able to talk to each other so so whenever you have issues that are you know that need these multiple implementations to to cooperate I think there is a a role for some sort of governance process there some sort of way of you know everybody who's writing these implementations to come together and say yeah we want to do this new thing on the peer-to-peer Network what should the message be called how should it behave and you know right now we have this bip process Bitcoin Improvement proposal process um that the last year actually the Bitcoin Foundation was putting money

into kind of talking to stakeholders and trying to figure out do we need a more formal bip process does the current bip process work for everybody to kind of Define new new protocol messages new you know consensus rules you know whatever um and unfortunately again kind of the Bitcoin Foundation had a hor year last year uh that work stopped because the foundation basically ran out of money to to fund it anymore but you know I do think that there you know that's something that we should be thinking about and we should be looking at Role Models like the way the internet is governed for you know how do we come to consensus on internet

protocols right if you want to improve SMTP right the mail protocol that we all use to send email to maybe support security how did that happen um you know how did the how did the people who are interested in that all come together to decide uh that it should happen and actually that hasn't that maybe that's a an anti- example because email security is terrible um and that didn't happen um but you know I think I think that's the kind of thing that we should be thinking about um and and you know that is a place where I think more formalism may be a good idea maybe maybe not I don't know I think that's debatable so when if we look at exist institutions uh which formalize standards on the

internet we have the we have Ian for instance we have the w3c um there's another one you mentioned earlier I can't remember which one should we be looking at for potential know guidance or to model and which one should be re which one should we be really uh trying to not emulate I I like the internet engineering task force ETF they have a pretty distributed uh model um where kind of people get together in working groups and anybody can join a working group The you know it's whoever shows up uh and then working groups uh produce standards documents that in the iatf case are

called rfc's requests for comments and then rfc's either become standards or they don't right I mean it depends on kind of what code people run which I also like uh you know there a bunch of ietf standards that nobody pays attention to because they turned out to be bad ideas uh there are a whole bunch of ietf standards that people do pay very close attention to you know including the standards for email the standards for you know I don't know all sorts of networking standards that we're all using every day that you know we we don't think about um we just use them because they just work uh and so I I like the ietf model um as a way of going going forward and

there's been talk of of you know perhaps the the Bitcoin bip process maybe parts of Bitcoin should become an ietf working group and you know that that's one possibility for you know how standards in the Bitcoin space evolve is that you know it we we create ietf rfc's that are related to bitcoin and you know maybe that's maybe that's a future path forward um again there's you know been debate about whether that's a good idea idea or not uh and whether bitcoin's ready to to do that or or not yeah I think it's an interesting idea if you think of this bip process or something like ITF if

that's like sort of separate from the reference implementation right so if it's just working group coming together thinking like oh what's the best way of going about something and then they maybe can do pull requests to all the implementations in use or or they just integrate it independently if they feel like this is something sensible and we're actually already seeing that with the bip process there are a bunch of bips that related to wallets that the reference implementation wallet doesn't Implement so you know we have uh HD wallets hierle deterministic wallets uh that are a bip that the you know reference implementation wallet doesn't follow right now which is a it's you know it's a bug in the reference

supplementation it really should um but the bip is useful and a bunch of wallets are are following uh are following that bip and they do interoperate because of it so you know that's that's a good example of where I think the bit process is is working pretty well I mean you know none of this is going to be perfect none of this is a Panacea uh just because you write a spec doesn't mean everybody's going to agree and and follow that spec but um but you know I I think it's better than just you know having no communication letting people go off in 100 different directions yeah know I agree one thing that has been brought up and of course we know that from other

open-source project is this benevolent dictator for life right so some projects have that like Linux or Python and I'm sure some others do that too what what do you think of that you think that's a good model for Bitcoin or maybe for for some of the implementations for Bitcoin or maybe so yeah so I I think I think it makes sense for an a particular code base particular set of code to have a clear kind of structure for how are decisions made you know who who ultimately makes the decision or is there some voting process that makes decisions or or whatever um and I think

you know Bitcoin core the reference implementation uh has had kind of this fuzzy um process you know when I was the lead maintainer of Bitcoin core basically the process was you know anything that there was broad consensus would go in if there was something that we just seemed like we couldn't come to consensus I'd make the decision I would just arbitrarily decide we're going to go this way instead of that way and and I think that that worked I was acting as benevolent dictator for Bitcoin core um when I stepped down from Bitcoin core and Vladimir became lead maintainer um you know he's not willing to have that kind of benevolent dictator

