{"id":"Ge1Cfq-0y5w","title":"Decision Discussion: Reducing Storage Cost","description":"Follow the latest from NEAR Protocol on, \n\nWebsite: https://nearprotocol.com/\nDiscord: https://near.ai/discord\nMedium: https://near.ai/medium\nTwitter: https://near.ai/twitter\nGitHub: https://near.ai/github\n\n#WhiteboardSeries #Blockchain #FutureIsNEAR","channel_name":"NEAR Protocol","published_at":"2021-02-01T00:00:00.000Z","thumbnail_url":"https://img.youtube.com/vi/Ge1Cfq-0y5w/hqdefault.jpg","transcript_text":"and we should be going live and awesome all right hello everyone thanks for joining the discussion about storage costs uh we'll start kind of digging in i will uh first start just kind of to cover the some context and some details of uh proposals uh which were posted on the forum and then we can kind of discuss pros and cons for each one and if there are any other ideas that we haven't had uh surfaced yet is this turn right for everyone okay all right so just to give some context right um well why we're here is the costs have been you know rising and became pretty high for some like specific use cases like uh people using multi-stake contracts which deploys a contract on each account or some folks are planning to launch uh products where you kind of need to create contracts for each um like group of people for example right so we have like splitting dial with you know each guild and each team or each squad even having their own kind of contract or i'm in base where each store has its own contract and so the current costs are prohibitively high for that type of uh also generally like account creation kind of becoming somewhat expensive and especially if people want to have fun that if some developers want to fund this in an easy way or we want to have some kind of faucet kind of somewhat independent of this but contract development has been somewhat cumbersome around storage taking where you need to handle manually like how much money is coming in how they like how much money coming in for storage versus how much money is coming in for like actual payment and things like that and then in case of cross contract costs paying for storage isn't harder where you kind of need to like keep track sometimes or like send pieces of storage uh sometimes you don't even know how much you get a spend on storage later so you need to keep some amount uh and send some amount so they're just all kind of cumbersome high level there's actually at least like as far as i know there's three ways to pay for storage uh in general so there's what i call storage taking and that's what we use right now this is when you lock some amount of money based token or other token on the on the account and a inflation of that amount or uh in some way pays for for the two validators or two whoever maintains the storage right so just to kind of the basic mask here right if you know if you have 100 near staying for a year on this account this is 100 years it's not being delegated or lent which but like specifically because you're not getting delegated the rewards that would before go to 600 near are distributed among validators and so it's a like somewhat indirect way but to kind of you know remove those coins from circulation and from uh staking to accrue more value to others or to whoever is taking so the second approach is storage trend so this is where uh if you store somewhat of bytes on an account the your balance is like decreasing block by block by some money and this is actually we have implemented before and it was a nightmare for a few reasons main one is because your balance is constantly decreasing which is super weird to people also also tooling needs to kind of know about what's going on because like the balance that's on chain and balance that's kind of like virtually computed is different and then the third option is just pay one time and you know store forever and this is kind of where really all the existing blockchains are uh like excluding few and like it has a very you know suboptimal obviously problems that when you pay once you and like kind of it only it only covers costs right now and it doesn't cover costs and kind of perpetuity right but the data needs to be stored in perpetuity and specifically for example ethereum has you know this model in like somewhat of like a form and you know the growth of state has been pretty exponential actually uh this well this is about the super latest data but like uh the gross is pretty exponential it actually like i think people analyze it's pretty you know a lot pretty well aligned with kind of usage of the network and so people don't really release data people people don't really kind of delete stuff so we have been running stores taking right for a few months and i mean i don't have i'm collecting hard data on this but i don't think we have had that much deletion happening either right so even though right we like the storage is pretty expensive we lock in here right the idea is that it will get released back if you delete the storage rate but that doesn't really happen because like most of the stuff is not designed around that uh we have a lot of confusion around ui where you know there's like two balances like this kind of this minimum balance that you cannot use which means there's like another balance which is what you can use there's a total balance and we also have all the lock up stuff on top of it and staking so you actually have like a lot of concepts just to explain like how much money you have right and that's been very confusing um and costs are high right uh so this has been kind of a problem uh for for various reasons so i want to mention kind of one of the approaches here which is like to approach to that problem like pay once you know start forever is actually to not store data on validators right and like generally the cost right the cost of transaction because it's a one-time thing it doesn't include the you know storage costs that kind of keep goes through in perpetuity and one way to solve this is to actually not require validators to sort store it in perpetuity in the first place and this is actually kind of similar to saying well actually the valdez will not store much right or very little but we're going to push everything into kind of indexers and you know like nodes that will index data outside of outside of consensus um there's still a lot of problems i mean the folks in the museum have been working hard on kind of figuring out how to how to how to make this work uh i think the current approach generally is like you need to have like full nodes or like application specific nodes which will produce blocks pretty much and then give them to validators so you kind of have a lot more like you know minor extracted value things which like kind of by design and a few other things and at least like from my perspective one of the bigger issues is like it adds latency because now you kind of need to like as you're sending a transaction it needs them to get all the data around this ideally it should be for formed in a block right away because otherwise like if you have two transactions to the same contract right and like they may not merge because like the state is different and stuff like this so it needs to be formed in the block and that blocks goes to validator and it goes back and the next transaction can only arrive right so um we're not going to discuss this because it's like a major major change and still a lot of research needs to be done but i just have to mention this yeah actually uh we have a plug here we're going to do a whiteboard session with uh downcrowd feast this thursday on uh stateless validation uh and see what's uh ethereum's plan is for it yeah yeah that would be awesome to hear um but we're not gonna at least like in short term i don't think we're gonna consider that but it's definitely super interesting as it solves some kind of more conceptual problems cool is there any more context we should discuss actually before going to the proposals well you can also mention pros of having states taken which uh blockchain state is kind of reliable now and can be exploded so so you mentioned the problems but at the same time like if we say storage is free then we'll have problems so we should say why we even did this yeah yeah so i think that the two benefits of storage taking right why we did it in the first place and why like it is on on what i call on paper really good idea is twofold right one is it's it's a little bit better alliance than the kind of usage patterns of the blockchain with uh like base tokens value accrual so pretty much as people use for example c20s or other things right the storage usage expands because you create new addresses and stuff like this and given you using these resources like for example on ethereum right now it's not really proportionally kind of house the storage expense to value acre like to how much helium is being spent on uh on doing this and so storage staging actually like you know proportionately locks that amount of uh like base stock and towards a kind of a bit of usage and kind of yeah the flip side of this guys that's kind of controls how much uh storage you can spend and because it is charged in base token right it has like a very explicit like as the storage amount increases the supply like available supply decreases so uh there's also kind of a mechanism around to control that people will not store too much so this is where the kind of the conflict comes in or like because yeah like from perspective of developer you want to store more from perspective of the system you want to kind of have somewhat a fixed amount of storage in the system so yeah so there's oh anything else actually i think some uh there is some other which has a different system where there is a fixed total amount of storage and there is a storage auction where there is a market where people can buy and sell the storage so the market sets the price for storage yeah yeah so there is a pretty much a way to kind of price storage more dynamically i think the when we explored this and when you know looking at some of the implementations of this the main problem was that like people like to speculate on things and so the there was a lot of kind of unnatural things happening on those markets even though like presumably markets should be somewhat uh you know they should be optimal but they're not right so people were buying out storage for no reason and then um driving up the prices and like selling selling selling them at high price uh kind of just to make the resource more scarce oh i'm not saying that this system works well i'm just saying that there is one more approach that this actually actually exists yeah so you're right yeah yeah i mean yeah i think generally there's like a little bit more yeah mechanic mechanism design approaches which are more yeah as evgeny said like around kind of creating auctions for for storage with various models um and i think eos has been has been doing that model uh as as as one example but yeah so so far it it's been hasn't been shown to uh work very well so here's like few specific proposals that i would like to discuss but we can also discuss some something in between or some other options so the simplest one is just decrease the cost by 10x and so uh kind of the reasoning there is like the calculations i was doing was like based on specific parameters and there's definitely like a 10x there to reduce as i was like rounding up a lot and uh like using a ver like kind of a one dollar near price as a benchmark and so this would you know reduce things like by 10x so a like our accounts are 100 like basic account uh is hundred bytes so this is without keys so that's uh that would go to one uh thousandth of a year and then kind of hey a somewhat standard 300 kilobyte contracts would be three near all right so that definitely would help add to perspective what it is like for two other metrics so we can compare this to average transaction like average function call for gas and also the maximum state you can acquire if you like lock entire near supply into the storage so currently we have fashion 10 terabyte for the total state that you can acquire if like full near is locked for storage and it was already slightly higher than where it we wanted it to be but kinda unrealistic to reach and this proposal essentially bumps it to 100 terabyte which is not completely insane but kind of on a border of like getting hard to to exist if someone actually acquires so much storage which is probably not realistic as well um and the second metric is the cost of a transaction so the cost of a transaction uh it's better to measure in tera gas so something basic is around 7.5 tera gas and 7.5 targets is um i think it's around 0.0001 or like or seven years so it's like one ten thousand of a cent so it's like ten times cheaper um or like 100 times cheaper than 100 bytes right something like this so so our cost to to write this data is probably around um from gas perspective is 100 times cheaper than to persistently store it or like to keep it in locked state if i'm not mistaken yeah i i was kind of didn't get to that math but yeah um yeah well one thing i i do want to mention that like evaluating full supply map to storage is definitely not realistic because of this like if even if a lot if we imagine that half of the supply locked in that the price of near like just because of supply demand right will uh be like outweighing a lot right and i'm not even saying that you know we already have like actually this the supply that's outside of staked and locked funds is pretty much almost non-existent in the first place at this point right so um but yeah i mean it's definitely definitely uh worth noting that yeah even with this with this lowering it's still pretty uh just the storing like you need to lock yeah about at least yeah 100 times more in years and a fee to yeah to actually like process it so so this goes into option number two we lowered by hundred x and i mean the justification here is mostly around like so this kind of these calculations don't really account for you know price decrease in storage um and like the calculations in general were like pretty basic um they don't also account for like contraction in the supply as well and kind of the like supply dividend there so this is really trying like i wouldn't say that the math at least like fully worked out and i have like a full model here but uh like kind of accounting for storage degrees right for example using models that we've used uh for like future storage price and also uh do it like kind of evaluating contraction on the on the support on the circulating supply as well so i think what is what is the what is roughly the curve they're using for this um i mean are they expecting it to exponentially become exponentially more uh expensive or like what you are yeah oh i said yeah i have a paper somewhere but yeah i think it's roughly like uh it becomes cheaper like every two years or something uh so yeah some kind of exponential decrease in price to store one byte like for for hard drives per se uh i was actually trying to pull the data from like a cloud provider but like i only found ec2 uh costs not not the storage uh costs but anyway i mean we can like we can pull the mass more more accurately later i think like the negative side of this this actually becomes you know a little bit too cheap on the side of like really sliding the data right so now if you want to store you know something like 100 gigabytes right um this becomes now like only hundred thousand near right so this and hundred gigabytes is kind of current state of helium uh and so this this is definitely cheap and from the perspective of um costs that like people would incur and you know definitely requires that we like start charting things reasonably quickly to to be able to to kind of capture it and i think this is where one of the kind of questions that you know we've been discussing a lot but haven't really finalized is like how to balance sharding on transactions throughput versus the shouting of storage so if there's a lot of kind of storage sitting there which is not super active uh but like it still makes sense to have shards which are reasonably evenly split on on the mountain storage now eugene you were pretty against the 100x decrease on a from perspective of the classes i mean like spamming as well yeah i think it's still like i will probably comment after all three proposals from like my perspective but yeah i think it's just getting too dangerous to to go like straight 100 x all right well so i mean these three proposals are actually kind of in the like doesn't mean one doesn't actually exclude another uh so the third one is yeah later on on top of 100x to actually burn the fee and then like merge the charging into gas charge uh also this burn and charge can also be like separated so the reason for burning is actually first of all if we burn we can actually and for example not