eth_multicall Meeting, Dec 30, 2024

Transcript

heyo C said that he's not going to join I guess we don't really have anything if the people who are actually do doing something are not joining yes that does seem like a less than ideal meeting but I guess now we can decide what to do with the gas estimation stuff it's easier to decide when the people are too true I would feel a little bit bad deciding without them though let's see when I wrote the line there then happen to have the time join

anyway oh suddenly I can make it look at that what's your stance on it would you add it to the current method or make the I I guess you want the V tool to have it yeah my stance is um I would like I would like there to be a way to get both the gas estimate and the call results at in the same call I think that is very important beyond that my opinion is get much weaker on the specifics like I don't care a whole lot about the specifics as long as that's achieved yeah I wouldn't want to modify the current back because it it should be

Frozen at some point and I think it should be should be frozen already but then I don't know what the E policy do they are people annoyed if there's V1 and V2 version because I don't those only exist for the signing this message signing I guess and there's huge amount of those I think there's like six ways to sign a message on eum yeah the whole message signing is a mess oh well I think there's somewhat good that there's a new all the time newer version but then still people are using that just signing by hash like I was surprised that fuel worked like that and they wasn't using the newer methods but maybe there was some cryptographical

reason there or something like that I'm not sure I'm guessing not I'm guessing it's just maybe they didn't know any better or since it wasn't even signed with etherum so it wasn't that standard either what do you mean what sign of ethereum there's that c standard that if there's a website that want you to sign the website with with the wallet then there's a standard on how you should do it but they weren't using that you're talking about the prefix or you talking about the hex en coding uh I mean this say the do to that se that's a standard onto how to how do you sign to

website with an ethereum wallet oh you're talking about for signing with ethereum I see yeah and they weren't using that standard yeah I mean they may may just not know that's not uncommon yeah well the all the option since they doing GK stuff and that kind of it might be that that those signatures are somehow harder to do or something like that I'm sure so there could be a good reason to do it but I don't know I would assume a good reason I'm very pessimistic these days but I think those signing methods are probably they were all done by

metamask so it was in a way somewhat easy to coordinate because they are all in the wallet level so one wallet could easily just add a lot of those methods but with this it simul bit harder because all of the clients need to implement it for it to work right but would be fine with that we do the V2 version and then after sometime then we do V3 version but then I'm not sure what the comedy thinks about that do they think that's a good idea or not like it I think it would be way somewhat cool that you always add one feature and then you would call it the new version so then kind of moving from one version to another is somewhat easy to

do yeah I my guess is people will generally be a fan of versioning just because previously we've had no versioning and it's really frustrating for everybody so everybody's suffered the consequences of not doing this um and that kind of helps get people aligned and in agreement with yes we should do some kind of versioning my guess is everybody will have their own opinion on when the right time to version is and all that stuff and then one challenge is since there's no way to check with what what are supported so then we would if there's like four each simulates then we would always need to call the fourth one and then check if that's kind of support

if not then use V3 and if that's not supported then then use V2 and that kind of yeah that's a solvable problem if someone just goes and actually does it basically need capability detection so when you first connect to the wallet one of the things you do is ask the wallet you know what what are you capable of what are your abilities and then once you know the wallet capabilities you can then make decision on how to interact with that wallet some wallets for example like a ledger its capability set is very different than adjacent RPC endpoint that has no wallet at all in fact they don't overlap at all in that case I think that should be a Jason RPC feature and not in a way wallet feature because I think with for

example with those sequences they only support two function calls so it would be nice to be able to call them that what you support and then they would say that you can always send transaction to this end point yeah I mean if it would be cool if it's Jason RPC feature I don't think Jason RPC is going to add it and if they did it would have to be a new version of Jason RPC spe why it has to be new version we couldn't just add add new Jason RPC method that does no we could but then maybe maybe I misunderstood what you're trying suggesting I'm you had speak about wallet so that the wallet would Implement that feature though it would be maybe the wallet name space but I think it should be in the E name space that you could somehow or maybe it could

