The Evolution of Ethereum | Tim Beiko

Transcript

if it just happens that somebody like you know sold their Punk for like one Heath instead of 100 East um in this block should like that 99 East Arbitrage go to like you know you as a validator just because you happen to be in that slot or should it be distributed like more Broad what's up everyone welcome back to another episode of zero X research today is February 27th and today we have a great interview lined up with Tim Bako a a long time contributor to the ethereum protocol and he leads the ACD calls for ethereum and of course we discuss all things ethereum but before we get into the interview we're joined by blockworks research analysts zero X Pebbles and ren for a little segment of Hot Seat cool

Throne to discuss the latest Market happenings I'll kick it over to you pibbles for for your hot seat or cool Throne yeah so I've got a cool thrown for fpis which is frax's CPI picked stablecoin and governance token is fpis they just launched staking which is vfpis on February 20th and they've already locked down 35 of the circulating supply for an average lock time of two years right now the apy is like 200 percent pay down fpis tokens so um we're just seeing like a pretty large token Sync here and you have fpis sitting at literally a 40 million dollar market cap

versus Frac shares at 800 million dollars so uh this is a super interesting one to watch because you've got FBI sitting at like a 87 million dollar market cap and that's the CPI pegged stablecoin and they've been farming profits I think they have like five million in profits from the past year so they're about to launch liquidity incentives governance and all that good stuff so probably a flywheel about to happen yeah it's definitely a good flag what is the actual mechanism that keeps uh fpi pegged to inflation because I know there's that chain link Oracle that it uses that takes on uh like U.S inflation but how does the the actual

stable coin stay pegged to the Oracle price so it does it gradually like basically every 30 days they add a new CPI reading and then the Amos will like gradually adjust the price and they have a surplus of collateral to take care of that so um it's been pretty much up only since it started last May but um you know as inflation starts to shrink some then you're going to see a lot more Surplus in the collateral which is just more profits okay so does it take the Amo profits in order to increase that peg in line with inflation so then more profits will be able to go to fpis stakers as the inflation rate comes down do I have that right yeah

that's that's how I understand it right now so it's really important that they keep at least a hundred percent collateral in case inflation starts going up because then they would just have to Mint a bunch of the governance token and dump that so that wouldn't be super cool is there sort of like a long-term average inflation that feis is able to support say like if inflation is two percent annualized the fed's magic number what fbis be able to support that over time in like into perpetuity or is that number like this based on like facts profits so I guess it just depends on what on-chain yields are right now like they have like 80 million of protocol and liquidity and

they just needed that into convex so they're getting convex curve and fractures so hopefully um on-chain yields can stay good enough for them or else it's gonna be a bad day for fpis holders right yeah because I guess like fpis is ultimately the backstop to that Peg so worst case scenario you end up minting and dumping fpis uh but yeah that's that's like a seems like the more fragile edge of some of the frax products but this is this is definitely another interesting one yeah I think that's why they launched a separate governance token for it so it wouldn't mess with fractures at all because this is pretty experimental stuff yeah Sam who you got in the uh Hot Seat cool Throne this week all right

mine's like the most easy one of the week base coinbase is L2 on the op stack uh that will in theory be interoperable with other op chains over time with with the same security guarantees as ethereum once fault proves go live um people think they're gonna be monetizing this by extracting Mev or taking the the Delta between um L2 fees that are generated and L1 gas consumption I I don't think that's really correct because they said in their blog post that they won't be extracting Mev until the sequencer uh job function is actually properly decentralized and then if you look at any other L2 like optimism or arbitrum today the fees that they actually get that's the Delta between uh the the L2 gas fees paid and the L1 settlement cost

is is pretty marginal so I think really the way that they're gonna monetize this and to kind of like confirm this thesis of mine I hopped on the coinbase app actually for the first time in a while and they have this little web 3 uh I guess uh toggle at the bottom right and you can actually interact with like Lido you can use Ave you can use one inch you can use Quick swap like all these different daps that we use like pretty often on chain you can use the coinbases apps user interface to interact with these apps and I'm sure they're taking a pretty big fee from the users interacting with that I haven't actually used it myself so can't confirm that but I could see a world where they're kind of like the UI to onboard people on chain without even knowing it and giving

people exposure to all these different apps that are on bases ecosystem so I think that's actually how they'll how they'll monetize it um over the long term so I think the main takeaway here is it's bullish on-chain adoption it's bullish op it's bullish coinbase uh to fight the regulatory battle from a centralized sequencer perspective it's bullish coin stock and more than anything it's it's hyper bullish eth so super exciting announcement something we're all watching super closely and I know everyone's been reading everything about it on Twitter so I guess I'm curious uh first if you guys have any thoughts on this that you want to share but then also it kind of feel like another narrative forming is that the Cosmos app thesis is kind of graduated onto ethereum L2 so I'm curious if you guys have any thoughts there

yeah no surprise you'd bring that last point up um but I think it's really exciting and I feel kind of dumb for not seeing this coming a long time ago right like Finance has been experimenting with l1s for a while now you know that buying smart chain BNB chain and we've kind of seen like the evolution of that and we've also seen like Finance be able to profit from this and really create like a vibrant ecosystem that then like you know they're Venture investing in uh and it kind of spins its own flywheel and it's pretty cool to see coinbase doing this on ethereum though because I think you know someone tweeted today like bass stands for uh building and scaling ethereum I'm like okay that's pretty sick uh and it just kind of gives Credence to you know today ethereum is

the tvl Hub that's where the attention is it's where the usage is it's where the innovation has been happening but tomorrow you know you never really know what that will look like um but for like every day kind of ticks through we see that the rope-centric road map that ethereum is building uh continues to really be the most likely outcome um you know I don't know if the cosmos is dead yet I think that's it's probably a bit of a reach this early and we'll have to see you know like what actually gets built on on base and how users Drive adoption to that platform but definitely bullish the op stack that's a huge like win for the Credence of what they're doing um you know we see synapse building over there as well so there could easily be some synapse Synergy between those two

protocols and really just kind of excited to see like this module that already continued to get built out but I agree like this is huge for coinbase it's probably huge for onboarding users into crypto as a whole I'll take some uh some concerns about bass with how optimism took off really because they were just shoveling token incentives of people to build over there I'm curious how coinbase is going to actually you know bring people over there to build apps and then also how are you going to get these people like your average Joe to go on there like look now I can buy you know the The Fringe convex stuff that like they have no interest in buying in the first place so it could

end up just like coinbase nft potentially hope not but it could and then on the uh the modular Roll-Ups thesis you've also got unidex who is probably going to announce this week that they're launching an optimistic roll up and then there's another one called bite Masons their token is oath they're also planning to launch a roll up and this is pretty bullish for like any sort of evm bridge including synapse Stargate Etc I think the counter argument there is base is saying Hey We're Not Gonna We're not gonna have a token so we're not going to throw incentives at you what we're going to be able to throw at you is users you know we're the most used platform we're going

to start channeling energy into our coinbase wallet and um that's how we're going to onboard people and if you want to build an app that our users can use put it on base so I think that's how they're going to kind of approach that situation but you're right like not having a war chest of token incentives will definitely be something that they've talked about and have to have a game plan for how they're going to overcome that like Beyond like migrating the liquidity over and also another thing I think about is that they days I feel like increases the vector for permission as Innovation right if you are a D5 protocol that's historically being very scared of trying to do something with like either like U.S entities or U.S users right like doing that on base would seem like this sort of more regulated way to do so and

also another thing that base can provide is sort of like kyc aspect right obviously the chain itself is decentralized but I wouldn't be very surprised if coinbase launches like decentralized identity nfts to allow D5 protocols to sort of run regulated investment platforms where you can only provide access to say accredited investors or qualified purchases right like if you are going to launch a protocol like that on like the entire ecosystem of blockchain networks chances are you're going to want to launch it on base just either a get that institutional branding or we actually get access to that U.S user base right and I think another thing from the monetization perspective