role so he will not he's more conservative he will not uh merge things that don't have consensus um which I don't know watching that Dynamic um it really means that that you know changes can be vetoed right it just takes a a a a loud develop to say I don't like that change um I'm going to veto it basically and I don't think that's actually healthy I think it I think for a software project it does help for a software project to have a a a clear like way of of making decisions and coming decisions um and I think core

recently has been having problems with you know how does it come to decisions what should the governance model for core be um and you know that's that's one of the reasons why I think we're going to see more Forks of of uh Bitcoin core is is is just I think people need more kind of uh uh it it's better if there's a clear process as opposed to you know some fuzzy you know if it's not controversial then it's okay um I think you know if we want to if we want to remain Innovative I think it's important uh that there be room to take risks and that

uh and that somebody who's contributing you know can at least have some idea that you know their contribution will will will come to a decision at some point it just won't be in limbo forever which I think is you know one of the big problems with contributing changes to the core is it's just it's hard to you know there are certain changes that it's unclear whether there's consensus or not and so they just kind of hang around for months and months and months uh which if your developer is really frustrating frustrating yeah I know I mean I think one one SE that so I was I was giving a talk uh two months ago something about the block size thing so before that I read through the whole Bitcoin Dev mailing list for like on the

topic basically since it started I mean I'm not up to date anymore on like what the discussion is but it did seem like it was just sort of uh yeah there's no process of reaching a decision right so it's just back and forth and anybody shouts in and people get angry with each other and it's just it obviously isn't isn't really working that process and then somebody says no the block size really should be lower and you just kind of shake your head but um yeah um and you know having been in the position of lead maintainer and knowing that kind of this whole multi-billion dollar industry kind of sort of rests on your shoulder because you're in charge of the code that everybody is running um that's a

horrible place to be right I was just going to say that yeah I mean there must be enormous pressure Tois pressure to make the right decision um which is part of the reason why I think we need to evolve beyond that so that no one person or group of people feels like the whole weight of this system rests on their decisions so that there is more room to make decisions to innovate and to even make mistakes um and just the the impact of those mistakes will be smaller if you have a more diverse system where maybe not everybody is running the same version of the code on the network there are lots of different versions of the code if anyone breaks it's not that big a deal um because you know the whole system

doesn't break right it's just more diversity really does bring robustness in this case and Bitcoin is sort of unique in in the fact that I mean if you look at Linux for example okay there's there's Linux um there's a Linux kernel and then there are multiple distributions of Linux those all use the Linux kernel but essentially you know you can make your own distribution and and it can be significantly different from the from the I guess reference implementation which you would probably call the Linux kernel I'm not like super well versed on OS design but but I don't think that the what is his name lonus tals or the lenus is actually pronounce it I'm told um I I don't think he feels

this pressure to uh to make everyone happy and that you know there's a whole industry counting on him I mean maybe he does but certainly the economic um the economic repercussions of his decisions are probably not as important as they are with Bitcoin yeah well they're certainly more indirect and I think he has more kind of checks and balances just because it's a more mature infrastructure between him and you know people losing money right so if there was some Colonel bug that you know allowed uh that allowed a hacker to get in right that would be really bad right

and that has happened to to to Linux um I don't know if it was linux's fault or not I mean probably you know there is some bug that caused a bunch of people to lose money because their servers got hacked um it's a little more indirect than you know some bug in the reference implementation wallet that you know let your funds get stolen it just I don't know it feels one step removed if you're working on on kind of like operating system level software as opposed to financial software which is really what we're doing yeah so we mentioned earlier that uh and we've talked about this with um with Mike Hearn that uh there was a potential Fork to Mike's uh client which

is Bitcoin XT uh can you talk to us about um what the status of that is and more importantly why that decision is why you're considering that fork sure so yeah Mike has had a a a fork of the Bitcoin reference implementation called Bitcoin XT for a while now where he added some features on top that he wanted in there that he needed in there that uh were controversial for whatever reason and and weren't added to to bitcoin core um he he like me believes that we need to scale up and the easiest way to scale up is increasing the maximum block size um

so the the latest release of Bitcoin XT will support bigger blocks and specifically it supports um my proposal bip 101 uh which increases to an 8 megabyte block size immediately and then scales up over time so that that maximum block size uh uh grows with the you know grows with technology um we've been working on that uh the last several weeks both Mike and I uh binaries uh should be available very soon um so you'll actually be able to download it and run it it will behave just like Bitcoin core so if you're running Bitcoin core you can drop in Bitcoin XT and you won't notice any