be in kind of more overlevel it could you could also cury if there's other other methods that just e methods there as well so then wallets would return the wallet methods and the E methods for example but the Chas RPC should also return the methods they are supporting I see I thought you were I misunderstood I thought you were talking about something else um but yes I would I agree that this should be a ethereum name space is the appropriate place for it I don't know if it's appropriate because the wallets you probably want simul well maybe there could be a another method in the for wallet name space as well or should be different method or or or not different method either you add one of these for

each name spaces or you add one that kind of you can cury all the name space the same one it would be one there there' be no reason to do multiple that would just be silly then it's somehow silly that's in E Nam space even though you can Cy all the names space really does well uh right you can do we can do new name space the name spaces aren't um anything special they are literally just how people decide to name methods that's it there's no um they have no other implicit meaning I think in get for example you can dis disable some completely name space so they have that kind of like I

guess regs or something starts with this string and they are disabled or or these are enabled yeah yeah you're right to and then if you add it to news name space then it might be disabled by default well I don't know anyways I think that this has been talked about in the past and metamask is one of the people that would like to do it like they um Dan has pushed for it in the past or at least push forwards maybe the wrong word Dan has um suggested we do something like this and I agree with him so getting Buy in is easy someone just needs to literally do the work of defining how this the capability system

works but if metam really wanted that they haven't done that work mean I think for them it would be the easiest thing to do because they can just make a spec and then Implement themselves and then then that's already most of the community supporting it already because the metamask has it yes I mean that would be ideal um that is not the current situation I think they could easily dominate this kind of thing through yeah they have the power to push I think there's some method but I'm sure it's used or something nobody implemented or something like that I don't remember if there's ever

ANP generated I know we talked about it in the past and everybody's like oh yeah we should do that it's a good idea I think I have seen seen something about this before but maybe it's a forum plus that people discuss but no one actually made an e about it possibly he Lucas we have no actual Engineers here today just just a bunch of people talking about things thanks I I feel a bit engineer still I used to be an engineer once yeah well then I took an arrow to the

knee Lucas are you familiar of method where you can check the capabilities of Json RPC or or wallet I have think I have seen some I or something about but I'm not sure sure uh something that is not actually Implement anywhere but I think there's some spec about it somewhere I remember seeing something like that maybe I don't remember anything to be honest don't know can EIP RPC capabilities but thinking that if we great the is simulate two then it would be nice to have that kind of method that

they could ask if if a CH RPC is supporting one or two or or any of those no idea I feel that's a thing that has discussed everywhere but I found it weird if no one has been pushing it any at all forward uh there were few yeah ideas about standardizing Jason RPC even some initiative some time ago but didn't go anywhere it's hard to find people to do the work it's not exciting work yeah it's the least exciting work

ever and it actually needs a lot of like structure and M client implementation right so the worst worst case scenario I think the frighting the spec is somewhat fine but it's always difficult when you need to get everyone to agree on it and that's the kind of the most annoying part of the work you need to kind of get buying from everyone for it and that's the hardest yeah unless someone uh really takes that into heart and just goes through all the clients and does it in all the clients I don't it will be that

easy that that's very difficult because you need to know so many languages and so many projects and so many coding Styles and you need really a Superman to do that yeah but did did that did happen uh when we had transient stor up codes for example oh someone did it yeah yeah yeah there were initiative I think by Unis swap or someone that's Moody from Unis did you guys did it actually make it or did it never actually gets pushed through well it's on it's life it was it the one one guy or a team that did it yeah it was one one guy Moody um

from Unis swap who just really really wanted it and he went and implemented in a bunch of clients and he put a bunch of work into it and then like he got pushed even after he did all the work he got rejected by cevs like for a year and a half or something he was really for one hard Fork I think it was a couple of hard Forks that he was pushing one or two I think one but I'm not sure or maybe two he he was really frustrated because he's like I literally did all of the work all they have to do is hit go and they will not hit go but the end result is it's uh wrongly priced and not us so it should be repriced did GV need to audit it and

that was the difficult that they didn't want to do or I think the testing was the hardest part I would say uh but yeah I think the mod didn't do the testing not all I'm not sure yeah uh Ser testing yeah uh and the the end result it's it's not priced really well and that's not used so it should be priced but is unop using it no not sure what was the method uh it's up code which up code uh T Lo T

