210 - Endgame 2.0: A Guide to Vitalik’s Ethereum Roadmap with Mike & Dom

for me I think the most important feature is censorship resistance in my mind being the ecosystem that prioritizes censorship resistance above all else is the most important feature because it's the most differentiated compared to web 2 in Canada your property rights you don't think about them until during the protest the trucker's bank accounts are turned off everything seems fine until it's very not fine and I think ethereum prioritizing censorship resistance above all else is is the Right Way Forward welcome to bank where we explore

the frontier of Internet money and internet Finance this is Ryan Sean Adams I'm here with David Hoffman and we're here to help you become more begas what is ethereum's endgame one of the most important computers Humanity has ever made David and I have argued before is ethereum this is a property rights system for the internet so what's happening next what's the road map what is the end game this is the sequel to an episode that we recorded just over two years ago with vitalic it was also entitled endgame and I think this

episode represents the most comprehensive overview of ethereum that's probably ever been recorded this is the next 3 to 5 years of ethereum's future strap in this is a long episode but I think every minute is worth your time towards the end of last year vitalik tweeted out an updated diagram to the ethereum road map this diagram vitalic made a little over two years ago which was the main motivation for endgame the first episode that we did with him and it is just this completely comprehensive overview of the six different Paths of the ethereum road map

all happening in parallel uh it was the main subject of the last episode and since he updated it because you know two years have progressed uh the road map diagram has two years of progress in it he updated it so I took that tweet took it to vitalic and say hey vitalic do you want to update the episode that we did two years ago or do you want to elevate some newer researchers inside of the ethereum foundation and FAL was very excited about that and so he nominated the two guests that you see on the episode today Dom and Mike uh dothy has been on the podcast before our uh me

burn and blob space episodes Mike has also been on the podcast before as my co-host uh for a um Igan layer reaking episode that we did with three ROM so both past guests uh and they return to walk us through some of the deep parts of the ethereum road map Dom and Mike are great they are the David and Orion of the ethereum research ecosystem guys so strap in enjoy this episode we'll get right to it in a minute but before we do we want to thank the sponsors that made this episode possible including the

place you can convert your Fiat to Ether that is Kraken our number one recommended exchange for 2024 bankless Nation I'm super excited to introduce you to both Mike and Dom researchers at the ethereum foundation both Mike and Dom are working on various parts of the ethereum road map both our previous bankless podcast guests and both are overall great guys that we are excited to have on the show here today Mike Dom welcome to back bankless thanks for having us thanks for having me so don we'll start with you uh where what on the ethereum road map have you been focusing on uh Mike is about to uh

answer the same question the r is very very robust so Dom just like introduce your area of focus as it relates to the etherum road map yeah I've been all over the road map in terms of educational content uh mostly The Surge with the the upcoming EAP 4844 otherwise I participated a little bit in the concept of MV burn like intr in the concept and working toward that with Justin beautiful and Mike same question to you where along the road map are you focused on yeah I'm a little bit all over the

place too so mainly have been working in the me space which is called The Scourge in in vitalic road map and yeah just want to kind of thank you guys again for for having us on the show Big Time listener to the show and I actually have the vitalic road map as my phone background so it's been that background as long as I can remember yeah it's been that for like probably two years now so to be hosting the or to be like the guest on the the show for this episode's really a treat so actually part of the story for how this episode came together is it's been a year since our original

episode uh with falic covering over the same thing some updates to the ethereum ro road map have happened and vitalic actually actually nominated you two to take the Reigns over the Guides of the ethereum road map so uh it's a very large responsibility also very high honor well maybe there was just no one else available I I was going to ask Mike if uh if having the the road map is your background if if like that is kind of Haunting for you is that inspiring to you or is that a little bit of like

pressure at all times I think it it's been it for so long that I don't think about it actively anymore you know like once once the you see the image enough but it definitely is nice to be able to like if you're in a conversation with someone be like hey this is on the road map like check it out like we're thinking about this when they ask you why aren't the devs doing something just show them the background of your phone yeah exactly woke up feeling bullish Dom how about you do you have this printed in real life somewhere or I just printed it for this episode but I got it bookmarked I look at it all the

time I refer to it I'm I'm looking forward to the tattoo uh that you guys can sequentially uh like update over time okay guys Denon is upon us uh when this episode goes live it'll be about the week before danoun so this is the very near-term upgrade to the ethereum layer one protocol uh it has some um some updates that are related to much many parts of the road map some prerequisites for future uh upgrades uh let's start there we're going to get into the famous urges which is this diagram that vitalic put together that

now is the one that we're talking about on Mike's uh Mike's phone here uh we're going to get into that one right after this but first since the denune upgrade is so close um let's talk about that what are the most significant eips ethereum Improvement proposals that are inside of denune uh and how will it change ethereum Mike I'll throw this one to you yeah sure I guess probably makes sense to highlight two of the eips that are in this Fork because those two are kind of the most visible in terms of how they how they change the network so the kind of main star of the show is EIP