that Sam mentioned that going into coinbase advice you can sort of abstract away all of like the decks or like the D5 protocol front ends people are very lazy right like there's a reason when metamaskovs first came out my first reaction was there is no way people are paying 0.875 to consensus just to do a solve through the metamask UI but today metamaskovs has done 25 billion in volume and consensus has just gotten 200 million in free money just from people being super lazy which like always blows my mind right I can't wait to see like the next Blue market and how much consensus makes they're probably gonna make like one billion dollars a year just from doing nothing just with a better myself so yeah I'm quite bullish

on the bass and actually on this note um tree of alpha just put out a news article or a newsletter that says Robin Hood gets SEC subpoena over crypto listings and I think all of the regulatory actions in the past say three months have been pretty bullish coinbase whether that's the crack in staking the USD or this Robin has Sofina right like coinbase seems to be the only one with enough Firepower to fight the SEC and probably the only ones with like a good enough like Financial standing or regulatory like standing to fight the SEC and everyone else is just gonna get pummeled into the ground and I think a lot of people on a team agree that like base is the regulatory hedge too right like if everything goes to the ground

like Gary against her wants to burn like coinbase down they still have base in network and that can be for example like the App Store of like D5 protocols for institutions or something like that yeah strong agree with everything you said there and it just blows my mind every single week how quickly narratives flop in crypto like if you would have asked me or anyone else in crypto two three weeks ago before this announcement if you're more bullish on arbit charm or optimism I think 9 out of ten people would have said arbitrum but you ask people today I think it's completely the other way around uh so pretty wild and kind of sobering to always think about that but one other thing I'll leave the the listeners with is there's a commemorative nft that you can mint for free uh which I feel like it's an

absolute Hail Mary play there's already almost been 400 000 Minted as of Monday February 27th but I feel like it's a good idea to to Mint that just in case in the future it leads to some lucrative airdrop or something of some sort but uh rent on that note I know you got a A hot seat for this week so I'll kick it over to you yeah perfect timing on a note of the narrative shifting from to optimism overtime had a slight PR disaster so to say in the past over the weekend so as a lot of you know overtime runs a centralized sequencer right it's not decentralized at all it's just off-chain labs they have a private mempool and transactions aren't made public until they are mined in a block which means that if you're a me researcher looking for opportunities you

really want to connect to the sequencer to know what transactions is processing so that you can put in your next transaction to capture any Arbitrage opportunity right and so what the centralized sequencer relay does is that it sends out these transactions to a random connection that's connected to it and so basically what's happened is that you get searches which spam the relay with like hundreds of thousands of connection requests just so they have a higher chance of being the one that receives the transaction that went through the sequencer right um and so basically what happens is option Maps when okay this isn't good for us we're paying like a lot of fees and trying to

maintain these like connection requests to the relay and one of the techniques for architecture performs the idea for these Searchers to mine a nonce and then from there depending on the nonstad you mine your score must be greater than or equal to a Target specified or else the connection will be refused so that's sort of like a gating mechanism for connecting to the video and for a lot of the people basically this doesn't make sense right you you don't want to give anyone a latency advantage over anyone else but while moving the work off the arbitrim server you've created some like other really random incentive mechanism for other people to basically put a lot of effort into mining nonsense and

there's probably like a lot of better ways to do this I think yeah I mean when we get to into like you know how do we make l2s more efficient how do we make them more decentralized this is definitely like the area of Cutting Edge research in my opinion and I wish we had Westie on this week to bless us with some some knowledge around uh what the current landscape of that research looks like but you know to me like these are just going to be problems that continue to arise right like if we look at l2s today they are not fully baked out they are not fully decentralized most of like if you look at uh you know optimism have fraud proofs live yet today so like these are we have like baby l2s if you will and getting to the states where they're highly optimized and I kind of

have their full potential is kind of like ethereum right like we are still building on unfinished protocols and there's going to be these like uh hiccups or roadblocks that come up along the way where existing Protocols are going to have to like completely reshape the way that they've built because we've discovered a more efficient way to do something and we're seeing this we frequently see this on the base layer um and so it's not really surprising to see this on l2s but uh you know it's just like we see the light at the end of the tunnel with this Grand Vision of these like l2s that are highly efficient and scalable but like getting exactly to that end goal is still a bit of a road unknown yeah I think the entire ethereum road map is going to change before they figure out how to decentralize

sequencers and then somehow ethereum just processes all the transactions they just the most random shit's gonna happen and then LTS are going to be forgotten about Westies it's honestly a fair take though and it's I'd be curious to see like in the future how much you know because everyone's kind of going for this evm equivalence and who knows how much of this technical debt that ethereum is creating that l2s will also have to carry into the future so I think there's a real argument to be made for like modular Roll-Ups such as the fuel VM or um you know there's a lot of them out there so that's definitely something to pay attention to going forward Sam I don't want to put you on the spot here

but uh you know I know you pay really close attention to like the the progress of ZK EVMS and uh l2s as a whole and so does arbitrim have any of the ZK Tech on its roadmap or is it like strictly trying to go a different route here yeah from what I've mostly recently read is they're like super open-minded to any kind of tech but right now they're still focused completely on optimistic roll-up designs and I do know that their next uh upgrade I can't remember what it's called but it's focusing on basically bringing like C plus plus like rust other development languages more from the web 2 space into the arbitrum ecosystem to onboard more developers so it's a completely different approach than optimism's approach and in trying

to be super modular and encouraging people to Fork their code so it's going to be interesting to see how that pans out but at the same time like optimism is more I feel like application focused and generalized and then arbitrum is really leaning into that D5 heavy ethos so I feel like they are going after two different markets but I just don't know whose strategy is going to win long term but I know optimism has some stuff in their docs for super chains saying that they're willing to use uh validity proofs or ZK attack for across uh chain communication between different op stack chains so I imagine arbitrum sometime down the line will also adopt similar Tech Dan I know you've got a killer cool Throne or hot seat actually sorry I

naturally want to put you in the cool throne for the work you put into this but uh hand it over to you right now to explain what you uncovered on Friday last week yeah so hot seat the Wormhole exploiter if you guys remember from about a year ago almost to the day the Wormhole the bridging protocol was hacked and they had 120 000 eth stolen from them uh so at the time that was about 300 million dollars and then if you flash forward to fast forward to the 21st of February this year um that's money was actually counter exploited by jump and so it's pretty interesting to see that how this development went down so um The Wormhole exploiter had this 120

000 eth that he was kind of Shifting around ethereum and you know plugging into various applications doing different things but most recently he minted a bunch of rap staked eth and then opened up two Oasis vaults which is a borrowing platform built on top of maker Dow and on with these vaults he was like essentially going lever long uh uh each staking derivatives so re through rocket pool and rhapsaked through Lido um and with these two vaults he opted into Automation Services that were offered by Oasis so essentially a user of the Automation Services can set certain buy or sell triggers where when the collateral hits a certain price the

protocol will automatically like buy more by taking out a loan against your collateral and kind of like continuing your looping So based on certain price action a user can kind of create a more customized Vault and it really tailored to their own needs and desires of how they want to go lever along a certain NASA so that's exactly what this Wormhole X Twitter was doing taking a massive lever-long position on rap stick teeth and little did the exploiter know the automation contract that is used is actually an upgradable proxy contract and so it kind of makes sense why if I was automating a vault I would need to give control of my Vault to a Smart contract right like if I want to do something like I want a certain trigger

is hit then the automation service needs to act on my behalf and I'm like giving it permission to do that so the automation service is inherently sign away uh privileged to act on my behalf on My Vault and that would have been fine if it was an immutable contract but because it was upgradable the Oasis multi-sig it's like a 4 of 12 multi-sig with Oasis team members has the ability to change the code of the automation service contract and they kind of did exactly this so uh a trigger was hit on February 16th that kind of like alerted it was a massive a massive buy it was like it was like a 12 or 12 and a half or so million dollar wraps State youth purchase that like was shattered through