return it later this three this easily justifies uh reducing the cost by like 20 x um now it removes a lot of kind of mentalities that people want to free up the space like the benefit of fearing up to space some of it can be like worked back like what ethereum has for example it gives like a gas charge back so like if you remove if you remove data you get back some gas while you're kind of running um and merging into gas is really about improving user experience right like the contract development is way too complicated and you know users and developers like paying two separate fees attaching balance plus gas fee is kind of complicated uh the the yeah so like we can just increase the storage right operation cost uh and so that actually increases that like i actually did actually yeah i looked at the storage cost and like kind of increases that gas usage by 100x that specific operation to do that the i think the the kind of the important piece here is indeed the like that you know on one side costs are low but it's still actually like if you compare to our gas usage it's actually pretty high uh so it will bump your gas usage especially when you're deploying like large contracts or like upgrading large contracts um but it definitely improves kind of experience and you know is more familiar for developers for existing blockchain covers uh yeah and like one option around how to burn is actually like it can still be charged in near per bite so like not in gas actually uh so like to not be affected by cars price and your charge pack also can be in near per bite so just get like like we actually record what's the gas price for this transaction you just get back that amount of gas proportional to how much gas like how much gas price you paid for this gas uh and actually refunds right now would convert it back to near so this all kind of like this uh a few of these formulas in the in the proposal on the forum i mean that's pretty much it for the kind of the proposals i had that uh curious to hear what other folks are thinking about this so also proposal zero like how how it actually addressed without this change basically so for the fungible token the way we designed it is we created explicit account registration api so basically any action you do on fungible token do not require you to pay for the storage except for initial registration and the way it works is like you have a standard that's saying how much tokens you need to attach for this particular contract if you want to start interacting with this so you first attach this amount by calling it that creates pre-allocates essentially your account storage assume that it's fixed so it would not work with like old contracts which has dynamic storage per user and then once you're done using it or you don't want to use it you can claim the storage bag by essentially closing your account on this contract um so this is kind of work around with the storage fees that you have for lockup the main reason here is if the account is one-to-many which means multiple users use this account or a contract which is majority of contracts supposed to be like this then to prevent kind of drainage storage and sible attack you require payment and this payment can originate either from you or from someone else who want to send you assets so it creates some problems still because now you need to think about whether you register it or not but this is usually done on the ui side an example is like unislope you go to unislocket checks that you register on a token that you want to receive and then if not it's saying you need to register first you register by like collocating your account then you can buy token by unisloop um that's a background like how it works right now but this still doesn't oh sure yeah but the problem is that still today deploying contracts very expensive right and the example is the 2fa like people just have 35 like near lot there that's very suboptimal yeah yeah so and for replicated contract the solution would be you have some kind of reusability of a contract binaries where the asset is like the storage that needed to store it and deploy it burned once instead of locked and then all like validator will be able to keep it in cash and you will be just reference it by hash um so this piece is far from being done but that will address the problem of replicated contracts such as tf2fa lockup contracts um maybe nfts or anything that is sharing the same instance of the code so if we do this uh are you suggesting that uh there is no need to reduce the storage cost uh i'm suggesting that potentially yes i still like the idea of dropping a 10x i don't see immediate abuse if you just go 10x because it's still expensive to store large amounts not as expensive as right now and may create certain problems originally but still um reasonably priced but i'm saying that like longer term we hopefully will arrive in a place where contracts are cheaper by optimizing some stuff maybe not only know std but rewriting certain libraries or rust contracts are cheaper assembly scripts contracts are already relatively cheap and given 10x reduction will go to like a very low price per unique contract for non-unique contracts it's still fine because it's a one-time payment once you finalize your binary pretty much yeah i mean i think that contract is definitely kind of the current big pain but i think the you know the user storage is also there it just kind of slides a little bit unnoticed because of the contract storage um like the problems we've seen right with wallet is when people kept adding access keys so i think generally like this the pattern is that people are not and not ex like hey nate the the the people are not uh planning to remove stuff from blockchain right it's not like it's not a pattern of like you know using limited capacity on blockchain it's more about you know keep storing more stuff i think that's kind of the one of the problems where like the storage taken is just like misaligned with how people are using the stuff um i just also wanted to share the this is rviv's pricing data so it's yeah it's not really exponential but it's kind of i feel like there's like actually uh a change in in opening progression yeah so like i wouldn't but but alex also shared the costs for aws and so this looks uh it went down to x over the course of seven years on the slide from december 2000 into december 2016. yeah i do want to mention this is a s3 which is a little bit different from like like i was looking for ebs which is like active because i definitely agree that like a long-term storage like we there can be doing a lot of stuff there but like active storage that validators need to have hot for executing code that's that's i feel like we're limiting factor i i mean um i i don't know whether alex is alluding to this but ceremony has this regency's plan which it's like they kind of just having the basically separation of hot story from code storage and validator only need to keep like uh started from uh some recent number of blocks and then they kind of relocate the old data to some code storage that they have some separate providers and you can curry promising well so this is good i mean this is going into like more stateless stuff i think in general like i think like if if going stateless it makes sense to keep the hot stuff like contracts and some like recent data on validators anyway and then just yeah like leave all the stuff that hasn't been touched in a year to some to somebody else uh so like you actually need to include in a transaction data if it's like too old so i just want to jump in and see from the developer relations perspective having those high storage costs is really crippling for developers in terms of growth and user adoption and i think that like this is not the time to be like limiting the data when we're just in the in the first small stage like maybe if you want to set some kind of like rail guards so that like insane things don't go in there so like at a certain point it kind of starts going exposing i think that's fine but like i'm in favor of the like 100x reduction because i think that we should be encouraging people to build we should be encouraging users to come on we shouldn't say like if you want to play with something on here by the way it's ten dollars to like step into the ring right and that was never the original intent but somehow we landed here and i think this is a way to address that in the immediate future and uh just to follow up on what danny said this is nema uh in the case of mint base right now i think it's like over 70 yeah 75 to deploy so even a hundred x reduction is a significant fraction of one dollar like i i one would expect near to cos actually lower than that so uh even 100x is not like i think it's not enough but it's it's more manageable right it's because of the model of replicated contracts so yeah i think farming basis probably uh yeah having having global storage would yeah basically just address this problem yeah so uh thor for mint base here as context we're currently deploying contracts that require 35 near in a factory pattern so that translates to roughly 70 dollars i worked a bit today on integrating the no standard pattern um but it looks like that's still in development because it doesn't look like it's completely covering features so for instance cross-contract calls are not currently uh something that you can do with the no standard pattern so we're looking for all sorts of ways to reduce contract size yeah so i mean like generally why i would suggest going for 100x right now is actually more reasoning that they need provided that's like right now we're not really kind of need to be perfect and we just we just we just need to deliver developer experience uh and then start like we can we can improve things like how exactly we're storing stuff like on the back end over time like again stateless stuff or whatever can be improved um over time on the back end as as the need of that arises um and i mean that's kind of why i mentioned stateless validation in general because like i do think it's a solution and i mean in some shape or form it's not needed right right now because like that's not actually a problem in the first place bowen you were talking about that we'll have it all solved when we figure out the cross uh the reduplicate of code that would help a lot but i also read that it kind of uh doesn't really work with sharding is that i need to read more um but i mean it works with it works with starting it's just that uh we're it's not i mean we have not started implementing anything anyway i think the design is not finalized i would say so uh it might take us some time to get there it's definitely not the uh the fastest thing for example we can do the storage price reduction in a very short period of time but for this it would take much longer got it so no std is not an option for us but we will have to wait for you guys to manually reduce the storage cost just as a hard-coded thing and then this duplicate code might get fixed maybe a year from now is that oh i think i think it's less than a year but it would uh probably be several months yeah i mean no it's nice to do i'm assuming i mean we'll like just get the features you need as well so i mean if there's like specific things that doesn't work it's just like missing implementation of the couple functions in sdk yeah i think thor was saying we just implemented the no std on a side project just to see if what would it would reduce and it reduced us by like one one year token oh no uh we can't even compile that that's correct uh that's incorrect because we we wouldn't be able to implement it because the store contract currently uses cross-contract features which are not currently implemented in the near sdk pure uh library so i i don't have an estimate about how much it would reduce the store that is safe to say but it could reduce it by a lot by little it's just that the uh no standard library is not complete it was here i mean from my poop well all right i mean there's a separate like like if you reduce storage under attacks you cannot bump it back it will break contracts that already there so that's why i don't uh advocate for trying to go 100 x because if we like if people start abusing it us for for doing this uh then it will be um very unfortunate state where potentially we cannot really address it with any like reasonable actions such as like the only solution is like delete this contract bump the storage break don't have contracts do a hard fork so you don't want to go there yeah i agree with that so can we actually talk more about the the last proposal of burning uh just burning tokens for storage might be interesting idea well i mean i think it still includes a kind of 100x reduction so like from perspective of like if we're thinking of an attack surface right the the attack is the same it's just like instead of locking that money you're kind of burning that money so that's the difference but i think that's a pretty big difference if you're talking about that much money like whether you just lose it or it's just temporarily locked no no even if temporary lock it's not a problem like you can create a state you can grow like archival state right now much cheaper so it's a persistent state so you you're basically locking it until the network goes down like whatever it is so it does it will not address like you you may not be able to recover the funds so so from this perspective it's not does not attack change the vector of tech so can we address the challenge that eugene brought up about the abuse of storage because i think the legitimate use of storage for projects versus the potential for abuse are two different subjects so is there a way for us to set up a threshold or some guard rails to protect the network from that kind of abuse maybe with an exponential rise in the cost after a certain stage while still giving enough space for developers and users to benefit from a significant rate reduction so like a tiered approach a general answer is no because if someone wants to abuse they will be abusing the chip storage and the expensive storage will be left to the to the good people also like prioritizing work on like replicated storage i think will bring huge benefits and differentiator for near it essentially creates um not stateless but it allows us to have like incredible stuff that we haven't tried using before short-term going 10x does not fully solve the problem of like creating this type of replicated contract that mean base is trying to focus on um but it certainly helps to bring it down not not exactly the to the point where it's as cheap as no no duplicated contracts so again i guess like the the thing that so i think like everybody agrees that we want to lower at least by 10x so let's let's set that that bar at least on this call uh we'll see i mean if if anybody watching uh let us know but but uh i guess like yeah the kind of main question is like if if we lower more than 10x like there's a bunch of stuff that's coming to to reduce kind of forces patterns like yeah multi-c contracts will also get reduced so from my experience now sdg brings like at least uh like from from just like pretty basic contracts it like cuts out like 200-ish kilobytes and then the i mean and if if we also remove survey uh maybe even more um so it's it's probably like reasonably assumed it like a five five x reduction from the current contract size was known as cd a very minimally sized repo i did the no std setup that is linked and it reduced my contract size from 43 kilobytes to 37 kilobytes that was a very small just kind of dummy repo with like one function in it so i assume that scales quite a bit when you add more functionality into it but for just the dummy setup it only reduced mine by roughly six kilobytes off of an already very small contract so so the mains like the the the one that's in pure it still uses survey and third is pretty heavy so i i like pretty much unhooked cert i well i actually did not use that no std thing i just like uh went with raw rpc like raw sdk calls and without uh any json parser and it can lower and so like we just need a very minimal json parts that it works uh but anyway we can follow up on this like and and uh investigate that separately um right but but i think the the the question still stands here it's like somebody that does want to you know like upload the whole i don't know like one terabyte of data engineer so one question that i have is like actually how much gas will that take to store that in the first place right because like if it's out if it's actually like stone like uses up a lot of gas anyway i'm not sure this is actually like a really the problem for an attack like right now for for each bite we we are charging let me find the parameters like do you agree with this logic in general that i think our gas is like 1000 or 10 000 times cheaper to to to store for the current pricing so so if you do reduction