MEV timing games, Layer 2 growth and more - The Daily Gwei Refuel #694 - Ethereum Updates

Transcript

hello everyone welcome to another episode of the delway refuel where you C the latest news etherum ecosystem I'm your host and this is H the 27th of November 2023 all right everyone let's get into it so I think today's episode is going to be a little bit ranty cuz I was going through the content that uh that I was getting ready for today's episode and I noticed that there is a lot to talk about uh on each of these kind of pieces of content there's not that much like in in terms of how much content I have to talk about but each thing is pretty important I think so to kick things off I'm going to start off with something that's got to do with me so Tony here who I've talked about a bunch on the refi before he builds the meev do uh meev boost. p webites and

other websites and just does a bunch of research on ethereum he tweeted this out saying pp. org or P2P org which is a ethereum staking service provider has announced that it will call the get header uh later in the slot trading off chain stability against more me profits this is exactly the centralizing behavior that we shouldn't tolerate as long as we as users have the choice about where to stake out Assets Now I guess there's a lot to unpack here and I'm no expert on me or anything like that and I'm no expert on what a ethereum slot looks like cuz that's what Tony's talking about here so you recall that a slot in ethereum is 12 seconds people refer to them as as blocks but a block the block portion of the slot only takes up a you know a a a tiny amount of

the overall slot or a short amount of the overall slot uh so yeah it's called a slot instead of a block those 12 seconds now what P2P or is doing here is that they're basically delaying um when they're calling get header uh in the in the slot in order to maximize the meev Boost profits because if I'm explaining this in plain English it basically means that by delaying when they call this they basically force people to bid up the uh I guess like um uh the amount that they're paying to extract me which means they could inch out more profits out of the the total slot here now this leads to things like chain stability it's a centralizing force and it's definitely not something that we want to see playing out uh on the network now there are multiple kind of tracks here

in dealing with this with this sort of thing and the technical name for this is called timing attacks or timing games which Cas by here who does research at the ethereum foundation uh put together a thread on which I'll link in the YouTube description below but I want to make clear that this is not something new this is something that has been in research phase for quite a while now it has been well known that can happen for quite a while now and we've basically been relying on each of these kind of staking service providers or just validators generally or note operators generally to not do this right because it's it it right now as far as I know there aren't any in protocol mechanisms to prevent this uh and there are I think some clients that have things uh implemented to prevent this but it is an

optional thing because it's not in protocol as long as it exists at the client level it is always optional whereas when it's in the protocol it's not optional you have to follow that uh and this is actually quite a difficult uh thing to work around and it falls into the larger Road mapap about around around me and kind of mitigating the negative externalities of me and I actually said last week I was on uh on on a panel or solo panel where I was talking about a bunch of different things with ethereum at the event that I was at at Sydney in Sydney and I basically said that uh something that I've said on the refu a bunch of times the me unchecked meev I should say is really the blockchain killer right unchecked meev no matter what blockchain it is like pretty much every blockchain

suffers from this issue unless that you have a fully private layer one uh where me is really hard to do if not impossible because you can't really do much M if you can't actually see what you're trying to M at least that's from my understanding but essentially the reason why I call it the the blockchain killer is because of things like this right we've seen the negative externalities play out around censorship and around uh the different Builders the block Builders and the relays censoring uh certain transactions uh we've seen it play out in the potential for there to be reorgs we've seen it play out in what I'm talking about today with these timing games right and there's are various different ways it can play out now as I said there is a huge part of the road map dedicated to dealing with me and and all that entails but this is

not unique to ethereum in any way I think I want to make that clear because people will talk about m in the context of ethereum uh because of the fact that ethereum has the most me to extract because it is the most active ecosystem in terms of the value that's flowing on the ethereum Chain now as I said that this means that other ecosystems are just smaller and if they had that same value flowing through they would have the exact same issues and they would have to deal with it in similar ways to the way ethereum is attempting to deal with it but it is not an overnight thing it is a long road map it is a hard problem to solve because the the reason why it's a hard problem to solve is because you're basically fighting against economic incentives here or essentially the uh I wouldn't call them

