EIP1559 Expert Panel | Tim Beiko, Hasu, Micah Zoltu, Barnabe Monnot, Matt (LightClients)

Transcript

bankless nation welcome to another state of the nation episode this is a special episode we have lined up for you because it is eip 15 5 9 week that's right the week has finally come the deployment of eip 1559 david we have so much packed in to this week but let's talk about the first event which we are hosting right now live streaming this on youtube you can catch on the podcast as usual later uh what is the panel that we have lined up today yeah we have tim baker who is one of the guys that really stepped up to get eip1559 over the finish line and then stepped into the all-core devs coordinator position with uh with all of ethereum

and so tim uh came to us and said i really want to ask some experts some some expert level eip1559 questions and so that is what is going on today tim uh pulled in who he thinks uh has uh just the best understanding as to what eip1559 is from the expert perspective so he is asking the questions that ryan i don't think you and i would have been able to even think up uh we're not smart enough man like we're not smart enough no we were smart barely smart enough to handle the uh the eip1559 introduction with with hassu that we had a few months ago and by the way guys if you're looking for a primer on 1559 we want to refer you to that episode take a look at the bankless archives we'll include it in the show

notes as well but this is a deeper dive because this panelist has the eip1559 rockstars you mentioned uh tim baco who are the other panelists david yeah we have a a bunch of uh uh core client developers we got uh barnaby a monot we have a light clients aka matt uh we have hazu uh who uh is uh i would like to say that hazu came over to the ethereum world because of eip1559 maybe that will be a question in the show and then we of course also have micah zoltul who is a major contributor to ethereum guys you can ask questions on youtube tim is is going to be um asking questions from his own roster of questions but after the conversation

dave and i are going to pop back on hopefully ask some of our own questions and address some of the questions in youtube so ask those questions we will be documenting those as we go guys this is just event one though because the actual deployment of eip 1559 happens we're gonna say what what block number is that block number 12 million uh 965 000 i believe if i got that correct david so what time is that going to be we don't know it's on the ethereum clock schedule so it depends on block speed but it looks like it might be around 5 00 a.m pacific time on thursday that would be 8 00 a.m eastern on thursday august 5th so you know david and i are going to be up and what are we going to

be doing yeah we're going to be live streaming with the boys over from eath hub and we're also pulling in some classic ethereum community members we got hudson joining the call i just got a message from danny ryan saying he'll be there uh trent van epps is gonna be there um tim i believe tim is also going to be there so uh it's going to just be a community call of people all watching the block of eip1559 getting mined uh and so that stream will actually be starting i think roughly around 4 30 a.m because we think that the hard fork is happening roughly around five so we want to be there a little bit early to make sure that everyone gets a chance to watch it happen live so david's gonna be up early not as early as our e2 staking live stream but still pretty early i'm

glad i'm on the east coast for this one last event for you later that day on thursday we have an eip1559 bulls panel we're talking about all of bullish things that eip 1559 could mean for ethy asset we have squish chaos dc investor james wang we are live streaming that at 2 pm the afternoon of august 5th so lots to catch up on the best way to just catch up on everything is to subscribe to the youtube channel bankless youtube and then click notifications on david one last thing and then we are going to get to the sponsors and then the panelists that is pool together they're doing some really cool things uh right now david and this is like celebration of scalability summer with pull together

so if you don't know pool together they are a no loss lottery so that means you deposit your uh eth it could be your your usd your stable coins in you don't lose any of it you only have the possibility of gaining uh they they actually use some of the d5 protocols to generate interest and then pay those out as prizes on a weekly basis so every week there's a lot of draw super cool protocol super cool way to save you can see right now if you deposit in some usdc there's a weekly price of 50k that you could earn if you click into that they also have some loot box prizes too in every single one of these these could be um you know other coins as well so super exciting the thing i'm most excited

about david is no gas fees because it is also layer two summer scalability summer and gas fees are much reduced with some of these right now side chain uh type deployments and polygon and others and i think roll ups will be coming soon so a lot happening there david anything to add no that's pretty comprehensive ryan guys just get in the pool uh the link is in the show notes bankless dot cc slash pull together relocate your savings account to pull together all right we are going to get to the sponsors right before this panel these are the sponsors that made this episode possible balancer is a powerful platform for flexible automated market makers typical amms just have two tokens inside of one

liquidity pool which can lead to fractured liquidity across the many pairs in d5 with balancer you can access the full power of multiple tokens inside of one single amm which unlocks an entirely new playing field of possibility this makes balancer an awesome building block for so many different use cases balancer pools can make asset indices but instead of paying fees to portfolio managers balancer lets you collect the fees from traders who use your portfolio for liquidity additionally balancer smart pools can be programmed to have properties that change according to pre-determined rules such as changing the swap fee based on market conditions or even liquidity bootstrapping pools which can help you launch and distribute your token with day one liquidity at bank list we use a

liquidity bootstrapping pool to sell our bap t-shirts too much success balancer v2 brings powerful new features that makes your money work even harder for you in v2 idle tokens are capable of generating yield in defy without sacrificing liquidity in the pool using balancer's asset managers balancers of vault architecture lets you trade between balancer pools at a fraction of the cost versus trading on other platforms balancers mission is to become the primary source of liquidity in d5 by providing the most flexible and powerful platform for asset management and decentralized exchange dive into the balancer pools at app.balancer.fi gemini is the world's most trusted

cryptocurrency exchange i've been a customer of gemiini since i first got into crypto in 2017 and it's been my main exchange of choice to make my crypto buys and sells gemini is available in all 50 states and in over 50 countries worldwide and on gemini there are markets for over 30 various different crypto assets including many of the hot defy tokens and it's one of the few exchanges that has liquid die markets gemini just launched their earn program where you can earn up to 7.4 percent interest on 26 various crypto assets if you're tired of paying fees and defy or you don't want to worry about defy exploits but you still want to earn interest on your crypto assets gemini earn is the product for you another product i'm stoked to get my hands on is the gemini crypto back

credit card which gives you three percent cash back on all of your purchases but paid to you in your preferred crypto asset when i get my gemini credit card i'm going to make sure that i get my cash back in east so whenever i buy something i get a little bit of eth bonus back to me at the same time you can open up a free account in under three minutes at gemini.com go bank list and if you trade more than a hundred dollars within the first 30 days after sign up you'll be gifted a free 15 bitcoin bonus check them out at gemiini.com go bankless one okay well good morning or evening everybody um welcome to this 1559 paddle uh that i'll be running for the next hour or so

um yeah so we have a lot of people here with uh pretty extensive experience with eip1569 um as you all introduce yourselves um one kind of first question i'd like to ask you is what was the moment that you kind of got 1559 and what i mean by this is that it's a pretty complicated change uh it tends to take people like a lot of pass through it to to feel like they they fully understand it um so yeah as you just give a quick description of yourself just walk through kind of what what made you get 1559 um [Music] matt do you want to go first sure hey everybody i'm matt i go by light clients on the internet

um i work for a small team called quilt we do a lot of core protocol r d and i was one of the main contributors to 15 for 1559 and go ethereum i don't think there was one specific point whenever 1559 like specifically clicked for me i think there were kind of a couple points and i originally read vitalik's paper a few months after he wrote it the pricing resources paper and i kind of at a high level understood you know 1559 is going to act like this dampener to dampen the spikes and gas prices during periods of high demand and that was something that sort of resonated fairly quickly through discussions and i kind of tucked that away for basically two years until i really

started getting evolved again in the end uh the end of 2020 um and so that was kind of one aspect i would say the other aspect of that was the the ten rough garden paper i really liked tim rufgarden's uh earlier papers on axiomatic approaches to block rewards and so when he wrote about 1559 i was like this is a really good um a really good paper describing all these things but there was there's another part of 1550 to my 1559 that didn't click for me until a lot later and that was how it affects the security of the the protocol and i was lamenting to some people like i didn't like i'm not really the biggest fan actually of this ultrasound money meme especially for core developers i think that it's

it's not really my place to you know propagate these memes and so i was lamenting to some people about this and they made this good point that we're moving to a world of proof of stake and proof of stake we have this economic requirement of ether to kind of have value to provide security to the network and we have a data to a degree in proof of work but it's becomes there's a lot less transactional cost of mining and only miners and stuff and so at that point i kind of you know back down a little bit from this ultrasound money meme the the community can can have that meme if they want i think it's it's done it's been a valuable resource and pushing 1559 forward not my place to perpetrate it but i think that the value of ether being something that's stable is is really good for securing

the network thanks for sharing that was yeah quite the history um barter bay you want to go next so hi everyone um very happy to to be here so my name is barnabe mono i'm a researcher at the robust incentives group which is a research group funded by the ethereum foundation we mostly do things that look like game theory a bit of economics modalization simulations and actually my experience with one five five nine started uh like everyone else when i was just trying to understand it and the way i understand things is i just try to simulate them and see kind of how they play out and

this is really kind of when it clicked for me and i think the major insight that i got and that i feel okay this is when i actually understand 1559 is understanding why the dynamics of what we call the priority fee or sometimes the tip are completely different from the dynamics of the gas price as they are today which is something that many people i think still have a bit of confusion around like why aren't we just shifting to a model where okay there's a tax the base fee and kind of people still need to compete on that tip to the niners and once you get this insight you i think you you really unlock um everything that makes one five five nine