4844 this is a big part of the the sharding the data sharding road map for scaling ethereum in terms of throughput of data um this has been in the works for a long time and yeah we've kind of successfully hard forked three test Nets up to this point and blob blobs which are these new data um data objects that flow through the network are are going on each of those test Nets and so there's high confidence that we're ready to do it on Main net I think the date was set yesterday for May or sorry for

March 13th so yeah that's it's very exciting very um very timely to record this yeah and I will say that with the content of that uh this topic 4844 blob transactions blob space is a podcast in of itself which we've done with dom uh a number of months ago so uh if you want to like dive down into that part of the road map uh there is a previous bankless podcast it'll be linked in the show notes uh for people to understand blob space it's a very big deal um but that's not the only thing that's in Denon there's also EIP

4788 what is that what does that do it's uh basically adding the beacon rout inside the evm so in layman's term it's having the the evm be aware of what's going on on the consensus layer on the beacon chain and the main benefit for that is relying Less on trust and more having trustless protocols for proving stuff against the beacon chain so you can imagine like something like Lido or rocket pool today they have to have oracles that tells the evm what's going on with their staking balances and stuff

like that but after this EIP is rolled out you can have the contract be guaranteed to have it right about the balances and such so you don't have to to rely on trusted oracles as much that would be the main benefit yeah another another protocol that really benefits from this is anything that has to do with reaking so igen layer kind of depends on some information about the consensus layer right like you have stake there you need to know if people got slashed you need to know if that stake is being withdrawn all of these

kind of data points are are stored in the beacon chain side of things which is the consensus layer and accessing that data from the execution layer is important um for igen lay in particular so that's that is all fully enabled trustless with 4788 so guys that is March 13th so that is uh that is coming right up that is the next ethereum upgrade we call it the hard fork and there are some hard Forks that are uh being scheduled after that but of course it it takes a while to kind of formulate all of the various features that those things will will

ship but uh now let's let's go to the meat of the conversation which I think today is an opportunity for us to zoom out and see where we are in the road map of ethereum the end game of ethereum if you will so we've recorded an episode just over two years ago this came out in January uh 2022 with vitalic butterin uh it was called endgame and for the first time this was on the back of him for the first time putting together the diagram

the diagram we were just talking about that Dom you have printed out and Mike you have as the background of your phone and this is the very first time and if you're watching uh this in video form you'll see it on the screen right now this at least for me I think for many in the ethereum community was the first time we saw it sort of organized and stacked in terms of ethereum's road map for a long-term future and this is essentially the end game of ethereum multi- uh hard Fork multi- year I don't know maybe multi- deade I'm not sure how

long this will all take to uh to play out but we've made made some progress on the last two years and this road map has uh adapted and evolved and at the time when vitalic kind of like put it out there he gave each category of features a name and these are what has become known I think as the six urges all right so if you if you hear us talking about urges that's what we're talking about we're talking about the these horizontal swim Lanes of ethereum functionality that are all grouped and accomplish some

purpose so I think we're going to go through this episode and talk through each of the six urges maybe one by one and have this be the sequel from from our uh endgame episode that we recorded over two years ago but in order to set up the context for this I think Mike and Dom you guys have to explain the highlevel urges because each of them are named differently they all end in urge and they have a specific theme Mike could you kind of take us through the at the highest level what what um what is

kind of the central goal for each of the urges like take us through those one by one yeah for sure I think I'll do the first three and then pass the second three off to Dom so for the merge um I guess I like to think of it as the kind of continuation of proof of stake Evolution for ethereum so the merge itself happened in September of 2022 so this is kind of like a year and a half ago but that doesn't mean that like the proof of stake mechanism that we use is in its final form so figuring out how to

make it most robust and most like long-term and stable is is the key goal of the the topics in the merge The Surge is all about da and Dom is a super deep expert in this too so when we go deeper in he'll definitely be invaluable but the idea here is that ethereum should provide da that the the kind of rollup Centric road map depends on right as rollups scale ethereum that's a huge part of of the VIS for for how we get transactions way cheaper how it's