difference it will behave exactly the same uh as Bitcoin XT uh because it's essentially the same code base just with a few changes on top um including this you know support for uh larger blocks um the roll out of Bitcoin XT I think uh you know we talk about stakeholders and we talk about who's important in the ecosystem so uh you know the way bip 101 works is it requires a super majority of miners to to support it before it actually happens before we allow larger blocks and before there would be any blockchain Fork um even before we get to

there it's really important to get big merchants and exchanges um on board so that they're ready for Big Blocks and that's actually uh my NE next task is to Circle back around and talk to bit Po and coinbase and bit finex and you know all of these other kind of big players in the space to say you know are you willing to support bip 101 um if not why not you know is there anything you can do to help make the transition easier um and really to get those players on board first and have them come to consensus that yeah this is what we want to do this is the code that we want to support um until that happens I'm not going to be trying to convince the miners to uh

to support bip 101 or to produce bigger blocks because it doesn't make sense for them to produce Big Blocks if the transaction in those blocks are just going to be rejected by coinbase by exchanges by bit pay by whoever um so it's really a process of you know making sure that we have the the uh what some people call the economic majority kind of the the the big players who are making lots of transactions make sure that you know there's a general kind of rough consensus that yeah okay Gavin we'll go along with uh we'll go along with bip 101 seems to be the most reasonable compromise um and then you know assuming that we get buyin from

them uh go forward and and and get buying from the miners but you know the first step was actually getting code that we can test we can make sure that it you know behaves the way we expect it to behave um and that we can ask people to start running because you know people the way people will Express their support is by actually running the code um so how exactly would that work so you you would go to to bit pay in coas ETC and then they will start uh I don't know if they switch to XT or if they just uh add that code to support bigger blocks to their existing uh you know maybe coinbase own implementation or bit pay Zone implementation and stuff uh and how will

you know for example when the support is is enough is is there is there some way that for example they signal that um yeah we're going to be asking um big companies if we can put their logos on there will be a uh Bitcoin XT website um and we're going to be asking them if they you know were willing to put their logo as you know we support bigger blocks uh and again you know we'll have to talk to them about do they want to support XT do they just want to support bip 101 or you just want to support the idea of we need to scale up I don't really care what it is but just bigger so I don't know maybe there'll be

multiple categories of support but again until I start talking to these these companies um and people and and finding out kind of what they want to do um you know I won't know uh how that plays out um we'll see it it'll probably be somewhat chaotic and messy and and there won't be a great way of of knowing whether we really do have um you know Buy in um and I'm sure there will be some companies that are like no you know we think one megabyte blocks are great forever uh maybe at least I know of at least one small company that will say that so I I somehow missed this bit proposal I mean I knew that you were

proposing 8 megabytes but I I hadn't read through this uh this bit can you explain how exactly the function this Max block size function will work and how block size will increase over time oh for bip 101 um yeah I can cuz I I know exactly how it works I wrote the code of course um um so uh bip 101 proposes that after a a super majority of miners have said that they will support Big Blocks and that super majority is set at 75% so they they express support by putting a a different version number in the blocks that they produce so we can look at the blockchain and see that 750

out of the last a th000 blocks said that they would support bip 101 and so once that happens that sets a trigger in the code the code then Waits 2 weeks to give everybody a chance to upgrade so if there are those 25% of minors who who aren't supporting it they have two weeks notice that says you know this change is going to happen guys you better get on board otherwise you're going to be forked off the blockchain um once the super majority plus the twoe grace period happens then the the new rule for the maximum block size kicks in and the old rule was a block could be up to 1 Megabyte big the new rule is that a block can

be uh 8 megabytes or larger where the larger is measured by how far away are we from January I think it's January 11th 2016 so January 11th 2016 was picked as an arbitrary start date um and then there's this the maximum block size is is set based on that starts at 8 megabytes on January 11th and then it doubles every 2 years and so January 11th 2018 assuming that this change rolls out and miners are on board uh the maximum block size will be 16 megabytes and 32 megabytes 2 years later and 64 and so on all the way up to a really scary sounding number of uh 8 gigabytes