like a step change in the way that the protocol is uh doing the the gas market and so yeah i hope we can talk a bit more about this but i'll leave the other guests thanks for for sharing um micah you want to go next yeah sure so i can actually show you guys my where i learned how eip1559 worked let me share my screen here real quick so uh back in july 2018 metallic wrote this uh article on e3 search forum it talks about basically what ended up being eip1559 and you can see here this is you know 12

am or whatever on july 3rd and then like i don't know six hours later micah shows up and i'll paraphrase so you guys don't have to all read this so it basically says come on vitalik why are you being an idiot this won't work there's no way this will work this is not new different than we have right now we can do better um then vitalik shows up and again i'll paraphrase a little bit and he basically says oh my god who is this guy i've never heard of him before he clearly doesn't know what he's talking about uh and he tries to explain some stuff to me but you can you can tell from his tone that he obviously doesn't think i'm going to get it i'm beyond help um then you know a few hours later again this is july 4th so we're going back and forth pretty

quick here um i basically just again tell him that he's wrong obviously and this cannot work and then he again tells me no you're wrong and we just basically go back and forth back and forth and then micah mysteriously disappears for like three months um and then uh i eventually come back to this thread actually maybe it was like a year later or something or another thread years later and i came back guns blazing and i was like oh my god everybody needs to get on board with erp 1559 this is the best idea ever hopefully no one remembers what i said in the first place um so yeah so if you're curious about my learning experience go ahead and look at this link on the youth research forum and you can see how

micah the person who is now probably the biggest proponent of vip1559 originally was the biggest naysayer and just spoke aggressively against everybody who thought this was a good idea wow do you remember what made you change your mind between kind of those two those two um it was just i mean vitalik ended up being right i was wrong and so uh he he patiently explained things to me and i just didn't get it and a lot of the things i didn't get were the exact same things that people who i explain eip15592 don't get and so i can appreciate where they're coming from it's um like i was there i was that guy who was like no this this won't work you this

this is not you know the mechanics are wrong um but once you actually dig in and really let it sink into your brain the mechanics actually do work like it it does make sense um but it's not obvious at first and the specifics are kind of nuanced and maybe we'll get into that later in this call but but yes it's the same reasons everybody else thinks it won't work wow that's really turning of of the tides um cool i want to go next uh sure yeah i'm hasu hi hi everyone um i'm a research collaborator at paradigm i'm also co-host of the uncommon core podcast and still general editor at terrible insights which i helped

bootstrap um my first touch point with erp-1559 i would say was in early 2020 when i did a month long deep dive with georges um who also um well also collaborated with later on on many other kids on um erp-1559 and we did what was basically a deep dive on all possible block space market designs that are employed in the crypto space so across all projects pretty much we look both at existing proposals and at um or basically existing implementations as well as stuff that has been proposed and try to understand

why it has been like what are the goals of the proposal and and for historic stuff why has it been discarded and um a lot of stuff that has been discarded has been broken so it was very very interesting to see all of the different stuff that you can propose and then like work through on your own why is this a broken proposal why does this not work and this is where we stumbled across erp 1559 which was barely starting to get any traction in here i would say um and yeah we like i was i would say mostly thinking about bitcoin at the time and not very much about ethereum and so my first instinct looking at this was well this is an over-engineered mess

so it has so many goals right described in the original post and so many aspects of it like so there's the fee burn there's like the elastic block size and then the code the protocol quotes a fixed price but they still have this tip option on top and it was so many things and it tried to be so many things at the same time at least what it seemed like from like reading the goals but then i tried to point out why it's actually broken and i had a lot of experience a lot of training at this because for months we had been dissecting all kinds of blog space protocols and i couldn't and like within like a few days i realized uh yeah this is actually like all of these um moving parts are

actually necessary and it just affects so many things inside ethereum that's why you can say it has so many goals but these are like really oftentimes like downstream effects right like like like like the fact for example that we burn eve is you could say like so the people who use ether's money they like that it creates a disinflationary pressure on the price of eve but then like people who care a lot about security like that it's immense eat as the fee paying asset in in the block space market even if the user pays in any other currency like even if you get extracted away on the user side which i think like we are all for in the long term but the miner still has to

keep inventory of ethan still has to burn eve so it has like all of these very nice um properties yeah um so yeah that that was basically my my experience with yeah p1559 and um so this culminated in in our post like analysis of erp one five f9 for derivative insights um again with georgia's um yeah and this um i think this for many people was the uh well the starting point hearing about this proposal because we we used to get a lot of reads uh at the time and um i think this this was really like for us when you you had like this this crowd of people who were really interested in and following it and asking for more and more and more

content on on it and um yeah i've been in so many twitter discussions and and so on um because so many people had the same questions right that i that i had um before it clicked for me and i think it was really good training like just engaging with people on twitter and on podcasts and explaining it to them like 100 times uh engaging with the different questions so yeah here we are yeah it's it's pretty impressive how almost everybody here or i think basically everybody here started off as a skeptic and um slowly got got converted to uh yeah to like 50 59. um bartemaeu you mentioned in your answer you know like the the new mechanism is is is obviously going to be

better for for uh gas usage and whatnot and you've also mentioned to me in the past that once uh it goes lives you want to actually track that and and try to quantify the impact um so can you kind of give give us a bit of detail of like what are you expecting to see what metrics are you expecting to change once we launched 1559 what like what would you consider a success what are you know what are some things that would kind of scare you if you saw them once once 1559 went live yeah that's a great question and i think my motivation here is really driven by the fact that i spent all this time doing simulations trying to think through okay how do things work when they're actually launched um

you can't really get these answers from let's say so we have all this test net which are awesome to test that the code is working clients understand each other but they don't really tell you so much about like the economic properties of the mechanism or really what happens when like there's actual money on the table so we know cannot go maybe too wrong because it is live in some version on some other networks but the things that i will be very much like looking at and that i think are super interesting to to keep track of the first one is kind of looking at if you're thinking of the chain you have this matrix block by block and looking at the gas used so you have

in one five five nine doubly sized blocks which allow to have like kind of this damper when you have demand shocks uh trying to understand well are the blocks getting too full so one question we often get is how often does one 559 revert to a kind of auction uh on the tip and hassu and georgios kind of made this meme of the 95 percent of the time like things go right five percent of the time maybe there's too much people at the same time and we need to to have this auction and in in that case it's it's the mechanism still works but maybe there's more instability on the on the user side and so one question is well is it really 95 5

can we quantify how often 1559 is pushed in in this kind of regime the second thing that i'll be looking at and it's it's certainly kind of dual to the to the gas used is actually the series of base fee themselves so we wrote a paper with some of my co-authors we've seen that most of the time basically so the basic update rule is kind of a multiplicative rule that depends on the gas that was used in the previous block these update rules they are really it's actually a fairly simple one and let's say even naive but it's it's good that it's naive in some sense because it makes it somewhat robust to whatever the

protocol can throw at it but we know that there are regions where that update rule can be maybe too slow to pick up the actual like if the demand is changing really fast you want the base p to catch up with this as fast as possible and in other regimes it can be also too fast and when it's too fast it just kind of flip-flops around a value without necessarily getting to that equilibrium and that can lead to instability in the sense that you can see blocks which are empty because basically kind of jumped up and everybody is priced out and then it goes down everybody wants to go in and you you can have this so we know that in theory that's something that can happen in practice maybe i mean we i'll be very curious to see if we if

we actually see that happen and one reason that we might see it happen which is maybe the third thing that i'll be looking at is also kind of how um so users send their transaction via their wallets and their wallets are using oracles to kind of determine the parameters for one five five nine you have two parameters the max fee and the priority fee and depending on how these parameters are set it's possible that you can find different patterns in in the way that the transactions are first dealt with at the transaction pool level so how they kind of get ordered by the clients but even when they reach on chain uh yeah what do you observe do you observe these oscillations do you

observe that maybe the oracles can be tuned uh given the data that we have on it like we often say okay you should take this kind of random constant as a as a oracle default we will change it when we know a bit more about the mechanism when we have the data and so i think it will be very important to to kind of back test how the oracles have done especially at launch where we have this mixed regime some wallets will be still sending legacy transactions so like the old format and new wallets or same wallets but new versions will be sending more 1559 native transactions so kind of picking apart uh yeah how wallets are doing can we backtest it can

we understand uh how to make them more optimal perhaps these are definitely the things that we i'll be looking at well thanks and so to summarize at a high level um three things how often do we actually get uh this reversion to the first price auction so basically uh 100 full blocks um what how does the base re evolve do we do we oscillate smoothly or do we just kind of you know go from full increase to full decrease um and how well do oracles adapt to set uh basically the the the priority fee for transactions right yes excellent scenario yes nice

um and uh micah i think you wrote what was uh at least for me the best explanation of like why 1559 actually improves ux for end users um and i know you spend a lot of time in in several discord servers kind of dealing with users who say have their transactions pending and stuck today um so can you take a few minutes to actually walk it through from like a kind of a a new user to ethereum um coming in like why is 1559 going to provide a better experience for them to get their transaction included than than the current system uh sure let me go find an article i wrote so i can read it real quick and remember what i said