attackers but the people doing the meev right and people benefiting from the me their incentive is always to get around any roadblox you put up to extract as much profit as they can or as much economic value as they can uh from their role on the network now because of this and because that's basically your adversary there or that that's what you're competing against as a protocol researcher and a protocol engineer and a client engineer essentially what you need to come up with is ways to make uh this uh impossible to do and as far as we know right now there's no way to totally eradicate meev from the network or totally get rid of meev from the network even under like fully private transaction conditions where the entire layer one is like fully private I think there are still ways to to do this I

mean as I said like I I'm not an expert on that front uh and V in of itself is just a huge field that requires you know all of your attention to basically be on top of it all but from my understanding and from what I said before unchecked meev like if you just let it go go to town on your chain will kill end up killing your blockchain because essentially what happens is that it just totally centralizes the chain all the value gets extracted out of it and in the worst case the chain stability uh becomes so bad that the chain just becomes unstable for regular users no exchangers want to deal with the chain no big service providers want to deal with the chain no one wants to build on it uh and you basically just have like this Barren Wasteland essentially it'd

be like a bunch of uh people coming through taking everything like for example if you want to use um some mythology here it'd be like a bunch of Vikings running through a town taking as whatever they could pillage and then uh and then leaving right and leaving the town in Ruins and no one wants to be there anymore you know everyone's everyone lost all their stuff no you know no one's coming to save them sort of thing uh and that's kind of what it looks like if you're you you leave meev unchecked which thankfully ethereum is not doing right as I said there is that large road map um dedicated to dealing with meev but it is a very very difficult problem it is a very hard problem um which is uh which which has attracted some of the brightest Minds to work on it but it's kind of a cat and

mouse game it is not something that you can just solve overnight and uh in terms of timing around how long this is going to take to play out in terms of getting a lot of this fixed and getting a lot of this rectified and making it so that it doesn't lead to things like chain uh instability and censorship it's it's years guys like I I I would have to say like minimum 2 to 3 years is to get the bulk of things that we want in but then it is it goes far beyond that like it's it's I think it's always going to be an issue until we get like a fully snar ified layer one which means like a fully private layer one which I think is definitely going to happen at some point but it could not it could be a thing that's that's 10 to 15 years out so I

think during that time waiting for that to happen we're really just kind of putting Band-Aids on the network or essentially kind of plugging holes in the boat so to speak uh because it's very very hard it is very hard like it's and then another I could use is that imagine that you're on a a boat with someone that wants to sink the boat and every time they they you know they put a hole in the boat you patch it but then they just put another hole in it right they find another way to put another hole in it no matter what you do that's kind of what it's like and uh yeah it's a very difficult problem but anyway that's the the first thing I want to talk about and as I said you can see Tony's tweet I'll link it in the YouTube description below as well as cap

Casper's tweet thread going into exactly what timing games are why they're bad why they lead to chain stability and he also links to a bunch of research around this as well which I'll link in the YouTube description below for you to check out all right rant number two is about a tweet that I came across today here talking about uh client diversity uh on ethereum now in this screenshot you as you can see on my screen is a list of all of the Lio node operators all 28 of them I believe um and what execution clients they are running so gu is in blue nethermind is orange Basu is yellow and Aragon is green now of course you can see that most run gu here but what this tweet highlighted was the fact that

even the teams building other clients like not building gu like nethermind are still running gu as their majority client you can see it here it's been highlighted that nethermind is running what it looks like 70% uh gu and only 30% are nether mine so of course you know people were like why are nether mine not just running 100% of their own client not just for client um uh client distribution or client diversity reasons but also just to put like their confidence behind their own own client right and I felt a little bit bad reading through some of the comments here and some of the the the replies in the thread because it felt like a pile on onto the neine team and to a lesser extent on some of the other cordev teams and I felt like it was getting a little bit toxic when I was reading through the

replies and I didn't feel like it was actually fostering an environment to actually discuss why this is the case and I have my own kind of take on this and why this is the case so when we're talking about client diversity or or I guess like the clients that uh different node operators are running you have to split this up into solo node operators so people like me or people like you if you're running a Solo solo staking setup there and people like the Lo note operators that are responsible not only for any of their own assets that they're staking but of the assets of the entire Lo uh kind of um uh Network so to speak right uh and this isn't just limited to obviously to lier it's for every staking service provider out there so if you've been running a client pair for a while