accessible to to many more people and rollups depend heavily on this data and scaling ethereum in terms of scaling the data itself is is kind of a key piece of the puzzle so so Mike just just to recap so far we've got the merge which is really focused on kind of like consensus and consensus activity needs to happen post merge right it wasn't it didn't just end at the merge and then we also have the surge which you said is Da and that stands for data availability right which is the the second layer of the ethereum stack if if consensus and settlement is kind of the innermost

layer what we're talking about in The Surge is data availability and that is important as well yes yeah absolutely yep and then the the third lane is called The Scourge and the the kind of new idea here is is how do we deal with the long-term implications and um the long-term reality of me right so me is this this hot topic that we've been addressing over the past year and year and a half and figuring out how to avoid

kind of the negative externalities associated with super um sophisticated actors extracting as much as they can from ordering transactions in the ethereum protocol that's the scourge and I think it's called The Scourge because I think it's easy to view these meev actors as kind of bad guys like trying to steal from regular users who are just interacting with the chain I don't I don't know if that like I think it's more of the pneumonic thing that like it's one of the urges but yeah think the scourge is because it kind of came out of nowhere and is impacting the protocol in a really meaningful way and we're

trying to kind of defeat that that boss I guess and and we should say The Surge the one you were talking about right before the scourge right uh the reason it's called The Surge I would imagine Mike is because we're we're surging our ability to support uh bandwidth trustless bandwidth like uh transactions per second is that is that the Genesis here yeah yeah I think so again I think this is mostly a pneumonic I I like to think of it as just like the data the data scaling of ethereum and surge like surging forward of of data into the into

the block space all right so we've got the merge we've got the surge we've got the scourge uh take us to the high level for the last three swim Lanes here da so first we got the Verge which the main thing is veral trees but also the the unifying theme is that verifying blocks should be easy even though we got sophisticated actors due to me which we've kind of accepted as being an unavoidable reality of the endgame so we make it very easy to verify blocks with things like veral trees and leveraging

zero knowledge Texs like Starks and snarks and basically the whole point of The Verge is we have super light clients as that are as trustless as full nodes are today and very easy to run uh either on your phones or has just and said on a a previous banket podcast is you can verify the whole chain on your smartwatch once we have all the zero knowledge magic into the blockchain so Verge for veral Tre yeah basically the main item is verle trees I think Burge

came from from that carry us to the next one Dom what comes after this the Burge after that it's to simplify the protocol as much as possible by eliminating technical debt mainly via history expiry so that the cl clients can be much more simpler and easy to code and understand by not having the technical Debs from previous hard Forks once we have like Fast syncing and stuff like that and we'll come into more technical details later on and the last item is the

Splurge which is just miscellaneous everything that's like good to have but doesn't really fit in any of the other urges go in there so like tweaking EIP 1559 so that it's a better algorithm for pricing the layer one evm and stuff like account abstractions and all the other deep cryptography stuff that we have that we are currently researching but not ready to implement that's going to go in there as well so when I look at this um graphic at a high level and just look at all the urges I actually kind of

see the first four as adding new features adding properties adding capabilities and then the fifth which is The Purge is actually removing technical debt removing things cleaning cleaning up the fridge uh and then the the last one the sixth is kind of like a miss category is that a fair categorization of everything yes and then one thing I think is actually worth pointing out about all of these swim lanes that Ryan's um called them is that they are all happening in parallel I think intuitively as people are looking at

this diagram they are thinking okay first merge then surge then Scourge and it's more like all of these things are more or less happening in parallel like actually in the Splurge the very last one the miscellaneous is EIP 1559 which happened you know centuries ago uh and so like all of these things are happening in tandem in parallel these are all progressing forward um uh both in uh both with relationships with others there are some cross threading going on but mainly independently they're all progressing forward we should ask vitalic to make the next one

in a different order just just keep everyone on their toes yeah I another thought I have as we get into this and and sort of explain the the urges too is um this is not basically the EF doing all of the development here right I mean so what we're talking about this this road map is the entire ecosystem of ethereum core developers client implementers even like like rollups in layer twos are getting involved C can you just go over like broad Strokes who are the participants

actually doing the work on all of the urges and making U ethereum happen because just the EF is just a kind of a tiny you know fraction of those almost like the the the coordinator of all of this activity the activity happens well beyond the bounds of the EF I would call that the entire ecosystem is bringing this road map uh to to to life here so how how would you um describe who's working on these urges yeah that's that's a really good point Ryan and I'm really happy you brought it up because I think one of the most unique things about ethereum is just like how

