Builder centralization concerns, Layer 2 growth - The Daily Gwei Refuel #459 - Ethereum Updates

Transcript

hello everyone and welcome to another episode of The Daily gray refuel where he kept the latest news in ethereum ecosystem I'm your host Ian Mr Sano and it's the third of October 2022. alright everyone let's get into it so we have an update here about the arrested tornado cash developer a developer Alexa pert save here that I've spoken about in the past now I think Suzuki here did a really good job of explaining exactly what the update is so they say update there are still no charges but they're continuing to hold him and are now apparently planning to auction all of his property while he sits without being charged without with any crime very normal country so that's the I guess like tldr of um of the update here and it's really shitty I mean he still hasn't been charged with any crime at

all uh and his wife is saying that uh pert service property will be seized and auctions such as I believe his car um I'm not sure about his house or anything like that but still again like the insanity Asus said was that he's stuff he's going to get auctioned off and he's still sitting in jail given that they I still haven't charged him and as I mentioned last time the Netherlands can halt someone up to for up to three months or 120 days I believe without charging them but then obviously they have to charge them or else um they have to let them go now I can't remember exactly when the 120 days is is up but uh yeah I mean I think he's already been in there for quite a while now um but uh you know it's kind of like it's scary because they could wait till

like the last second and then just charging with something but what are they going to charge him with like is he actually involved with money laundering is he actually involved with enabling things for uh for for kind of like terrorism financing and things like that I mean I don't think so but we don't know for sure right um but at the same time uh it's still like even if like someone is involved in that stuff you'd think that they would they would charge them with this and you think they wouldn't be able to hold someone in prison for that long without charging them so that alone is is pretty crazy right uh but on top of that as well I would say that I think he Alexi here is just in the wrong place at the wrong time because he is Russian right and obviously the Western Europe does

not have a very favorable view of Russia at the moment for obvious reasons and because Alexei uh was part of the tornado cash Dev team and because tornado cash was uh used by North Korea uh to move money around and because uh North Korea is sanctioned and because Russia is sanctioned as well there's just kind of I guess all these things are being put together and it's not a great place for uh I guess Alexa to be just just based on that at least politically right we all know that a lot of these games are political games it's just not great so that's the latest update on him uh it sucks uh I'm not sure uh whiffy's gonna get out without charge I'm not sure if they're going to hold him the whole time and and charge him at the end we're going to have to

just wait and see what happens here but generally uh not great to see this and I really do think it's bizarre the Netherlands that the Netherlands is able to actually hold us on for this long without lodging them I believe in the US it's much shorter than that uh I think you know most countries actually it's much shorter than that the fact that it's three months is is pretty insane to me all right so obviously there's been a lot of talk about Mev lately uh and especially Mev boost and different relays and censoring versus non-censoring relays and I've talked about it on the refill a bunch of times before we have a new little kind of Sankey diagram from Tony here today that shows Builders to relays so the Builder addresses on the beacon chain and then

the real air that they're using obviously flashbots has a pretty big dominance over the beacon chain uh sorry on any on the me V boost Mev boost relay side right now I think they account for over 80 still uh with a mantra others like blocks route even block Nate even manifold also having some market share but flashbot is definitely by far the largest one and unfortunately the flashbots relay does censor uh transactions and specifically the tornado cache addresses uh and I guess the oh fact ofac sanctioned addresses that are associated with that so I think the reason why there's been so much discussion is because of that fact because of the fact that people look at this and they're like okay well that

means 20 25 of all ethereum blocks are being sorry contain censorship properties within them and that is right right like that at the inclusion level at least and I've spoken about this before where there's two types of censorship there's the inclusion level censorship and then there's the censorship uh when it comes to a testing to blocks and I explained it last week I'm not going to rehash it here uh but it's been a big discussion uh since I've I've talked about it last week it's continuing to be a big discussion I think the funny thing is that some people are using it as a way to show that ethereum's moved the proof of stake just centralized it more and I don't think that's true at all this was happening under proof of work and there were already mining pools dropping or at least censoring tornado cash