D5 I had like 100 something logs on the transaction and so white hats picked this up and if I actually like searched through Twitter to uh after I was like found this whole analysis and uh this transaction actually got flagged a ton of times but nobody figured out that it was like the very first step in uh like uncovering this massive counter exploit and so with these white hats basically saw this massive transaction and they went oh this guy is using an automated upgradable contract like there is a like we can counter exploit them and so sometime between this the 16th and the 21st of February Oasis was became aware of the fact that they could essentially deploy this counter attack

um and we don't know if like Oasis decide like was approached with this and like hey like we're not going to do that and that's what triggered this court order that eventually came or maybe the Oasis kind of took the approach of well if you're going to do this like we're going to need a court order to kind of like cover ourselves I have no idea about the the internal conversations that went on there all we do know is five days after they originally found the possibility of this exploit the high courts of England and Wales came to Oasis who is headquartered in uh in the UK and said hey like you have to you know take all the necessary action to enable jump to retrieve These funds and so on the 21st there was an additional signer added to the Oasis multi-sig we

still have not confirmed the uh the true owner of this but in you know in my personal view it is of all indication that this was indeed jump and so they were added to them Oasis multi-sig executed a series of transactions that retrieved the funds and then were essentially kicked off the multi-sig and so all in all this additional signer was only live on the Oasis multi-sig for about an hour and 53 minutes but in that time they were able to execute a transaction that swapped out the automated Services contract code and enabled them to essentially uh open a new Vault that is controlled by the multi-sig transfer the debt and collateral position from the Wormhole

exploiters Vault to this new Vault and then they had full control of this collateral in debt and so jump then sent transactions to this additional address they paid down the debt to the tune of about 80 million dollars and then withdrew the about 200 or so million dollars of collateral so net net jump had to pay 80 million dollars to retrieve about 220 so they ended up about 140 million dollars better off um and this was like a huge fine so I I don't know I actually stumbled into this and had no idea this was the case I was like doing some analysis on wraps trans wrapped staked transfers and noticed that like the largest transfer of all time was like you know 12 or 36 hours

before I was actually searching for it I was like oh that's weird like who's just slinging around 200 million dollars of eth in one transaction uh but it was actually the transfer between uh the like holder contract and a sender contract and so uh if you we published a free report on blockers research that like goes line by line on what happened through these transactions uh it's pretty hard to like like verbally walk through but uh so sorry if I just kind of like mumbled through that but if you read the the free report which we can link in the show notes again it just goes line by line of exactly what happened um and it really doesn't get into any of the second order implications which is kind of the more interesting part right because at the end of the day

I've been seeing like a ton of like oh d5's dead on my timeline and in my view like that is just really not what happened here like there was a very specific set of events that needed to occur for this counter exploit to take place so you know first of all there was a multi-sig controlled by a known set of individuals through the Oasis team and they were based in a regulatory friendly jurisdiction in the UK and further to that they controlled an upgradable contract that the Wormhole exploiter opted into using and gave that contract control of its assets so there's like five things that needed to happen for this uh this eververse exploit to be possible and most importantly none of

this was possible without the permission of the Oasis multi-sig so they essentially had to facilitate this from happening and were asked to do so or ordered to do so by a court so I think this is the first time a court has ever ordered specific on-chain transactions to occur and then they have occurred but it's like this friendly reminder that you know just because if you're if you're a private company building on blockchain Rails you're not just exempt to the like the laws of the world or the laws of the area you live in um even if you are like building a protocol but again and more importantly here like Oasis is essentially a private company that offers on-chain Services it is not decentralized it does not have a token and it has you know its governance

is the team making decisions so it's more so like it should be viewed as a private startup than anything else um and it's all the other takeaway here is like it's not that upgradable contracts are inherently evil they have a great purpose in D5 and it's not that multi-cigs are inherently evil either it's the combination of a multi-sig controlled by a known set of individuals that have the ability to change the code of this upgradable contract that is really like this centralization risk here and like the way I view decentralization is like it's it's a spec it's a spectrum it's not a zero and a one and it also spans through time like this is the end goal we need to build towards not the place we just begin at when we start um and so you know there are ways that this could have been mitigated but at

the end of the day like this isn't something everyday users of D5 need to be worried about unless they stole like 120 000 or 120 000 eth from one of the largest and most sophisticated trading firms in the space like if that's not you then you probably have nothing to worry about here yeah honestly you killed that that uh analysis Dan I thought that was really good like even like makes sense verbally uh okay but I guess it makes sense too because you were pretty much balls deep in that for like 36 hours straight so that tip to you but I I just Echo everything you say uh I mean the I think it's important to distinguish between maker Dao and Oasis like they're not the same team they're not the same product and they're the maker down Maxis are going to come out and attack if you

don't distinguish between the two so that's the only other thing I would add yeah for sure for sure it was uh it was nice to get some some sleep over the weekend because when I found that I was like oh my God like I think I like Googled or not Googled Twitter search like every combination of like jump and exploit I'm like surely somebody's found this like I was not the first person to stumble into this but one thing led to another and sure enough and you know like there's definitely a lot of viewpoints to have but at the end of the day like this was pretty based like jump got 300 mil still from stolen from them a year later they still haven't given up and they they figured out a way to like reverse exploit it and take it back like that's that's sick yeah I guess if I had

to add anything to uh both Dan and Sam have said obviously a lot of us in crypto really even like the whole decentralized permissionless sort of like system that we're building but I've also always felt that as long as we're able to build a better Financial system that's better cheaper faster even if a lot of that is centralized that's a partial win for what we're trying to do already if we're able to reduce the average fee for 700 billion dollars of remittances every year right from six percent to say one percent or even 0.1 percent which I think is definitely possible on a blockchain network that's a win if we're able to sort of allow people to exchange money or take out a loan at lower rates in developing countries to build their own like

business or start their own Farm the that's a win okay not everything has to be decentralized obviously like I feel like the best scenario is that there should always be a decentralized and permissionless option for the 0.1 percent of you that I really wanted um but I think like dancing right we live in a world we live in a society there are laws that we have to comply with like you can't just go around and stealing money even though like code is law quote unquote yeah I I agree though like there will I think there very much so will be that decentralized value that does exist on chain um but I agree like that doesn't need to be everything and that shouldn't necessarily be the focus of everything but um I'm excited that people are very dedicated towards building that because

it is something that doesn't exist today and it does have product Market fit like people that should be something that people have the ability to opt into based on like your you know your risk preferences and desires so excited to see that you know that's still something that we're building towards but again doesn't need to be everything all right I think that's a good time to call it let's kick it over to the interview with Tim FICO after this ad I cannot recommend enough for you guys to all check out blockworksresearch.com if you go over to the research Tab and toggle free research you're going to get access to some of the best free reports in the industry and if you want to subscribe to blockworks research you can do so using Xerox research 10 at checkout in order to receive 250 off and you can also sign

up to our free newsletter if you want to just get a little taste where we give Alpha on governance dgen trade ideas Market commentary charts of the day Etc kind of get you caught up to speed on everything you need to know in the Market within 5-10 minutes give us a follow at blockworks res blockworks res on Twitter we'll release our new reports during the week and even if you don't have access to the reports you're not a paid subscriber you can still check out the topics we're writing about and get a little bit of a brief insight into what the contents of the report is about you want to know a little bit more how we think on the data side of things head over to our Dune public account we have four dashboards live there for free the revolution will not be quarterly