apparently it was a great argument um if i remember correctly so basically right now what we see so i for the for those who don't know i run a little side business where um i offer so support services outsource support services to ethereum applications on it on on discord and so i have a lot of insight into what problems users run into because my team is constantly helping these users deal with those problems one of the most common problems we see with from users is stuck transactions or pending transactions same thing different different names and so a user does a thing on uh on ethereum like they go to uniswap or they go to whatever app they're using they go to

use it they plug it into in the app it pops up metamask and they say okay and they basically choose the defaults and they think that their transaction they just say this little pending spinner shows up and they're like okay my transaction should be included soon the problem is is writing an oracle is very hard so figuring out what the right gas price is is a very hard problem today and so that that app so metamask is doing its best to try to choose a gas price that results in two two things at the same time one the user isn't overcharged so they don't set a gas price that's really high and then the user ends up paying more than they needed to and two a gas price that is not too low that the user's transaction doesn't get included

when gas prices are stable so when the demand for block space and ethereum is very stable over time meaning like the exact same number of people want to be included as are submitting you know spread out over let's say 15 minutes then this this problem is pretty easy you just say okay what was the you know look at that previous blocks okay they're all around you know 20 or whatever so 20 is what we'll set this problem becomes much much harder when the demand for ethereum is increasing for for any reason so it increases because there's an ico or a new nft sale or just because people in the us are waking up or because it's a weekend or a weekday or you know it's the second thursday of the month and it's payday in you know azerbaijan or whatever you know whatever it is that

causes gas demand of ethereum to increase um is very complicated like this is the whole world of people doing things and for some reason they have all decided to do things on ethereum right now and so what happens is the user shows up fills in their men fills in everything hits submit and a gas price is set by metamask using oracle that usually works pretty well but this time it didn't because it estimated a gas price based on history and demand is increasing and so the gas price keeps going up and up and up and up but the problem gets worse so the user has a stuck transaction you think okay well they can just increase their grasp price the problem is the user doesn't know this the user just knows their transaction is stuck so they what do they do like okay i'll refresh the page

so they refresh the page but now the page doesn't remember that their head of pending transactions so then it just prompts them again hey what do you want to do on the internet and they say oh i want to swap some coins so they go and fill in the form again they swap some coins and hit the button and it still doesn't work so they try another browser and they repeat the process and now by the time the user comes to us for support or comes to somebody for support they've got like six transactions out there if we're lucky someone helps them that can walk them through the process now of canceling all those transactions because the user only one do one swap but they've got six swaps pending and so someone will walk into the process of cancelling the transactions which is not super hard but

it's you know it's kind of technical and users don't really understand what's going on so it's kind of blindly following people on internet who are giving them advice and we all know that's a terrible idea um and so like that's that's the happy path where people are following advice of random strangers on the internet to fix their problem the unhappy path is they don't go seeking help soon enough and what ends up happening is all six of those transactions go through and instead of buying you know five east worth of dogecoin they now have bought 400 east worth of dogecoin and they do not want 400 eth worth of dosh coins it turns out that drop tanks the dogecoin price or whatever it was they were doing like the thing they're doing is not the thing they wanted to do and

so now you have a very unhappy user who now is seeking help and like how do i undo this okay so the first thing they try of course is okay let me i'll swap back and so they swap back but you know for whatever reason there's huge demand and they get in the same situation so they you know swap back 20 times and now they're you know shorting doge or something i don't know like the point here is the users get into these really weird situations when things don't behave as they expect and with one five five nine the hope is that we can kind of solve for the most part solve this whole class of problems by just making it so that one root problem so one root problem in this scenario was writing an oracle for the gas price is hard predicting what the future price of

gas is is a really hard problem because you both don't want to go too high and you don't want to go too low and so you can't just like over bid because then they cost the user too much money you can't just under bid because then you're not stuck and so that's a really hard problem and so 1559 will hopefully make it so when the user goes to goes to metamask metamask can follow a very simple algorithm that will 95 percent of time or hopefully even more 99 time for lucky you know we'll get it right and will not result in users in that situation and when it gets it wrong we want it to fail gracefully so in the ideal situation that 95 99 that we're hoping for what's going to happen is the user will be included in the next block and they'll

pay exactly like the minimum they need to pay and that was it in the kind of worst case hopefully worst case scenario the user will have to wait a little bit but they'll still get a good price like they still won't end up overpaying and in the you know worst case scenario they you still i will acknowledge you can still end up in this kind of pathological pending transaction situation but the hope is that becomes a much much more rare thing that it doesn't happen to like you know 100 users a day it's happening to you know one person every couple of months get stuck in this situation and that's a significantly better situation just from usability and making it so users experience with ethereum is simpler and and all this just because we can solve that oracle problem easier because we

now have the data and we now have this second price auction thing so you don't need to try to do complicated auction stuff yeah that was a very colorful uh overview of of the situation and i think it's one of the things that gets missed a lot like when we talk about 1559 people say like it'll improve the ux um but um yeah that's often like a very abstract thing uh so yeah this helps kind of put it in perspective um moving up to matt uh you've worked on like not only 1559 but on a ton of other kind of pretty uh important eips like

your team worked on account abstraction last year which eventually morphed into 30 74. you've implemented 1559 and you're one of the people who has background both kind of with the smart contract level and the core protocol level which is which is rare usually people are on on one part of the stack or the other um so how should smart contract devs ability on ethereum kind of think about 1559 and what you know should they change anything about it how should they how should they basically yeah change their practices or approach 1559 as it gets deployed [Music] yeah i think that's one of the cool things about 1559 is like micah was saying there's all these ux improvements

but for the most part smart contract developers get to just take advantage of all of these things for free like there's been a lot of work by core developers and researchers up front to try and make sure that 1559 is a change that's safe that it's actually works the way it's intended to and a lot of other burden has been shouldered by these wallet teams like developing uh how they're going to choose these priority fees and interoperating with new rpc endpoints and so now the smart contract developers kind of get this free ride they just get to take advantage they get to continue building their their smart contract to take advantage of a lot of this work that people have been doing over the last few years there are a few uh nuanced cases where 1559 really does

come into play and one of the interesting ones is that the gas price op code will begin to behave slightly differently it's not really something most people are actually going to notice but now for 1559 transactions the gas price opcode will return the effective gas price of the transaction so that takes into account the base fee it's the actual amount that's being deducted from the account paying and if for the most part people who use the gas price opcode this probably isn't going to be a problem there is like a very specific class of people though that use the gas price opcode and pathological ways or expect that they could potentially have a gas price of

zero and that's something that won't be possible with 1559 anymore so you've kind of removed you know a few paths of execution that could have existed in the world so now if you ever had anything that was checking if gas price equals zero then do this that's not going to be possible anymore and so if you were a smart contract developer who decided that was a great idea then that's something you should consider there is a class of people i don't know if they actually have gas price equals zero but flashbots does generally require that these bundles are being included with a gas price of zero this kind of helps them from if their transaction is unclear they don't have to pay the gas costs that go into uh the transaction and so if you are someone who's planning to use flashbots

for something whether that's for mev or for account abstraction or user onboarding these are things that you should take into account and i'm not certain what flashbots has chosen to do about this but now you have to understand that you won't be able to have zero gas price transactions anymore so that's kind of you know how 5059 changes what exists in the evm today but 1559 does add this new base b functionality and an op code was added not a 50 59 but in a different eip 3198 and this base viop code is going to provide a really exciting new functionality to the protocol a smart contract developer should consider looking into and maybe using and there's a couple like places where this base

fiat code could be really valuable the first one is if you want to pay for some uh bounty to be executed on chain i know there's a couple of protocols that do things like this where maybe someone will say i want to have this limited order executed at this price and a relayer will decide to submit it on chain whenever that bouncy would pay out but there's kind of this difficult problem with like how do you decide what a fair price to pay for them executing that on chains because if you you know if you want to have a bounty that pays out like years down the line it's really impossible for you to know like is 10 way going to be a good price for that bounty to pay out to the user or is like are we going to be like you know 200 way and you can't really

use the gas price off code because then the person who submits the bounty could really just grief the user they could submit a 500 way transaction and there's no way for the execution of the smart contracts right now to get a fair idea of what the gas market looks like but base fee is basically a mechanism for having insight during execution of what is the current demand for gas in the protocol and so this creates like this really nice mechanism where you can pay for bounties more fairly and having this base viab code is also going to allow like new financial primitives around the gas markets to emerge i think we'll see people develop like future

mechanisms uh based on um the gas market people want to speculate whether gas prices are going to go up and go down and we kind of had this with gas tokens because you could mint a lot of gas tokens whenever gas was really cheap and then there was demand for it if you could sell them whenever the gas prices were higher you could call them in but there was a very significant transactional cost of using gas tokens and it wasn't really something that was a net good for the protocol in general and now this base fiat code is going to be something that we can use to much more efficiently develop gas market primitives so i think that's something that interesting that smart contract developers should should look into and consider the last one is really for

these like layer two scaling solutions there are a lot of fraud proof based solutions roll ups uh even state channels have some sort of like replay mechanisms to play out states and right now you kind of have this this uh this world where we just choose a static amount of time to give for people to submit their fraud proofs on chain and i know like optimism says like one or two weeks is this is the challenge period that you have to submit a fraud proof and that's to avoid you know minor censoring for a long period of time what we might be able to do with base fees we might also be able to look at the actual load of the chain and determine how