and everything's going good and you're trying to maximize profits as well because of the fact that uh as a note operator that's your duty to try and maximize profits in order to uh to be competitive and in order to stay as part of the Lio note operator Set uh because obviously the rewards are juicy for you as well then you're not going to just change what you're using because it's not just as simple as a solo Staker like myself for example changing what I'm using and if anything goes wrong it just affects my setup if anything wants to go wrong this affects like the entire L Network because all of the losses are socialized across the entire L not operator set not just the node or the validators that um you know each each individual not operator is running but everything is socialized so if you're

offline or if something happens um and I don't know in the worst case scenario you get slashed or anything like that right or you leak your validator keys out anything Got to Do With It uh it gets socialized so it is definitely not something that you want to do lightly and on top of that each of these note operators has their own pipeline their own kind of De devops pipeline here uh for how they do staking as part of lier and and just how they are as as a nde operator here so if you've got something that's running well uh and even though you know that it's not the optimal case because you want to definitely um uh contribute to client diversity uh you're not going to to to rock the boat too much and you're not in a rush to rock the boat that's the first thing the

second thing is the fact that up until recently I would say that gu is still the best client right to to run it is still the most reliable and and obviously it being a super majority is not good for the network it's because if there was a bug in gu then uh we would potentially finalize on a um a buged chain but when you're a core Dev team like nethermind for example they know all the limitations of their client right they know that um they they know that they probably know their client is not as I guess performant as gu or maybe even the Bas Aragon team is in the same boat there so because of that they also don't want to change because I mean when you see how the sausage is made you become bearish on the sausage right doesn't matter if it's an amazing

sausage to everyone else you're still the one seeing how it's made you know everything that could go wrong with it you know all the all the kind of like um under the hood stuff right so when you do know that uh it it becomes a bit difficult to change there now I do believe that nethermind as a client is up to scratch with gu at this point in time but it historically hasn't been and nethermind has been a lighter not uppr for quite a while now so those are the two main things I just wanted to cover there but also as I said I don't think we should be shaming any of these teams we shouldn't be shaming core developers into running other other clients here or basically saying to them hey why aren't you running your other client are you not confident in your client because

there are so many other reasons why they would be doing this as I just outlined as well there and on top of this when it comes to client diversity L is actually not too bad right it is not not great I mean you can see how much gu is here but if you want to talk about percentage share of the network versus gu dominance um you can look at if we go to rider. network we can actually look at the percentage share of the network um and see which is most dominant here you've got coinbase now rated on network will only show you the consensus client distribution here coinbase is doing okay I mean not the best they've got like 50% prism 38% Lighthouse and 11% uh teu which is fine you know it's generally okay and there's no um I guess like uh

uh there's no confirmation on this but I'm assuming here I'm making a big assumption but I think it's an educated assumption based on what I know I think coinbase is running 100% gu and they have 50% of the network so their entire node operators sorry all their validators are running off of gu right whereas with lato it's more varied as you can see in this screenshot here it's still heavy gu but it's more varied so if you really want to shame someone or encourage a provider to change to something other than gu you really should be telling coinbase this you really be reaching out to coinbase and saying to them hey you know you're running majority gu like or we're assuming that you're running all gu you should change right you should change to NE mind you should change to some of these other clients there and it's not

just coinbase you know maybe even binance is running all gu and and Kraken and other providers out there and and same with ligher obviously you should be telling these other node operators that hey you know you should change to other clients as well but it shouldn't be something where you shame them it shouldn't be something where you make them feel uh kind of like backed into a corner or you make them feel bad we should really just Foster a more healthy environment uh when it comes to this sorts of stuff so that was my kind of main issue when reading through some of the replies here I don't like it when we shame people that's not what the ethereum ecosystem and ethereum Community uh to me there but hopefully that kind of explains that for you and kind of my general view on that there so

that's rant number two done now taking a pause from the ran here just a highlighted tweet from Proto Lambda today uh where he said wasted a lot of time researching wallets they're all bad so obviously starting the Tweet off a little bit uh a little bit funny there but essentially this is a great tweet uh it's not a tweet threat it's just one tweet going over not all of the wallets that exist in in ethereum today but a lot of the wallets that exist in ethereum today and Proto's uh experience with them as well as a very very long and comprehensive wish list of features that he wants in wallets now he does make the kind of Distinction here that the wallet that he wants and the features that he wants is probably not going to be what the majority wants right the normies the people who don't