star it's temporal temporary storage a transient storage of course transient yeah transient yeah trans storage mhm uh I'm not sure it's yeah cheaper storage you just use that did the never mind or maybe he he just had someone doing that so solid is not using that I think if it's in the solid no no it is it's just people are not using it in their contracts because it's overpriced oh but is just cheaper than normal storage I think so but still I I don't know yeah Ben uh tried

to uh explain it to me and we even I even tried to make our guys Benchmark it and and propose some repricing but we didn't do that yet uh yeah for uh for us I think he collaborated with someone else to to implement it and never mind because I see it's it was not moody it was someone else that did that PR so yeah [Music] uh and to be honest there was a bug that we uncovered sometime later that was a consensus

is but well it wasn't really cons well it wasn't really consensus issue because it was uh wrong in all the clients no in our client uh I'm not sure it was if you kind of restarted or retried the block it might go through but it was like kind of um kind of bad so that testing is is hard yeah well I'm not saying that to to condemn the implementor because it was hard to to grasp that this bug can cure H which

is very frequent in in the in the nodes yeah uh yeah I guess we don't have anything else I don't have anything since not everybody's here to discuss the guas uh estimation stuff yeah I think we should make a V2 method that has it I otherwise the same than we1 ah V1 V2 I'm fine with doing that in V1 but V2 is also

fine I feel we should freeze the spec at some point and then it's it's always after the next feature we add we freeze it but we never do okay I'm fine with and anything to be honest anything then I think M have a lot of ideas how to make ethereum better no one likes my ideas for how to make ethereum better we like them but we don't like to implement them but one of M's ideas is that you don't Implement anything anymore yeah just aify stop so my least like idea by everybody

how I will make money like earn for the living for a living go build something newv tool I mean if you guys don't have if you if you guys can't figure out what to program just let me know I have a laundry list of things that I would like to build in my lifetime I give you a you know there's the far we go in in in ethereum development there are more and more initiatives more and more Eeps and more and more parallel streams that people are working on so kind of doesn't work like this yes that's my number one problem with ethereum I think

that's a sign of success right um it's the problem is that I mean yes when you're successful you keep people tend to keep doing the thing they theyve gotten that success and you keep doing it and the problem is is for a financial system you don't want a financial system that's changing out from under you at once every six months like you don't want your your bank to be like reimplementing all of its internal stuff every six months and having to watch it you want to be able to put your money somewhere and just you know trust that it worked Worked six months ago uh I somewhat disagree because we were very rarely do breaking changes

I mean so it's I mean it's not the the breaking changes aren't what aren't what worries me the the fact that there's four different clients is actually quite reassuring like as far as building software that doesn't have consensus bugs and I think the client diversity is the strongest thing going and you guys ethereum does have a good testing team and so I'm not really worried about bugs what I'm worried about is people changing just how ethereum works like contracts breaking um for example which has happened like yes the Cavs generally don't do breaking contract changes but they have it's not like it never happens it does and it's just they they go through and you guys look and see okay we're only going to break you know 32 contracts let's go ahead and do it seems worth it but if one of those contracts

is someone's you know life savings or whatever um and then this is especially hard when you have counterfactual contracts where the contracts aren't even onchain it's just like some offchain contract that may be deployed later and then now you can't deploy it later stuff like that is what worries me and then of course monetary policy changes those was wor me as well I'm not yeah I'm not not eager on making too much changes there and I don't think there will be in the near future and then you cannot make a layer two that has aifi unless if the layer one is aifi that's also one interesting feature that you could get with deification yeah we're stuck with these constantly like there are a few layer

twos that I've talked to that they I I asked them can you guys aify and be an actual layer too where you don't have um anyone in control totally permissionless and aside the second reason why they say they won't do that or can't do that after the whole we're worried about bugs issue is they're like well ethereum can change and as long as eum could change we can't aify like we can't relinquish keys because ethereum can change out from under us and break and so we have people unwilling to build actual layer tws because they're similarly afraid that the L1 could change and they have to be able to adapt to that okay but what's the what's the solution just get