transactions so it doesn't matter if it's like or proof of work this is this in this context is strictly an Mev a related issue where these relays exist uh because of the fact that Mev exists and it's profitable to be part of these relays and to use them and to use MAV boost to you know on your validator to just get more uh APR out of it right we all know that uh so there's that uh but in response to this there's been a bunch of I guess talk about how to prevent this and vitalik came out with any research post uh over the weekend titled how much can we constrain Builders without bringing back heavy burdens to proposers which is the tldir is that we want to make it so that Builders uh are constrained in what transactions they

can and cannot include in a block uh and I guess there's been other Solutions like this worked on in the past such as PBS and stuff like that but uh I think the main idea here is that basically as vitalik says here one natural response to the risks of Builder centralization is to try to constrain the power that Builders had so instead of Builders having full Reign to construct the entire block if they win an auction Builders would have a more limited amount of power this power should still be enough to capture almost all MAV that could be captured and it should ideally still be enough to capture other benefits of PBS but it should be weakened to limit opportunities for abuse the this idea is sometimes called partial block auctions instead of auctioning off the right to decide everything in a block auction off

the right to decide some things whether some things could be much more nuanced than for example the Builder chooses the first half of the block and not the second you could give the Builder the right to reorder prepend a pin and you could even constrain the proposer this post gets into some possible ways of doing this and some of the trade-offs that result so there are some ways of this such as inclusion lists which I believe is similar to see our list which has been proposed in the past are proposal suffixes fix the proposal suffixes with pre-commitments and stuff like that and there's a discussion in here that you should give a read to as well I'll link it in the YouTube description but it just goes to show that there is a lot being done around this issue around this issue of censorship and around uh mev's role in

it and and the way we kind of give the responsibility or I guess the power to builders to include whatever transactions they want into our into a block or not want into a block so these are the Technical Solutions but there is also a lot of social Solutions I covered the Mev watch website last week on the refuel that I would consider a social solution which is basically shining a light socially on the issue and saying hey you know flashbots has a dominance here and there's you know promoting censorship on ethereum consider changing over to a different relay you know here are your options so that is one social way of doing it but there's also an outcry from the community generally you know I think uh there's been a bit of talk over the weekend about the fact that if a relay they can't offer

censorship resistance like if it can only offer a censored relay then it shouldn't even exist in the first place and I think that especially when it comes to flashbots people have been saying that well okay they're a us-based company and the reason why they're centering is because of that well then they shouldn't run a relay at all they're not even actually making money from their relayer as far as I'm aware they're running it as public goods so to speak so if they're not making any money from it then why not just shut it down instead of encouraging censorship on ethereum right like there are there is other relays out there they did they did open source the code so that people could build other relays so if they really do care about the I guess public good of ethereum if they care about ethereum as a network and keeping

censorship resistant then they should shut down their relay right that's the argument I'm not saying they should or shouldn't I haven't made up my mind yet I'm erring more on the side of I think it is their responsibility to shut down their relay if they can't offer censorship resistance because I think that's much more important than anything else but I as I said I'm erring on that side it's not like I have this strong stance just yet because there are obviously trade-offs to that the trade-off being that we shouldn't just rely on altruistic actors here because flashbot is only one actor what if another one comes along with a sense8 censored relay and that gets a lot of market share should we ask them to shut it down and then what happens if they say no right which is they entirely can say no it's in their power to do that

well that's why we're working on the Technical Solutions as well because we want to make it very hard if not impossible for uh block Builders to censor things and uh if we do that then we solve a lot of these issues and we also want to eliminate as much mov as possible a lot of smart people think that we can eliminate it or eliminate most of it but it will take some time so in the meantime we have to deal with these things just like we had to deal with client diversity we have to deal with relay diversity as well and just like we had to deal with um uh the I guess like uh social slashing and and talking about that and having hard conversations about that in terms of I guess like proposes building on top of the wrong chain and things like that like not attesting to certain

