Ethereum staking, Minimum Viable Issuance and more - The Daily Gwei Refuel #669 - Ethereum Updates

Transcript

hello everyone welcome to another episode of the delway refuel where up the latest news in the ethereum ecosystem I'm your host and thiso the 10th of October 2023 all right everyone let's get into it so apologies for not doing an episode yesterday just kind of ran out of time and wasn't able to do one but I'm back today of course and there's a bunch of things to get through mostly all kind of ethereum core Dev related actually which is uh which is pretty cool because it all goes back to the discussions around staking and Lio and all that sorts of stuff obviously the stuff that and I covered on the drive-through episode that came out on the weekend if you listen to that one there that was a pretty good episode we got a lot of great feedback on that but to kick us off today I'm going to start

with a tweet thread here from Christine Kim who updated us on the latest all core devs call so really I guess the the meat of this update is an update on when devet 10 will be launching for denune now it she says here the on the timing for the launch of devet 10 not all developers were confident that the next devet could be launched in the next two weeks week so this is from October 7th so I guess next two weeks being up to October 22nd or or 23rd or so especially given the fact that client teams are still discovering and resolving bugs on devet 99 so this speaks to the larger Trend over the last maybe month or so where I where I basically said to you guys that denune is definitely not going to happen this year now uh it's going to

happen probably early next year and that's simply because testing is taking longer than anticipated because 4844 is quite a big change to the network and uh the testing has has has uncovered a bunch of bugs obviously they need to get squashed but also performance related things as well um because obviously there is uh still a worry amongst the core developers that too many validators were added too quickly on the network and that's why we did that other EIP that's coming in denune the um the I guess like the the churn limit cap right even though that the staking queue right now is is almost empty I don't know if you guys have been paying attention to it but I believe there is uh less than 3,000 validators left on the Queue so yeah pretty much almost empty which is good to see obviously from a Network

Health perspective as well I think that there is definitely diminishing returns to adding more validators at this point um I think that what I would like to see is maybe the same amount of validators but just shuffled around and I'll talk more about that um more about that later when I talk about all the uh the lto stuff but yeah that's the the meat of this update here is that we can probably expect Dev Net 10 maybe hopefully towards the end of the month uh into November and then hopefully we still get that first test net launched in like mid to late November there first test net fork in mid to late November there but really at the end of the day as I said to you guys denune is not happening till probably Jan Feb at this at this stage I'll keep you guys updated on timing around that

but that was the meat of this uh this kind of thread here and Christen Kim gives a bit of bit more information as well which you can check out we'll link it in the YouTube description below for you to do so right obviously the Talk of the Town lately lately being maybe the last couple months especially has been all about staking uh and specifically staking centralization or kind of like Lido's dominance on the ethereum network and there's been been a bunch of podcasts there's been a bunch of debates there been a bunch of blog posts written about this but two that I wanted to focus on in particular were was from vitalik and Mike uh Mike n here now these two are relatively recent posts I mean they just came out over the last few days but I think they really cover all the bases so to speak right and

obviously as I said I did an episode on this with Eric Connor over the weekend where we talked a lot about Lio uh stake centralization you know what the actual issues are how they can be rectified what the you know what the kind of state of play is right now all that sorts of stuff so you haven't listened to that episode yet I highly recommend doing so but the first blog post here is from vitalik and this post is less about I guess the current kind of um what the current staking ecosystem looks like and more about what we can actually change at the protocol in order to improve the ethereum proof of stake protocol now this is titled protocol and staking pool changes that could improve decentralization and reduce a consensus overhead now obviously I highly recommend checking the this blog post out for yourself but it really it fits

into the same same theme as what has been become known as two- tiered staking so uh the TAC defines this as there being a staking model where there are two classes of participants those two participants being note operators and delegators now obviously we already have that within ethereum today you know you have the node operators the solo validators the Rocka pool node operators the LI o node operators and then you have the delegators right the people who delegate their stake to one of these node operators whether it's a exchange or rocker pool or or Lio they're delegating their stake now this is not done in protocol this is done extra protocol right because ethereum is not a delegated proof of stake uh it doesn't have delegated proof of stake as its consensus mechanism it just has this