care about all these extra features they just want to kind of like swap a token on unisoft or something like that they're not going to care about about all of this but this still is a really really great uh tweet that you should probably definitely bookmark that basically goes over the wish list of what maybe the advanced users would want in a wallet and the different wallets out there that are available to you and I think from my read Proto was the biggest fan of Rabbi so uh rabie he he tested here rabby desktop um uh and he was the biggest fan of that and I've tested it out as well and honestly I'm I'm a pretty big fan of of of rabby too I think it's a it's a great wallet there but different wallets have different purposes right I've always said that metam mask is like that all reliable

where you know what going to do what it's going to do but it hasn't got all the fancy bells and whistles it also is lacking in in in certain security features and pop-ups and warnings and stuff like that uh and then there are a bunch of wallets like that basically and then there is the kind of more recent wallets that have come out that you may not trust as much because they haven't been around for very long uh but Al but that they do have all those extra features they have those warnings those bells and whistles something like rabby right for example but I just figured it was worth highlighting this uh tweet because it's really great and and the wish list is very comprehensive and you can actually see what it looks like for an advanced user like all of this sorts of stuff is definitely Advanced user

stuff because Proto is a I mean he was an ethereum core developer and now he's a core developer at optimism so he definitely is an advanced user of of ethereum here but still I think it it's definitely something that you should give a read to so I'll link it in the YouTube description below uh for you to check out all right on to the last rant of today's episode and this will round out today's episode it's basically all about layer 2 so I came across this really great tweet from Jesse eckle which basically Echoes a lot of what I've been saying over the past last few years both on the refuel in my newsletter on Twitter but he starts off the Tweet by saying there's a lot of criticism going around regarding ethereum having too many layer 2os this criticism misses the entire point of layer 2os the argument

for a layer two scaling strategy is an argument for the free Market's ability to solve problems efficiently it's ethereum's App Store moment where innovators and entrepreneurs each get to build their vision for ethereum's future and let the market decide and then he lists out a bunch of different l2s and what their Vision looks like and this is is bang on guys like I probably couldn't have said this better myself to be honest this is exactly what ethereum's rollup Centric or L2 Centric road map is all about we pushed out all of the scaling Innovation and it's not just scaling it goes beyond scaling it's like privacy Innovation uh innovating with different um virtual machines and different programming languages but we pushed all of that out to the free market because essentially what had been happening for years prior to that is that it was pushed out to layer other

layer ones obviously because ethereum layer one is slow to move hard to change we had this kind kind of a massive amount of layer ones that would just get spun up and they had all these other bells and whistles and everyone was like you know this is going to kill ethereum blah blah blah whatever so really it was just like okay well well if this is going to be the case this is what this if this is what the free market does let's just bring that home to ethereum let's just do it so that layer one is the settlement layer and layer twos become the new alternative layer ones where essentially you can experiment to your heart's content and you still get to inherit the um good properties of ethereum right in a properly constructed rollup you still get to inherit the security uh the decentralization you

still get to have a best-in-class bridge all that good stuff there and you also get to have um the the easier bridging for users to the l2s and then ethereum as an ecosystem ultimately benefits from this and no one's calling a layer 2 an ethereum killer which is also great from a social layer perspective um but it really is about that free market it's about that bizarre model where essentially everyone gets to compete fairly as a layer two everyone gets access to the same ethereum layer one block space like no one is uh is kind of kind of favored right because it's a permissionless platform and we get to bring all that Innovation uh to ethereum without having to I guess put up with the fact that layer 1 moves very very slowly because if you remember the last hard Fork was in April for layer 1

ethereum and that was only to enable uh big and chain withdrawals really we pushed everything else out and the next hard Fork is looking like March so almost a year after the previous one obviously to bring big a big upgrade to the network in EIP 4844 but that is way too slow to move if you want to innovate on things like scaling things like privacy things like other virtual machines whereas at layer 2os they can move as fast as they want yes of course they're more centralized but so were all these other l1s out there right they were they were centralized they were easily upgradeable but that's the whole point when you're trying to move fast and innovate and and basically compete as part of a of a free market you need to be able to do that whereas ethereum is not competing on that front