blocks there's also a chatter about social slashing when it comes to inclusion now I actually don't think that we sh we we should slash validators uh because they're not including certain transactions into blocks uh just just because of the fact that we've given them that power as part of the protocol so it's an in protocol power that we've given them and also if it's not degrading the uh user experience for most of the network then I think that the Damage Done from a social slashing is not worth it so for example when would I be in favor of social slashing on the inclusion layout well at the inclusion level well it would probably have to be something like censoring uni swap for example so censoring a popular app on ethereum let's say the majority of the network decides to not include

uni swap transactions in their blocks at all for whatever reason well at that point I would be can I would consider or at least I'd be in favor of a social slashing of those validators because they are materially impacting the user experience and materially damaging the ethereum networks are a credible neutrality right and censorship resistant properties but when it comes to a I guess like something like tornado cache which is used by a small subset of people and the and the thing is it's at the moment it's only 25 that is engaging in this so instead you know even if it was 50 you'd still be able to get your transactions in within 24 seconds instead of 12 seconds I don't think the damage or the collateral damage that would that would be incurred from a social slashing is worth it in that

context so it's always a context thing uh and that's why I think that it's a very nuanced discussion it's not just a black or white if we see any censorship at all we need to slash these validators no and when it comes to not attesting to blocks well if it's just a minority if it's only like a few validators well they're just going to kick themselves off the chain they're going to walk themselves onto an alternative chain and they're going to lose rewards anyway but when it becomes I believe the threshold is 33 of the network then it's a big issue and then we need to be ready to socially slash them on the canonical chain so that they can't to come back and start doing whatever they want on that chain because they've effectively forked themselves off when they start a test not attesting to certain blocks and

building on top of a different chain so that's a different scenario that is a scenario that's actually materially impacting the chain for for everyone right uh whereas inclusion censorship is a lot more nuanced a lot more murky and definitely requires both social and Technical Solutions to to fight off but thankfully those are both in the works uh and in the works in a big way and have been for a while this isn't anything new we've been talking about this stuff for a long time now so it's not like we've suddenly been like oh wow okay we didn't foresee build decentralization in ethereum I mean yeah we did we foresaw it years ago and and the issues with Mev were also foreseen years ago so anyone telling you that we're only cluing into this now is completely wrong you can literally find

plenty of literature out there of both Mev and Builder centralization especially around Mev being talked about for a very very long time uh you know years I mean two three four maybe even five years I think so obviously not anything new there but that's a research post here from vitalik you can go take a look at if you want to learn more about what he's proposed here all right so uh ashine he had a great thread around app chains over the weekend so he said app chain takes a hot again let's recap app specific L1 is high issuance for low security no automatic City hard to be most decentralized as Pacific L2 roll up low issuance gets high security

crosswallup toxicity Atomic City through the L1 piggybacks L1 decentralization and he continues this thread I'll link a YouTube description it's only a short thread but basically this is what I've been saying and what I said last week where I think that an app specific L1 a sovereign L1 that isn't as specific things uh kind of what like Cosmos is trying to do what you need to do for that is you need to have your own token which you're issuing for security so you're issuing it to pay validators to keep the network alive right to keep the network going and in order to do this you need to dilute the holders of the token then essentially the cost is born on on the holders because and the bias of that token because they're the one paying off for

the issuance right uh and you're not going to get the security of ethereum nowhere near the security of ethereum you won't even approach like one one thousandth of the security of ethereum that you you'll basically get an extremely tiny portion from the validators that you're able to spin up unless you can somehow get that that token uh to be a very high value but even then it depends on what consensus mechanism you're using it depends on the distribution it depends on a number of different things right um and as as ashine says here no automation Thomas City I'm not going to say this word right I I keep screwing it up but uh basically what he's saying is that you can't do uh I guess across um L1 transactions in such a way where the user experience isn't degraded and