how many people are trying to execute things on chain and potentially expand those fraud proof periods during periods of high demand and so that might allow these challenge periods to become more efficient so they can kind of become elastic with how much demand is in the market for block space thanks yeah that was pretty comprehensive um and one thing i'll add is uh you mentioned uh you know using the base fiat code to build gas derivatives um it's also worth noting that london has another eip 3529 which will make the current gas top tokens uh unworkable uh because of how it changes gas refunds um so yeah if you are kind of a heavy user of gas tokens

um that's not gonna be possible after london and so um yeah looking into ways to build these gas derivatives off the base yap codes uh is is probably worthwhile i should probably also mention that if you're more of a full stack developer for ethereum there's actually a fair number of changes that came to the rpc with london and so this isn't something that if you're a pure smart contract developer you may interact with but if you are a someone who maintains a back-end system for a d5 application you should consider looking at i think tim you and trent put together this this comprehensive guide of the the things that are changing with london uh we could probably link that in the show notes but there are a handful of changes and i'll mention like one the if you're

calling this rp these rpc endpoints that receive transaction objects there is a gas price element of these transaction objects and normally that's just the gas price of the transaction but with 1559 we now have these two different elements that kind of relate to the fee market we don't have a gas price but for backwards compatibility reasons we're continuing to support gas price and so that value would be equal to the max fee for per gas before the transaction is mined and then after it's bind it will be equal to the effective gas price of the transaction and that's something that you should be aware of if you're a developer interacting with these things um so yeah take a look at this this document that tim and trent put together yeah good point um yeah a lot of applications are not just smart

contracts and they'll actually have to interact with the the node itself um and changing gears completely um there's been a lot of like talk about minor revolt around 1559 and the fact that the base she burns the east which which obviously lowers their income um and i think cassio you put together alongside with georgios i believe kind of one of the best explainers of why it's in miner's best interest to support 1559. um so could you kind of quickly summarize that that argument for us and yeah what you how you think kind of the game theory plays out here yeah i mean so you already touched on it right so why would miners even be uh interested in blocking the ip1559 or

somehow like revolting against it it's that uh the eip will burn some of the fees that are generated in the block space market right and at this point i would ask you like how into how much detail you want me to go um because it would cover like um stuff like mev and uh how that plays into how much it's being burned since i understand these might be like later questions oh go for it yeah okay yeah okay so um basically the first analysis that we did was first to find out how much is actually being burned because we thought that before that point there was this very well sort of naive understanding that like all the transaction fields that exist today in ethereum um would end up

being burned and so like all other fee revenue that miners make would go down and they would be only left with the block subsidy but fortunately at the same time was when flashbot started to get a lot of traction sort of the the private relay for miners and they um they were very transparent with their data and they started classifying a lot of sources of mev minor extractable value and made public how many of the transactions um in ethiopium today are sort of transactions that fight for for mev stuff like liquidations and front running users and and so on and arbitrage most of all um

and so we we took that data set from flashbots and tried to map it over the total miner revenue in order to understand how much of what miners make today is actually from med and how much is from regular transfers and the reason we did that is that um to understand how much is being burnt you can think of sort of the the block the list of transactions there's not actually one market but actually several markets so the the market to be included like in anywhere in the block is actually a very different market um from the one about being included at a very specific point in the block for example like at index zero like being the very first in

the block so you can be the first to do an arbitrage between unisop and sushi swap for example so there was a very important insight that these are two different markets and in the priority market basically erp that 1509 doesn't change anything um this is this is a market that will always happen um so between those who want to submit their transactions get their mind in a specific slot and the miners who control the ordering and and this is money that couldn't possibly be burned like because if the protocol set some value for that then the market would just move off chain for that um yeah and so we realized that all of the mev

that is extracted in atm today will be like untouched by erp-159 and wouldn't be burned and we tried to to simulate that we found that like maybe 30 percent um of the fees that are generated today are actually fees that are uh from regular transactors who don't care about where their position in the block is and probably 70 or more is actually nowadays just from uh mev extracting transactions so uh this was like an important insight for us because it showed that it would be like much less impactful on miners and

i think this also helped sort of to diffuse um the opposition from miners a bit um because if you i mean you you you could argue that it's in their best interest for miners to like adopt any change that sort of the users want um and it is to a degree but you can still create disruption in the mining market and create like perverse incentives there if you end up reducing the revenue that you pay the minus by too much over too short of a period of time and um so it was it was quite good that um you know the the reduction of revenue turned out to be much smaller and um also that the that like the uptick the uptake from the market of

this update was so positive that arguably contributed to uh like how ethereum and ether is the asset was perceived in in the the the the public basically right so because for miners it matters not just how much revenue they make they nominated in eve so they don't just care about their denominated fees but of course they also care about uh so the other side of the equation how much can they sell that e4 in dollars so um so both of those were uh i think very important so i think you have this one side which you just discussed that the the reduction of fees is actually not that large um and and second that miners are

structurally long eve um like they have a lot of future revenue at stake and so generally they are incentivized to go along um with what users want and what they think but users and developers and and all of the parties building on an ethereum what they think is best for the ecosystem because that is what maximizes both the value of if and so the economic activity that ends up generating uh all of this mev and all of this transaction activity thanks yeah um i guess yeah this is one thing i'm curious to hear other people's thoughts on this because estimates will vary wildly you mentioned you know the current fee market is you can think of

the gas price as as combining two things the right to be included in a block at all and then your positioning in the block and revenue in a post-1559 world from just the right to be included in the block is what's associated with the base fee that's what gets burned and then revenue from uh positioning or right of inclusion when there's there's too much demand is is what goes into the priority fee um and you estimate there's basically like a 70 of this that would end up in the priority fee and therefore not be burnt and then 30 would would uh represent the base fee um i'm curious if anyone else here has thoughts on on that breakdown do you all agree this is kind of what we should expect or um yeah anyone

have strong opinions on this i would add maybe one thing because this isn't this does not ever elect refute what uh so what barnaby said earlier that you have different regimes right where the demand is very high to get into a block and that's why you end up with the tip auction so this is something that happens in every block even if the block is just 10 full you are still going to have competition to be in the top spot in that block that has otherwise zero base fee yeah and zero tips and yeah i guess a way to think about it it's almost like if you had two blocks and one of them had just room for one transaction and then you know the rest of the transactions is almost like a separate block yes um yes this is exactly the right framework thinking of

it not as one block but actually as several blocks stacked on top of each other yeah yeah and there's always gonna be obviously kind of one transactions worth of demand for that sort of premium block and then the demand for the rest of the block space varies based on just general demand for ethereum um yeah i'm curious do people generally agree with that split the people have kind of differing opinions so i i don't know the exact number breakdown um the thing to keep in mind for for everybody is that the mev or minor minor extractable value is highly variable and so we will see some blocks that have relatively little like it's not very interesting and you'll see a

block that's got like 100 east in it and so um in that 100th block as hasu kind of described like the person who paid 100 to get to the front of the line they really really really really want to be in the front of the line like they're willing to pay a hundred each to be at the front of the line um they are going to pay a hundred each to somebody to get the front of the line like if either that's a minor through gas price or it's a miner directly or it's some off-chain payment channel or it's you know a cash dropped at a underneath the mailbox like they will figure out a way to when you've got underneath you're throwing around you'll figure out a way to get the money into the hands of the person who gets to decide who goes first

and so um so yeah so from that standpoint i totally agree with everything has to said like the you can't avoid that you can't make that go away you can you can drive into the dark which just makes it harder to see and harder to record but it will always be there and so uh but that means result regarding 70 versus 30 percent um it's hard to say like uh there's many people who have done research on this and come up with different numbers i think a lot of it comes down to like what time frame you look at like if you look at something like back when things were going crazy what was it like six months ago or whatever nine months i don't know how long ago was um like that's very different than you know

a boring week in the middle of summer you know and so um just keep that in mind with with any sort of stats like this that it really depends heavily on what time frame you look at like if you look at like a one a five year time frame it's very different than a one year time frame which is very different if you look at 2018 versus 2017 like there's so many variables here it's really hard to come down to make a concrete number um but in general i think casual is generally correct that there there is still a lot of value going to miners and will continue to go to buyers as long as they are the ones choosing block order any thoughts

i mean i don't have much to and i think it's it's quite clear but i i do have in mind this idea yeah basically looks like today's gas price or like average median gas price let's say priority fee is mostly going to be nominal like the minimum you need to send to be included by the miner except maybe when you have kind of like these tip options and then you have this med process which i think so maybe one thing is the 70 30 figure it kind of hides the fact that as mikael pointed out an ev is extremely variable so you have much more variance perhaps in revenue for minors and and this is perhaps something that's

yeah worth just pointing out yeah and i guess talking about mev um i think hasu your analysis was maybe one of the first ones uh related to 1559 that actually took me into account um but if you look at a lot of previous analyses um including for example tim rufgarden's paper um or even some previous papers that were written on how kind of high transaction fee blockchains are less stable and more prone to short reorgs than high uh block reward relative to transaction fee chains um none of this kind of took nbv into account that they

were all kind of written in the in a pre mev world um so i'm curious you know are there other spots where you know aside from the actual just minor reward um you think mev can have an impact with 1559 oh was that at me i thought anyone yeah yeah oh yeah yeah i don't really have anything uh to say at the moment so if anyone else has any thoughts maybe just to yeah point out that by design 1559 really addresses more of a problem of pricing inclusion in a block and it's almost completely agnostic to to the problem of position so inherently there's not so much