the gas is still not as expensive to store like a gig of data terabyte is unlikely it'll probably like take longer so yeah so starting one bite well one bite is kind of hard starting the megabyte is i see it's only like one three hundreds three one hundredths of an here per megabyte all right yeah that's pretty small amount um is there any ways we can we can for now for example yeah i mean what uh danny's suggesting like can we change the price based on kind of how much i mean i know this kind of sucks but like as an intermediate solution uh change the size based on based on how much is taken then you create an auction right if you're changing the price back to the higher then you can allocate storage now yeah then you can sell it off well what's the like main problem outside of my mean base if it's only for mean base then there's clearly two solutions that we should tackle one is reusable contracts and we can try to follow this next second is minimize the size of the contract which we're already doing just focus on this a little more well i do think that kind of current experience is somewhat cumbersome right where you need to send around the near two calls and like account into into smart contracts as well for storage um like it's it's really easy and and i can tell you people don't do this and then you know their contracts get in the state where like it actually doesn't like the new transactions don't cover storage so somebody needs to like donate near into it to cover storage right um like that that is pretty you know like because it's not enforced it was kind of the model right it's like it's on the contract developer right now to handle it uh that that's what's why i kind of suggesting the third option of like burning like integrating into gas and burning it but like it needs it the cost needs to be way lower for that to actually work i mean you were you were doing that math at some point right otherwise like the gas costs are not mapping very well i would say this is not just mint-based i would say min bass are the ones that are being probably most impacted by this right now but for example for atomic swaps we're trying to write something where a new user is onboarded when they can't find existing near keys and to enable them to do atomic swaps i think it was like something like minimum five near based on the contract size which means that in order to enable anyone to come acquire near we have to figure out how to get every single user like almost 15 usd to like just let them come step in or even just like the cost of acquisition of like a user if we want to bring them through a traditional method of like a full near name into the ecosystem versus restricted key access is really really high so we're essentially like cutting off our legs well before we have started running an indirect implementation at the same time like our like accounts are not limited tfa is an issue but again it's addressed through reusable contracts account and like restricted access key is in one year you can deploy 50 of them but at restricted accounts per year that we don't get to acquire the user and bring them into the entire ecosystem it means we restrict them to that one application which again is fine for a use case but it's not our collective best interest as a protocol that isn't that would like not to be an empty playground so there's two ways to address this one of them is what you've gained upon suggested which is on cryptography side of the key it is like 2fa that only restricts for certain actions second is what matlocker is doing this keys on boarding where a restricted key is converted to a full access key or like full account by using implicit account models so both of them do not require any contract to be deployed and the storage is like limited to the size of the keys plus the base of the account yeah but i think that kind of the reasoning is and that like we want more like people using named accounts right because they are better for user experience you know they're better for like social um things as well and like we can yes we can use simplistic amps and we should but like anywhere where we can convert people into like named account like that that'll benefit right and right now like it is kind of like it is expensive and it's also like it's it's it's a it's it's it's complicated like that that's like from my perspective actually complicated there's a lot like like i think we're reducing by 10 you know it saves the sales expense um but the complicate is a complicated part but the other part we haven't discussed is you know if in case near you know price actually uh changes like the you know all of this kind of converted to stable currency is also would uh increase in price right so i mean that's the thing we we kind of try not to think of just because it's like not not really able to to analyze here but um like it is something we should be kind of considering as well yeah and in ethereum price changes dynamically right as the as the token itself does to prevent that crazy spike from happening so well if ethereum ethereum is a bad example because because of like they actually not like so so i mean that again the problem is people don't want to release storage like that's that's a real problem and like i actually think if if we acknowledge that right then like this is where we need to figure out like what's what's the next steps like i mean not not for right now but like how we're gonna be dealing with yeah massive amounts of storage that's going to be like accumulating uh so yeah so i mean useful storage then current model is kind of like doesn't address it if it's useless storage then the current model is kind of addresses it but useless you mean spam or or whatever users no it's just like storage that's not needed on chain and stuff like this it kind of discourages you to to create the storage but at the same time allows you to claim it back if it's you if you somehow recycle the same amount of space by replacing the existing storage which is like all fungible talking do and like some other contracts can be designed similarly but do you want to i mean this is like just adding more and more requirements on kind of contract developers that's what i'm saying it's like it's just more and more cumbersome to develop which is kind of what we're not supposed to do no it's a trade-off between the price like the price of this action towards the hardware so like on our side the compute and like writing into state is fairly cheap coming from like actual estimation of how much it's cost so the only limit is um validators nodes if you go like too crazy on the decreasing side then it means you'll need to do optimization of the storage very like very fast focus on state syncs and like stuff that actually like if you have a terabyte of data and one shard then you need to be able to access it very fast at the same time you need to make sure you can replicate it very fast for the new node like when you're on board in a node so you need to optimize network sync and access right now it's kind of slightly less of a problem but if you go this this pause um it will be like immediate problem that you'll need to start working on what do you mean by this pass like if you reduce let's say 100 x then you'll need to start focusing on terabyte of state very soon well how soon are we talking because like terabyte of state is it should start well you don't know right because people can just take the voltage i mean even with the current price this would take like a milli million near to to do right that that's that's a lot by you know by any definition android x or uh million of stake in here so if someone just wants to do it for kicks they they can well that's that's why like the second thing is yeah look with 100x i would suggest us to switch to burning it because like for for that that would be one of the reasons for sure and also like just making everything simpler and but i think that you know eugene's point was that that's the staking and burning doesn't make a difference if you if you're kind of doing it for attack purposes because like you're betting on on a failure not on uh that you know you just do it and then get it back so so like if we assume terabyte of actual you know useful storage because people are launching all kinds of stuff on near then i mean that that's kind of no but but they can't but they can't attack using a smaller amount on archival nodes as eugene suggested they just for example i i just like use uh 30 300 000 here and then i i just you said that i deployed that amount of data and then i deleted them but then the archival knows this is good yeah yeah but that i mean that's the current state of that that's currently like you can just store a lot of data in in the transaction in transactional data you can literally send like blobs of data and transactions like each transaction is a megabyte you know send a thousand transactions per block each one is a megabyte that's like a gigabyte well i don't know but that's not going to fit it can be addressed on like preventing query of archival nodes on like historic state for like execution this way you you only need to care about the staging for cable not so like giving blocks away so if you want to run this you also have to query indexer which is designed for like tracking some part of the state or like it it can be addressed like transactional state is different from executing something on historical note and demand and that's why you don't need to like keep it in state and that's why this current model makes sense because you only keep the current state and state like maybe last and blocks in state until it's finalized and then you delete it from your uh quick access and only keep it like as a view where you can shift from a past similar how gas turbo gas optimizes this uh yeah i think we're over time maybe you can yeah well so like i i do i do want to like what what like i mean get given i think eugene well and bowen potentially like are against hundred x reduction like what what would be a reasonable like if what do we need to commit to to make hundred x happen now uh for next whatever six six to twelve months on uh kind of on the back end improving to to make this happen i think solving contract size is one going 10x lower right now like on the next upgrade is also going to help and the third is making a replicated contract a priority which is like we discuss this and it's not as terribly like implementation-wise so well no i'm asking you if if if we do 100 x like you're saying we'll need to deal with like one terabyte you know storage syncing stuff like this that's kind of more like what what do we need to do to get 100x to to commit to doing 100x now well i i just don't see how like going now when some stuff is still unstable gonna help us in the short term like outside of the developer i think we'll have more problems on validators potentially if someone like just deploys four megabyte contract because it's cheap now oh it's too expensive to run problem like the megabyte contracts are becoming more expensive to run right so really well we actually have pretty good like currently we have somewhat linear independence in the contract size and yeah but that moves to deploy so you only pay once one thing i was going to suggest with the also at the replicating contract as a priority uh so every mint you do every nft you create you have to add storage right uh what we can do if we want to like fill stadium seats if we do ticketing for a stadium to deploy the contract if it's really tiny that's really nice but if we wanted to do it for a stadium we could do a thing called lazy minting where basically uh the customers mint and so they would pay the storage per piece so we don't want the stadium to have to pay hundreds of millions of dollars for the storage of minting all 300 000 tickets right but lazy minting would would kind of solve that that piece but the store deployment uh being ridiculously cheap would would really help so if that helps with the but you just move i mean the cost is still the same right you just moved it to page users i think like at least like from perspective of cost we want to reduce costs for everyone right i mean it's the same as like if like what matters like for eugene what matters doing right with the demo to get access keys and just accounts work right that actual access key paying for that is still like it's not zero right and i understand the point is just saying here's a way my brain is thinking about what our custom what customers are going to be going after and so you guys can take that information go well this is one of their ways of actually using this technology uh how would that help with storage processing so i appreciate it yeah yeah you're right um okay well i mean i think like we should go with 10x either way it sounds like everybody's on board uh i mean for specifically for mint base and like multisig and those things uh we should like definitely uh get the no std sdk going uh i i would really like to see like can we bump it yeah 200x reduction like what what would we need to handle like can we commit to like going just like partially stateless for example in a year or something like this uh to do that um because like i i i mean as as danny pointed out i think like right now is the time to definitely grow the usage versus like trying to be super uh you know careful versus like this hundred you can do twitter and chain and like you'll face problems of like people doing crazy stuff on chain that shouldn't be unchained so because of how cheap the compute we also like you have to bump the computer like excess of the storage on the gas price and that will affect other contracts because they rely on cheap rights like our functional token transaction will get more expensive if you bump the price of writing so right now overwriting is super cheap and this is why lots of transactions are possible but if you go that every operation increases storage per system storage then we'll have different issues danny is also asking if we can do something in the middle like 20x well it doesn't affect as much like we currently trading some stuff that should be addressed differently then you do you want to like do you have specific reason for 20x um yeah i think just in terms of a short term patch at least that brings down the acquisition numbers of new users and some of the smart contracts down to like previous more reasonable levels and it so it gives us a bit more of a patch so like that would bring like minting a contract down to like three and a half dollars getting a new user you know down to like under a dollar you know just it makes it makes the overall acquisition cost for new applications more reasonable again i don't i understand the challenges with um having to think about edge cases and attack vectors and the attack surface that we're opening so like i i respect that you're trying to keep that in mind um i just want to be an advocate for our developers and our users so that we don't end up keeping everyone out just by trying to be very vigilant about the bad actors that we also block the good actors from participating so i think that if we did this 20x reduction it would be a short-term patch for us to think about all the other challenges and all the other ways in which we can address this in a way that's reasonable and thoughtful and like protects the network long-term but also enables us to move forward and and to keep things manageable for the for the developers the applications and the users who are playing in the near space right now and who are building and who need a right now solution yeah i'm pretty sure that like we should address every problem with a proper solution instead of just like saying let's go cheaper like it's similar argument to gas where the price of gas can be dropped to zero because we have inflation and it brings a tax well i mean the gas is still like not like somewhat of an auction right so like as as there's more usage it will like the gas price will increase um so like it's just we have enough capacity and you know sharding is like allows to increase capacity on the on the meat from like you know they level to uh just like this okay well let let's rub this up uh we're definitely over time um i mean i like i'm i'm if we can justify 20x let's do that i think that's pretty kind of it still was was in reasonable of like calculations that were originally done on this uh and i mean yes we should definitely do the kind of longer term stuff that's not a question i think like it's worth thinking again like my main problem is the current like usage is cumbersome and the the way people use it is more to store kind of forever um then as a you know store store and remove kind of thing so but yeah let's let's let's go with just uh kind of this fixed reduction for now and then uh post what post a more longer term uh roadmap on this going forward people are asking actually what how long would it take