reported so definitely check those out and let's kick it over the interview all right we are joined here by Tim baiko a prominent uh contributor to the ethereum community uh he's kind of built up a solid reputation by Leading the the ACD calls all core devs calls for the ethereum community um welcome on Tim thanks for joining us thank you so much for having me yeah I think it'd be great just because I personally haven't heard that many interviews with you uh myself so I would love to just kind of get your back story on how you got into crypto and and why why you picked kind of the ethereum community is the the one to stick with yeah of course um so I first got introduced to crypto through uh Bitcoin and it was kind of weird um I had a friend who was always uh into like weird

internet things um and at one point the two things he was into were uh Bitcoin and the other one was like buying Iraqi dinars off eBay because um there was like some sanctions things in the the price like the currency had gone down and his his view was like you know when the US sort of left Iraq um the the price would go back up but she couldn't like buy them on a FX exchange you had to like buy physical bills um so he was into both those things at the time um so I I bought a bit of Bitcoin because it just seemed logistically simpler um but like equally weird um and and sort of got me down the rabbit hole where like I didn't I I

didn't sort of go all in on bitcoin right then I think this is like 2013 or 2014. um but sort of knew about it you know and and roughly kept track of it um and I think you know over the years I sort of saw like when ethereum happened and the thing that sort of got me uh really interested in ethereum is when uh the Dow launched um so the Dao was like the first big kind of killer app on ethereum and the idea was to have a sort of VC fund where the token holders um would like buy these Dell tokens and like vote on which Investments the Dao um would make so basically you would send your Ethan get some Dow tokens and it would like pull your eth and you could you know choose which which projects to invest to as a Dao

um so that sounded really cool I bought something to invest in the Dao and then literally I think two days after um the Dell got hacked um so somebody found a way to like take all of the East out of that smart contract um and so that was really kind of what pulled me in really deep into like uh ethereum where like I was very dramatic moment where people were trying to figure out like okay you know can we stop this hacker and then uh there was uh eventually because of how the Dow worked there was like a time lock with how the funds uh like to withdraw the funds so the hackers funds were sort of sitting there for a month and people decided to like make an upgrade to ethereum which would uh take the hackers funds away and return them back um

and and so that sort of all happened and after that there was this whole ethereum classic split so these people who disagreed with this this this change uh sort of started running the original chain again and then as a user if you had coins um like on chain you would have to like split them send them to two different addresses so like it would be it would be safe um so that's really kind of got me um as someone with like mount a ton of you know blockchain experience Beyond like having bought Bitcoin and bought eth like really down uh and and like interacting with this stuff um and after that after the Dao sort of happened an ethereum classic happened uh live view is like I wasn't quite sure if like ethereum would work in that like if this is a

platforms for smart contracts and like you build a smart contract on it then lose all the money like um it's like is this like a good product in a way um so I was still paying attention but a bit less and I think later that year like about six months after that you started to see other projects uh you know with like this first round of icos come on and they were doing more than just like um basically pulling money so there was uh at the time Gollum wanted to do some like GPU stuff um there was this other one uh melon Port which was kind of a Vex fund trying to build on ethereum I think seeing like that there was a bit more diversity in terms of like what people were trying to build sort of got me back in um and then the next year was like

obviously 2017 there was this huge Ico boom um and and I then like as a user of the chain it was pretty clear that like there was huge demand for the network uh but the infrastructure was terrible like you know the transaction full would get like clogged for like days and stuff when there was this these like popular icos it was just really like a painful experience to use the network um and so that's kind of what got me to want to work on ethereum so I I wasn't super confident in any of the specific projects at the time and so I didn't didn't want to like you know risk it on like any one of those so I was like working on the protocol itself um seems like something that would be valuable like there's clearly a ton of Demands for it even if all those icos

fail I'm sure you know they'll be more like a year from now there'll be different projects using it um and and from there it sort of took me a while to get a full-time job in the space I was a product manager uh but I wanted to work under protocol which is not a product so um yeah took a while to find to find something that was like that overlap um but luckily in 2018 consensus stood up a protocol team and uh hired me as a product manager as part of that so I started working um on their their ethereum implementation the basic client um so I did that for two two and a half years um and then the protocol is sort of managed through these uh these these calls which are open to anybody who's contributing to ethereum and they're called the all all Cortez calls so I

attended those because I was I was working on one of the implementations and they were run by this guy Hudson at the EF and then in 2020 or 2021 uh Hudson had been doing this for like over five years wanted to move on and I sort of stepped in and and took over uh for him there um and that's what that's what I do today still so basically I I run these calls where the different contributors to ethereum come and like talk about you know potential changes to the network um technical implementation details and all that stuff yeah and that that's a great little background there and one like a interesting point where we left off there is kind of how you know ethereum governance works as a whole I feel like to a lot of people even power users of ethereum they still don't fully

understand like what that process looks like of like taking an idea uh and pushing it live to an upgrade that hits the network so can you kind of walk us through what that looks like yeah of course and I think the best way to think about like uh you know the the governance processes it's almost like there's two like uh parties or or groups and they both need to end up agreeing so just the people who write the code right like the the court ads the software developers um but then there's everybody who runs a note on the network um and so when like the whole you know what people think of like the whole like governance process is basically only the first half of it so say you come uh you want to propose a change to ethereum we have templates to do that they're called

ethereum Improvement proposals you know you have to like explain what your changes what it does uh write a specification for it uh write like a reference implementation write some test cases and whatnot so you go you know you do that um once you have a specification that's like an indecent spot most people will like come on these calls and try to get feedback uh on it from from like client developers um most of the time there's something wrong but there's specification you know they've missed a bug um I think this is the thing that's like very different uh with ethereum than say writing other software is that the network like can't go down and so like there's a huge amount of scrutiny on the the security side of things um so most of our conversations and

these all core Dev stalls are basically like there's a security issue with this thing how do we fix it um and so most of the process of like somebody wanting to propose A change is actually like working through those things and and kind of iterating on that and then um at some point uh you know we decide we want to like make an upgrade to the network um and there's different like reasons for that like but generally there's at least one proposal potentially a set of proposals that like we think makes sense together um and by we here I mean you know the different teams of like core Developers agree makes sense um and there's no like formal Vote or anything here so we use what we call like rough consensus um and the idea is that

we we have like this very open process where anyone can come and make proposals client teams you know over time sort of agree both conceptually to stuff but then like as they implement it they're like what is a set of proposals that like are coherent together that you know add value are not so big that like it's impossible to test them and like you know feel confident in the quality of um and then they basically pull out a a software release which says you know if you run the software as of this uh time the the rules of the network will change but this is kind of where you know and then you know we'll put the ethereum foundation will usually put up a blog post like saying like you know hey this

is a proposed upgrade these are like the changes going in and this is when it would happen um but this is sort of where this sort of quote-unquote power that you know the the like ethereum governance has sort of stops because if nobody chose to run this software then like nothing would happen right um and and so you need basically all of the validators but not only them right like even people who are non-validated those you can think of exchanges or you know just individuals you need all of these people or at least the overwhelming majority of them to also run the software um otherwise the network will like split right like um and and or not upgrade at all you could imagine the core developers put out an upgrade that nobody wants and you know the network

just does an upgrade because nobody's writing that software um I think this is the part that's like very different from ethereum than other blockchains which have stuff like on-chain governance or you know like formal votes um is like it does require like this like amorphous group that is like the ethereum community to then upgrade their software um and what you see in practice is like you know most of the time there's an upgrade that's proposed people actually run it and upgrade a software and it feels like this continuous process um but I think it's worth like highlighting that this is this is kind of because I think client developers anticipate that like the community could refuse a change that you know they would dislike and client developers would be

the one who would have to deal with like the technical Fallout of that like if the network Forks you know it's it's obviously a bunch of technical work uh that they'll be the ones handling um so there's like this proactive you know like uh this proactive thought of like okay what are things that the community will actually want or or like um but you could imagine a case where it's not that like imagine the client developers decided we're all going to mend ourselves a million each obviously the community would refuse that um and and so like you know that is the case for like the upgrade probably wouldn't go through um but I think it's yeah it's easy to forget that there is like this two-way kind of power Dynamic there and and we've seen this happen in the past with