interaction between the two like the base fee is kind of yeah this minimum fee to enter the block and so that really has nothing to do with what happens at the top heavy like long tail of this high uh gas fees so yeah that's maybe one reason why mev and 1559 don't interact that much the one place where they do is now these mev strategies they need to pay the base fee whereas before they were able maybe to submit more like zero gas price transactions so but this is not a huge um i think difference yeah i would second that i think it just makes sense just even for the sake of it just drive from uh like the the notion that the eip159 does not fix mev

um because for a long time uh this was a popular talking point i would say um and um yeah it took like i think it's just important to say it again and again and again like there's not really much overlap between the two apart from what what wannabe said um mika you unmuted yourself do you do you want to say anything i was gonna say basically the same thing uh erp5159 does not fix um censorship attacks like censorship attacks are still a thing and they are still a risk this is where moving to proof of stake but that's for a separate call 1559 will not make those go away magically

uh yeah i i do think this is a yeah an important point to make the other thing 1559 doesn't fix is high gas prices in case anybody still believes that um so yes oh yeah that's that's the other good one yes um um cool i guess uh bring it back a bit more to the the technical level um matt one thing that was interesting is you actually implemented 1559 and geth and you're not part of the get team itself and so you kind of ended up writing the reference implementation the implementation and kind of the reference client um can you talk us through like what that process was like how you know yeah as like a non-maintainer you got this this pretty

massive change uh included in the court in the code base yeah so i think like the journey for working on 5059 for me started in early 2020 when we started working on account abstraction and so the my team quill we spent about six seven months working on account abstraction we wrote an implementation in in geth of this eip and we were benchmarking all these things and during this time there there was like more discussions this summer about how to have typed transactions in ethereum and this was something that micah was was proposing and i was really interested in type transactions because

that was a way to path forward to having sponsor transactions in the protocol and that's something that like is really exciting for me because i think that's a big ux problem with ethereum and so i was all on board with type transactions and it also helped account abstraction and this is kind of whenever i started getting more interested again in 1559 because 1559 was another type of transaction and so i sort of volunteered in the fall i said you know it sounds like type transaction is something that's useful let me just implement it and geth and so i pinged the go ethereum tip martin from the team and i said hey uh is it okay if i just implement 2718 for you and he said sure go ahead knock yourself out this was kind of like earlier in the cycle of berlin and it

wasn't 100 certain it was going in and so they didn't want to waste cycles on dealing with 27 18. so i and it was a pretty big change in the code base because it basically added this discriminator to like it basically broke this transaction type up in the go theorem code base and now you have like multiple transactions and kind of this new abstract type for transactions and so i would kind of like ping them every once in a while and say hey does this does this look okay and and they would you know respond and say yes this is good this isn't good and generally that would happen i mean a lot of this stuff happened in the open through through github and through pr's and so i think that if people are like watching this this uh live stream and they're interested in core development like a

lot of the times it's just a matter of opening prs and trying to become part of that process so i did 2718 and that really set me up for working on 1559 and i think tim i don't know if it was started with like you kind of like pinging me and you're saying hey are you interested in helping with 1559 and i think i was kind of like rebuffing you in the beginning because there we were working on some other things at the time and 1559 has really never been something that like i personally was very very excited about because it was something that made a lot of sense but i was more interested in things like sponsored transactions and so i kind of like rebuffed in the

beginning saying oh there's other things we'll work on and so after i think a little bit i just realized that you know there's only a handful of people right now who can implement these kinds of changes in go ethereum the ethereum team at that time was like doing a lot of other other things they were working on snap sync that was like a really big thing they were trying to launch and so it was like this is something that i should do this like this isn't what i want to do necessarily but this is like for the greater good of the the ecosystem so i need to become a part of it and like it worked out really well i really had a good time uh implementing and so i implemented most of 1559 and like one or two weeks the core consensus changes were

like relatively straightforward and i kind of like put a pr together and shared it with the go ethereum team and was able to start syncing with the 1559 test nets at the time that adele from the basu team had put together and so the actual core changes were not super significant they touched a lot of parts of the code but like generally it wasn't a crazy change and then basically for the next six months it was a matter of all of the auxiliary pieces all of the testing and i don't think that it's i don't think a lot of people understand you know just implementing that's like 10 of the work and you know i'm already standing on the shoulders of giants all the people who have done the research vetting vetting the idea in

general like i think 1559 is one of the most rigorously analyzed eaves that we've ever had going to the protocol on all uh aspects from you know original design research to engineering to testing to fuzzing and so i for about six months period was just helping implement all these other things right working on the mempool implementation changing the way that the miner works and um a lot of the the work lately has been rpc and endpoints and 15 and 50 1559 changed a lot of aspects of the rpc and i think for me like the biggest takeaway from working with 1559 is that

we need better facilities for discussing changes in the ethereum protocol and if you look at the 1559e but basically just implements a really stupid ethereum client in python because that is kind of like the best way of describing these changes and so i think the similar thing with this are the rpc endpoints it's really difficult to just discuss in pros and through one-off eips how these things are changing it's a lot simpler to follow a more traditional software engineering process and say here's a new release of the rpc etc yeah thanks and i guess a quick related shout out that you are working on an actual full spec

of ethereum uh executable spec so that'll be really helpful to have um yeah shout out to my team sam and somo they've really been grinding out we're gonna replace the yellow paper with the python implementation so that's very exciting yeah that'll be great uh micah i see you have you went off mute for a second uh yeah i was gonna say that just to echo a little bit of what client said in the middle there um i think a lot of people assume that you know a platform that's managing 200 or 300 billion dollars or whatever probably is being built by a company the size of microsoft and has you know 40 000 people working on it and so surely one developer won't help um in reality this platform that is managing you know

hundreds of billions of dollars has far far far fewer people working on it than microsoft has and yes one more developer definitely can't help like there there's lots of places you can contribute like matt said one of the greatest ways to get started is just find something that you can make better in just a small way and submit a pr and just make it better you know if if you're not a developer there's other places you can help as well you know eip editing is a lot of just kind of grammatical and clerical work like just kind of making sure things flow through a process you don't need to be an expert in you know some arcane language in order to do that lots of places people can contribute if you're interested yeah i definitely agreed there the

amount of people who make this possible is shockingly small um yeah i guess yeah before we we move on to audience questions uh i have kind of a final two-part question for you hasu so yeah you were actually pretty new to the ethereum community uh [Music] and i could say you know 1559 was almost your entrance in into this community um so i guess first i'd be curious you know to hear you just like talk through your experience of kind of advocating for this massive change um as someone who who wasn't like deeply embedded in the community um and then kind of the second part would

be is is 1559 something you would want to like campaign to see elsewhere uh possibly even on bitcoin in the future okay um so yes i would say that yeah yeah p1559 was sort of my my entry into the eve community i i had been e-folder before that and i was still following it um but yeah i really got my start with the ip159 and with d5 when i really started to appreciate what you can build on ethereum and yeah i mean how what what that affords you basically the trade of study here makes because as a bitcoin are you always focused on the trade-offs that the theory makes but you like some people don't see all the good stuff that buys you right

the ability to have non-custodial permissionless finance have a sandbox for developers to play around and i think we will see amazing things um happening as a result of that in in the next decade um and yeah i would say that it's hard like to trace back exactly like um where that started like that that turning point for me but it was around that time definitely like both the ip159 and d5 and seeing you know just writing research and just it was like taking up so well by the community and like the just the feedback was so good like so many people were open to collaborate it was it was just i mean it's it's a second meme but it was really like everything that like people praise about the human community so i could

really like confirm that this was definitely true for me right that it you have a collaborative spirit and um everyone's really open to talking to you like no matter who right can be like really high profile people i never had the the impression that like any there were any closed doors um in that sense so uh yeah i think uh that's why nowadays i would definitely say i'm more or at least just as much at home in the atm community as i'm in the bitcoin community and going back to sort of my intro um when georges and i did this um deep dive into block space market protocols this this was at least partially informed by sort of my

like a deep-seated fear that i have um about what happens in bitcoin when the block subsidy goes away because until this point uh transit like the demand to transact hasn't really been enough at any like longer period of time um to pay for security in bitcoin so there's really like two uh possibilities here like i either the demand picks up and like we we still have this like systemic problem but it's sort of it's sort of not not not relevant because people there's not enough demand to transact and people are going to um to pay these fees or we make some change to the block

space market mechanism or to the to the monetary um yeah to the monetary policy basically and yeah this was sort of the original what set us out on this quest to to find um what's possible and i don't like so first of all i would like every chain that has um a block subsidy i would like recommend to adopt eip159 and i think we are seeing it almost everywhere now like almost every new um ethereum competitor starts out with efp159 implemented i think we know i've seen like at least eight chains seven eight chains that have it have had it before ethereum which is kind of funny but i mean that's just how the thing how things go right

when you have these like smaller more nimble um chains where there isn't a lot of at stake and you can just you know just do it so for for them i would definitely recommend it but this shows sort of why it may not be possible in bitcoin at least like as the designers right now because eip15f9 burns the non-priority part of the transaction fees which in bitcoin is 100 of it basically because there's no med or almost no med and so you would burn all the transaction fees simply speaking and we already don't have enough transaction fees so you you kind of rely on there being a perpetual block subsidy that can pay for