stuff built on top now of course there was a lot of talk about potentially enshrining this at the base layer but there are trade-offs there so italic come up with this uh it's not like a change that is probably going to go into the protocol but he's come up with a bunch of ideas around how we could actually do a kind of hybrid delegation within ethereum where node operators still have to post cider will still have a risk of of obviously getting slashed and all that but delegators have a much easier time of moving their delegated stake around so I think one idea in here was that basically you wouldn't need to go through the execute if you wanted to redelegation Bringing Down the cost of

those of those switching costs there and then he also compared things in the blog post like basically the cost to attack and this is what I was talking about before where where we're kind of about this diminishing returns where adding more validators isn't really adding more security to the network so when we the the main focus should be on obviously making sure that the node operators are as distributed and decentralized as possible so that no one uh node operator no one kind of like governance umbrella on top of node operators like lier for example has too much power in the nwor here but yeah that's kind of like the high level stuff there and there's much more in in this post but I'll I'll let you go read this post for yourself if you haven't yet but my general thoughts on this uh from what I've read and

comments and that I've read and discussions that I've had with people is that I think people read this and they took this to mean that this is what's actually going to happen in ethereum proof of State this is this is what this these are the changes that are going to be implemented that's not how you should read this this is basically just a blog post that vitalik has put out with a bunch of ideas that people can begin to discuss and begin to to talk about and begin to maybe potentially create something workable that actually goes into the protocol eventually but in no way are any of these ideas in vitalik's uh blog post even being considered right now for inclusion in the protocol they are just ideas and I think that I got a little bit of frustrated reading people's comments because people are

acting like you know this is new this is the new ethereum proof of stake you know this is what we're going to work towards U and they were getting kind of some some people that I know were getting kind of of upset about it because they're like oh is this is this ethereum abandoning its uh you know current proof of stake design for a more I guess like delegated proof of stake design and that's not what it is at all as I said this document alone is is just a bunch of ideas but even the ideas in this document don't propose moving ethereum to to delegated proof of stake like you see on other chains it proposes a new and novel and kind of like hybrid approach to it where essentially we still get the properties of um ethereum proof of stake as as it exists today

while also making we're also removing kind of I guess the security and decentralization uh concerns around the delegation of stake right so anyway you can go give this blog post a read for yourself I'll link in the YouTube description below now fallowing off of this was another post that Mike Nea put out over the week actually he put this out yesterday I believe um or today even um depending where you are in the world but this blog post I think is probably the best thing that you will ever read at least that has been published to date about Lio and about what the actual attack vectors are you guys know that I've been complaining for a little while now that much of the Lio criticism never actually details what the attack vectors are it's just Li's at 32% that's bad you

know we need to get Lio down which is fine like I don't think that's necessarily kind of like a wrong thing to say but for the everyday person they see that and they're like okay why is it bad like what can Lio do you know what's the attack vectors here and Mike nea's post here honestly is amazing and should definitely read it if you haven't yet I'll link it in the YouTube description below but it is very comprehensive and even if you don't want to read the whole post you can scroll to the bottom and there's a tldr at the bottom of what the of what the kind of um five Shades of Gray of Lio look like as Mike calls it here where there are different kind of uh vectors that people think that can be launched as like used as an attack and

then on the right hand side of it uh you know WID lighter would do this and then the reality is in the middle cuz I've always said that where there's a lot of nuance here where people say oh well you know Lio is you know is is only composed of 30 note operators if they collude they can attack the network okay but there are consequences for that and also to get 30 n note operators it's actually going to be 37 soon by the way but to get them to collude um is is really difficult to begin with right so you need to basically qualify these statements you can't just say something and then assume that everyone knows what you mean by it you actually have to qualify and that's why this post is so good because it doesn't just go over node operator attacks it or potential