stagnant then you will be replaced I am totally okay with that I am very much not a one change rule them all guy I a big fan of ethere like Bitcoin was around it was great it was useful it still is eventually it will be superseded by ethereum probably ethereum is great ethereum is useful eventually will be superseded like nothing lasts forever I guess what my general philosophy on life um that doesn't mean you can't have useful things along the way and I think ethereum is a super useful thing and Bitcoin was a super useful thing and whatever is next will be a super useful thing and in the meantime um ethereum can be useful and valuable I will admit the one big problem with this is the whole thing

where the system is only secure if its native token is valuable that's a hard problem to solve I wonder for layer two layer twos we could add an OP code that could kind of identify which hard work we are on and then the layer twos would be oied during that hard for but when the hard for happens then maybe there's some kind of board that can decide what to do on those hard FKS but then you could they could only modify things on hard FS it would be better than the current situation so the idea is that their upgrades are limited to when hard Forks occur I mean it still lets them upgrade every six

months so you still have aaable contracts yeah but as user you know that the hard f is not happening in the way you are safe and then then every time hard Focus happening then you have to kind of reconsider your options yeah re audit everything every hard hard for problem is hard Forks every six months like I used to actually audited all the changes and you know review everything make sure it's up to date and it just got exhausting I just can't keep up with ethereum's like I have too many things going on in my life these days to be able to keep track of what ethereum cevs are doing like I didn't even know the transi off codes went in for example it's how out of touch I am okay so we need to figure out how to scale M

I guess so that M can still keep up with everything sure if I had if I had one of me who's dedic who is dedicated just to tracking ethereum changes making sure everything was okay and not bad and fighting against the things that are bad um then yes then I can spend the other mic could then go do other things that mic wants to do okay but I I don't like the alternative that we are just getting stagnant I mean you can always build something else like this is the thing I don't understand maybe you can help me understand what is wrong with just going and building something else like ethereum 2 or you know whatever I start it will start from scratch right it's

very likely to fail for example right so like is it just a um like from marketing side right not from the technical side from marketing adoption side there's no guarantee it will you know change anything in in real life right it will be a successful product while if you make ethereum 10% better that's a huge benefit for everyone at now right everyone went silent on the pl there's no need for Native currency

would your position be different uh just a use like a useful datab base that didn't have native currency people could build the network effect Network effect is what Lucas is talking about you need to some when you make a new one you somehow need to do marketing to get the same network effect in the new chain as well and that's very expensive I guess my my my CLA weak claim here is that the only reason you need to have a big Network for a database is when your database is hosting a currency if you just have a database like you know people can have MySQL databases and postgress databases and everybody can have their own it's okay like you can have 10,000 different databases out in the world and they don't all need to be one Mega Global dat

base and I think that's because the problem here is is that ethereum is trying to be a currency and for a currency to work you really need everybody to use it and there needs to one and that's the problem I have not been able to solve I don't know how to get rid of the currency and still have a decentralized database I don't think that's the only reason because you still need to have the decentralized database and the way we do it is that we have a lot nodes so they you need to have a lot people people to support it and that's also one reason why you also need Network effect but we don't even have a lot of noes anymore this the second thing I argue about all the time which is that no one actually runs a node there's infura quick node Alchemy and then like 12 devs

that's ethereum Network these days I guess Cloud Fair's got one there's literally like you can count the number of people that are running nodes and it is not a big number in terms of actual unique actors which is what actually matters and so we don't even have the benefit of this hypoth medical big Network in terms of decentralization we have you know like I said dozen people I'd be more convinced if we actually had you know everybody running a note on their local machine and maybe portal network will help with that I think we also have all the sexis or other sexis running on Alchemy on those that's some something that I don't

know but I would imagine that those sexis would be running their own notes because that somehow feels a lot safer when they're managing customers funds you would hope I don't know maybe I think coinbase probably runs nodes you're right and Kraken may run their own nodes that seems reasonable to assume I think they all have the validators so I think they would kind of have well they also might might be buying that service from somebody else I'm thinking about what you said and trying to come up with a repost uh easier one we don't really need all people running noes we all potential for people running notes so if anyone becomes dishonest you can just someone