to to introduce this it changed to 10x or like 20x yeah basically will go um the next uh upgrade that will do if if it's really urgent uh which i don't think so we can push it like you know almost immediately um but but i don't think it's like i i don't think that this is threatening the uh the house of the entire network at this point so uh yeah i suggest we wait until the next release so to explain that urgency a little bit uh our big pitch to all of our customers on ethereum was it's too expensive it costs over 80 to 80 to deploy a store when we go on test net or even mainnet and we deploy and it's the same price they're kind of kind of look at us and go uh why'd you do all that so so it's it's going to be a pr nightmare for us a really big one so there's there's urgency i mean does that urgency dictate that we should deploy tomorrow for example uh all right what's wrong like what is the like yeah what is the timeline you're looking at is it like are looking a week or like a day or like two weeks or what is so we're trying to release on test net on uh for east denver when we do our thing on saturday if everything goes the plan that's test nets that gives us a month of basically a couple weeks of testing the heck out of it then the audits will start two weeks after that and then may net month month and a half from now um but there's definitely going to be the pr piece of wow it actually deploys this it costs uh 40 near to deploy on testnet is not going to be the same on on main net and we'll have to say yes why did you guys move everything to near and we'll have to say well there's a plan to make it cheaper that's the okay so basically i'm just i'm hearing that uh uh we should deploy this within a month is that correct east denver is in four days that's when he's going to pr oh i see uh well then okay i'll have a discussion with eugene later today to see but but if i hear it right we need to deploy on test net within a day and then on the main net within the month right give or take yeah yeah i mean that's probably a month and a half uh as long as we can go well as long as yeah okay well we can definitely deploy this on tesla within four days uh or just oh thank god yes that'd be great because then we can see it in action and it won't be oh we think it might happen that would be okay okay yeah we can we can definitely yeah we can i i guarantee you we can deploy this within four days on tesla yeah you know how things always take longer than they should so a month month and a half to you know as it goes okay sounds cool yeah i mean the the main net is like you know atomic swaps and multi-sticks that's where that's who's waiting on it on main net kind of or i mean i don't know about atomic spells but multi-sticks are waiting on this right now so yeah we have atomic swaps coming in about four four to five weeks so it would be helpful to have that yeah i think i think a mouse is like quite reasonable uh we're going to have another release anyways so yeah okay sounds good all right thanks everyone uh thanks nate and thor for joining um and providing feedback yeah thanks for having us very clear thank you for having this minute awesome","transcript_synced_at":"2026-11-04T04:19:23.993Z","segments":[{"start_seconds":0,"text":"and we should be going live","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=0s"},{"start_seconds":18,"text":"and awesome all right hello everyone thanks for joining the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=18s"},{"start_seconds":25,"text":"discussion about storage costs uh we'll start kind of digging in","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=25s"},{"start_seconds":34,"text":"i will uh first start just kind of to cover the some context","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=34s"},{"start_seconds":40,"text":"and some details of uh proposals uh which were posted on the forum and then we can kind of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=40s"},{"start_seconds":49,"text":"discuss pros and cons for each one and if there are any other ideas that we haven't had uh surfaced yet is this turn right for","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=49s"},{"start_seconds":58,"text":"everyone okay","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=58s"},{"start_seconds":65,"text":"all right so just to give some context right um well why we're here is the costs have been you know rising","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=65s"},{"start_seconds":74,"text":"and became pretty high for some like specific use cases like uh people using multi-stake contracts","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=74s"},{"start_seconds":82,"text":"which deploys a contract on each account or some folks are planning to launch uh products where you kind of need to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=82s"},{"start_seconds":90,"text":"create contracts for each um like group of people for example right so we have like splitting dial","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=90s"},{"start_seconds":97,"text":"with you know each guild and each team or each squad even having their own kind of contract or i'm in base where","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=97s"},{"start_seconds":104,"text":"each store has its own contract and so the current costs are prohibitively high","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=104s"},{"start_seconds":111,"text":"for that type of uh also generally like account creation kind of becoming somewhat expensive and especially if people want to have fun that","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=111s"},{"start_seconds":119,"text":"if some developers want to fund this in an easy way or we want to have some kind of faucet kind of somewhat independent of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=119s"},{"start_seconds":129,"text":"this but contract development has been somewhat cumbersome around storage taking where you need to handle manually like how much money is coming in","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=129s"},{"start_seconds":137,"text":"how they like how much money coming in for storage versus how much money is coming in for like actual payment","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=137s"},{"start_seconds":145,"text":"and things like that and then in case of cross contract costs paying for storage isn't harder where you kind of need to like keep","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=145s"},{"start_seconds":152,"text":"track sometimes or like send pieces of storage uh sometimes you don't even know how much you get a","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=152s"},{"start_seconds":159,"text":"spend on storage later so you need to keep some amount uh and send some amount so they're just all","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=159s"},{"start_seconds":166,"text":"kind of cumbersome high level there's actually at least like as far as i know there's","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=166s"},{"start_seconds":173,"text":"three ways to pay for storage uh in general so there's what i call storage taking and that's what we use right now this is","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=173s"},{"start_seconds":181,"text":"when you lock some amount of money based token or other token on the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=181s"},{"start_seconds":188,"text":"on the account and a inflation of that amount or uh in some way pays for for the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=188s"},{"start_seconds":196,"text":"two validators or two whoever maintains the storage right so just to kind of the basic mask here right if you know","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=196s"},{"start_seconds":204,"text":"if you have 100 near staying for a year on this account this is 100 years it's not being delegated or lent which but like specifically","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=204s"},{"start_seconds":212,"text":"because you're not getting delegated the rewards that would before go to 600 near are distributed among validators and so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=212s"},{"start_seconds":220,"text":"it's a like somewhat indirect way but to kind of you know remove those coins from circulation","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=220s"},{"start_seconds":227,"text":"and from uh staking to accrue more value to others or to whoever is taking","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=227s"},{"start_seconds":235,"text":"so the second approach is storage trend so this is where uh if you store somewhat of bytes on an account","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=235s"},{"start_seconds":241,"text":"the your balance is like decreasing block by block by some money and this is actually","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=241s"},{"start_seconds":248,"text":"we have implemented before and it was a nightmare for a few reasons main one is because","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=248s"},{"start_seconds":256,"text":"your balance is constantly decreasing which is super weird to people also also tooling needs to kind of know about what's going on because like the balance that's on chain and","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=256s"},{"start_seconds":265,"text":"balance that's kind of like virtually computed is different and then the third option is just pay one time and you know store forever and this is kind of where","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=265s"},{"start_seconds":274,"text":"really all the existing blockchains are uh like excluding few and like it has a very you know","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=274s"},{"start_seconds":284,"text":"suboptimal obviously problems that when you pay once you and like kind of it only it only","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=284s"},{"start_seconds":293,"text":"covers costs right now and it doesn't cover costs and kind of perpetuity right but the data needs to be stored in","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=293s"},{"start_seconds":300,"text":"perpetuity and specifically for example ethereum has you know this model in like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=300s"},{"start_seconds":309,"text":"somewhat of like a form and you know the growth of state has been","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=309s"},{"start_seconds":316,"text":"pretty exponential actually uh this well this is about the super latest data but like uh the gross is pretty","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=316s"},{"start_seconds":323,"text":"exponential it actually like i think people analyze it's pretty you know a lot pretty well aligned with","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=323s"},{"start_seconds":332,"text":"kind of usage of the network and so people don't really release data people people don't really","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=332s"},{"start_seconds":339,"text":"kind of delete stuff so we have been running stores taking right for a few months","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=339s"},{"start_seconds":346,"text":"and i mean i don't have i'm collecting hard data on this but i don't think we have had that much deletion happening either","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=346s"},{"start_seconds":354,"text":"right so even though right we like the storage is pretty expensive we lock in here right the idea is that it will get released back if you delete","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=354s"},{"start_seconds":361,"text":"the storage rate but that doesn't really happen because like most of the stuff is not designed around that uh we have a lot of confusion around ui","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=361s"},{"start_seconds":370,"text":"where you know there's like two balances like this kind of this minimum balance that you cannot use which means there's like another balance which is what you can use","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=370s"},{"start_seconds":378,"text":"there's a total balance and we also have all the lock up stuff on top of it and staking so you actually have like a lot of concepts","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=378s"},{"start_seconds":385,"text":"just to explain like how much money you have right and that's been very confusing um and costs are high right uh","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=385s"},{"start_seconds":394,"text":"so this has been kind of a problem uh for for various reasons","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=394s"},{"start_seconds":401,"text":"so i want to mention kind of one of the approaches here which is like to approach to that problem like pay once you know start","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=401s"},{"start_seconds":409,"text":"forever is actually to not store data on validators right and like generally the cost right the cost of transaction","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=409s"},{"start_seconds":417,"text":"because it's a one-time thing it doesn't include the you know storage costs that kind of keep goes through in perpetuity and one way","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=417s"},{"start_seconds":425,"text":"to solve this is to actually not require validators to sort store it in perpetuity in the first place and this is actually kind of similar to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=425s"},{"start_seconds":433,"text":"saying well actually the valdez will not store much right or very little but we're going to push everything into kind of indexers and you know like nodes","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=433s"},{"start_seconds":442,"text":"that will index data outside of outside of consensus um there's still a lot of problems i mean the folks in the museum have been","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=442s"},{"start_seconds":450,"text":"working hard on kind of figuring out how to how to how to make this work uh i think the current approach generally is like you need to have","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=450s"},{"start_seconds":458,"text":"like full nodes or like application specific nodes which will produce blocks pretty much and then give them to validators so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=458s"},{"start_seconds":465,"text":"you kind of have a lot more like you know minor extracted value things which like kind of by design and a few other things","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=465s"},{"start_seconds":474,"text":"and at least like from my perspective one of the bigger issues is like it adds latency because now you kind of need to like as you're sending a transaction","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=474s"},{"start_seconds":482,"text":"it needs them to get all the data around this ideally it should be for formed in a block right away because otherwise like if you have two","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=482s"},{"start_seconds":490,"text":"transactions to the same contract right and like they may not merge because like the state is different and stuff like this so it needs to be formed in the block","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=490s"},{"start_seconds":498,"text":"and that blocks goes to validator and it goes back and the next transaction can only arrive right so um we're not going to discuss this","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=498s"},{"start_seconds":505,"text":"because it's like a major major change and still a lot of research needs to be done but i just have to mention this yeah actually uh we have a plug here we're going to do","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=505s"},{"start_seconds":514,"text":"a whiteboard session with uh downcrowd feast this thursday on uh stateless validation","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=514s"},{"start_seconds":521,"text":"uh and see what's uh ethereum's plan is for it yeah yeah that would be awesome to hear um but","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=521s"},{"start_seconds":529,"text":"we're not gonna at least like in short term i don't think we're gonna consider that but it's definitely super interesting as it solves some","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=529s"},{"start_seconds":538,"text":"kind of more conceptual problems cool is there any more context we should discuss actually before going to the proposals","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=538s"},{"start_seconds":546,"text":"well you can also mention pros of having states taken which