decentralized both the the road map work and kind of governance over the whole process is you know like there's this allcore devs call that's led by Tim and Danny these are members of the EF but you know all these cevs are are contributing on different eips and like they're the ones actually building the the software they're kind of in the weeds in the technical details so like so we can talk about it you know what I mean so like there's so much value being created and added by like people across the ecosystem that that's so much a team sport and I think just the scope of the

road map shows you that like no single team could could do that all on its own you know like this we're building something that is like much bigger than than what a single team could could ch ship quickly so it's kind of weird and cool like anybody can get involved too right it's just being built out in the open like in kind of a completely permissionless way so like the destination is somewhat known but it's not entirely clear how we get there and everyone kind of like brings their part to the table I mean I don't know that I've I guess to be fair I haven't been

involved in other large open source communities like I I haven't um you know seen the Genesis of Linux as a project kind of like being born and bir into ex existence but this is just not how commercial software development works like a company like Facebook does not build software out in the open to this extent I don't know if you've ever seen anything like this or you you just now you kind of like take it for for granted or this is kind of like the way ethereum is built and that's like a lot of the experience here but does this strike you

as as weird odd cool in any way it's beautiful what makes a beautiful do just everyone just kind of shows up and works on what they want and the pivot to the modular blockchain is very interesting because then we have like these little parts that are connected Loosely but all come together to make a like a beautiful blockchain ecosystem in all these various aspects and you can see like all the developers on the all core they have calls that all argue with each other

about priorities and what should be done and like how it benefits one team and like stuff like that thing there's almost like something quas organic about it right it almost feels like sort of a a life form uh in a weird way I mean like we don't have to get um to too too deep on this metaphor in the episode but just the way it's growing is uh incredibly cool and uh yeah beautiful is a great word for it Dom I I want to ask another question about this before we get into the individual urges in this what's the actual point like so I think

I'm trying to recall the episode I haven't listened to Al endgame episode in a while but uh there are some core values as to the why we are accomplishing all of these different urges and I'm wondering how you would articulate those core values is it about decentralization is it about censorship resistance like what are all of these urges actually working towards in the in in kind of like the fullness of this road map yeah I mean it's it's a good question and I think after we go through

the urges we'll kind of take another step back and and hopefully like summarize clearly what the what the long-term vision of the of the ethereum ecosystem is from from our perspective but yeah for me I think the most important um the most important feature is sensorship resistance right like in my mind being the ecosystem that prioritizes censorship resistance above all else is the most important feature because it's the most differentiated compared to web 2 right like we anything that's centralized is like by definition

gate kept by some controlling entity and I actually think think of censorship resistance as almost like this Black Swan thing right like it doesn't matter and and to kind of just paraphrase that it's it's like censorship resistance isn't important until it's the most important thing right so you know in in Canada your property rights are like kind of you don't think about them until during the protest the the trucker's bank accounts are turned off you know it's like everything seems fine until

it's very not fine and I think ethereum prioritizing censorship resistance above all else is is the Right Way Forward MH ethereum has always taken the longest term time Horizons uh to its advantage to its benefit to its detriment as well uh sometimes that leaves room for for other people to to cut corners and move faster but this has always been something that's attracted me to ethereum is thinking about things that their logical conclusion and working backwards from there um let's dive into the first urge the merge and the merge itself the event is actually in the rear

view mirror and so this thing is starting to like wrap up for the most part still some things to check off the Box uh just to reiterate the merge is all about very robust proof of stake consensus while preserving the abilities of solo staking by all means um but Mike maybe you can kind of uh update us on where we are in the merge how much complete are we and what are the boxes left to check in the merge how do how do we get this thing to 100% complete yeah so I think the the first thing to

highlight is this idea of finality right so the way that the current ethereum consensus layer works is there's kind of this two-phase process by which blocks are produced and then they're finalized by this other thing which we call a finality Gadget and you know the the reason for this is is kind of historical so I won't get into the details but the idea here of the fact that a block can be produced but not finalized kind of is is a weird intermediate state that um

that is kind of non ideal so the endgame in terms of a proof of stake system is this thing called single slot finality which means that instead of the blocks being in this weird kind of produced but unfinalized State the minute they're produced they get finalized in in the base case and this is this is especially important because you know settlement assurances are what the blockchain is trying to provide and when a block is is produced but not finalized there's a chance that it could end up not in the chain right so your transaction could be

on the chain and then what we call a reorg Could Happen meaning that block gets removed from the canonical chain and suddenly that um that settlement assurance that that was from provided by the blockchain for that transaction is like no longer there so yeah single slot finality is like the key endgame goal here and that's that's a big part of what all of these kind of sub projects within the the the the merge are contributing to famously there was this non- finality event um not terribly long ago 3 four months ago maybe longer six