else can run the noes and be done with it right that's enough well except for so you have a few problems one is the censorship problem infura actively sensors and it is a real problem for users what do you mean by sensors infura sensors like everyone from everyone from Venezuela is censored by infura someone in Venezuela can run the node node problem people people in Venezuela can't run a node because the cost of running a node is too expensive people in Venezuela you have to use click node Alchemy or infura and infura is what's baked into literally every tool um applications all bake INF fura into their app and these days they don't even

let you use your wallet connected wallet anymore which is incredibly frustrating for us we run to it all the time um that's you know that that's a potentially solvable problem getting applications to actually be built a decentralized way is maybe solvable um like running running node is not that expensive now don't don't get me wrong it is when you're in Venezuela and you can't afford to buy a machine just for ethereum like users cannot run nodes on their home computers is basically my the argument I always make you build a machine that runs a node and it probably cost you $500 or so um you cannot run it on like a home computer like that you've already got you have to go out and buy something specialized Hardware no one

has four terab drives on their home okay okay you potentially need a drive right and we are bringing we are bringing for Force to to handle it for force will help and portal network will help a lot I mean there there there are things on the road map that could help address this problem and I'm looking forward to seeing if they do like I'm not totally pessimistic I think in a year or two this will be mostly obsolete you are more Optimist this obsolute eum or this argument great uh the storage well maybe not obsolete but reduced a lot because obsolete would have to go to stateless to be obsolete that's

longer yeah um is there some State rent or something what solves the state growth problem well not the state growth problem not the state gr problem but running a node with without the state stateless solves that right so you you still can't do everything because you need the state to do some things but you can at least interact with the blockchain in really decentralized matter but who is storing that someone still needs to store them someone needs to run portal also portal also helps you with that because it can give you give you the part of the state right so there are ways of that we are developing that will will handle that or

I I would say I think of portal I'm not sure if portal launched the state Network part they were to launch it this year I'm not sure if they finally launched it uh need to look into that but uh yeah uh the idea is there the hist network is already there and working I know because we we have an integration with it uh not sure about the state Network and then if we have stateless clients and state Network work through portal kind of I'm not sure if it's fast enough really to be useful the state network but if it is fast enough to be useful then we're kind of done yes if nothing else changes the

thing I'm worried about is that as soon as we get any Breathing Room in terms of State people are going to be like okay let's 10x the gas limit let's you know add 10x blob limit and we're just back to where we started 10x in gas if you have uh statelessness is not a big deal it just CPU which we have really lot of oh to build blocks it's it's more problematic but yeah and I I'm not worried about building blocks I I don't mind block building requiring heavy Hardware that that doesn't bother me really at all what I worry about is so let's say we we get portal Network and is great and it does everything it promises to do um and we get veral trees so we can actually

have proofs that are not insanely big yeah or something similar okay something similar what whatever it is um and so people can now um run a fully validating client stay up to date with head without needing multiple terabytes of hard drive space right you will complain about bandwith now about what band yes that's where I was going because some then people will say okay we we now have breathing room let's 10x the gas limit and now in order to get the proofs necessary to validate a single block you're going to need to download you know a gigabyte of data just to valid not gab exaggerating I'm yeah I'm exaggerating with the gigabyte

the point being is that from a we're again we're on home networks with people with just using their regular da daily driver machines they don't have a dedicated bandwidth for this they're doing other things they're browsing the web they're watching Netflix they're playing games they're you know listening to Spotify whatever it is they're doing with their computer is a home computer for doing home things and we can't expect them to have be willing to dedicate you know 100 megabits to getting blocks every 10 seconds that have you know 200 million gas in them and with proofs for the whole thing like that's huge and so we're tra so yes if we get all the stuff that we just talked about that'd be great I'd be very happy I'm very worried that the immediately following that or immediately before

that we're going to spend all that gain we get got on more gas or something else well to some extent we will spend uh you know but there there is some resource requirements that we have to go for right there always will be yeah so I mean we the we have to set some lower bound like I I don't think it's reasonable to say okay you can run ethereum on your phone like that's not not a Target the the target I used to think that everybody was kind of in agreement that we were targeting was people's home computers meaning computers they already had in their home they can run an installer and then ethereum client will