attacks it goes over a governance attacks as well it goes over things like what the um what the I guess like Optics look like too because this is something that Eric discussed in particular over the weekend that I don't think as as is as important as the other stuff but it's still important where like the Optics are just bad you know when someone says oh well ler controls 32% of of the um the uh the validators on ethereum from an Optics perspective yeah okay that doesn't look great right but if you actually dig into it and dig into the Nuance it's a lot more Nuance than just that but again most people won't do that so from an optic standpoint it it it does lower probably ethereum's uh kind of like credible neutrality and decentralization in some people's eyes

but if you actually read this post from Mike and understand all the Nuance you know that one an attack is really difficult to pull off for l two it doesn't really benefit them in most scenarios right three they they're actually not just one monolithic entity yes the governance is is quite centralized and probably the governance vectors are the um the the the uh the most concerned for attack here are uh but uh the governance itself can't force nde operators to do anything it can put pressure on them but it can't actually technically force them to do anything so that's another bit of nuance there and this is all stuff that Eric and I discussed obviously on the on the podcast as well but Mike just basically puts it into a very very nice blog post you can send to people basically this is why I love blog posts like this because

you can send these to people and you can basically say hey this is the state of L this is what's up with lto you know especially if they ask for I guess more more clarification around it because they probably see people on Twitter and on YouTube saying you know Lido's got this massive share of the network uh you know it's it's bad for ethereum for such and such a reason but they want to dive deeper they want they want to understand okay why is this bad for ethereum what are the attack vectors what are the chances of these attack vectors what is the damage done here um and so on and so forth and even Mike in his post I believe says that really the I guess riskiest thing about Lido's dominance and like the the worst thing about it is the ldo token governance because L the

ldo token has an atrocious distribution it is incredibly centralized which means the governance is incredibly centralized which means that any power that Alo token holders have over Lio is is is basically de facto in my mind centralized power now thankfully right now as I said the token holders do not have the power to force node operators to do anything technically they can force it via social pressure which is a lot harder than actually just forcing it via VIA technical change and there are a few other things that that um that ldo holders can do or ldo governance can do uh but there but there's Nuance off that as well where there is a committed road map from ler whether they deliver on this or not is another thing but there is a committed road map that that

they've put out where they're basically trying to take away power from governance so that governance really just becomes this dumb thing where maybe it just ends up being the thing that polies um the addition and removal of node operators based purely on technical Merit not based on anything except you know is this node operator actually performing or are they shutting down because they've been forced to right so so in the US maybe you get shut you get shut down because you're forced to and you're not allowed to run this product anymore stuff like that right more of a administrative task rather than a how can we leverage our centralized power in order to profit from it kind of thing right so essentially just taking away that that Vector there but anyway I won't bore you anymore with my takes on this you can go give this uh blog post

to read for yourself I'll link it in the YouTube description below for you to do so and actually the last thing I'll say on on this in particular is that commonly what we want to achieve in ethereum and what we want to achieve with basically everything that takes advantage of the ethereum network is a symbiotic relationship right and this is what was so special about EIP 1559 is it created a symbiotic relationship between the applications that use the chain and the eth holders where the eth holders benefit because eth is burnt and the applications that use the chain um pay obviously the users still pay the fees but they also get a better user experience with 1559 and then um most of that Fe fees is burnt which obviously benefits eth holders which are the ones securing the network to begin with and

and providing this infrastructure for people to use the applications so when it comes to ethereum staking when it comes to different entities having different Market chairs we want them to have symbiotic relationships but the problem is is that some of them can get to the point where they're parasites and when they get to the point where they're parasites that's where it becomes a problem for not just the ethereum network itself from a technical standpoint but from a social standpoint and I think a lot of people believe right now that lier is not in a symbiotic relationship with ethereum it is a parasite on ethereum because its stated goals is to grow as big as possible and and that is not a symbiotic relationship because in a symbiotic relationship you don't have the state goal of growing as big as possible you have the stated goal of making sure that