months ago um and this this happened because we don't have single slot finality correct can you explain explain a little bit more of like the mechanism of how single slot finality actually works yeah sure so the the kind of notion of finality that we use today is this idea that a block will be considered finalized if more than 2/3 of the stake in the consensus layer votes for that block to be on the Chain right um the the reason why blocks take a long

time to get finalized right now is because we have this kind of second time Horizon which is called Epoch there's 32 slots in an Epoch and um the the validators a slot is a block a slot is the opportunity for a block to opportunity for a block okay and is usually filled by a block sometimes people miss their block proposal and so a block doesn't go there but there could have been exactly yep and so the way it's it's divided up now is you take all of the stake and all of the validators and

you split them evenly over the 32 slots so like as a solo Staker I only am sending a vote once per Epoch rather than once per slot so that means it takes all 32 of these Epoch to collect enough stake up to get getting to that 2/3 threshold for us to be able to call a block finalized so that's kind of this two-phase thing and that's why it's um not instant finality in the way that other um blockchains might claim that how does this impact uh like just the

topology of the network maybe not topology but is there increased messaging uh increased bandwidth usage why can't why can't we have this right now what what is the constraint or is it just more of just an engineering thing yeah super good question because this leads directly into the the next Point here which is this whole um max effective balance thing um and max effective balance it's a very technical topic so I don't think it's it's too important to get into the weeds but the the key detail here is that it results in there being a very large number of

validators in the network so right now we're at 925,000 validators in the network each of those validators has 32 eth Stak but the reason there's so many is because that 32 eth is the hard cap for each validator to have you cannot have more than 2 e inside of a validator you also can't have less you have 32 yeah exactly and what this results in is basically coinbase who runs like 15% of the ethereum network by in terms of like eth denominated units they also

have to run like 100,000 plus validators even though those validators are all kind of controlled by a single entity so it's almost of likely of single computer right so like I don't want people to think that there's different computers there's 32,000 like computers all over the world there's probably just a small handful of computers that are running many many instances of validators correct yeah exactly so kind of the the reason why we can't just do single slot finality today is because 900,000 validators have to

would have to cast a vote during each slot so instead of throughout the course of the epoch you'd have to do that within one slot right now that's a 12C time window and verifying those signatures just is not possible in that time in that 12 seconds right right like you have to kind of go through do the cryptography check that they they voted for this block and check that that signature is valid and that whole process is not possible given the current size of the validator set which is why we can't just do single slot finality as is today because we already

have the algorithm we want to to implement it's just the the number of validators that's the limiting factor okay and uh this is because there just too many messages going around because too many validators exist even when it's many many validators exist all on the same computer how does max effective balance change this right so max effective balance allows validators to consolidate onto fewer kind of entities from the perspective of the consensus layer right so like as you were describing before coinbase might be running let's let's just use 10,000 for

an easy round number they could be running 10,000 validators on a single computer so this is kind of 320,000 ether of stake all controlled by a single computer that could be repr presented in theory by like a single validator with a single signature so instead of having 10,000 signatures to verify if if the effective balance of that validator could be 320,000 eth then that could be represented by a single signature that has a lot more weight than each of those 10,000 individual signatures right so the idea here is to

um instead of every signature having the same weight and you have multiple multiple signatures for a heavy validator or for someone who has a lot of stake you have a single signature that that is kind of backed up by a lot more eth and that signature kind of carries more weight as a result of this change in the effective balance so this uh upgrade is called max effective balance the max balance is talking about that 32 number we can increase the max effective balance of a validator from 32 to a higher upper bound what is that upper bound or is the proposed upper

bound right now we have it set to 248 eth just because we like powers of two and because it's kind of a intuitively fine number um but yeah there's kind of no there's no real reason you couldn't take it much higher than that okay but Mike is this a this is not a centralizing force because they're already like if you're a coinbase or something you're already running all of your validator instances on the same machine anyway and you still are essentially like one entity here um am I

missing something or does this have some kind of like downside with respect to uh a disadvantage to for solo stakers relative to the the larger whale stakers nope it doesn't change anything yeah it's and that's kind of the beauty of it one one important thing to note here is that the minimum effect like the minimum balance to become a validator would still be 32 e so right now the minimum and the maximum are both the same all we're doing is change is keeping the minimum at 32 and making the maximum much higher so there's it's really an accounting detail and like we don't even