it gives an example in each thread which I'll read in a sec uh and then hard to be most decentralized I mean again again you're not going to ever be as decentralized as ethereum it's going to be pretty much impossible for you to get to that point or even close to that point then we can look at our specific l2s slash Roll-Ups well low issuance gets high security and even in this kind of frame here you don't even need any any issuance all you need to do is pay the L1 security fees or the L1 gas fees I should say to secure yourself with ethereum right obviously there are other things like sequencer decentralization and being able to be you know economics is economically sustainable but the costs are vastly lower than they are for an app specific L1 you do get that cross roll up transactions through the L1 and

you do piggyback off L1 decentralization so the example that uh uh ashin gave here was that uni swap could run their own secret sorry he said in his thread uh blockchains are great because you can Outsource your application state to something more reliable than yourself and you can offload the cost of these resources to the consumer transparently rather than paying an AWS bill that will explode if your website goes viral so unisop could run their own sequencer that batches trades and redistributes the Mev building decentralizing and maintaining uni Chan uni chain on the other hand is way more work and risk for Less return in conclusion app Roll-Ups tick app change X so I'm sure you've probably heard discussion around uh something like a uni chain like if

universe will still spin up their own blockchain would it be their own L1 or would it be a roll-up and the benefits for for doing that the big benefit is redistributing or capturing the Mev that the ethereum layer 1 captures right now instead of L1 capturing it would be I guess uni chain capturing it but it would there would still be better off doing their own rollout rather than their own L1 for a number of different reasons like I I think that there probably will be a uni chain right that makes a lot of sense but it doesn't make any sense to me for it to be its own Sovereign L1 especially around those security concerns and especially around if the goal is to I guess increase the value of uni Swap and then by uh uh by extension increase the value of uni well

then you're not going to do that if you're diluting the value of uni by having to pay it out to people to validate and secure the network so if you do it as a roll-up your costs are considerably lower and the more of the value flows to token or at least if that is the goal right so I thought this was just a really great thread here which polymer actually expanded on in their first blog post about Roll-Ups in in a while here since May uh which you can give a read I'll link in the YouTube description below but it basically expands on what ashine said in his tweet thread which I think was a great read it's not too long you can go give it a read but it gives an example it gives a number of examples around this a number of uh I guess like pros and and I guess

list the cons as well and and and more details on everything that I was just talking about so highly recommend giving that a read as well but last thing I wanted to say about this as well is that I talked I spoke about uh last week about stacker Network and how they're building a roll-up SDK so you could easily spin up your own Rob and then I said that Cosmos has had this for a while for their app chains where they had this Cosmos SDK that people like using because it's very easy to use so there is obviously demand to spin up your own chain here as an app but I think that the problem it has become is that people may look at roll ups and be like okay well how can I become my own roll-up right how can I have my own app

chain as a roll up and they look at the existing Solutions of which there really isn't any right there is here and there I mean polygon has their own SDK but it's it's become more of like a white label service and it's early days uh with that as well uh and obviously stacker networks early days too but once we have these things spun up and they're actually able to be used by the masses so to speak or by the masses of developers so to speak I think we're going to see more of this stuff happening and I do truly believe that the roll-up model he's just vastly superior to the app chain L1 model it just it makes so much sense economically socially uh security wise like technically I mean socially is a big one too because if you are roll up on ethereum you inherit all of

ethereum's uh community and network effect but if you're your own L1 while you don't really inherit that you're part of a different ecosystem so I think that obviously you know some people argue and say well that's not that big of a deal because uh we don't have that many users blah blah blah maybe it's not a huge deal for users but I think it's a huge deal for developers inheriting the ethereum developer Community is very very powerful not just on the solidity side of things but just the theorem developer community in general is huge you know and I think that that's a huge advantage that Roll-Ups get over these alternative l1s as well so yeah go give this thread a read from ashine and this blog post from Pioneer I'll link them both in the YouTube description below all right so speaking of blog posts you