you remain in symbiosis with the uh the thing on the other side basically the the host essentially right whereas a parasite is not concerned with that a parasite just wants to grow at the cost of the host it doesn't want to live in symbiosis wants to basically drain the host for everything that it's got for its own benefit so a lot of people view Lio like that and you know what they've stated that their goal is to grow as big as possible and honestly I can see why people believe that that it's a parasite because of that and I I don't necessarily disagree with that but that's why I've also said in the the past that more competition will actually uh stunt L's growth which it actually has for the last 15 or so months and reduce it too and uh I strongly believe that that will be the case forever

basically I don't think think that social pressure alone is enough social pressure without action is pointless if the social pressure leads to action that is what's most important and and that I think is is exactly what the different Protocols are doing now I mean I have a tweet up here from Super feiz he shared an eth Finance post uh that he made where he basically said that he's uh very committed to this uh vampire attack that that d.va is doing and I've talked about this before about how they're taking in St and then once they once they launch the diva protocol they'll um they'll automatic swap that out for their own LST here but it's not just supporting Diva it's supporting every staking protocol out there that is committed to decentralization obviously you know

right now there are centralized exchanges that have market share and that that have less market share than Lio but there are arguments on that side as well people say well is it better to STI with Lio or is it better to stick with centralized exchange because with the centralized exchange it's literally just them it's not 30 different note operators right it there's not not much Nuance there really it's just them um and and they especially coinbase you know with their with their validators is based in the US which is even worse because what if they're forced to do something malicious right so really we want to just be uh especially the the smaller protocols we want to be encouraging people to use them no matter what they are whether it's Diva or swell or Stak wise or rocker pool of course but any of these ones that are committed

to being decentralized to making sure that they're not uh they're not basically becoming a parasite on ethereum and and want to be in a symbiotic relationship with ethereum is the the stuff that we should definitely be focused on at least in my view here now talking speaking of coinbase and their stake hild Dobby actually was able to unveil that coinbase has a lot more e stake with them than previously thought so rider. network still shows coinbase at around 9% of of market share here but according to Hill Dobby coinbase actually has 14.1% of market share so they have 14.1% of all staked eth on the network now I wasn't exactly surprised by this to be honest considering that Queen Bas has notoriously been hard to track on even

on chain and track their activity on chain but also as I mentioned on the podcast with Eric Queen base right now if you're a US person or even a a European person or anywhere where coinbase is coinbase is the by far the easiest way to stake your eth it's even easier than Lio so their market share does not surprise me at all uh in in in this respect here in saying that I believe coinbase has actually been going down since withdrawals so I think it was up at maybe 16 or 17% factoring in their their market share at time of withdrawals so it's gone down by 23% here which is great to see obviously um because we don't want uh uh any centralized exchanges to have large market shares either I mean I I'm honestly on the the side of I would rather have Lio at probably what they

are right now than have a centralized exchange anywhere near that because as I said before at least with Lio and I know there are still concerns around this and I'm not trying to sugarco anything here but at least with Lio you have the separate node operators right with coinbase you have one node operator that's coinbase they have 14.1% the uh the uh the the largest market share that a lighter not operator has is I think 1.6% I think it's all equal and that actually goes down over time as more not operators are added so if we're talking about the hierarchy of of what's good and what's bad I still consider centralized exchanges the the worst to to stake with uh from a decentralization um perspective right because they are

always going to be one note operator whether it's binance or coinbase or Kraken they are all under one entity they are all under one company whereas with Lio yes there is that governance concern and I'm not trying to minimize that but in terms of the actual node operator set behind it Li is a million a million times better than than these centralized exchanges um and obviously Rocka pool is even better than that because rockol has many more node operators than than even Lio does here but again there's Nuance there because within Rocka poool there is the issue of uh people not wanting to take on RPL uh risk price risk in order to have to spin up note operators and that's being addressed in upgrade sorry in order to spin up mini pools and that's being addressed in some future upgrades for rocker pool and as I've mentioned before