uh","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=546s"},{"start_seconds":554,"text":"blockchain state is kind of reliable now and can be exploded so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=554s"},{"start_seconds":562,"text":"so you mentioned the problems but at the same time like if we say storage is free then we'll have problems","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=562s"},{"start_seconds":569,"text":"so we should say why we even did this yeah yeah so i think that the two benefits of storage taking right why we","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=569s"},{"start_seconds":577,"text":"did it in the first place and why like it is on on what i call on paper","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=577s"},{"start_seconds":583,"text":"really good idea is twofold right one is it's it's a little bit better alliance than the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=583s"},{"start_seconds":592,"text":"kind of usage patterns of the blockchain with uh like base tokens value accrual","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=592s"},{"start_seconds":599,"text":"so pretty much as people use for example c20s or other things right the storage usage expands","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=599s"},{"start_seconds":607,"text":"because you create new addresses and stuff like this and given you using these resources like for example on","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=607s"},{"start_seconds":614,"text":"ethereum right now it's not really proportionally kind of house the storage expense to value acre like to how much","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=614s"},{"start_seconds":621,"text":"helium is being spent on uh on doing this and so storage staging actually like you know proportionately locks that amount of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=621s"},{"start_seconds":629,"text":"uh like base stock and towards a kind of a bit of usage and kind of yeah the flip side of this guys","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=629s"},{"start_seconds":637,"text":"that's kind of controls how much uh storage you can spend and because it is charged in base token right","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=637s"},{"start_seconds":644,"text":"it has like a very explicit like as the storage amount increases the supply like available supply decreases so uh","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=644s"},{"start_seconds":652,"text":"there's also kind of a mechanism around to control that people will not store too much so this","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=652s"},{"start_seconds":660,"text":"is where the kind of the conflict comes in or like because yeah like from perspective of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=660s"},{"start_seconds":667,"text":"developer you want to store more from perspective of the system you want to kind of have somewhat a fixed amount of storage","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=667s"},{"start_seconds":674,"text":"in the system so yeah so there's","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=674s"},{"start_seconds":681,"text":"oh anything else actually i think some uh there is some other","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=681s"},{"start_seconds":690,"text":"which has a different system where there is a fixed total amount of storage and there is a storage","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=690s"},{"start_seconds":698,"text":"auction where there is a market where people can buy and sell the storage so the market","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=698s"},{"start_seconds":706,"text":"sets the price for storage yeah yeah so there is a pretty much a","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=706s"},{"start_seconds":714,"text":"way to kind of price storage more dynamically i think the when we explored this and when you know","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=714s"},{"start_seconds":721,"text":"looking at some of the implementations of this the main problem was that like people like to speculate on things and so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=721s"},{"start_seconds":730,"text":"the there was a lot of kind of unnatural things happening on those markets even though like presumably markets should be","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=730s"},{"start_seconds":737,"text":"somewhat uh you know they should be optimal but they're not right so people were buying out storage for no reason and then um","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=737s"},{"start_seconds":746,"text":"driving up the prices and like selling selling selling them at high price uh kind of just to make the resource more scarce","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=746s"},{"start_seconds":756,"text":"oh i'm not saying that this system works well i'm just saying that there is one more approach that this actually","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=756s"},{"start_seconds":763,"text":"actually exists yeah so you're right yeah yeah i mean yeah i think generally there's like a little bit more yeah mechanic mechanism","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=763s"},{"start_seconds":772,"text":"design approaches which are more yeah as evgeny said like around kind of creating auctions for for storage with","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=772s"},{"start_seconds":780,"text":"various models um and i think eos has been has been doing that model uh as as as","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=780s"},{"start_seconds":788,"text":"one example but yeah so so far it it's been hasn't been shown to uh work very well","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=788s"},{"start_seconds":797,"text":"so here's like few specific proposals that i would like to discuss but we can also discuss some something in between or","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=797s"},{"start_seconds":805,"text":"some other options so the simplest one is just decrease the cost by 10x and so uh kind of the reasoning there is","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=805s"},{"start_seconds":814,"text":"like the calculations i was doing was like based on specific parameters and there's definitely like a 10x","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=814s"},{"start_seconds":822,"text":"there to reduce as i was like rounding up a lot and uh like using a ver like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=822s"},{"start_seconds":828,"text":"kind of a one dollar near price as a benchmark and so this would you know reduce things","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=828s"},{"start_seconds":836,"text":"like by 10x so a like our accounts are 100 like basic account uh is hundred bytes so this is without keys","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=836s"},{"start_seconds":845,"text":"so that's uh that would go to one uh thousandth of a year and then kind of hey a somewhat standard 300 kilobyte contracts","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=845s"},{"start_seconds":854,"text":"would be three near all right so that definitely would help add to perspective what it is like for","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=854s"},{"start_seconds":863,"text":"two other metrics so we can compare this to average transaction like average function call for gas and","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=863s"},{"start_seconds":872,"text":"also the maximum state you can acquire if you like lock entire near supply into the storage","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=872s"},{"start_seconds":881,"text":"so currently we have fashion 10 terabyte for the total state that you can acquire if like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=881s"},{"start_seconds":889,"text":"full near is locked for storage and it was already slightly higher than","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=889s"},{"start_seconds":896,"text":"where it we wanted it to be but kinda unrealistic to reach","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=896s"},{"start_seconds":904,"text":"and this proposal essentially bumps it to 100 terabyte which is not completely insane but kind","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=904s"},{"start_seconds":912,"text":"of on a border of like getting hard to to exist if someone actually acquires so much storage which is","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=912s"},{"start_seconds":919,"text":"probably not realistic as well um and the second metric is the cost of a","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=919s"},{"start_seconds":927,"text":"transaction so the cost of a transaction uh it's better to measure in tera gas","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=927s"},{"start_seconds":935,"text":"so something basic is around 7.5 tera gas and 7.5 targets","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=935s"},{"start_seconds":943,"text":"is um i think it's around","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=943s"},{"start_seconds":952,"text":"0.0001 or like or seven years so it's like one ten thousand of a cent so it's like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=952s"},{"start_seconds":962,"text":"ten times cheaper um or like 100 times cheaper than 100 bytes right something like this","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=962s"},{"start_seconds":972,"text":"so so our cost to to write this data is probably around um from gas perspective is 100 times cheaper than","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=972s"},{"start_seconds":981,"text":"to persistently store it or like to keep it in locked state","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=981s"},{"start_seconds":988,"text":"if i'm not mistaken yeah i i was kind of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=988s"},{"start_seconds":996,"text":"didn't get to that math but yeah um yeah well one thing i i do want to mention that like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=996s"},{"start_seconds":1003,"text":"evaluating full supply map to storage is definitely not realistic because of this like if even if a lot if we","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1003s"},{"start_seconds":1011,"text":"imagine that half of the supply locked in that the price of near like just because of supply demand right will","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1011s"},{"start_seconds":1018,"text":"uh be like outweighing a lot right and i'm not even saying that you know we already have like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1018s"},{"start_seconds":1026,"text":"actually this the supply that's outside of staked and locked funds is pretty much","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1026s"},{"start_seconds":1034,"text":"almost non-existent in the first place at this point right so um","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1034s"},{"start_seconds":1043,"text":"but yeah i mean it's definitely definitely uh worth noting that yeah even with this with this lowering it's still pretty uh","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1043s"},{"start_seconds":1052,"text":"just the storing like you need to lock yeah about at least yeah 100 times more","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1052s"},{"start_seconds":1060,"text":"in years and a fee to yeah to actually like process it","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1060s"},{"start_seconds":1069,"text":"so so this goes into option number two we lowered by hundred x","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1069s"},{"start_seconds":1076,"text":"and i mean the justification here is mostly around like so this kind of these calculations","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1076s"},{"start_seconds":1083,"text":"don't really account for you know price decrease in storage um and like the calculations in general were like pretty basic","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1083s"},{"start_seconds":1091,"text":"um they don't also account for like contraction in the supply as well and kind of the like supply dividend there","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1091s"},{"start_seconds":1099,"text":"so this is really trying like i wouldn't say that the math at least like fully worked out and i have like a full model here","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1099s"},{"start_seconds":1107,"text":"but uh like kind of accounting for storage degrees right for example using models that we've used uh for like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1107s"},{"start_seconds":1115,"text":"future storage price and also uh do it like kind of evaluating contraction on the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1115s"},{"start_seconds":1123,"text":"on the support on the circulating supply as well so i think what is what is the what is roughly the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1123s"},{"start_seconds":1130,"text":"curve they're using for this um i mean are they expecting it to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1130s"},{"start_seconds":1139,"text":"exponentially become exponentially more uh expensive or like what you are yeah","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1139s"},{"start_seconds":1145,"text":"oh i said yeah i have a paper somewhere but yeah i think it's roughly like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1145s"},{"start_seconds":1152,"text":"uh it becomes cheaper like every two years or something uh so yeah some kind of exponential decrease in price","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1152s"},{"start_seconds":1160,"text":"to store one byte like for for hard drives per se uh i was actually trying to pull the data from like a cloud provider","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1160s"},{"start_seconds":1168,"text":"but like i only found ec2 uh costs not not the storage uh costs but anyway i mean we","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1168s"},{"start_seconds":1178,"text":"can like we can pull the mass more more accurately later i think like the negative side of this this actually becomes","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1178s"},{"start_seconds":1186,"text":"you know a little bit too cheap on the side of like really sliding the data right so now if you want to store","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1186s"},{"start_seconds":1194,"text":"you know something like 100 gigabytes right","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1194s"},{"start_seconds":1199,"text":"um this becomes now like only hundred thousand near right so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1199s"},{"start_seconds":1208,"text":"this and hundred gigabytes is kind of current state of helium uh and so this this","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1208s"},{"start_seconds":1214,"text":"is definitely cheap and from the perspective","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1214s"},{"start_seconds":1222,"text":"of um costs that like people would incur","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1222s"},{"start_seconds":1231,"text":"and you know definitely requires that we like start charting things reasonably quickly to to be able to to kind of capture it and","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1231s"},{"start_seconds":1240,"text":"i think this is where one of the kind of questions that you know we've been discussing a lot but haven't really finalized is like how to balance","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1240s"},{"start_seconds":1248,"text":"sharding on transactions throughput versus the shouting of storage so if there's a lot of kind of storage sitting there","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1248s"},{"start_seconds":1255,"text":"which is not super active uh but like it still makes sense to have shards which are","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1255s"},{"start_seconds":1262,"text":"reasonably evenly split on on the mountain storage","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1262s"},{"start_seconds":1270,"text":"now eugene you were pretty against the 100x decrease on a from perspective of the classes","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1270s"},{"start_seconds":1276,"text":"i mean like spamming as well yeah i think it's still like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1276s"},{"start_seconds":1284,"text":"i will probably comment after all three proposals from like my perspective but yeah i think it's just","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1284s"},{"start_seconds":1291,"text":"getting