security um and you don't have that in bitcoin so i think unless that changes which i see is very unlikely i think it's unlikely that we will get we get to see any sort of proposal where transaction fees are burned or the the um or the protocol itself sets the price of transactions i would support other proposals um for example uh fee smoothing mechanisms um where sort of part part of the fee is paid forward into a security fund for example that can then like when demand transact is high generate more fees than necessary so you start paying some of those maybe into a fund and

uh and later then when block subsidy has declined you start consuming the so at least buys you like maybe a couple more years or so until this becomes serious problem so yeah that's that's a proposal that i think would be more fitting for bitcoin and one that i would campaign for yeah that's super interesting and you can even imagine like a naive implementation of that where if fees are on average high enough to secure the chain but maybe not on a per block basis then like a 15 59 mechanism where the fees get sent over say the next 100 blocks or something can help you convert this kind of volatile income stream into something that's a bit more uh predictable um as a security budget

yep yeah um yeah so everyone will have done 1559 before ethereum and ethereum will have done it before bitcoin um i think this is a good place uh to open up for community questions um david you want to take it back absolutely guys that was an absolutely fantastic panel i'm gonna have to rewatch that myself to make sure that i caught everything so thank you guys all for being here i need to get ryan back into the show because we kicked him out before just video logistics uh so we're gonna take community questions if you are watching on youtube we have been logging some of our favorite questions in a little document but this is your last chance to get new questions in there we're gonna cut to sponsors for a quick two minute break i'm gonna get

ryan back in here and then we're gonna start going through some of the community questions that we got through youtube so here we go as soon as i find it bankless is proud to be supported by uniswap uniswap is a new paradigm in asset exchange infrastructure instead of a cumbersome order book system where trades are matched with other humans uniswap is an autonomous piece of software on ethereum which is what ryan and i call a money robot no human counterparties or centralized intermediaries just autonomous code on ethereum input the token you want to sell and receive the token you want to buy something brand new in the uni-swap ecosystem is the uniswap grants program is now accepting applications for grants

we have been saying this for a while and will say it again dows have money and they are in need of labor if you think that you have something to contribute to the uniswap dow apply for a grant to uni-swap just look at the size of the uni-swap treasury it's almost 3 billion dollars this mountain of capital is looking for labor do you have something of value to contribute to the uniswap dow no matter how big or small your idea is you can apply for a uni grant at unigrants.org and help steer unit swap in the direction that you think it should go that's exactly what we did to get uniswop to be a sponsor for bankless and you can do the same for your project thank you uniswap for sponsoring bankless the ave protocol is a decentralized

liquidity protocol on ethereum which allows users to supply and borrow certain crypto assets ave version 2 has a ton of cool features that makes using the ave protocol even more powerful with ave you can leverage the full power of d5 money legos yield and composability all in one application on ave there are a ton of assets that you can supply to the protocol in order to gain yield and all of those same assets can also be borrowed from the protocol if you have supplied collateral here you can see me borrowing 200 usdc against my portfolio of a number of different defy tokens in eth i'll choose a variable interest rate because it's a lower rate than the stable interest rate option but i could choose the stable

interest rate option if i wanted to lock in that interest rate in permanently v2 also features the ability for users to swap collateral without having to withdraw their assets trade them on unit swap and then deposit them back into ave with ave users can do this in one seamless transaction saving you time and gas costs check out the power of ave at ave dot com that's aave aave.com hey guys we are back the panelists are still on we got some of your questions that came in by youtube we're going to ask them david man it was fun to watch as a guest on one of these bankless panelist episodes uh what a fantastic group of individuals uh you know tim and crew managed to pull together and uh yeah what were some of your

impressions i guess yeah my my impressions are there's just an entire massive playing field of just thought around eip1559 that like i've i've thought i was paying attention to eip159 but like there's it's one of those things that it has rabbit holes in of itself and so like i said right before we cut sponsors i'm gonna have to actually re-listen to this for my own benefit rabbit hole inception rabbit holes within rabbit holes yeah and i you know a few interesting things to me i tweeted some of this out it's like everyone on the panel seemed to start as skeptics and then they turned into supporters that's kind of interesting um you know a panelist made the question like client made the question eip1559 is one of the most rigorously tested eips to go to the

protocol um we heard hasu's story about uh moving into the ethereum community as part of protocol so i think this this represents a lot of what's great about the ethereum community the level of analysis the permissionless nature of it it also represents some of the challenges like um i don't know how many times the panelists said hey there's there's not that many of us actually doing this super important work like come join us come submit some prs like we need your help in order to continue uh this this whole research project and implementation project we call ethereum anyway uh lots of stuff to dig into let's get to the questions though and i'm going to start here because one of the panelists um actually may have thought they showed up to the wrong

medium because i pronounced this eip 1559 and i've just been told that might not be the most canonical way to represent it so actually somebody in uh our discord channel asked this question this is oddba11.eth maybe these panelists can tell us if it's pronounced one five five nine or fifteen five nine or as i have taken a habit of saying 1559 guys which is it what is the correct correct pronunciation of eip blank i won't bias anyone by saying what i think it is uh like client you have any thoughts on this yeah at the beginning of the show i you know it was just settling in and i heard

you say welcome we're going to begin speaking about 15 5 9. [Laughter] i missed everything after because i was like oh no is this this is the wrong call [Laughter] i've actually heard people say 15.59 in in a while at least i feel like it's 15 59 that's the i don't know tim probably talks about tim probably has dreams of different ways of pronouncing because the optimal way of pronouncing 1559 but to me it's always been yeah 50 59 that's what we do on bank list yeah

that's like actually the right bankless branding maybe that's just me i don't know any other thoughts panelists i mean we're talking about naming obviously i have thoughts my personal favorite is the 1500 eip since the start of ethereum that's usually when i'm in casual conversation that's how i'll express myself to users sometimes if i'm you know in a more business environment i will say the fee market change for the ethereum chain um you know it really depends do you guys think that we're going to need to call this differently in the future after it's implemented or is kind of eip1559 here to stake is it is it just stuck i think that the boats

of calling eeps by not their number sailed a few years ago oh my god i mean even stronger than that other blockchains call it eip 5059 which always make me laugh right i don't know how many times i've had conversations with people like you know david and i like work on the the layers here the the social layer the the meme layer of ethereum uh if you will and like can we come up with a better branding for this thing and no we threw around ideas we had names we tried some things nothing stuck like absolutely nothing and so eip 1559 it is i think for eternity where where are we out on bips like can

we just put some like no op dips in to get to bip 1559 [Laughter] i don't know uh what bip are we on what oh 100 400 so i don't know how it works in ethereum but bibs and bitcoin aren't given chronologically um [Music] so they are like the number goes up but uh like the first part of the number is is reserved for like special things like it says what kind of what type of bit this is so what what kind of what part of the system it touches basically as long as number goes up bitcoiners are happy we know that it's true david you want to ask the next question

yeah maybe turning to more serious questions a question from qb on youtube is what happens when people send the old transaction types after eip1559 is switched on um maybe i'll throw that one to tim sure um basically what it does is behind the scenes it'll treat your legacy transaction as yeah behind this uh matt said in the chat of our funds are burnt i don't know so behind the scenes it'll send your your trends it'll convert your transaction to a 1559 one and what it does is it sets your gas price equal to both your max fee and your max priority fee um and what this means for the user

is that when you send a 1559 transaction you basically get a refund for the difference between your max fee and the base fee plus the priority fee so if you put a very high max fee you'll never pay more than the base fee and your max priority fee and that makes it very easy to figure out you know to set a transaction because you can just set a high a high max fee and know you'll get included and get a refund for legacy transactions um your max fee is equal to your max priority fee so that means that any difference between the base fee and your gas price basically will go straight to the miner or the validator um so it just means you kind of don't get that refund for extra fees paid um you can still you know if you're if

you're savvy kind of estimate exactly what's like the right gas price and and how to get that into a block and that's basically you're back to the problem we have today um but this is super important because we you know even though it might be slightly more expensive for users we never want to be in a spot where like somebody is using a wallet that doesn't support 1559 and now they can't interact with it with ethereum um or even uh another edge case is like you signed a transaction a year ago before 1559 was even like considered for mainnet and you had like this you know note to yourself that you were gonna send this transaction on the network in two years and that transaction should still be valid right uh so yeah uh legacy transactions will still will

still work fantastic anyone want to add on to that before i move on to the next question hearing none i will move on to the next question uh this comes from mr green and i'm going to actually amend this question just a little bit um mr green asks what happens with your gas during a cancelled or stuck transaction after the eip 1559 goes live and and i'll add on to this like is what how does the nature of a like pending transaction that gets like quote unquote stuck how does that change once the ip1559 goes live um uh i'll just throw that one to the to the open anyone want to take that uh so so if you submit a transaction after one five five

nine and um you got very unlucky because you happen to submit it right when there is a massive increase in demand so today you get stuck if there's any increase in demand when you submit it with 1559 you you still can get stuck but only if there's a huge increase in demand and we're talking like 2x increase in demand over a very short period of time and it's sustained so that's another key is that's this that demanding increase has to stay increased like continue increasing um for a century in order for you to get stuck if you do get stuck then you are back in the situation you're in today so we have very similar mechanics where you have to um submit a cancellation or you'll have to submit a replacement um and so so that problem still does technically exist i want to