run in the background and so ipfs is a good example of this I have ipfs desktop running and I've had it running for years and it doesn't I don't even notice it's there doesn't consume too much bandwidth doesn't consume too much um drive space it's yet I'm contributing to Network I'm sharing files I can use it like that that's the ideal that I would like to shoot for and I think I'm pretty alone in that I think most of the people in cev these days um don't agree with me that that's a Target demographic who should be able to run a node I disagree because you know I'm running the node on my own Hardware maybe not constantly but for long periods of from time to time and it's fine my Hardware maybe except my bandwidth is not not something

special not anything special uh a computer you do other things with yeah yeah yeah I develop on it okay and um I'm guessing you bought a hard drive for it though right yeah that's the only thing I have a hard drive and and do you think that you'd still be able to do that if we got 44s and portal Network and everything and proofs and every time you want to validate a block you had to download a full like 100 million gas worth of proofs yeah I think so that's not that you know that's not that big big amount I think uh 30 million was uh ballpark

of less than megabyte so that would be few megabytes or way less than megabyte I don't remember um but it was in kilobyte size so even downloading uh few megabytes so for example with epbs the latency games are also more important but if epbs we have a lot lot bigger window to do that and I think there with some also with some I don't remember I'm not sure but okay I'm not sure about that well I thought with epbs the latency games uh like stop being that important potentially but I'm not sure if that's

correct so like for stakers or for home users for producers yeah for producers like I said I don't care about producers can make them require no no but the the problem is that producers withhold the blocks till till late right so we have a short time to download it to attest to it but this there's I think there's something in the pipeline that should fix that so they don't have incentive to to wait till the last moment so that will help with the band with I I don't think that would change anything in terms of people running just valid full nodes or validating clients well uh it does change that if you have

uh one second to download validate the block and push out the station or you have six seconds to do that that's change on when I when I say valid better nameing for this when I say validating CL just a client that is following head and validating all the state changes so they're it does it does on the band with site because you have lot more Headroom and time to do that so you're band withd requirements are lower if you're not a testing you just want to follow head I just want to follow head okay uhhuh sorry about that I got

disconnected yeah sure okay I I I see what you mean okay it doesn't make big changes there yeah okay what I'm worried about is just that you've got if you've got let's say I don't know 25 megabit home network for people in you know third world countries in Germany because apparently Germany's internet sucks um I don't care about Germany much I'm polish so that's my mortal enemy okay oh Polish people I'm joking but they you know they can Fox the blocks if they want so the the question is is how

much I think I think it is certainly possible to make changes to ethereum such as portal Network such as 44s such as brle trees or something similar Etc and have it so the Target demographic that I I used to think everybody agreed was our Target can run nodes and that would be people who are running on their home computers I'm not sure that's true if and let me rephrase that I guarantee you that's not going to be true if the cevs don't agree that is our Target demographic if everybody thinks our Target Target demographic are people who build custom PCS to run ethereum then we will use up as much space as we can to hit that Target demographic like we'll

hit whatever Target demographic we we choose to go for and the Target demographic that cordev seem to be going for when I talk to them is not the home user it is people with specialized Hardware who go out of their way to run and maintain an ethereum node so I somewhat disagree all right because like I said we we are testing for example with uh with this kind of user in mind we are setting gas limits or gas pricing uh op code with that user in mind um the whole like statelessness uh is to actually alleviate a lot of burden of that user uh huge amount of Burden from that user

so the the problem is we also want do scaling and why not we do want to scale uh we want to scale blobs we want to scale execution right both uh and I think the ma major problem in the future is bandwidth in that in that regard and I somewhat agree there this is the major problem the major problem is what bandwidth internet connection band with yeah and so the do you all other all other things have potential Solutions right all have they're not here yet but that's why we don't want to aify the ethereum because

we want to solve those problems right so is it correct that you your honest opinion is that um the core devs care a lot about making sure that ethereum we trying to make ether make it so ethereum can run on consumers daily driver PCS like the computers I'm not sure if I'm talking about Hulk or def but I know that GE team for example has this as one of the top priority and that's why they also uh their kind of beef with the researchers is that researchers are pushing for scaling a lot and that kind of is somewhat compromising on this especially if depending how we deliver