too dangerous to to go like straight 100 x all right well so i mean these three","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1291s"},{"start_seconds":1299,"text":"proposals are actually kind of in the like doesn't mean one doesn't actually exclude another uh so the third one is yeah later on on","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1299s"},{"start_seconds":1308,"text":"top of 100x to actually burn the fee and then","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1308s"},{"start_seconds":1314,"text":"like merge the charging into gas charge uh also this burn and charge can also be","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1314s"},{"start_seconds":1322,"text":"like separated so the reason for burning is actually first of all if we burn we can actually","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1322s"},{"start_seconds":1330,"text":"and for example not return it later this three this easily justifies uh","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1330s"},{"start_seconds":1336,"text":"reducing the cost by like 20 x um now it removes a lot of kind of mentalities that","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1336s"},{"start_seconds":1344,"text":"people want to free up the space like the benefit of fearing up to space some of it can be like worked back like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1344s"},{"start_seconds":1352,"text":"what ethereum has for example it gives like a gas charge back so like if you remove if you remove data","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1352s"},{"start_seconds":1359,"text":"you get back some gas while you're kind of running um and merging into gas is really about","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1359s"},{"start_seconds":1366,"text":"improving user experience right like the contract development is way too complicated and you know users and developers like paying","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1366s"},{"start_seconds":1374,"text":"two separate fees attaching balance plus gas fee is kind of complicated uh the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1374s"},{"start_seconds":1385,"text":"the yeah so like we can just increase the storage right operation cost uh and so that actually increases that","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1385s"},{"start_seconds":1392,"text":"like i actually did actually yeah i looked at the storage cost and like kind of increases that gas usage by 100x that specific operation","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1392s"},{"start_seconds":1403,"text":"to do that the i think the the kind of the important piece here is","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1403s"},{"start_seconds":1411,"text":"indeed the like that you know on one side costs are low but it's still actually like if you compare to our gas usage it's actually pretty high","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1411s"},{"start_seconds":1419,"text":"uh so it will bump your gas usage especially when you're deploying like large contracts or like upgrading large contracts um","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1419s"},{"start_seconds":1427,"text":"but it definitely improves kind of experience and you know is more familiar for developers for existing blockchain covers","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1427s"},{"start_seconds":1437,"text":"uh yeah and like one option around how to burn is actually like it can still be charged in","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1437s"},{"start_seconds":1444,"text":"near per bite so like not in gas actually uh so like to not be affected by cars price and your charge pack also can be in near","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1444s"},{"start_seconds":1452,"text":"per bite so just get like like we actually record what's the gas price for this transaction you just get back that amount of gas proportional to how","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1452s"},{"start_seconds":1460,"text":"much gas like how much gas price you paid for this gas uh and actually refunds right now would","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1460s"},{"start_seconds":1467,"text":"convert it back to near so this all kind of like this uh a few of these formulas in the in the proposal on the forum","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1467s"},{"start_seconds":1476,"text":"i mean that's pretty much it for the kind of the proposals i had that uh","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1476s"},{"start_seconds":1483,"text":"curious to hear what other folks are thinking about this so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1483s"},{"start_seconds":1492,"text":"also proposal zero like how how it actually addressed without this change basically","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1492s"},{"start_seconds":1500,"text":"so for the fungible token the way we designed it is we created explicit","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1500s"},{"start_seconds":1507,"text":"account registration api so basically any action you do on fungible token do not require you to pay for the storage","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1507s"},{"start_seconds":1516,"text":"except for initial registration and the way it works is like you have a standard that's saying how much","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1516s"},{"start_seconds":1525,"text":"tokens you need to attach for this particular contract if you want to start interacting with this so you first attach this","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1525s"},{"start_seconds":1533,"text":"amount by calling it that creates pre-allocates essentially your account storage","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1533s"},{"start_seconds":1540,"text":"assume that it's fixed so it would not work with like old contracts which has dynamic storage per","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1540s"},{"start_seconds":1549,"text":"user and then once you're done using it or you don't want to use it you can claim the storage","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1549s"},{"start_seconds":1557,"text":"bag by essentially closing your account on this contract um so this is kind of work around with","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1557s"},{"start_seconds":1566,"text":"the storage fees that you have for lockup the main reason here is if the account is","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1566s"},{"start_seconds":1575,"text":"one-to-many which means multiple users use this account or a contract which is majority of contracts supposed","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1575s"},{"start_seconds":1582,"text":"to be like this then to prevent kind of drainage storage and sible attack","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1582s"},{"start_seconds":1590,"text":"you require payment and this payment can originate either from you or from someone else","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1590s"},{"start_seconds":1598,"text":"who want to send you assets so it creates some problems still because now you need to think about","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1598s"},{"start_seconds":1605,"text":"whether you register it or not but this is usually done on the ui side an example is like unislope you go","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1605s"},{"start_seconds":1614,"text":"to unislocket checks that you register on a token that you want to receive and then if not","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1614s"},{"start_seconds":1621,"text":"it's saying you need to register first you register by like collocating your account then you can buy","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1621s"},{"start_seconds":1628,"text":"token by unisloop um that's a background like how it works","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1628s"},{"start_seconds":1636,"text":"right now but this still doesn't oh sure yeah but the problem is that still today","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1636s"},{"start_seconds":1644,"text":"deploying contracts very expensive right and the example is the 2fa like people just have","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1644s"},{"start_seconds":1651,"text":"35 like near lot there that's very suboptimal yeah yeah so and","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1651s"},{"start_seconds":1658,"text":"for replicated contract the solution would be you have some kind of reusability","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1658s"},{"start_seconds":1665,"text":"of a contract binaries where the asset is like the storage that needed to store it","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1665s"},{"start_seconds":1674,"text":"and deploy it burned once instead of locked and then all like validator will be able to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1674s"},{"start_seconds":1683,"text":"keep it in cash and you will be just reference it by hash","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1683s"},{"start_seconds":1691,"text":"um so this piece is far from being done but that will address the problem","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1691s"},{"start_seconds":1699,"text":"of replicated contracts such as tf2fa lockup contracts um maybe nfts or","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1699s"},{"start_seconds":1707,"text":"anything that is sharing the same instance of the code","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1707s"},{"start_seconds":1713,"text":"so if we do this uh are you suggesting that uh there is no need to reduce the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1713s"},{"start_seconds":1721,"text":"storage cost uh i'm suggesting that potentially yes i still like the idea of dropping a 10x","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1721s"},{"start_seconds":1730,"text":"i don't see immediate abuse if you just go 10x because it's still expensive to store","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1730s"},{"start_seconds":1738,"text":"large amounts not as expensive as right now and may create certain problems originally but","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1738s"},{"start_seconds":1745,"text":"still um reasonably priced","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1745s"},{"start_seconds":1755,"text":"but i'm saying that like longer term we hopefully will arrive in a place where contracts are cheaper by optimizing some","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1755s"},{"start_seconds":1763,"text":"stuff maybe not only know std but rewriting certain libraries or rust contracts are cheaper assembly scripts contracts are already","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1763s"},{"start_seconds":1771,"text":"relatively cheap and given 10x reduction will go to like a very low","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1771s"},{"start_seconds":1780,"text":"price per unique contract for non-unique contracts it's still fine because it's a one-time payment","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1780s"},{"start_seconds":1789,"text":"once you finalize your binary pretty much yeah i mean i think that contract is","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1789s"},{"start_seconds":1796,"text":"definitely kind of the current big pain but i think the you know the user storage is also there it just kind of slides a little bit unnoticed because of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1796s"},{"start_seconds":1805,"text":"the contract storage um like the problems we've seen right with wallet is when people","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1805s"},{"start_seconds":1813,"text":"kept adding access keys so i think generally like this the pattern is that people are not","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1813s"},{"start_seconds":1821,"text":"and not ex like hey nate the the the people are not","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1821s"},{"start_seconds":1829,"text":"uh planning to remove stuff from blockchain right it's not like it's not a pattern of like you know using limited capacity on blockchain","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1829s"},{"start_seconds":1837,"text":"it's more about you know keep storing more stuff i think that's kind of the one of the problems where like the storage taken is","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1837s"},{"start_seconds":1845,"text":"just like misaligned with how people are using the stuff um i just also wanted to share the this is","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1845s"},{"start_seconds":1853,"text":"rviv's pricing data so it's yeah it's not really exponential but","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1853s"},{"start_seconds":1861,"text":"it's kind of i feel like there's like actually uh a change in in opening progression yeah so like i wouldn't","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1861s"},{"start_seconds":1868,"text":"but but alex also shared the costs for aws and so this looks uh","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1868s"},{"start_seconds":1876,"text":"it went down to x over the course of seven years on the slide from december 2000 into december 2016.","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1876s"},{"start_seconds":1882,"text":"yeah i do want to mention this is a s3 which is a little bit different from like like i was looking for ebs which is like active because i","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1882s"},{"start_seconds":1890,"text":"definitely agree that like a long-term storage like we there can be doing a lot of stuff there but like active storage that validators need to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1890s"},{"start_seconds":1898,"text":"have hot for executing code that's that's i feel like we're limiting factor i i mean um i i don't","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1898s"},{"start_seconds":1907,"text":"know whether alex is alluding to this but ceremony has this regency's plan which it's like they kind of just having the basically separation of hot story","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1907s"},{"start_seconds":1915,"text":"from code storage and validator only need to keep like uh started from uh some recent number of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1915s"},{"start_seconds":1922,"text":"blocks and then they kind of relocate the old data to some code storage that they have some separate providers and you can","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1922s"},{"start_seconds":1930,"text":"curry promising well so this is good i mean this is going into like more stateless stuff i","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1930s"},{"start_seconds":1938,"text":"think in general like i think like if if going stateless it makes sense to keep the hot stuff like contracts and some like recent data on validators anyway and then just yeah","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1938s"},{"start_seconds":1947,"text":"like leave all the stuff that hasn't been touched in a year to some to somebody else uh so like you actually need to include in a","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1947s"},{"start_seconds":1954,"text":"transaction data if it's like too old so i just want to jump in and see from the developer relations perspective","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1954s"},{"start_seconds":1962,"text":"having those high storage costs is really crippling for developers in terms of growth and user adoption and i think that like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1962s"},{"start_seconds":1969,"text":"this is not the time to be like limiting the data when we're just in the in the first small stage like maybe if you want to set some kind","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1969s"},{"start_seconds":1978,"text":"of like rail guards so that like insane things don't go in there so like at a certain point it kind of starts going exposing i think that's fine","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1978s"},{"start_seconds":1985,"text":"but like i'm in favor of the like 100x reduction because i think that we should be encouraging people to build we should be encouraging users to come on we shouldn't say like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1985s"},{"start_seconds":1993,"text":"if you want to play with something on here by the way it's ten dollars to like step into the ring right and that was never the original intent but somehow we landed