should go give a read uh nor swap here from optimism has uh published a blog post first uh called Bedrock versus Nitro so basically doing a technical Deep dive on optimism bedrock and arbitrim Nitro which you can find Linked In the YouTube description of course and as I said this is technical uh it's not that technical it's it's high level technical but still like you probably will uh don't don't feel bad if you don't understand everything in here uh but I thought it was a really great breakdown of the differences between the two obviously coming from someone who works at optimism uh so I think he tried to be as unbiased as possible in this blog post and from my first read of it uh it came across that way so kudos to northshop for this

um but uh definitely go check it out I'll link in the neutral description below and I think you're going to be seeing more of this it it's kind of clear that optimism and arbitrim are the leading evm compatible Roll-Ups right now uh it remains to be seen if the ZK VMS can catch up to their leads so far I think that they can but as I've said before both optimism and arbitrim can easily become ZK EVMS in the future and they already have the community they already have the ecosystem so they definitely have a head start here but we'll have to see we'll have to see how it shakes out there's plenty of ZK VMS going live relatively soon which I think are definitely going to make the game a lot more interesting uh and speaking of growth of optimism you can see here that

Michael from the optimism team here shared that the transactions per month on the optimism network is actually at an all-time high again so 5.1 million transactions happened on the optimism network uh during September sorry yeah during September which is pretty cool considering that I believe September to be I guess the depths of the bear market so it's cool to say that optimism is still growing in the bear Market but it's funny when you see this because 5.1 transaction a million transactions for the month ethereum does one what to 1.2 million transactions a day ethereum layer one which is crazy because that would mean that ethereum does over what 40 or 40 million plus transactions uh per month at the higher gas fees right

than these Layer Two so you can see that ethereum layout once Network effect is still so huge its usage is so huge and it's basically capped in the amount of transactions it can do because the gas fees are always going up because the gas usage is at a hundred percent whereas optimism arbitrim a lot of these Layer Two still have plenty of capacity out there for people to use and people to take advantage of that's why and you're going to say that continually steady go steadily go up as more and more of these apps get deployed to these networks as newer apps come online and as I guess like the market heats up again that's a big thing the market heading up again especially as the ecosystem exists today but it's not a strict prerequisite I think you can see here that these things

can still grow in a bear Market they can actually onboard the existing users into their networks and leverage that until more and more new users come in I mean right now obviously we don't really expect many new users to be coming in we are truly really in the depths of the bear Market it's actually hilarious when I go on Twitter now like it's so bare I mean there's the same people every day I'm not seeing anyone new on there which is fine like I actually enjoy these periods right but uh you can you can kind of feel that like compared to like even six months ago it's just totally totally different which is which is kind of funny when you think about it because even if you go back to last year everyone was like that's a super cycle you know we're going to this this and

this we're going to do this this and this every you know we have mass adoption blah blah uh and I at times uh I I think I don't know if fell for it is the correct terminology but I at times believed that because it like the bull market lasted for what felt like forever uh but then obviously it all came crashing down everyone got humbled and went and and the people that are around now are the ones who actually care about the technology the builders are still building things are still getting funded which is actually pretty crazy and it's a main difference to 2019 which I've mentioned before this is what the current period feels like and I love the fact that we have layer 2's live and improving all the time because the next time the bull market rolls around yes ethereum R1 fees are going to be high

but people will just be using the l2s you know but we're still going to get people complaining about our one gas fees uh that's always going to be a thing but we're going to be able to say hey you can do everything you do on L1 on these l2s go you know here's how to bridge to them you can use one of these centralized exchanges or you can directly on board and and you know go do all this sorts of stuff especially if by the next bull market we have Proto dank sharding live it'll just be so Dirt Cheap to use these things that it'll be I'll be very surprised if we still see a lot of people complaining about gas fees as much as they did uh last year right uh but I mean they probably will uh there's always a new cop of people that

come in don't understand how things work it's just the cycle of crypto right at the end of the day um but one last thing I want to talk about when it came to scalability here was this great tweet thread from James ho who said evm block space increasingly abundant with L2 scaling optimism and arbitrine each flipping Avalanche in Daily transactions September 22 transactions 5.7 million for arbitrim 5.2 million for optimism and 4.8 million for Avalanche sea change slowly but steadily so when it comes to these evm compatible chains equivalent chains layer twos are always strictly better Avalanche cannot compete the Avalanche C chain I should say cannot compete with optimism and arbitrim on user experience it cannot