things right so that's kind of one of the point of conflict right uh I think researchers want to scale more but the cord devs actually are are trying to make it real you know because they are the ones that support this right yep and do you think that the uh core devs are going to win in the long run or do you feel like the well we can just say we W implement it so we have the final site right yes but the I guess the thing I worry about this gets to my the argument I'm always making with people on um on capture where like it used to be

you know gu and parody and then nethermind and and Basu Basu has always been kind of the corporate version corporate client so they're a little bit outlier there um but then we we got Aragon and they are they're not corporate but they definitely do not share guest position on who the Target demographic is at all and that's fine you okay they're building their client for different Target demographic but they on the other hand they are doing great work with archive right so they kind of democratized archive notes they're they're doing great work so like as from an engineering perspective I think they're doing great things the thing that worries me is that now you have so start with the addition of

Aragon you have one group at the table you have two groups at the table now who aren't pushing for the Target demographic of end users that being Basu and Aragon B's always been kind of ambivalent though so I'm not too worried about them just philosophically um but then you got reth who joined recently and they are very pretty strongly not like ethos aligned with the let's target end users they're much more like thinking about things like MV and VCS and businesses and things like that and so my worry here and then you have you know Peter leaves and you have Danny leave and so over time I'm not sure if Peter left but just hius for an unknown amount of time I think he will be back maybe

maybe what I'm getting at though is to be honest uh giving you heads up I'm also planning uh two to three months leave in the summer but I'm I'm I am I am cing it at that right so it's not longer but I want to just do a proper reset and I think that's reasonable some energy what what worries me is that we have over time the set the aggregate ethos of cevs has been shifting and I am not at all confident that in the long run cevs are going to continue indefinitely to care about the end user as the Target

demographic and so you know like you're right right now there are still people in the cevs that are fighting for it but like you mentioned re so research used to also be very aligned with cevs like they were a lot of the same people and these days that's not the case anymore and this is just like a slow thing that's been happening over the course of a fairly long period of time and I just I'm I'm just not at all confident that we're going to end up maintaining that ethos of okay let's target end users let's make sure these things run on people's actual machines so you can maintain censorship resistance if someone's trying actively trying to censor you um okay like I said there are different ways of censorship resistance right getting you access to the blockchain at all or getting your transactions in so

you're saying again this to get access because the second one is actually uh being solved by the fossil uh inclusion lists second second version yeah that I'm I'm more optimistic on on that one we we did we do have a bit of a sense for that yeah yeah yeah there's spec for that yes yes and Shar uh it's I will look it up for a second uh somewhere on the internet yeah yeah yeah exactly uh I know you're talking about there was a breakout room if you want to uh look at it oh yeah let me there's a good

researcher I think there's an EAP already somewhere or it's being written they want to get that into Osaka uh I'm I'm reasonably optimistic about the um inclusion list being able to mostly address the censorship issue we're seeing right now on mainnet and honestly the censorship issue has almost kind of semi Ben addressed on its own it's still a problem but um yeah it's bigger it's wor better yeah I think it's a more problem in potentia than than actually being so big problem in reality but uh we want to address it to yeah you know not go away completely right instead of

instead of uh hoping that it just doesn't get worse yes exactly exactly exactly yeah like I'm agree I'm agree with that the the one I'm less confident that will be solved is the um ability of people to run their own nodes especially as we we we know that the centralized nodes will censor people because they do today like that is a real problem for people but you know if it censors enough people I I don't see why someone won't invest those few hundred dollar into an infrastructure that will address that like it's not thousands of dollars it's not 10

thousands of dollars it's few hundred dollars potentially right um yes I mean the plus the bandwidth yeah plus the bandwidth which for a lot of people that's just isn't again the bandwidth is the hardest things to solve at the moment to be honest from the resource point of view right I do agree that this is the and but it it is the most like controversial topic between researchers and cordi at the moment the bandwidth issue yeah yeah I think that's the number one because the rest even if it's not Sol of now like we are actively working on solving it and it's getting there it will you know it