here","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=1993s"},{"start_seconds":2001,"text":"and i think this is a way to address that in the immediate future and uh just to follow up on what danny","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2001s"},{"start_seconds":2009,"text":"said this is nema uh in the case of mint base right now i think it's like over 70 yeah 75","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2009s"},{"start_seconds":2017,"text":"to deploy so even a hundred x reduction is a significant fraction of one dollar like i i one would expect near","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2017s"},{"start_seconds":2026,"text":"to cos actually lower than that so uh even 100x is not like i think it's not enough but","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2026s"},{"start_seconds":2033,"text":"it's it's more manageable right it's because of the model of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2033s"},{"start_seconds":2040,"text":"replicated contracts so yeah i think farming basis probably","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2040s"},{"start_seconds":2049,"text":"uh yeah having having global storage would yeah basically just address this problem","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2049s"},{"start_seconds":2059,"text":"yeah so uh thor for mint base here as context we're currently deploying contracts","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2059s"},{"start_seconds":2066,"text":"that require 35 near in a factory pattern so that translates to roughly 70 dollars i worked a bit today on integrating the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2066s"},{"start_seconds":2074,"text":"no standard pattern um but it looks like that's still in development because it doesn't look like it's completely covering features so for instance","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2074s"},{"start_seconds":2082,"text":"cross-contract calls are not currently uh something that you can do with the no standard pattern so we're looking for all sorts of ways to reduce contract","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2082s"},{"start_seconds":2090,"text":"size yeah so i mean like generally why i","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2090s"},{"start_seconds":2103,"text":"would suggest going for 100x right now is actually more reasoning that they need provided that's like right now we're not really","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2103s"},{"start_seconds":2112,"text":"kind of need to be perfect and we just we just we just need to deliver developer experience","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2112s"},{"start_seconds":2119,"text":"uh and then start like we can we can improve things like how exactly we're storing stuff","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2119s"},{"start_seconds":2126,"text":"like on the back end over time like again stateless stuff or whatever can be improved","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2126s"},{"start_seconds":2132,"text":"um over time on the back end as as the need of that arises um and i mean that's kind of why i","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2132s"},{"start_seconds":2141,"text":"mentioned stateless validation in general because like i do think it's a solution and i mean in some shape or form","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2141s"},{"start_seconds":2149,"text":"it's not needed right right now because like that's not actually a problem in the first place","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2149s"},{"start_seconds":2158,"text":"bowen you were talking about that we'll have it all solved when we figure out the cross uh the reduplicate of code","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2158s"},{"start_seconds":2165,"text":"that would help a lot but i also read that it kind of uh doesn't really work with sharding is that i need to read more um","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2165s"},{"start_seconds":2172,"text":"but i mean it works with it works with starting it's just that uh we're it's not i mean we have not started","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2172s"},{"start_seconds":2181,"text":"implementing anything anyway i think the design is not finalized i would say so uh it might","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2181s"},{"start_seconds":2189,"text":"take us some time to get there it's definitely not the uh the fastest thing for example we can do the storage price reduction in a very","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2189s"},{"start_seconds":2197,"text":"short period of time but for this it would take much longer got it so no std is not an option for us","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2197s"},{"start_seconds":2205,"text":"but we will have to wait for you guys to manually reduce the storage cost just as a hard-coded thing","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2205s"},{"start_seconds":2213,"text":"and then this duplicate code might get fixed maybe a year from now is that oh i think i think it's less than a year but it would","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2213s"},{"start_seconds":2220,"text":"uh probably be several months yeah i mean no it's nice to do i'm assuming i mean we'll like just get the features you","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2220s"},{"start_seconds":2229,"text":"need as well so i mean if there's like specific things that doesn't work it's just like missing implementation of the couple","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2229s"},{"start_seconds":2237,"text":"functions in sdk yeah i think thor was saying we just implemented the no std on a side project just to see if what","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2237s"},{"start_seconds":2245,"text":"would it would reduce and it reduced us by like one one year token oh no uh we can't even compile that","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2245s"},{"start_seconds":2252,"text":"that's correct uh that's incorrect because we we wouldn't be able to implement it because the store contract currently uses cross-contract features","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2252s"},{"start_seconds":2260,"text":"which are not currently implemented in the near sdk pure uh library so i i don't have an estimate about how","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2260s"},{"start_seconds":2268,"text":"much it would reduce the store that is safe to say but it could reduce it by a lot by little it's just that","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2268s"},{"start_seconds":2276,"text":"the uh no standard library is not complete it was here","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2276s"},{"start_seconds":2283,"text":"i mean from my poop well all right i mean there's a separate like like if you reduce storage under attacks","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2283s"},{"start_seconds":2292,"text":"you cannot bump it back it will break contracts that already there so that's why i don't uh advocate for trying to go 100","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2292s"},{"start_seconds":2302,"text":"x because if we like if people start abusing it us for for doing this uh then it will","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2302s"},{"start_seconds":2310,"text":"be um very unfortunate state where potentially we cannot really","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2310s"},{"start_seconds":2317,"text":"address it with any like reasonable actions such as like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2317s"},{"start_seconds":2325,"text":"the only solution is like delete this contract bump the storage break don't have contracts do a hard fork so you don't want to go there","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2325s"},{"start_seconds":2334,"text":"yeah i agree with that so can we actually talk more about the the last proposal of burning","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2334s"},{"start_seconds":2342,"text":"uh just burning tokens for storage might be interesting idea","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2342s"},{"start_seconds":2351,"text":"well i mean i think it still includes a kind of 100x reduction so like from perspective of like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2351s"},{"start_seconds":2359,"text":"if we're thinking of an attack surface right the the attack is the same it's just like instead of locking that money","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2359s"},{"start_seconds":2366,"text":"you're kind of burning that money so that's the difference but i think that's a pretty big difference if you're talking about that","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2366s"},{"start_seconds":2373,"text":"much money like whether you just lose it or it's just temporarily locked","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2373s"},{"start_seconds":2381,"text":"no no even if temporary lock it's not a problem like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2381s"},{"start_seconds":2388,"text":"you can create a state you can grow like archival state right now","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2388s"},{"start_seconds":2395,"text":"much cheaper so it's a persistent state so you you're basically locking it until the network goes down","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2395s"},{"start_seconds":2403,"text":"like whatever it is so it does it will not address like you you may not be able to recover the funds so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2403s"},{"start_seconds":2411,"text":"so from this perspective it's not does not attack change the vector of tech","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2411s"},{"start_seconds":2419,"text":"so can we address the challenge that eugene brought up about the abuse of storage because i think the legitimate use of storage for projects","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2419s"},{"start_seconds":2427,"text":"versus the potential for abuse are two different subjects so is there a way for us to set up a threshold or some guard rails","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2427s"},{"start_seconds":2435,"text":"to protect the network from that kind of abuse maybe with an exponential rise in the cost after a certain stage while still giving","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2435s"},{"start_seconds":2443,"text":"enough space for developers and users to benefit from a significant rate reduction so like a","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2443s"},{"start_seconds":2452,"text":"tiered approach a general answer is no because if","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2452s"},{"start_seconds":2460,"text":"someone wants to abuse they will be abusing the chip storage and the expensive storage will be left to the to the good people","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2460s"},{"start_seconds":2476,"text":"also like prioritizing work on like replicated storage i think will bring huge benefits and differentiator for","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2476s"},{"start_seconds":2485,"text":"near it essentially creates um not stateless but it allows us to have like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2485s"},{"start_seconds":2494,"text":"incredible stuff that we haven't tried using before short-term going 10x does not fully","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2494s"},{"start_seconds":2504,"text":"solve the problem of like creating this type of replicated contract that mean base is trying to focus on","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2504s"},{"start_seconds":2513,"text":"um but it certainly helps to bring it down not not exactly the to the point where","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2513s"},{"start_seconds":2521,"text":"it's as cheap as no no duplicated contracts","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2521s"},{"start_seconds":2536,"text":"so again i guess like the the thing that so i think like everybody agrees that we want to lower at least by","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2536s"},{"start_seconds":2544,"text":"10x so let's let's set that that bar at least on this call uh we'll see i mean if if anybody watching","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2544s"},{"start_seconds":2554,"text":"uh let us know but but uh i guess like yeah the kind of main question is like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2554s"},{"start_seconds":2562,"text":"if if we lower more than 10x like there's a bunch of stuff that's coming to to reduce kind of forces patterns like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2562s"},{"start_seconds":2570,"text":"yeah multi-c contracts will also get reduced so from my experience now sdg brings like at least","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2570s"},{"start_seconds":2576,"text":"uh like from from just like pretty basic contracts it like cuts out like 200-ish","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2576s"},{"start_seconds":2584,"text":"kilobytes and then the i mean and if if we also remove survey uh","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2584s"},{"start_seconds":2590,"text":"maybe even more um so it's it's probably like reasonably assumed","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2590s"},{"start_seconds":2597,"text":"it like a five five x reduction from the current contract size was known as cd","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2597s"},{"start_seconds":2605,"text":"a very minimally sized repo i did the no std setup that is linked and it reduced my","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2605s"},{"start_seconds":2612,"text":"contract size from 43 kilobytes to 37 kilobytes that was a very small just kind of dummy repo with like one","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2612s"},{"start_seconds":2620,"text":"function in it so i assume that scales quite a bit when you add more functionality into it but for just the dummy setup it only reduced mine","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2620s"},{"start_seconds":2628,"text":"by roughly six kilobytes off of an already very small contract","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2628s"},{"start_seconds":2635,"text":"so so the mains like the the the one that's in pure it still uses survey and third is pretty heavy","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2635s"},{"start_seconds":2642,"text":"so i i like pretty much unhooked cert i well i actually did not use that no std thing i just","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2642s"},{"start_seconds":2650,"text":"like uh went with raw rpc like raw sdk calls and without uh any json parser and it can","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2650s"},{"start_seconds":2658,"text":"lower and so like we just need a very minimal json parts that it works uh but anyway we can follow up on this like and and","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2658s"},{"start_seconds":2665,"text":"uh investigate that separately um right but but i think the the the question still stands here it's like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2665s"},{"start_seconds":2674,"text":"somebody that does want to you know like upload the whole i don't know like one terabyte of data","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2674s"},{"start_seconds":2682,"text":"engineer so one question that i have is like actually how much gas will that take to store that in the first place right because like if it's out if it's","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2682s"},{"start_seconds":2690,"text":"actually like stone like uses up a lot of gas anyway i'm not sure this is actually like a","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2690s"},{"start_seconds":2696,"text":"really the problem for an attack","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2696s"},{"start_seconds":2710,"text":"like right now for for each bite we we are charging let me find the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2710s"},{"start_seconds":2720,"text":"parameters like do you agree with this logic in","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2720s"},{"start_seconds":2730,"text":"general that i think our gas is like 1000","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2730s"},{"start_seconds":2737,"text":"or 10 000 times cheaper to to to store for the current pricing","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2737s"},{"start_seconds":2745,"text":"so so if you do reduction the gas is still not as expensive to store like