which uh freaks people out because I think people aren't good at imagining what our technology will look like in 20 years um and I'm pretty confident that in 20 years 8 gbyte blocks will not be a big deal uh just because I've seen the last 20 years of technological Pro progress um and it's amazing so this voting system you said that it it has to be 75% of miners but if you had like one mining pool that had um a large portion of the hashing power and they were mining more blocks uh doesn't that put them at an advantage yes I mean yeah it is Hash

rate based um so well I mean and that works on both sides right so a mining pool with 30% of hash power can veto this change right they could decide I like one megabyte blocks we're not going to change it I'm just the 30% of blocks that I create aren't going to support this change and and the change wouldn't happen um so yes I mean you know you're the the the when it comes down to actual blockchain Fork yes I mean it's it's a hash power vote the more hash power you have or control the the bigger your influence um and again you know that's why the task over the next several

months is going to be to uh you know talk to the economic majority talk to the big exchanges Merchants talk to users uh talk to the developers and try to get everybody on board um with this change so I also read I mean we don't have to spend a lot of time on this but I also read Jeff garak's proposal bip 100 uh which also has some sort of a mechanism by which the blocks increase but I believe that the values are voted upon every time there's an increase is that right yeah Jeff's proposal is that basically the miners get to decide kind of what the next maximum block size

should be and so there's kind of a regular vote I every I forget on the order of months isn't that more cautious than simply saying 8 megabytes and then we double that we double that we double that maybe I mean his he starts with the status quo so he has no jump up until the first minor vote um so it's more cautious there I mean theoretically miners could double up to uh he has a maximum of 32 megabyte block um could double up more quickly so if miners wanted to and they all were voting for the maximum increase every time you we could be at 32 megabytes more quickly than bip 101 um which would probably be okay right if the miners can are comfortable with that and they think the network can

support it then okay so be it um so yeah I mean on uh it it it I don't know it depending on how you look at it it's either less conservative or more conservative uh it's probably more conservative um I don't like it as much as as bip 101 I just I I don't really care what the rule is I but I really like predictable rules I really like rules that are I know exactly what the maximum block size is going to be on you know February 15 2022 right and and with bip 101 we know that right it's it's set down there in the spec we know exactly what it's going

to be in in much the same way that you know we know what the supply of Bitcoin is going to be on that date too pretty much right I mean you know satoshi's very predictable uh money supply algorithm for introducing new Bitcoins is kind of what inspires me to to to propose that we just have this limit be you know very predictable so that people can can plan right if you know what it's going to be it's much easier to plan in advance you can do some modeling of you know what I think the the the demand for transactions is going going to be you might be able to figure out what the fees are likely to be you know all those kinds of things I think are easier to do if you kind of fix this rule as opposed to kind of letting it be minor consensus

over time one question on that you know as we all know the the block reward hals every every four years and that to me seems to be not such a great design choice right because uh it it presents a lot of risk right all of a sudden you're mining well the revenue drop I have uh you know next year we'll see if something bad happens there or not right but you you might have that a lot of miners all of a sudden become unprofitable shutdown their mining Hardware hash rate drops all that sort of thing right I mean I guess Satoshi you know didn't necessarily think of that or just wanted to have something simpler instead of like so I mean I understand why he made the choices there but yeah like I think well

yeah I don't know if if you know I could go back in time I would probably urge Satoshi to make the block reward kind of continuous yeah every single block take the time stamp and use that to calculate kind of you know what the reward ought to be um so so don't you think that that would be better for the block size increase as well to have like a a continuous increase because I mean I can imagine in the future it is I mean b b 101 has uh the block size changes every single block so it actually is doing exactly that it's taking the time stamp that's part of the block header oh it is and using that to yes so when I say it doubles every two years it doubles but it's continuously right it's a

line so it is it is continuous it's not a sudden jump okay that makes that makes sense okay so perhaps bip 102 then should be that the block re Ward also follows that same pattern but changing the uh well I mean yeah no I don't think we should do that that would be too much risk for not enough reward right I mean it's not a big enough problem that the mining reward has and it's going to be less of a problem over time because the mining reward gets smaller and smaller over time right so uh I think I would I would be against you know that proposal yeah it it certainly would get create a lot of controversy and I think it would be almost impossible very very

hard to get through Reddit and I mean and you know I mean when we're talking about mining reward there are weird like incentives like if you can get more reward by tweaking the time stamp in your block maybe you try to tweak the time stamp in your block to get a few extra Satoshi reward you know there's there's stuff so um yeah I think we're stuck with uh are stuck with the the step function every four years for that so what exactly is the timeline on so you mentioned that actually the bigger blocks would come in early 2016 hopefully uh right if if all goes well so you you said you were talking to companies now and then you will try to