like ethereum classic for example even yeah and I think that's what that was like a pretty contentious split like there was like a pretty strong you know Community there um with the merge we saw sort of like a micro version of that where like there was you know a very small contingent of like proof of work miners who who didn't upgrade and like kept running their proof of work chain um but I think that was the case you know say ethereum classic was like I don't know some sort of like maybe 80 20 split or something like that the merge is probably sort of a 99 to one type of split um but you could imagine you know just a very bad proposal leading to like 99 to one in the other side where like the entire Community rejects it and we also have a bunch of cases actually of eips

that just never made it in to like an upgrade because of like community pushback and so there was one uh several years ago where like a big multi-sig got locked um and and sort of there was a request that unfreeze those funds and a lot of the community was against this and so even though it was a relatively technical technical technically simple change that actually affected some of the core developers it never made it into an upgrade because of like this anticipated Community pushback um there was another change uh Prague pal around like how proof of worked at proof of work worked um which would have like favored gpus over A6 and then again this was something where I think core developers were like quite sympathetic to it but

the community had like a strong negative reaction and the change just never made it into an upgrade um so yeah that's roughly how it works so I come to consensus on which change to push but the community always have the like has the last say to like actually run the software yeah that's a great high level overview because I do know it is a little like daunting in a way to like try and figure out how this is actually working like underneath the hood and a lot of people aren't aware like oh you just go to this website and you can kind of participate in the in the conversation but I guess I'm curious how what what methods of communication are the most effective in terms of getting everyone who's running a client on the same page like to get the merge pass through like that's pretty

incredible that all these people all around the world running clients like decided oh yeah okay I'm gonna upgrade at this time so how does that work and then also can you speak to kind of the importance of client diversity yeah of course uh so there's a couple of things to unpack there and just on the first thing you said like go into a website and seeing how it happens uh the ethereum.org team has done like a phenomenal job over the past year or two of like documenting all this stuff and they have a really hard job because they're like documenting things as they're sort of moving and changing but they do have a governance page I think it's like ethereum.org governance or something and it does roughly kind of explain what I just said and links you to all like the places where you can

participate so that's would be like the first place I'd I would send people who are just high level interested when we have so when we have uh actual Network upgrades like there's a couple different communication channels um like the the one that like people refer to first like the ethereum foundation has a Blog so like blog.etheism.org and you know uh there'll be a blog post there but I think beyond that uh all of these client teams right like they have communication lines with their users like if you're a user of like uh Basu or another mind or prism or Lighthouse um you know you go to their GitHub you go to their release page whatever like they'll have a release and they'll explain kind of the same thing saying like this is the release to download for prism for this upgrade and these are the

changes that are going in um so that's that's like another place um obviously you know I'll tweet about it personally um and then you know there's a bunch of folks like that'll from the infrastructure side that are either proactively reach out or like if they have some generic uh you know contact that email sometimes I'll send them an email just to make sure but um yeah you know I think uh it's also maybe like another one that's worth noting is we do have like this uh r d Discord server that's you know pretty open where people can join so there's a lot of folks like I don't know say from like inferior coinbase or Lido like these these people they're just like hanging out there to monitor things as it's as it's happening so they they have noticed sort of before the blog post happens

um and then even things like you know week in ethereum like in these newsletters they'll also do like a good job of advertising yep um consensus also has been like a good job historically of like promoting this stuff um yeah so I think there's like diverse diverse channels um and you know the the this the quote-unquote decision gets made on these public calls that like me and uh Danny Ryan also runs them um so this is like the Main Place anyone can go to there's a YouTube channel like uh I think it's uh ethereum protocol um this transcript there's notes so like if you just want to like know exactly when it happens and see it then like you can just follow the calls but then if we sort of repackage you know the contents of the call once and and try to put it

in a cleaner format for to send over to people um yeah oh and so yeah you mentioned sorry the last bit was client diversity um so this Center kind of can of worms but um ethereum is different than a lot of other blockchains and that um it's written as like a specification for which there's several implementations um and if you're not very technical like this might sound weird maybe the the best analogy I found so far is like a web browser so if you're like a user of the web uh you know I'm using Chrome right now but you can use Safari Firefox Opera whatever um all of these web browsers are like different clients right like if I go to riverside.fm on Chrome and you go on Safari we get served the same website um because they speak a common protocol

right like they make an HTTP request to the server they get you know whatever the website like should render and and then they have like a an implementation of how they should render HTML how they should render you know CSS and all those things um but Chrome and Safari can have like different features right like there's a bunch of different customizations and whatnot um and this is basically how ethereum is structured as well so there is a specification for the protocol um it's within like mostly in Python um but it just says you know like this is what an ethereum block should look like and when you have a transaction this is like what a valid and valid transaction should look like and it has like all these rules or for example like with the evm you know this is how much it costs to add two numbers together

this is how much it costs to add to multiply numbers and whatnot um so we have this specification which is like the thing that we agreed to change and and you know that like everybody sort of needs to come to consensus on but then each of these implementations so like uh you know we have both consensus layer clients who run the beacon chain and then execution layer clients who run like the evm they each have multiple different implementations uh so like on the on the weekend chain we have prism Lighthouse taku Nimbus load star on the execution side we have gath Aragon uh base two in another mind and all of those you can think of as analogous to web browsers that like you know they will follow the protocol rules

but they'll give you different features they'll optimize for different things um and this is really valuable for ethereum because it means that a bug in any one of those clients cannot like take down the network right like um so and we've seen this in the past actually you know in 2016 there was this big attack on ethereum that tried to like crash the network and at the time there's only two clients get in parody and the attacks would like find a bug in one of the clients and like exploit that but because we had the other you know people could switch over and keep running the chain and vice versa um and and I think this is like the basically the the really important property that if you don't have this imagine you know a world where we just had yes and nothing else if there was a

bug and get that say like I don't know minted a million East then the chain sort of can't stop like there's no one there's no other like source of Truth uh on the protocol and and that's really bad right whereas now you know if say there's there's a block where like you know get things a million eats should be printed but like Aragon basically Nimbus all sort of disagree um then you know it doesn't have to like scarred or protocol in that way right like we we just have like different implementations telling us oh actually this is this is wrong I'm and so there is a cost of doing this like there's like a complexity cost where like when we Implement stuff we need to like make sure that it doesn't just work and get but it works in Geth and all the other

ones and now since the merge you know we have these uh Beacon chain clients as well so every pairwise combination needs to work um so there's like a higher engineering cost to do that that's what it buys us it's like a resiliency and like security and that's like you know if any one of those clients or any one of those pairs doesn't work um then you know you have like 19 other pairs that you can use um or you know four or five other clients um that you know hopefully would not have to bug um because they're written in different languages by different people they have different you know testing processes and whatnot um so yeah that's how sort of that I guess that's how ethereum is structured and like why it's important is just because it gives us this extra degree of security um where if there's a bug you know it's

not the entire protocol but it's like a single client implementation that has that issue so when you think about things like um you know liquid staking how would how does that play into client diversity because this I guess the you know the net benefit would be that it brings in more uh ethereum into the like drives more stake to ethereum um but then I guess It ultimately it matters on how decentralized that validator set is uh that the liquid staking protocol offers so how do the core devs kind of think about how liquid's taking as a whole uh kind of shifts the security uh promises and guarantees of the chain I guess that's two ways to answer that question so first is like what did the human quartets think but there's also like how does the ethereum protocol in a way like anticipate or deal with that and it

might be easier to start with the protocol as like a foundation and then you know uh go to like the more Human Side um but like basically the ethereum protocol uh if you think of it as like just like a set of computers it's like this distributed network of computers that all has to agree every 12 seconds on like what state of the what is the state of the network um and obviously like because it's so distributed we can't expect like a hundred percent of the computers to agree all of the time like we have to be lenient to some degree that like you know some people will be offline we still have to like chug along and move forward and this is a decision that like it sort of this design decision that ethereum makes that like the most