anymore ethereum layer one is basically competing to be the best cament layer in the world which I think it is right now the best cament layer in the world or I guess like sent layer for for blockchains right now and and it continues getting better over time and it's and it's being optimized for that but when it comes to scaling and privacy and the other things layer 2 is are where it's at and as Jesse highlighted here each layer two has their own Vision optimism has the Super chain Vision arbitrum has the universe of chain chains Vision with a focus on app specific chains polygon has a vision of unified liquidity using their ZK Tech similar super chain to optimism eclipse is doing um you know L's svm as a layer 2 right uh blast L2 he says yeah it's mostly like a cash grab and it doesn't

have anything to do with l2s I'm actually going to rant about blast L2 uh in a second um but yeah I just thought that was a great tweet highlighting exactly what ethereum's L2 or rollup Centric road map is and why we went down uh that path there now I do have uh some L2 stats that I wanted to go over that I'm going to say for the end of the episode I did want to quickly segue into the last rant from of today's episode which was about uh the blast L2 now I say blast L2 like it actually means anything it doesn't I tweeted out today if it's not obvious by now blast currently has nothing to do with l2s people deposited funds to farm an airdrop that's it anyone can set up a multi- promise an airdrop and then say there will be a product later such as an

L2 a defi app a social app or whatever else it is and we've seen this in other areas too I mean as much as I love Igan layout right now it's literally just a contract that you deposit funds into they're doing that strategic on purpose in order to bootstrap the network right because they need to bootstrap the amount of uh e that's restak with them in order to bootstrap their products but at the end of the day it's the same thing the vast majority of people are depositing into there because they expect there to be an airdrop I mean I hasn't sent anything about an airdrop as far as I know but they expect there to be one whereas with blast they just basically took it to the logical conclusion of okay let's just tell people that there's going to be an airdrop let's just Market it as like

there's going to be an L2 because that's like the buzzword of the day and then people deposited whatever it is I think it's almost half a billion dollars now into the contract which is just a multi-seed contract it's got nothing to do with an L2 there is no Bridge there is no L2 there is no us usable thing the funds are locked and it's just like any other multisig like I could set up a multis right now right I could set up a 405 multi 305 whatever it is I could make myself all the signers for it or I could get other people to do it with me I could say to people hey deposit funds into this and uh there'll be a token at some point in the future and uh there'll like a def app that I'm building on top of it and you know maybe given that I

have a bit of a reach in the ecosystem I could probably get some funds into it right I would never do this of course obviously like can't think of anything more grifty to be honest to do to do something like this um but I wouldn't do this but I could and anyone can do this it's a permissionless platform no one can be stopped from doing this but it's got nothing to do with l2s right it's got nothing to do with the defi app it's nothing to do with a social app all it's got to do with is people farming an airdrop that's literally it people are just trying to make money and this applies to literally everything so that I mean this isn't going to be a long rant it's like just a mini rant there to continue on what I said yes uh not

yesterday Saturday about blast and I and I really it really irks me that people are are making this mistake here because I even had someone reply to my tweet and said you know this is true but that being said the L2 road map sets this up to happen again and again and again but again that's got nothing to do with the L2 R map it's got nothing to do with l2s I I replied and I said you missed the point of my tweet this has nothing to do with l2s anyone can set up a multi and Market the fact that that there will be any product later with a token attached and people will Farm it this is just the consequence of a permissionless platform so yeah that was the whole point of this tweet to basically say to people that L2 or not this is just people farming an

airdrop they don't care if it's an L2 all they want is the token and when blast comes out and says that there's going to be a token right and they have all these in uh this this marketing around that and they're very aggressive with it not only with the marketing but but but but with their investors obviously backed by Paradigm right and all of the the the influencers on Twitter are Shilling it and the referral links and stuff like that right of course people are going to deposit funds into this because they're trying to make money I like an that to people just buying a coin like if someone shiing a coin or if a coin's going up people are buying it it's just very hard to see how much money is actually flowing into that coin you can see it going up but you