will take those two or three years to get there but it's getting there and you don't think that when it gets there people are just going to um Lobby for using all that nice fresh bandwidth they've just well yeah they will try uh yeah definitely they will try you think they will not win uh I think there will be some compromise and we'll see where it's where we get there and we'll see let's see the numbers right let's not just I think this is where we disagree then I am less confident than you that uh okay let me let me let me get to the other other side of the of the

discussion so um would you like to ethereum to stay as is endlessly 30 million gas and uh uh I don't know free blobs I would I would be fine with that like I said I'm I'm a big fan of new blockchain later like I I wish eth2 was actually eth2 but it's like I said it's it's extremely unrealistic to to expect new blockchains coming out and be successful there will be few but it's it will take years for that uh yes and this this is why like the the

number one reason I have not built a blockchain of myself to compete with ethereum because I'm annoyed is um specifically because I know that launching a currency so having a functional currency requires Network effects and that is really hard to get I if I could figure out how to build a blockchain meaning a blockchain here meaning a censorship resistent and permissionless database shared database that didn't require currency I I would probably stop my other work and go build that um because I think that would be really useful that's like that's what I think what is great about ethereum is not the currency in my opinion it is the censorship resistant and permissionless database people can you can put usdc on

there you can put wpdc on there you can put whatever on there that is less interesting to me than having the shared Global database or globally shared database that is permissionless and censorship resistant well I think I don't know how to do that just the database you have also execution right that is somehow verifiable Etc yes there's I don't know what do you mean do maybe you put it on the under database already so yeah I mean I I put that in database like database can do um like even postgress can do what do you call it stored procedures yes like like it's pretty common for databases to be able to execute code as part of database operation and so for for me that's

that's the interesting part because no no other database in the world is permissionless and censorship resistant where anyone can write to the database anyone can read from the database and it's uh shared between users and so if if I could build that without a currency I would definitely go build that and then you don't have to if someone could solve that problem then you wouldn't have to worry about needing Network effects for your database your datab but what's the incentive of running this database then because it's got data that interests that of use to you or interest to you so like take ens for example ens doesn't need eth to be ens ens is a useful shared database and in theory ifs was actually permissionless and then it

would be like an ideally ideal name system where anyone could register a name have it point to something and U no one can stop you no one can censor your name no one can take your name away from you etc etc all the things ens is supposed to do you don't need eth for that like you need a shared Global database for that and that's a useful tool to that I wish we had or rather we do have it in the form of ethereum but in order for ethereum to work ethereum needs e so that's the part I would like to get rid of how would you how would you limit the the database that is exactly the problem I don't know how to solve this problem without a currency

and it frustrates me to no end things they finite right in infinite World you you could do that yeah sure in infinite World you could do it um re realistically I I don't know I mean something that makes it so you the best thing I can think of that maybe would work if someone can figure it out I have not been able to would be something where you don't have a native currency that is tied to the database itself but you have a mechanism for charging fees from users in whatever currencies like let's say usdc running on top of that that database um and then you can charge user for that and then distribute

that uh those transaction fees to use uh sorry to block producers in some way and so you wouldn't need native currency in that case you would just use whatever currency people want to use on the database and then if your database then its usage declines over time that's fine because the currency wasn't baked into that database it was just a currency that happened to live on that database and it can move and so the currency can leave the database behind and go to a new database later and importantly users can slowly migrate off of it safely got layer tools um I mean so people could move the the currency to Layer Two is yes but ideally you want something that's not bound to layer one even like there's the

big risk with with Bitcoin for example is bitcoin's aifi for the most part and over time as it there's a very big risk of a bank run Bank Run on bitcoin at some point in the future and a big risk of a bank run on ethereum at some point in the future and so when when these layer ones eventually get superseded by something new everybody kind of leaves and this you this happens with actual physical currencies too this is not unique to digital currency this is a currency problem if your currency is a global dominant currency and everybody leaves it because there's some something new or better out there uh that can be a very big problem for the people who are the

last ones to leave interesting we didn't see this with um the British pound the British pound used to be one of the top currencies it's not anymore but there wasn't a big Exodus and I don't know maybe we can figure out how to replicate that like British pound kind of wound down safely hey guys I need to finish sorry got a call okay bye talking to see you guys later bye