valuable property of the network is what we call liveness but this basically just means like keep producing blocks um other blockchains will will have designs where like you know if not enough people are online they'll just stop um but ethereum really values this idea that like you know the network should keep moving forward pretty much no matter what um and if everybody's like online all of the time then it's it's pretty easy to keep the network moving forward you know you just ask everybody what do you think about the state if they all agree you know move forward and lockstep um and but that's it like if if people start being offline that starts to be bad um and the way ethereum sort of is designed is that um there's like an exponential uh it's exponentially bad the more people

are offline and that like you know we have about 500 000 validators now I think give or take um if one of those is online is offline it's really not the end of the world so the penalty to them should be small but if you know ten percent twenty percent thirty percent go offline uh we should kind of scale the penalty because um this this means like the network is like struggling much more to come to consensus on on uh on the state of things and and the flip side of this is it's actually quite nice because it sort of encourages uh decentralizations and solo stickers so for example if I run you know like a node from from my house in Vancouver and I go offline and I'm the only one who go up it's offline I get a much smaller penalty than if like

the entire network has gone offline um and and the reason for that is like if every if like a large part of the network has gone offline you want to like have a high incentive for people to get back online and and this is why you sort of scale that that penalty um so uh you know the penalty basically scales with however many people are offline and once you get past the third of the network being offline it'll start scaling much more um because ethereum sort of assumes you need two-thirds of the network to agree on on the state of things to move forward so if we have less than that we're sort of stuck where um we can move forward but we don't have what we call finality so you don't have a guarantee that you won't uh reorg the chain you know back to an arbitrary

point we need about two-thirds of people to uh to make a decision on like what is the deepest the network would reorg um unless it was uh severity compromise um and so if you if you think about that like the the sort of high level thing that that like is incentivizing is we want to have failures that are Decor related from each other as as much as possible so like it's fine to have you know up to a third of the network you know going down um and in practice we've never had anywhere near that I think you know like five-ish percent for like hours is it's probably the worst case we've seen But like you know the protocol can tolerate up to like a third and still move along you know to to a large extent

um but that means ideally you never want more than a third to go offline and you know it would be better if like you get like much less than that and when you know parts of the network go offline they go offline at different times than other parts of the network um and you know in practice this means like the network sort of incentivizes you to not have like all the nodes on the same AWS region because if that one went offline your penalty as like a Staker you would lose way more money than like if you split your nodes and like even just two AWS regions um where like uh you know they wouldn't hopefully fail at the same time but then and then if you take this further it's like well if if you all have if you have like this coordinated risk on AWS you

know maybe AWS shuts off your nodes uh for whatever reason and if more people are on AWS um then they'll get a harsher penalty so that sort of forces you to move to like other places um and and this is kind of the how the protocol is designed it's like can we can we like push the the failure mode to be as decorated as possible such that like when something bad happens it affects as little of the network as as possible um and so I think generally that's also how core devs try to approach it like what are ways that we can make speaking like approachable from as many different groups and individuals um and this is why we have like such a big focus on things like keeping the node requirements slow and things like

that um and even you know the people often ask like why it's like 32 weeks that you need to stake and like why not 16 or like you know 64. um and this is basically based on bandwidth so like the number of validators on the network sort of create they each create some amount of messages that they need to send to everybody and so if we had like 60 Neath instead of 32 you would need to send twice as many messages and like 32 East is like roughly you know I believe it's like a 10 maybe 20 meg per second like bandwidth requirement so it's like accessible you know in in like a lot of places with a pretty reasonable internet connection um so it's like about finding this sweet spot between like how much Capital should you have to put versus like what

sort of you know technical equipment that you need because if the minimum stake was like one East but then you need a gig per second in bandwidth um you know you've created like another barrier to entry and I think this is generally like you know the the sort of problem client teams try to optimize for with regards to like staking access is like you know what's The Sweet Spot of like you know low capital and like low hardware and low bandwidth requirements and obviously you know price of ease is volatile so like the capital bit is the one that moves the most um I think that's sort of what we pay attention to um and similarly when people talk about like the block limits like wire blocks this big it's basically the same thing it's like you know the bigger the blocks

are the more bandwidth it takes to gossip them and the more storage it takes to store them so you need to find this sweet spot of like you know blocks being big enough to you know keep fees low but not so big that like then nobody can run node and you need to centralization um yeah so I guess I'll pause here but this is like generally just like how I think the protocol is designed how we think about the problem and like why sort of there might be some like counter-intuitive design choices in ethereum that that are very different from from other blockchains yeah that makes a lot of sense um yeah I appreciate that overview there's so much knowledge in there actually that like it's kind of like the basics but at the same time like a lot of people you know kind of move on to

the advanced stuff before they actually know the basics so I think it's great to to go over all that stuff uh in terms of the uh eth road map going forward uh what things are you the most excited about I know we got Shanghai coming up here pretty soon but beyond that like what would you say you're looking forward to the most in ethereum yeah so um I guess yeah I'll start for more wrath now and then sort of wander off in the future but like um the biggest thing we're working on right now is like you said Shanghai so this uh will bring uh Beacon chain withdrawals uh to the network which which is you know uh a very anticipated feature in case people don't know so right now if you stick on the beacon chain it's sort of a one-way operation you can't actually withdraw

your funds um there's a bunch of like historical reasons why it was designed that way uh and when we when we did the merge uh we wanted to keep it as simple as possible because there's already a pretty big change and so we didn't add this withdrawal functionality but now it was like number one priority to get it out after the merge um and we're pretty much you know finishing up on this um this morning we agreed to like Fork the first test nuts towards the end of February so like in the next you know month or two like we'll start moving to like test Nets and then maintenance and you know have have that out of the way um the next big thing that we we've been working on already is um what's called uh Proto Deng Chardon or VIP 4844 um and the idea there is to help make

layer twos more uh I guess cheaper for users um so layer twos uh are sort of how ethereum thinks about scaling to more users you know like we were just talking about you can't really scale the layer one too much if you want to maintain uh like no decentralization so layer twos are like these really interesting constructs which run sort of a separate ethereum chain run a bunch of transactions on it and they'll only post either like compressed data or proofs of transactions back to layer one um and only have to like run the actual computations on air one if there's like a suspected bug or fraud happening um so that means you can kind of uh

lower the cost of transactions because you're you're you're basically running out all of the transactions quote unquote off chain on the on the layer 2 Chain itself um and only running like a small fraction of them um on on L1 if and when there's there's like an issue um that said though because they post all these proofs and all of this compressed data back on chain this is still quite costly for users so if you transact on Layer Two today um I think optimism had had a dune dashboard which was selling something like 95 or so percent of the cost is actually posting your data back to layer one um and and and and it sort of makes sense because you know this data on

layer one needs to be downloaded by every node and then stored forever in the blockchain um so the idea with eip4844 is that we're going to create this new sort of data layer on ethereum that's kind of temporary storage for layer twos um so layer twos when they post back this data on there one they don't actually need it to stay forever on chain um they only needed to be there long enough so that if there was a bug or a fraud somebody could like dispute it and this is why I say when you withdraw from something like arbitrum or optimism you have to wait like a week for your funds it's basically waiting for somebody to say like you know something went wrong here and then assuming that doesn't happen to give you your money back um so effectively these are choose don't need access to that data for more than

like you know a week and you can imagine having it be longer than that you know for some historical archival purposes um so eip4844 is is going to introduce this this new data layer where you can store data where for I believe it's roughly two weeks give or take uh on the network and then pay a much lower fee than you would uh by storing it forever on chain um and so that's a big thing we're working on um so it'll enable layer choose to like go from posting their data back to another ethereum L1 uh forever to posting it to this Taylor layer which will be more ephemeral um and and pass those cost savings on to those users and that hopefully will help us you know scale the amount of people who can use their cues uh in in the near future