can't quantify the amount of money going into that coin there's plenty of coins out there where people uh buy you know half a billion dollars worth over time uh because it's going up or because they they think it's going to go up or or or something something other but we don't look at that because it's very hard to track that whereas with a multi you can just see how much money went into that multi as I said it's locked as well so it's just sitting there it's in I mean it's not technically just sitting there the eth it's it's being staked but I mean it's just sitting there as staked eth as as as as St eth uh and then eventually the people who deposited they're expecting to get an airdrop out of it so yeah anyway going to end the rant there because I wanted to go over

some of the actual L2 stats that I um that I've been looking at lately but uh moving on so L2 beat and grow the pie. XYZ are the my two favorite websites to keep track of what's happening in the L2 space you guys know L2 beat very well by now I've talked about it for a very long time now but essentially we're at alltime highs in terms of tvl tvl is not my favorite metric for a number of different reasons uh not just for l2s but like for literally anything in crypto but if you look at kind of what it looks like right now we've been up only for quite a while I mean since the start of the Year basically um and TV is obviously affected a lot by by um the asset prices and obviously as asset prices go up the tvl will go up but it

also shows that uh you know people aren't really withdrawing either and and and and people are depositing and and these and deposits are accelerating here for different l2s out there so that's the TBL uh then you can go to activity which is actually a more accurate measure not the best still I I don't think TPS is like the best measure out there but it's better than tvl I think and you can see that TPS wise it's it's actually been been pretty pretty standard and at about like what 50ish TPS for quite a while now across all of the l2s that L2 beat tracks but then it's been spiking recently and there's been more kind of activity happening here and as I've mentioned before not all TPS is created equal across different l2s and what um L2 be tries to do is try to normalize this uh and basically try to make it so

that it's equal but it's not an exact science but still it's cool to see that both tvl activity and a bunch of other metrics are are are trending up still for l2s even though I mean the Market's heating up but as I said we're in that transition period between a crab and a and a bull market uh it's definitely not uh you know full bull market or anything like that yet but it's going to be interesting to see what that looks like when the market does uh heat up a lot from from here uh now that's L2 beat and what they have now we can go to grow the p. XYZ and see a bunch of different metrics so they have way more metrics than L2 beat has um which you can go check out a link in the YouTube description below but in terms of the

layer 2 user base which is basically just tracking the number of addresses um not actual users because we don't know obviously um you know one user can have multiple addresses you can see that that's at all-time high right now now um and it's been at alltime high for a couple of weeks now maybe two 3 weeks now uh and you can see users per chain you can see what chain these users are on right now arbitrum uh and ZK sync error are are leading um and then you can go to kind of like a percentage breakdown as well of this so this is just the basic kind of addresses uh active addresses on on these different kind of chains here then there's other metrics out here out of here as well such as on the left here transaction count which I believe is also probably

at uh on on kind of like uh uh approaching alltime highs and then you can go to one of my favorite one which is the fees paid to layer one by the different layer twos out there I love I love monitoring this because you can actually see what these layer twos are actually paying to use ethereum and obviously most of this is burnt because of the IP 1559 mechanism it's not at all-time highs yet it was much higher during I guess like May of this year and this will also change dramatically post EIP 4844 because of the fact that essentially layer 2s get a fee Market that's reset to one gay essentially it's not going to be at the 33 gray that we see right now uh because L will no longer be sharing the fee Market with uh uh with the rest of the the network

there but still that's that's pretty cool to see and you can see like fees paid by users transaction costs different anals of block space there's so much here guys I mean I'll link this in the YouTube description for you guys to check out but really the tldr of all of that is that layer 2s continue to grow they continue to thrive uh and their future looks very very bright uh and yeah I can't wait to see what this looks like both post 4844 because are obviously going to be a lot cheaper more users and uh will come to the chains there'll be more apps built on those chains to attract those users and then as the market generally heats up obviously more and more people are going on board here and the network effect just continues to grow but I think on that positive note after raning for most

of today's episode about various things that's going to be it for today so thank you everyone for listening and watching be sure to subscribe to the channel if you haven't yet give away a thumbs up subscribe to newsl join the Discord Channel and I'll catch you all tomorrow thanks everyone