compete with optimism and arbitrim on the uh in um on security right because obviously I mean not right now I guess optimism and arbitrim don't have fraud proofs uh uh implemented yet they're not I mean they're still very centralized I'm talking in the end state of things Avalanche in its end state is the same as it kind of like is today where they've got their own consensus mechanism they've got their own security they've got their own validators all that sort of stuff right whereas optimism and arbitrine we know what it looks like when they do we have fraud proofs they get rid of their centralized controls and everything they fully inherit ethereum security and they have the better UI sorry the better user experience the cheaper fees uh because of that uh they're just the the better choice and especially because they inherit the ethereum network effect

which I don't really think Avalanche does beyond maybe the fact that because it's evm it's easier for the developers to get involved with it right is it for users to bridge across but if the choice is between the Avalanche C chain and optimism or arbitrim I think most people are going to pick optimism or arbitrim at the end of the day now at least from my perspective I'm biased of course whatever but that's my kind of perspective and the proof really is in the pudding here arbitrim has zero token incentives optimism has rather light token incentives uh Avalanche had a ton of token incentives but maybe not so much anymore but they had a lot right for quite a while and that's what led especially during the bull market that's what led to a massive spiking users there but the token incentives run out they've got worse user experience than

optimism and arbitrim do they don't have the ethereum community the Avalanche Community is non-existent now do you see anyone talking about Avalanche anymore I don't and that's because they were all just bull market people like they're not here anymore so taking all that together you can see why optimism is an arbor trimmer just clearly in the lead now and will definitely remain in the lead I believe uh and not just them other layer twos as well this is literally just two of the layer twos out of the 2030 that we have these days right and the Plenty more that we're going to have monolithic chains like Avalanche just cannot compete with that right especially when they're the same thing in terms of user uh in terms of kind of like being evm but they're not the same thing in terms of user experience like as I said the

layer twos are always going to be strictly better there and uh they're I think sometimes they're actually cheaper than than the C chain and post-prodang shine there definitely will be while inheriting ethereum security so putting all that together yeah you can kind of like see what I mean here but there's also James also continues this in his thread giving a bunch more stats here which I thought was very interesting you can go give this thread a read I'll um I'll link it in the YouTube description uh below there but yeah I mean just a final note on that bull market people versus bear Market people I'm sure a lot of you have noticed this by now especially if you're a first cycler uh you've noticed that that there's a lot of people that were around in the bull market that just do not exist anymore

like they're just completely gone and it may you know the communities that seemed very vibrant very Lively very I guess like healthy in the bull market I I essentially non-existent now because they were not actual communities they were just people who came opportunistically because they wanted to make some money they were first cyclists they had no understanding of the tech and to them ethereum was a broken expensive mess and Avalanche and all these other l1s were this cheap uh thing that solved all the issues and they could speculate on the token instead of buying eighth and maybe some of them believed because these tokens were cheaper they didn't really understand market cap it was going to lead them to get make more gains and all those sorts of stuff so you can see that now and I've seen this time and time again this

is I mean I I think I said this on the refill at one point where I said that these other networks they were popular last year I just the NEOS the Q terms the trons of 2021 of the of the you know the current cycle and there'll be new ones next I mean already know this new one's next cycle Aptos Suey say I mean they're already the new cop of layer ones that are getting uh bought up by VCS and they're getting ready to Turbo Shield them next bull market uh so yeah I mean I just think it's funny looking back on it how things just I mean history doesn't repeat but it Rhymes I think it's very aptia uh but yeah I'm gonna leave it at that for today on that note so thank you everyone for listening and watching be sure to subscribe to the

channel if you haven't yet give it a thumbs up subscribe to the newsletter join the Discord Channel and I'll catch you all tomorrow thanks everyone