there are other protocols coming online that are that are are not going to require those those sorts of things and they're going to be doing it in various different ways so yeah this is all coming together but I think that outside of L the next biggest concern is obviously centralized exchanges and in my mind a centralized exchange is worse than lighter I know that might not be a popular opinion with people but in my mind a centralized exchange is definitely worse than than Lio in saying that I don't think Lio should be anywhere near the market share they're at right now ideally we want to get Lio down to let's just let's let's say uh in a realistic world we want to get it down to 25% in the next you know 12 to 18 months or something like that because as I've said before it is a slow moving

ship when it comes to ethereum staking it's not going to happen overnight but the fact that we've actually been able to slow or basically stop lios market share growth yes they're still getting more stake in but their actual relative market share growth has has completely stopped over the last 15 months because they're they were at 32% in May of last year and they're at 32% right now so when you look at that okay we've stopped them so let now let's get the engines running on actively taking away their market share and and and making it so that they are down to 25% and then hopefully down to 20% and then hopefully down lower than that but it is an uphill battle it is a long-term battle it's not going to happen overnight but uh we're going to see how that goes there and also what's critically important to this

as well and last thing I'll say on this is that we we don't want laos's market share going down just because centralized exchange market share is going up we ideally want Li's market share going down because these other smaller protocols especially the the the decentralized ones are growing right like Rocka and Diva and all the others that I mentioned before uh and swell and and and stake wise uh and so many others uh we would rather those grow and take away from Lio than coinbase and binance grow and take away from lat kraken's very small right now so I'm not too concerned about about Kraken to be honest the the mo the most concerning is is coinbase but even then like there's still noce that falls off of that right where kbase is definitely very ethereum aligned as an

organization there's no rational reason for them to to attack the network or anything like that or to to leverage their power here um and some people will argue that coinbase actually doesn't even need to leverage their ST eth power they actually have more power because they are I guess like they're not technically the owners of usdc right but they are very closely tied with usdc and people believe that well the stable coins actually have a stronger say than the the stakers do I don't necessarily agree with that but yeah I mean there's also that aspect to it as well there but anyway you can go check out Hill Dobby's tweet Thread about this year good uncovering from him I'll link it in the YouTube description below for you to check out now falling off of all of this Anders allerson here who works um basically in

the economics uh research field at the ethereum foundation here has put together a 118 tweet thread all about ethereum's monetary policy and potential changes we can make in order to basically make proof of stake ethereum as uh secure decentralized and resilient as possible while keeping eth as the base money of ethereum rather than the base money becoming some kind of LST now you don't have to read it on Twitter there's actually a blog post that he did where he basically just took all his tweets and put it in this blog post so of course I'll link this in the YouTube description below for you to check out now because there are so many tweets here it's hard for me to really focus on any one thing that he spoke about but I will focus on the high level which is

what which is the term that he used which is minimum viable issuance now a lot of you may not know this but this is a the's unofficial I guess monetary policy that and this term was coined all the way back in I believe 201 19 I believe ethhub coined this term to be honest uh me and Eric and a bunch of others that were contributing to ethhub at the time and what this basically means is that the ethereum network network should only issue the minimum amount of eth it needs to secure itself basically and and and that's why it's called minimum viable issuance and we actually uh had this policy in proof of work where we changed the block reward from 58 to 38 and then to 28 because the eth price went up obviously and we were overpaying for security by a large margin and then we obviously clapped

collapse the rewards under proof of stake by 90% at time of the merge and this tweet thread that was put together by Anders goes even more into I guess like what the monetary policy can be changed to in order to still adhere to minimum viable issuance but make it fair for everyone now warning here there is a lot of math in this tweet thread there is a lot of there is a lot to unpack in honestly I'm still digesting it I'm still I'm still going through it and reading it back and forth here but as I said the general gist of it is is to make sure that we can achieve a kind of I guess symbiotic relationship between uh issuance of eth uh you know eth staking and eth as a money here so you can go check this out I'll link it in