get people to switch to XT I guess sometime this this fall or next year or well the next I mean XT binaries will be available probably this week assuming that we get we're actually getting get in building you know reproducibly building the binaries uh as we speak uh my machine's actually working on that um and so we we should have binaries out for XT this week um anybody who wants to who thinks that you know 101's a good idea and they they think that scaling up this way you know we'll be free to download and run those binaries um I would encourage people not to mine on it quite yet you know wait a little while uh if you're a mining pool or Miner um the test net bip 101 has a test

net uh switch over happening earlier it actually already happened so you know we could mine Big Blocks on testnet and you know when I'm not talking to companies about this I will be doing more testing to make sure that you know Big Blocks don't cause any problems on VPS machines on machines with you know maybe limited bandwidth uh you know all those kinds of things all that kind of testing is will be happening one thing I forgot to ask before so so let's say everything goes well and or or or or poorly in the in the minds of those against this but let's say the the switch to XT happens how will decisions be made in XT is that

are you going to step back into that lead developer thing is Mike going to do that Mike Hearn will be the benevolent dictator he will be the ultimate decision maker um I've said that I'm happy to help out with maintaining that project you know accepting pull requests um yeah I I frankly don't really want to be benevolent dictator of an open source software project it just sucks up a a time um uh but yeah Mike will be the ultimate you know kind of decision maker there and and you know part of the philosophy talking with Mike part of the philosophy for the XT project will be that decisions will will happen fairly

quickly so there won't be this you know hanging about in limbo will my pull request be accepted or not um so between Mike and myself you know we we will commit to to making decisions fairly quickly about you know whether a given is going to be accepted or not and how confident are you because Mike seem to be under the impression that like a lot of the major stakeholders exchanges and and payment processors were on board with changing to XT how confident are you that at least this part of the coming to consensus will will come true I think they're all on board with bigger blocks you know we'll need to talk to them about XT and the other changes that are in XT um there are a few other patches that might maintains on top of

XT and and I think they're fine uh you know one of them I actually wrote myself and submitted to core and then it became controversial and and didn't get accepted um uh but we'll see we we we'll talk to them I mean you know if you're big enough infrastructure company then probably you're compiling the code yourself and you might even have some patches that you apply yourself on top of Bitcoin cores code to I don't know do whatever you need doing um um and and certainly you know the the code for bip 101 is available uh and and you know I will be maintaining patches against all the major Bitcoin core releases um so it'll be easier to for people to to pull just

that change if they only want to run just that change and we'll see I don't know I need to I need to talk to folks and see what they're thinking cool I know a lot of people just yeah they just want there to be consensus they don't really care what it is I think that's that's kind of the default position for for a lot of people in the space and they wish the developers could just go into a room and I don't know arm wrestle until until we came to consensus um and I don't know I don't know how to get a bunch of independent-minded developers uh who have different ideas and priorities to to come to consensus it's it's a really hard problem definitely yeah that that is one thing that's been apparent so well gav thanks so much for coming on was was really

interesting talking with you and I think it's a a crucial conversation at sort of a a very important stage in bigcoin development now and and I do think it would be it would be a great thing if we can end up with a different implementation and and more diversity when it comes to cuz CU how that is run so and and of course I do think also larger blocks would be a laudable thing oh good yeah and and I don't know there's been lots of um extreme positions I think on all sides of the block size debate and I mean truth to be told you know Bitcoin will be fine no matter what uh I think it will evolve

and and you know it will be good I just think kind of the the path can be smoother if we can get past you know kind of the extreme sides and and and and come to a compromise and and you know it's not going to be perfect it'll probably make everybody unhappy uh in some way but uh but we'll see I don't know even if we don't again as I said I think Bitcoin will be okay cool excellent well Gavin thanks so much for coming on thanks for having me good talking to you and uh so one last thing so we've started this this t-shirt bribing bribery contest so these t-shirts so

once a week we do a uh Lottery sort of so if you leave an iTunes review then you will be in the draw to get one of those some of you have already won it and we'll send them out soon uh and yeah of course your opinion shall remain unbiased so you can say negative things as well as wonderful things um so yeah thanks so much so you know our podcast comes out every Monday so you can download it of course get it for your podcast app or also get the video on YouTube that's uh youtube.com/ appcenter BTC and well thanks so much for listening and we look forward to being back next week n