um so we don't have a timeline for that yet but it should happen sometime this year you know uh we're already working on it now that that's what draws is wrapping up I'm beyond that uh I think there's a bunch of other sort of big kind of problems or or areas that we're tackling um on the proof of stake side there's a few so uh the first is uh this idea of uh single slot finality so right now on ethereum when you to wait for finality or like you know being sure that the chain is is not going to reorgan the Past you have to wait about 12 minutes and it's kind of this binary thing where after 12 minutes you know that like uh it would take more than two-thirds of the stake to like reorg the chain and we

consider that safe um but it's kind of weird that it's like this binary Threshold at 12 minutes right it's like why wouldn't you get like half that certainty at like six minutes um and the idea with single slot finality is that it's to like sort of apply a more gradual sort of finality threshold so that like imagine you know you send a thousand dollar transaction maybe after one block you get like you know 50 million dollars of economic like weight behind it um you might be fine accepting that transaction right um but you know if you're sending a billion dollars on chain maybe you still want to wait like you know 12 minutes that's fine um but I would provide like much better ux for the average user who obviously doesn't send a billion dollars on chain um so that's one of them the other big

one uh is around Mev so this idea of proposer Builder separation which is something we've we've seen uh sort of play out now outside of the protocol um but basically um again just to put a bit of context the idea with Mev is that there's a lot of money to be made by choosing the right order of transactions in the block um and for people who aren't familiar at all you know like the most obvious example here is Imagine somebody's making a big trade on uniswap you know you might want to buy the thing they're going to buy before them when the price is low then they're going to buy it to push the price up and then you can sell right after them and you know collect an Arbitrage profit and this is like the most I guess common Mev example but there's a whole kind of world of that

um so people it's like this entire industry of people competing to build like the best most profitable blocks right like the order transactions in a way that they can extract the most value from um and when you think of like what validators or miners have done historically it's a bit of like a quote-unquote like Dumber process where they just have transactions lender mempool they'll usually sort them by the highest fee and like include that and you know propose the block um so we've seen sort of this segregation of like this very specialized like uh block building industry which yeah is optimizing for every possible sort of you know uh like Center value they can extract and then

we have these validators who like we want to make sure they can keep a relatively low uh Hardware requirement then they'll have to run like these compute intensive algorithms to figure out which which is the best block they're going to build but ideally you also don't want them to be at like a huge disadvantage where like you know only like big centralized validators have access to like these these uh profitable blocks but everybody else doesn't um so the idea with proposer Builder separation is you you know you make those two rules separate and you sort of create a Marketplace between the two of them where like if I'm a validator and I'm about to propose a block then there's sort of an auction between different Builders who will each like pay me for the right to like propose the block for me and you know extract

Arbitrage profits um and today this happens outside of the protocol so we have this thing called Mev boost um where which is basically you can think of as this marketplace where validators can choose to run the software and this will connect them to a network of Builders who can each propose blocks um but the fact that this is outside of the core protocol sort of creates all these trust assumptions on you know that that like Marketplace so that's something that I think over time as we understand how the like this Market structure develop it makes sense to have more of this live inside of the protocol and just reduce the trust assumptions um to make sure that like you know the there's not something where like the the dominant players in that Marketplace can

can end up sort of either censoring ethereum or extracting some um like disproportionate amount of value and so that's another exciting one um and then maybe a third one uh that like I'm personally excited about is um something we've been working on for a long time called uh like stateless ethereum um and the idea there is that uh again this comes back to like uh node requirements um ethereum has this concept of a state where this is all of the accounts and all of the smart contracts on the network at a current time right so it's not even like the historical balances and whatnot but it's like at this current block you know my address has X your address has y this smart contract has this data this code

um and this grows over time obviously like more people use ethereum there's new contracts that come in there's new users that come in um and it doesn't grow as fast as like the history of everything that's happened but it's still growing sort of relatively fast um and one future you can imagine is like you know it's going to be impossible for everyone to store this if it's too big so maybe you sort of break it up in pieces and people only have to store a subset of it and you can imagine a world where it's like you store a random subset of it you could imagine a world where like when you run a node you specify like oh these are my accounts these are like the smart contracts I care about and like just for this um and and but that means that if uh if

not everybody has like a full proof of this date you need to have when you make a transaction you can't just like send uh a transaction you would also potentially have to send people like a proof of the actual state which they can then verify so because if they don't have a copy of your smart contract say you know you're selling an nft they might not even know this nft contract exists so you need to send them approved that's like hey here's this nft contract on ethereum I actually have this nft you know at you know this storage slot in the contract and whatnot um and sending those proofs over the network is quite expensive so we've been working on just a a sort of database overhaul for ethereum that would help make it easier to do that and um and the basically the value there

would be you could have more and more applications but not force everybody to keep track of all of them and and therefore grow the node requirements um so yeah I I'll pause here but like I think yeah those are the three that I'm interested in just like how how do we make finality more Progressive come quicker how do we just keep Mev from being like centralized and just a small number of providers and make sure that like all validators can can have access to it and then how do we you know enable ethereum to like keep scaling and keep having more applications more users um without forcing every every node to store all of the state forever yeah yeah that's definitely a super exciting even just to think about how far ethereum's come and yet how bright the future still is on the development side of things and

so obviously a lot to unpack there but there's two things that really jump out at me I and the first being really love to get your take uh on the current state of l2s because of course l2s are going to play a very eccentric role uh in the future of ethereum but if we look at where they are today just like you know objectively assessing things you know optimistic still roll up still really don't have fraud proofs pushed live and the sequencers are still quite centralized so where do you think we are in the L2 evolution timeline and like kind of like what comes next like how do you think we get to these decentralized l2s yeah I you think in terms of trust assumptions like they are like much more trusted spinel ones today and people should know that like you know if you bridged

optimism arbitrum uh start net like and any other like they they all have like pretty significant trust assumptions and they're working on lowering them um and the two so so the two the two things you mentioned are like yeah decentralized sequencing and and and fraud proofs or fault proofs how they like to call them now um so I hear from LT teams that the fraud proofs will be coming in like the next year or so um I think we should hold them as a community accountable to that um that's probably the single most important thing and because it it sort of allows you to uh it allows you to dispute something that happened on the Chain you know if uh if something goes wrong in testing that mechanism is

really important um I think you know there's also other parts of centralization like for example you know like who can upgrade the smart contracts like uh if there is a new release like who who is like the group that like you know says this is the new arbitrum contract this is the new optimism contract um and that might be you know that might be something that like different ltus choose to like have different stances on like you could imagine a world where like I don't know l21 says like where as a company always going to upgrade these contracts you know we're sort of offering your product um but you know then there's full fraud proofs and whatnot like when you're using it right and this is kind of the model that's like something like uniswap

takes right like when you need a swap has like uh went from V1 to V2 or V2 to V3 um you know they obviously put out this contract uh but they can't force people to like use it right like the contract has to be better and you know that is like a an incentive to like use it like if it actually brings something something new so you could imagine like an L2 that's like doesn't have fault proofs and then like publishes a new version would would fall proofs um they might sort of do that centrally but you know provide a better product than I think you know that's that's potentially a reasonable approach um but you could also Imagine the opposite right like you could imagine that out to you which wants to have like a much more public governance process about that than like vote on like

version upgrades and things like that um and I think we'll see people experimenting in that like optimism is experimenting with this dual governance structure um of you know like having both token holders but also some individuals both and and see how like you can have an interplay there um I think on the ZK side as well there's a lot of uh there's a lot of so zks don't have like fault proofs basically um but what they do have is like uh the the provers uh for for for the cryptographic proof that they post on chain so like you know if it's one thing that like you know they post proofs of what happened on the ZK chain on L1 but if nobody can actually verify that like the the software that is