the YouTube description below for you to do so all right so we have ethereum.org Q4 2023 product road map which is now out on GitHub so I cover this every quarter that they basically put this together but if you are interested in getting involved with the ethereum.org uh initiative here and contributing to it or you can go check out their road map what they're going to be working on in Q4 so here they basically say they are working on redesigning uh their navigation redesigning uh parts of the website here uh they uh they're going to be doing a f wallets design improvements uh as well uh seed phrase page uh denune content updates of course and 2024 Road mapping and budget and much more so you

can go check out this little GitHub uh note here which I'll link in the YouTube description below but uh if you're has a lot of contributors guys like it's actually pretty crazy how many contributors they have uh considering that it may be like a really Insider baseball ethereum thing to contribute to the ethereum.org website right but if you actually go here you can see they have 1,363 contributors on their GitHub now of course not every contributor is going to be as active as every other one but if you click into the contributors section you can actually see which ones are active how active they are um what when you know when they've been active stuff like that and a lot of them have been active different periods of time which is actually pretty health healthy to see because it means they contributed in their own way and that's all they

needed to do they don't need to become full-time contributors right they can contribute in their own way make their impact and then move on which is what the majority of people do of course and then you have like the the top communs up here which of course are probably the the the kind of like what you consider the core team of ethereum.org especially the ones that work out of the ethereum foundation here but yeah that's just amazing to see for for a kind of like open source uh um website here or I guess like an open contribution uh website here for ethereum uh having that many contributors is a site to behold so yeah anyway I'll go I'll link this Q4 R map in the YouTube description below for you to check out all right so zerx designer has a really great little tweet video here

where he's teased something that I've been talking about for a little while so he captioned this with prove you meet the requirements of a rent application without sharing any personal information with zero knowledge and as you can see here this is what it might look like when you're applying for a rental you can show them your proof of funds uh and your credit score but they don't actually know what your proof of funds is and what your credit score is all they know is that um your credit score matches what they want and your proof of funds matches the minimum what they what they want as well uh in terms of kind of uh rent right so this to me is the future of kyc and AML and all this kind of stuff around identity because there should not be a world that we live in

where essentially your identity and all of your personal information is just stored on centralized databases that are constantly hacked constantly leaked where your most sensitive information is leaked right like your your passport details your your home address uh your your kind of like crypto addresses as well so and your email addresses so people can do targeted fishing attacks towards you all that sorts of stuff like it's just insane right so doing this in a privacy preserving way using ZK where you can still have those cryptographic guarantees that no one's fudging the system even though you don't have direct access to their data is exactly what we want to see in this world and this is one of the uh non-financial use cases of crypto that I think is going to have the

biggest worldwide impact it's going to take a while though because obviously this space generally is pretty slow moving because identity is a very sensitive thing um and these are obviously very sensitive details and we have to convince pretty much like potentially millions of different kind of I guess state governments and councils around the world to implement this implement this sorts of stuff but I do think that this is the end game for uh I guess kyc and AML and sharing information and stuff like that because it does it in a privacy preserving way while still guaranteeing that you meet the requirements that has been set for something like a rental application and I'm really really bullish on this technology and I've talked about it a lot before but this little video from zerx designer which I'll link in the YouTube description below shows this

nicely and shows what this could actually look like and shows how simple it can be as well and it can all be secured by face ID right by on your in your iPhone or by um ever Android users so that you basically have this on a secure chip a secure Enclave um and it's all just secured on your own device no need to store this data on centralized databases that are hacked all the time you just need this ZK proof or the zero knowledge proof uh to Pro to prove this to your counterparty and Away you go from there but that's really cool there uh but I think that's going to be it for today everyone 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 newsletter join the Discord Channel and

I'll catch you all tomorrow thanks everyone