make sure you know all the listeners are aware that that problem doesn't 100 go away the critical part is getting into that situation should now be very rare and so you again just reiterate same problem same solution um cancel or bump bump the price that your uh your gas price effectively but we really hope that if if we tune things correctly and that's what barnaby was talking about early on the call that after this launches we'll be watching so we can tune things and we want to tune it so as few people get stuck in that situation as possible and so what maybe you know right after launch in a few days we won't have gotten the tuning quite right and they'll still be you know significant portion of people getting

stuck um that sucks and didn't achieve one of our goals and so then in some future hard fork ethereum we'll just dial some parameters and do some research and tweak things and try to get better and so eventually we can hopefully get to a point where no one gets in that situation actually uh micah when you're talking about um you know tweaking things and uh changing things that that leads me to another question that was asked on on youtube and this i was baker i believe is the username handle um i hear bitcoiner is using this argument against ethereum that the ethereum foundation constantly changes ethereum is there any merit to what they're saying i think this is a reflection on the critique that hey core developers shouldn't change the anything uh economic anything to the

issuance policy at all and that is one of the virtues or merits that bitcoin has and theorem shouldn't make this trade-off any thoughts on that question is there any merit to this argument i actually want to start with the hassu who's kind of our you know resident bitcoiner on the panel sure i mean i would say there's some merit to it i mean i don't think i don't think there's any more to saying that the ethereum foundation chooses what upgrades get into ethereum i think that's not really how the erp process works and i'm not even sure that the atm foundation pays most client developers maybe that's something that the actual client developers here can shed some light on

yeah the short answer is they don't uh there's one team on each one and zero teams on each two that are part of the ef uh yeah thank you yeah thank you that's that's what i thought um so i would definitely say yeah atm's changing more and it's easier to get stuff implemented in ethio but the ip159 is actually a great example um it's it is a very big change like i wrote once wrote that it's the biggest change ever made to a an already deployed blockchain um like proof of stake will also be a pretty big change but i think erp-1559 is is to date the biggest and this is like a really cool thing to look at to see how much like how long the process took um and how many people

were engaged in the process and how much scrutiny it received and and it took a couple years to get a eip159 implemented and there were a lot of skeptics i would say for a long time um and it was just a process of you know eventual uh consensus in the community where i would say like those of us who wanted the erp just produce more and more uh educational material and analyses and and so on eventually like swaying some of the skeptics you know over to our side and at this point i would say it's definitely a consensus decision so it's not like there's somebody you know making a top-down decision that we want

to have this eip right right it's it's like it has so many supporters now and i i can't think of any uh skeptics anymore at this point really um so i would say yes so it's it's bad when things i mean there are there's value to things being hard to change but if it changes in the right way like if the right checks and balances exists exist then i think there's also a lot of good that can come from you know adding new features mike i think you've written posts on things like this what what's your take on the same question yeah so i think that there is definitely merit to this the this question is this the is it better to change or is it

better to not change when you're building a large financial system as people are familiar with the legacy financial system um it's well known that banks do not change fast like from a programmer's point of view it's always like a running joke that banks are running on code that was written in the 1980s and it's now hard for them to hire people to change that because everybody's dead who knows how to write code in those languages and there's a reason that banks are hesitant to change and it's because when you're dealing with large amounts of money especially other people's money you ideally don't want to break things like you don't want to you know make a make a mistake and then have people lose you know all their money or all your money or all of some investors money it's a very scary thing um and so from the standpoint of the the

argument that well bitcoin is stable um that's very reasonable like having that stability is very valuable people like stability people want especially in their financial systems they want comfort in knowing that things will be stable whereas ethereum is definitely moving fast we're changing things every six months we have a hard four that some of these hard forks are very big one five five nine is very big and that's a very different um paradigm and i think the the reason i like ethereum personally is because i feel like the space is so new and so young that it's too soon to calcify and it's too soon to stop changing like there's so much we can do and there's so many things that are right around the corner like the technology is just so much in

its infancy it's like having a kid and then saying okay you're two years old now is a good time to stop growing it's like there's so much opportunity whereas you know once they get to you know 20 or 25 maybe that's a good time to stop growing they've matured enough they can be useful they can be productive and they have learned enough and they have advanced enough that that's a useful tool and and so i think ethereum is not there yet i think eventually i will eventually lobby that ethereum should calcify like like bitcoin has i just don't think that's now i think that sometime after proof of stake sometime after sharding some time after you know zk roll ups on mainnet like like there's a number of things that i think ethereum should do before we decide that okay now's the time let's slow down let's solidify

let's make this a platform that people can trust is not going to change i just have a quick follow-up on that but before we do we want to thank tim baker who's got to go to attend another call and uh maybe help get eip 1559 out the door tim you're a fantastic moderator maybe future co-host of the banquets podcast who knows unlimited potential here sir who's going for everything you've done i don't know yet david remains to be seen thank you sir we appreciate your moderation yeah thank you so much for having me and for setting this up um yeah and to obviously all the other panelists i appreciate you all taking the time to come here awesome guys we still have a few more questions coming in if the panel can can hang on i just have a quick follow-up to

micah though uh you raised a really interesting point um do you acknowledge though that that like the the time for ethereum to calcify is somewhat subjective right it's like how how do we determine when the right time to calcify is like it's obviously not now immediately but isn't there always going to be new tech new features on the horizon to implement a hard question especially since there's so many participants i think that at the least like when ethereum's yellow paper was originally produced it said we're going to do switch to proof of stake so at the very least we should wait to calcify until after proof of stake because that was like a thing that was promised and everybody kind of presumably agreed on

that if you're going to use ethereum you acknowledge that this will eventually adopt proof of stake so so at the very least i don't think that it's super reasonable to argue for calcification prior to that after that things start to become a little more questionable because we have things like state experience and sharding and you know zk stuff like magic and those are things that you know weren't necessarily agreed on when ethereum was launched you know six years ago or whatever it was and so that's where things start getting a lot more murky where it's like should we um should we continue to build and iterate or should we calcify and maybe one of the options is ethereum just forks and we say okay proof of stake launched everything we promised in the yellow paper is done we're gonna like calcify that and that will be a supportive thing and we'll get

security fixes separately we're going to continue building off of that in this other direction and build great things i'm personally a big fan of forking as a form of governance i think it allows people to have you know different belief systems and different values and kind of live and self-segregate and operate in sandboxes that align more with their beliefs and so but i know a lot of people don't like that a lot of people would rather ethereum stay is one big thing when there's one ethereum and so i do think that we will eventually see you know big debates on this exact topic and i don't think there's a clean answer necessarily any other panelists want to add on to that before i ask the next question maybe to add to this i think the classification is is kind of a default

position in the minds of many people who maybe come from the bitcoin ecosystem like this idea that at some point the protocol just cannot uh keep changing um it's true that over time the more things are built on it the more difficult it is to make these changes but rather than argue for finding the correct point at which the protocol should classify i would argue for finding the process that makes us decide whether we should do it or not so finding what mika was describing with the checks and balances um like in my position i'm definitely not an immutabilist i joined the ethereum ecosystem because i like this idea of its ever-growing

ecosystem it's like this organism that just adapts to whatever is thrown at it i cannot tell what will happen in 10 years i mean quantum computers zk magic so i would definitely be more in favor of rather than thinking okay is it the time to classify just asking the question how do we make the choice if we if we have it in the first place absolutely oh go ahead i would pose um a different approach to the question i i don't think it's really necessary for us to choose when it's time to calcify i think this is also something that just happens automatically because every upgrade has a cost right i mean first you have to create social consensus which is

a difficult very difficult and a long process and then you have to like all other shareholders in the system have to upgrade their systems and so on and that's just so much uh involved and this is like this is very costly like a lot of companies and there are prefab projects and a lot of people have to spend a lot of developer hours um you know on making this up every upgrade in a human reality right so a human asks a lot basically from their users and the more users there are and the more businesses built on ethereum as you know part of the infrastructure stack and so on um the more people will be affected by every upgrade and the basically every upgrade needs to do more

to to pass their bar for still supporting it um and i think it will get it will just get harder and harder over time to make enough people want to upgrade and so it just classifies basically on its own matt you wanna you wanted to add something yeah i was just gonna add one other thing um in terms like the calcification one thing to me that like really makes a blockchain a blockchain something that's like goes beyond just a database is having ordered available data and where we're going with eth2 with sharding and having massive new massive quantities of data availability that is to me something that even if you

don't even if you want to do something that's not possible to do specifically on the ethereum chain if ethereum provides that platform for making data ordered and available you can begin to build different types of primitives using client-side verification rather than having things go through the evm and so i think that that's going to open up an avenue for people to still use ethereum but really be able to experiment like really crazy things that may not be possible on the base layer it's a great addition to that we just have a couple more questions to get through and then we will have gotten through all of them and hazu i want to point this one to you because it's a a little bit in the world of game theory and i think you might be best equipped but uh for the other panelists feel free