evaluating those proofs you know you have to trust them that there there's not like some back door malicious or not right like it could be a back door in a malicious case it could be a bug in the in the like uh non-malicious case but you sort of have to trust them like the code says what it does right and and that is something where like I think again like it's valuable to have the community put pressure towards like making that open source so that people can actually review um review the implementations um I think for decentralized sequencing um this is I think this is something that's almost like good from uh so it's good for a few reasons right like it's good from uh basically like a risk management

standpoint for the L2 and that imagine if there's an L2 where like the sequencer is only run by like one company um and you know their infra goes down then the whole chain sort of stops and that's really bad right like so there is some amount of like resilience that you can get from like you know decentralized sequencing um but I think there as well it's worth you know there's definitely different flavors that like l2s might explore so for example you could imagine something like a token based you know almost like staking on ethereum right like you put up your tokens and you you get a spot as a sequencer for some amount of time or something like that but you could imagine something different that's more like some sort of like committee or like multi-sig of like um okay like these are the six entities

that like run a sequencer and you know like we think that they're good and and hopefully they don't go down um or even like some mix of that right like you can imagine something where like not anybody could be a sequencer because you want like a high a high technical requirement for it but like maybe governance votes on like okay who are like the top 10 sequencers or something like that right um so I think hopefully we'll see like some experimentation there um but yeah generally I I agree that like you know it's it's fair to the man like of l2s like what you know if you're not trust us today at the verities like what are your plans to do it and then hold them accountable to that um and I think one thing I am though very excited about is like that we have

this like diverse set of teams trying different approaches um so once upon a time there was like this old ethereum scaling world map where like all the scaling would come from the core protocol itself um and this means like Klein beds and researchers would be the ones that have to like design and Implement everything and I think the fact that today like we see you know teams with such different approaches like already live and trying different things um is is actually really valuable like you know clearly like the client teams that exist on on L1 today like couldn't have done like one of those things well and now we see you know like five maybe ten different approaches uh to the problem and um I think that's really exciting so I hope

um yeah and I think in the next year we'll see sort of which teams are actually you know walking the walking and like like yeah like doing good on those promises to decentralize yeah I strongly agree with your your viewpoints there and you know as you see arbitrary and optimism kind of like really garnering some uh like user base as well as TBL and like you know that's the first thing that comes to my mind is like how can we encourage that like we push in the path of decentralization um but the other question I really had uh when you're running through like what's to come was around like the scourge section of the road map so uh you know part of that is really based around how do we create like credibly neutral transaction inclusion and I think a big part of that is

democratizing Mev and so one of the pieces that I saw on the roadmap was Mev burn which is a really interesting opportunity so kind of like love to get your take on like why that could help achieve that goal um and really like one of the things I'm toying with is you know should Mev only flow to stakers and validators because like that's the the extra incentive of participating in network security so I think mbv Vernon is still pretty contentious topic um I'm not even sure I oscillate between murder I think it's a good or bad idea you know depending on the day um but I think you know the the strongest case I can make for it is something like this where like um

if like I guess you know yeah to start from the highest like the highest level thing it's like should High Mev go to like the validator of a certain block right like you know if if it just happens that somebody like you know sold their Punk for like one eighth instead of 100 East um in this block should like that 99 East Arbitrage go to like you know you as a validator just because you happen to be in that slot or should it be distributed like more broadly right I think this is like the first question there um and again there's like a case for both sides but assuming you think like you know it should be sort of distributed to some extent but I think the next logical question is like okay

so then should it go to just you know the validators that are staking right now or should it potentially go to like all of the eat holders right as as a group or sort of a mix of that um and I think this is where Mev burn comes in um and this the idea with Mev burn is just like you know if if you get the the builders to sort of bid you know for for high uh high like profit opportunities they're gonna bid a lot of money for that and what do you do with that bid right like do you send it straight to the validator do you burn part of it or you know burn all of it and when you burn all of it which you're effectively saying is you redistribute it to all the etholders right because you're reducing the supply

um and um I think that the the thing that like so and you know maybe like the obvious argument there is like you might think it's like more fair to like uh to burn it so that like uh you know every like it's just redistributed and it's not it doesn't go to that single validator um there's also some uh some like security properties that you get in that like if you uh if you have like really high Envy blocks um people might want to like reorg the chain you know for example like this Punk example um you know there's 99 East to be made then like should I maybe like not see your block and just you know Fork you

out and rebuild with that transaction in my blog and and then the next validator does that as well and that creates like a really bad user experience right so like muv burn is a disincentive to do that because you're sort of you're not actually getting the 99 East you're getting you know potentially nothing if it's all burned or potentially just a small fraction of that um and I think there's another uh another thing that that plays into it uh that you know potentially makes me more in favor of Mev Byrne is um the question of how much validators does ethereum need um so like we were discussing like at the very beginning each validator on the network had adds an overhead in terms of like bandwidth requirements to run another validator so you could imagine a world where like if we have too many

validators um then we might start to crowd out like validators with poor internet connections and drop them off um and it's not clear that we need like this you know there's diminishing returns to like how much eats should be securing the network like uh at some point you know the the like incentives to attack sort of get unrealistically high and like the marginal validator doesn't uh doesn't add much value and we already have this sort of assumption in the protocol where like we reduce validator rewards based on how many validators there are so there's already this this curve of issuance you know that like tries to attract people if not enough people are staking and you know incentivizes them less if more people are going and the question is like you know should potentially that curve like

go negative or something like that or you know there's just a fixed cap or something there and I think if that was a direction that we wanted to go and that direction could potentially let us explore things like lowering the minimum State because you could imagine like you know if bandwidth gets better right like in in the next five ten years or something um then like okay say we need 10 to 20 Megs today with the current Stakes you know maybe in 10 years like we can double this to 20 to 40 Megs and it's like the same you know quote-unquote technical requirement um so you and we have to stake the the 16 or something like that but if we've like doubled or quadrupled the number of validators then we're sort of in the same spot so you can make an argument that there's like a a right number of

validators to have if you believe that then um I think muv burn makes much more sense because you're telling people you're sort of gatekeeping people from being validators right and then saying like look you know even if you're not a validator as a needs holder you can receive the Mev profits that the chain generates that sort of reduces the incentive to give to be a validator and you know and and you can get like you know quote unquote most of the rewards from being just an etholder um that I don't know that that makes more sense to me um but yeah I I don't think it's like a simple question um much more complex than like the 1559 burn so um the 1559 burn was very different in that like the reason we burned the fee in 1559 is just so miners

don't like mess with it and fill blocks with with Spam transactions and this is like a clear like it just distorts the entire like Market signal if we don't burn it and so there was like a clear rationale for it it's really hard to argue it's like the whole mechanism sort of falls apart in a pretty clear way if you don't burn it um whereas this I think it's like a much more nuanced answer like there's definitely a strong case to be made for it um but I think there's people who would argue like that you know like maybe like yeah you already of either of that block you're lucky you like got the Arbitrage opportunity and like you know you produce that block therefore you should keep those fees um and and I don't think it's an unreasonable position either um but yeah I I think we'll talk a lot

about that in the next uh couple years for sure yeah this is one of my favorite topics I know me and the rest of the guys on the research team are always talking about it so it was actually really interesting to get your thoughts because you raised a couple there that we hadn't discussed but uh Tim you've already been so generous with your time so and I know you're a super busy and highly sought after guy so do you want to just share with the audience where they can find you learn more about you know what you do and and then we can call it so Twitter's the best place uh I'm at Tim Bako there um my DMs are open so if uh you want to chat you can tweet at me or hear me there um yeah I've also been using forecaster as well uh recently so uh at Tim bakel

there as well um yeah those are probably the two best places nice well thanks so much for coming on we'll have to do it again soon yeah thank you for having me this is great