to also hop in on this this question comes from yuan uh miners uh might artificially reduce the block size after one five five nine up the update uh so that users are forced to be in a tip bidding war like we already are now is is this something that we can be concerned again about is there a topic of conversation to be had about minor block size manipulation um to maximally extract value for themselves have you thought about this yeah i mean this is uh probably the most obvious thing that you have to think about like with with regards to eip15f9 and how it can be broken by its participants right um so if we if we argue that miners don't want any fees to be burned then they could

make this a reality they they could just say okay we never mind blocks that are higher than the target because the blocks higher than the target on a mine then this is what makes the base we go up right and if blocks lower than the target i might then the base fee goes down so they mine blocks lower than the target basically goes to zero and then they keep the block size below the target and that's why this way you can basically keep the keep the base view at zero forever but if only some miners do this then um they will always have to so they basically have to leave transactions in the mempool of users that also pay fees um and i mean then any other miner can come

along and include all of these transactions in their block and they can include a lot of transactions because there's all the slack like all this extra room um in in in the block size in in eip1559 and they can scoop up all these fees and you know make more profit so i mean this shows that you can do this but you you need all the miners to cooperate on this so either everyone follows along or so that the miners who want to do this have to have to pressure the the non-compliant miners into compliance like for example by ignoring their blocks like when you have like 70 percent alliance of miners that wanted to do this then they could ignore the other 30 percent blocks um but it also comes at a cost right um so

if the any user who basically saw their transaction valid when like the 30 or confirmed when a 30 minus block were um was mined and then later becomes reorganized i think that's very bad user experience i think we've talked about this a lot in the last couple weeks because of you know all these discussions about short-term reoxene ethereum so basically that's the answer so miners can do it but they would need to be willing to make um short-term reorgs and basically enforce a minor activated software and once we once we are like in the realm of saying hey a minor activated software is possible then we also have to think uh okay so isn't this possible today and the answer is yes this is

totally possible today um if miners were to collude and like enforce a certain high a fee like a mandatory fee that they charge users for transacting on ethereum then they could extract a lot more money from users than users would be willing to pay in a fair market right because right now users pay the minimum that is necessary to get into a block but when when miners were to set the prices then users would pay the maximum they are willing to pay like this would be the correct sort of pricing strategy for miners um miners could multiply the revenue like if they were willing to do that but they don't they are not right because i mean they are in competition with each

other and um if if this property of a blockchain is broken and i think users would lose a lot of trust in in ethereum and um far fewer people would transact than the value of effort with dwindler and ultimately i think it's not in my miners best interest guys we're gonna close it out with this last question and i think this is uh for each of the panelists uh this is a question of timelines timelines in crypto in ethereum are uh interesting things now i remember when eip 1559 was first proposed you know some people were like oh it'll be here in a few months right others will like this will never be deployed to may net right like never it's not happening not five years not

ten years it's never going to happen um the truth it seems like turned out to be someone in the middle i want to ask this question of on timelines did the timeline surprise you um maybe first i'll turn to hasu did the timeline surprise you hasu i would say yes um so everybody in the bitcoin community so like well ethereum has all these future plans for changing the protocol but they never ship them it's just and they were right with regards to proof of stake and with regards to to sharding both of which were part of the here in white paper so like they had a lot of ammunition in that sense but

first ethereum shipped the beacon chain and now they are shipping erp 1559 and both of these have been important milestones and have contribute contributed a lot to convincing me that the you know the development and governance process is actually much more functional than i thought and um ethereum can deliver what it promised at least like how things are today so maybe in the past it was like more ineffective but but now i think we like it it's building a track record basically and this upgrade has helped a lot with that like client how about you on timeline i think whenever 1559 was originally proposed it was i was kind of a little bit newer as

the ecosystem and i was um very naive i thought that these things get proposed and then we upgrade the masternode and it's like very quick type of process but it you know i've like i think i've become like quite jaded in the last few years especially with some of our own proposals i've been working on this 30-74 proposal and i'm just like seeing that these timelines they are really like multi-year timelines and that's kind of i've had to shift the way i think and work to try and treat things as you know small packets of work over periods of years and you have to like kind of have a few things that you're trying to maintain so like general high level

goals the community is on board with and 1559 was one of those things and i wasn't really involved until kind of the very end of the process but i can kind of see how like different people have come in to play different roles whenever their roles were needed and i came in to play my role whenever it became more of an engineering problem and so i think that for a lot of changes in ethereum that's like a better mindset to think about it like it's a multi-year process and with the merge coming i think we're not going to see very many new features in ethereum until the merge happens i think everybody has sort of decided that we're going to do the merge this is the number one priority and nothing you know we're not going to focus on anything else until it happens and so that's something

that you know means that we're probably not going to see new we're like we're not going to see new things happen in ethereum for the next 12 to 18 months other than merge related things micah what's your take on timeline so i'm similar to uh to matt i was not young and naive back in 2018 and i thought surely this thing is well after i realized that metallica was smarter than i gave him credit for um then then i thought okay this is a great idea sure this will go through quickly um turns out that was not true but i think there's a good reason for this um when you're you know driving down the freeway with a bus full of children and you're trying to change the wheels on it you

want to take that process very carefully like you you don't want to just you know just you know tip over on the two side wheels swap the wheel out real quick and keep going you know just hope for the best this is not a yolo sort of situation and so four one five five nine there was a lot of you know a lot of people including myself and sounds like most people on this call were skeptical of it and so the process requires us to you know convince all those skeptics that this is actually a good idea and then once we convince the skeptics then we need to you know actually implement it and then make sure there's no bugs and you know there was there was a bug on roxton as reason in the 1559 implementation as recently as what was it two weeks ago or

something like that and so this isn't like a simple thing like with this is this is after massive amounts of testing and you know a month or so on test nets and then we found still found a bug so this is this is very hard to do and so i think that taking it slow is is the right right approach at this phase of ethereum's life cycle you know when ethereum was brand new and had 10 users sure we probably could have knocked out one five five nine in a week you know it's it's like it's not that hard just throw it out there see what it breaks and if it breaks you know restart everything um it's even worse for things like proof of stake because proof of stake had to you know build an entire do a bunch of research to actually you know

design a system that previously wasn't designed like there are chains that have proof of stake but they make trade-offs that ethereum's unwilling to make and so ethereum had to actually go through the whole research process and figure out how do we solve this problem that no one else in the world has ever solved before without making trade-offs we're unwilling to make and so they finally did that and then you have to build the whole thing and then you have to test the whole thing and then you have to convince everybody to you know actually upgrade and accept that whole thing and so um so yeah so i think it it was slow but i think there are good reasons for that and i don't think it means we've reached that point of stagnation we talked about earlier micah is that you use the metaphor changing a school bus wheel with a bunch

of children on it is that what it feels like to be an ethereum maintainer like just driving a school bus with a bunch of kids is that is that how you think are we are we the kids yeah i mean if if you see children as just dollar signs then yes it's basically the same thing is that how you see kids deeper question we'll say that for a different panel barbie i want to make sure that the five nine panel not the 15 that's what i signed up for uh barnabas uh take us home what do you think about timeline yeah i mean i don't have so much to add but um eav1559 was kind of my first project i

joined the ef in 2020 before that i was in a more academic background let's say things in academia move quite slow and so when i joined ethereum ecosystem and kind of started contributing i actually thought wow this is moving really fast like it's a huge system so many moving parts and people are willing to go through these big changes but it really it gave me like such an appreciation of the work that goes on behind the scenes especially like a very thoughtful kind of staging process through the test nets all the work that goes into making the clients work coordinating uh an ecosystem which just spreads in so many like different directions especially one five five nine i mean it touches the

economic engine it touches the client it touches wallets the way users interact with the network so it's a huge amount of work and i i would say yeah i i can't really say if it was slow of us because i don't have a good point of comparison but to me it feels like it's actually quite fast well guys it is here we could say that for sure this week it is happening eip1559 currently scheduled for thursday maybe around 5 00 a.m or so pacific time 8 a.m eastern time join us for that live stream where we will be broadcasting uh hopefully a celebration when the hard fork goes live panelists we want to thank you so much for availing your time to us this has been fantastic and i actually want to thank you on behalf of

the ethereum community for all of the work that you guys put into this we have been excited about eip 1559 for many reasons for a long time and you guys are part of the core team along with many others who are not on this panel but you guys have been part of the core team to make this happen we just want to thank you for all of your work on it thanks a lot guys thanks for watching the word thank you all right that's what we do we spread the word i gotta end with this of course written disclaimers none of this was financial advice is risky crypto is risky hard forks are risky fingers crossed you could lose what you put in but we are headed west this is the frontier it's not for everyone but we're glad

you're with us on the bankless journey thanks a lot and if you are watching the youtube one of the 426 people watching the youtube make sure to like and subscribe for the two eip live streams that we are having on thursday do not miss those thanks everyone hey we hope you enjoyed the video if you did head over to bankless hq right now to develop your crypto investing skills and learn how to free yourself from banks and gain your financial independence we recommend joining our daily newsletter podcast and community as a bankless premium subscriber to get the most out of your bankless experience you'll get access to our market analysis our alpha leaks and exclusive content and even the bankless token for airdrops

raffles and unlocks if you're interested in crypto the bankless community is where you want to be click the link in the description to become a bankless premium subscriber today also don't forget to subscribe to the channel for in-depth interviews with industry leaders ask me anythings and weekly roll ups where we summarize the week in crypto and other fantastic content thanks everyone for watching and being on the journey as we build out the bankless nation