a gig of data","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2745s"},{"start_seconds":2754,"text":"terabyte is unlikely it'll probably like take longer","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2754s"},{"start_seconds":2772,"text":"so yeah so starting one bite well one bite is kind of hard starting the megabyte is i see it's only","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2772s"},{"start_seconds":2781,"text":"like one three hundreds three one hundredths of an here","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2781s"},{"start_seconds":2788,"text":"per megabyte all right yeah that's pretty small amount","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2788s"},{"start_seconds":2798,"text":"um is there any ways we can","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2798s"},{"start_seconds":2806,"text":"we can for now for example","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2806s"},{"start_seconds":2817,"text":"yeah i mean what uh danny's suggesting like can we change the price based on kind of how much i mean i know this kind of sucks but like as an","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2817s"},{"start_seconds":2825,"text":"intermediate solution uh change the size based on based on how much is taken","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2825s"},{"start_seconds":2833,"text":"then you create an auction right if you're changing the price back to the higher then you can allocate","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2833s"},{"start_seconds":2840,"text":"storage now","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2840s"},{"start_seconds":2848,"text":"yeah then you can sell it off well what's the like main problem outside of my","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2848s"},{"start_seconds":2856,"text":"mean base if it's only for mean base then there's clearly two solutions that we should tackle","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2856s"},{"start_seconds":2862,"text":"one is reusable contracts and we can try to follow this next second is minimize the size of the contract which","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2862s"},{"start_seconds":2871,"text":"we're already doing just focus on this a little more well i do think that kind of current experience is somewhat cumbersome","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2871s"},{"start_seconds":2879,"text":"right where you need to send around the near two calls and like account into into smart contracts as well","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2879s"},{"start_seconds":2886,"text":"for storage um like it's it's really easy and and i can tell you people don't do this and then you know their contracts get in","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2886s"},{"start_seconds":2894,"text":"the state where like it actually doesn't like the new transactions don't cover storage so somebody needs to like donate","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2894s"},{"start_seconds":2900,"text":"near into it to cover storage right um like that that is pretty","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2900s"},{"start_seconds":2907,"text":"you know like because it's not enforced it was kind of the model right it's like it's on the contract developer right now to handle it","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2907s"},{"start_seconds":2916,"text":"uh that that's what's why i kind of suggesting the third option of like burning like integrating into gas and burning it","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2916s"},{"start_seconds":2923,"text":"but like it needs it the cost needs to be way lower for that to actually work i mean you were you were doing that math at some point right","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2923s"},{"start_seconds":2931,"text":"otherwise like the gas costs are not mapping very well","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2931s"},{"start_seconds":2938,"text":"i would say this is not just mint-based i would say min bass are the ones that are being probably most impacted by this right now but for example for atomic swaps we're","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2938s"},{"start_seconds":2947,"text":"trying to write something where a new user is onboarded when they can't find existing near keys and to enable them to do atomic swaps i","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2947s"},{"start_seconds":2955,"text":"think it was like something like minimum five near based on the contract size which means that in order to enable anyone to come acquire near we have to figure out how","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2955s"},{"start_seconds":2963,"text":"to get every single user like almost 15 usd to like just let them come step in or","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2963s"},{"start_seconds":2971,"text":"even just like the cost of acquisition of like a user if we want to bring them through a traditional method of like a full near name into the ecosystem","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2971s"},{"start_seconds":2978,"text":"versus restricted key access is really really high so we're essentially like cutting off our legs well before we have started running","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2978s"},{"start_seconds":2988,"text":"an indirect implementation at the same time like our like accounts are not limited tfa","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2988s"},{"start_seconds":2996,"text":"is an issue but again it's addressed through reusable contracts account and like restricted access key","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=2996s"},{"start_seconds":3005,"text":"is in one year you can deploy 50 of them but at restricted accounts per year that we don't get to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3005s"},{"start_seconds":3012,"text":"acquire the user and bring them into the entire ecosystem it means we restrict them to that one application which again is fine for a use case but","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3012s"},{"start_seconds":3020,"text":"it's not our collective best interest as a protocol that isn't that would like not to be an empty playground so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3020s"},{"start_seconds":3029,"text":"there's two ways to address this one of them is what you've gained upon suggested which is on cryptography side of the key","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3029s"},{"start_seconds":3036,"text":"it is like 2fa that only restricts for certain actions second is what matlocker is doing this keys on boarding","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3036s"},{"start_seconds":3043,"text":"where a restricted key is converted to a full access key or like full account by using implicit account models so both","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3043s"},{"start_seconds":3052,"text":"of them do not require any contract to be deployed and the storage is like limited to the size of the keys plus the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3052s"},{"start_seconds":3060,"text":"base of the account yeah but i think that kind of the reasoning is and that like we want more like people using","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3060s"},{"start_seconds":3069,"text":"named accounts right because they are better for user experience you know they're better for like social","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3069s"},{"start_seconds":3076,"text":"um things as well and like we can yes we can use simplistic amps and we should but like anywhere where we can convert","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3076s"},{"start_seconds":3084,"text":"people into like named account like that that'll benefit right and right now like it is kind of like it is expensive and","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3084s"},{"start_seconds":3094,"text":"it's also like it's it's it's a it's it's it's complicated like that that's like from my perspective actually complicated there's a lot like like i think we're reducing by 10 you","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3094s"},{"start_seconds":3103,"text":"know it saves the sales expense um but the complicate is a complicated part but the other part we haven't discussed","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3103s"},{"start_seconds":3110,"text":"is you know if in case near you know price actually uh changes like the you know all of this kind of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3110s"},{"start_seconds":3120,"text":"converted to stable currency is also would uh increase in price right so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3120s"},{"start_seconds":3127,"text":"i mean that's the thing we we kind of try not to think of just because it's like not not really","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3127s"},{"start_seconds":3134,"text":"able to to analyze here but um like it is something we should be kind of considering as well","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3134s"},{"start_seconds":3144,"text":"yeah and in ethereum price changes dynamically right as the as the token itself does to prevent that crazy spike from happening","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3144s"},{"start_seconds":3152,"text":"so well if ethereum ethereum is a bad example because because of like they actually not like so so i mean that again","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3152s"},{"start_seconds":3160,"text":"the problem is people don't want to release storage like that's that's a real problem and like i actually think if if we","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3160s"},{"start_seconds":3167,"text":"acknowledge that right then like this is where we need to figure out like what's what's the next steps like i mean","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3167s"},{"start_seconds":3175,"text":"not not for right now but like how we're gonna be dealing with yeah massive amounts of storage","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3175s"},{"start_seconds":3182,"text":"that's going to be like accumulating uh so yeah so i mean useful storage then","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3182s"},{"start_seconds":3189,"text":"current model is kind of like doesn't address it if it's useless storage then the current","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3189s"},{"start_seconds":3197,"text":"model is kind of addresses it but useless you mean spam or or whatever users no it's just like storage that's not","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3197s"},{"start_seconds":3207,"text":"needed on chain and stuff like this it kind of discourages you to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3207s"},{"start_seconds":3217,"text":"to create the storage but at the same time allows you to claim it back if it's you if you somehow recycle the same amount","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3217s"},{"start_seconds":3226,"text":"of space by replacing the existing storage which is like all fungible talking do","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3226s"},{"start_seconds":3234,"text":"and like some other contracts can be designed similarly but do you want to i mean this is like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3234s"},{"start_seconds":3243,"text":"just adding more and more requirements on kind of contract developers that's what i'm saying it's like it's just more and more cumbersome to develop","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3243s"},{"start_seconds":3250,"text":"which is kind of what we're not supposed to do no it's a trade-off between the price","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3250s"},{"start_seconds":3259,"text":"like the price of this action towards the hardware so like on our side the compute and like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3259s"},{"start_seconds":3267,"text":"writing into state is fairly cheap coming from like actual estimation of how much it's","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3267s"},{"start_seconds":3274,"text":"cost so the only limit is um validators nodes","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3274s"},{"start_seconds":3283,"text":"if you go like too crazy on the decreasing side then it means you'll need to do optimization of the storage very like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3283s"},{"start_seconds":3291,"text":"very fast focus on state syncs and like stuff that actually like if you have a terabyte of data and one shard then you need to be able to access it","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3291s"},{"start_seconds":3300,"text":"very fast at the same time you need to make sure you can replicate it very fast for the new node like when","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3300s"},{"start_seconds":3307,"text":"you're on board in a node so you need to optimize network","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3307s"},{"start_seconds":3313,"text":"sync and access","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3313s"},{"start_seconds":3321,"text":"right now it's kind of slightly less of a problem but if you go this","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3321s"},{"start_seconds":3328,"text":"this pause um it will be like immediate problem that you'll need to start working on","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3328s"},{"start_seconds":3335,"text":"what do you mean by this pass like if you reduce let's say 100 x then you'll need to start focusing on","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3335s"},{"start_seconds":3344,"text":"terabyte of state very soon well how soon are we talking because like terabyte of state is it should","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3344s"},{"start_seconds":3351,"text":"start well you don't know right because people can just take the voltage i mean even with the current price this would take like a","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3351s"},{"start_seconds":3359,"text":"milli million near to to do right that that's that's a lot by you know","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3359s"},{"start_seconds":3367,"text":"by any definition android x or uh","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3367s"},{"start_seconds":3376,"text":"million of stake in here so if someone just wants to do it for kicks they they can well that's that's why like the second thing is yeah look","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3376s"},{"start_seconds":3384,"text":"with 100x i would suggest us to switch to burning it because like for for that that would be one of the reasons for sure","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3384s"},{"start_seconds":3392,"text":"and also like just making everything simpler and but i think that you know eugene's point","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3392s"},{"start_seconds":3399,"text":"was that that's the staking and burning doesn't make a difference if you if you're kind of doing it for attack purposes because","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3399s"},{"start_seconds":3407,"text":"like you're betting on on a failure not on uh that you know you just do it and then get it back","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3407s"},{"start_seconds":3415,"text":"so so like if we assume terabyte of actual you know useful storage because people are launching all kinds of stuff on near then","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3415s"},{"start_seconds":3423,"text":"i mean that that's kind of no but but they can't but they can't attack using a smaller amount on archival nodes as eugene suggested","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3423s"},{"start_seconds":3432,"text":"they just for example i i just like use uh 30 300 000 here and then i i just you said that i deployed that amount of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3432s"},{"start_seconds":3440,"text":"data and then i deleted them but then the archival knows this is good yeah yeah but that i mean that's the current","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3440s"},{"start_seconds":3448,"text":"state of that that's currently like you can just store a lot of data in in the transaction in transactional data you can literally","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3448s"},{"start_seconds":3457,"text":"send like blobs of data and transactions like each transaction is a megabyte you know send a thousand transactions per block","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3457s"},{"start_seconds":3464,"text":"each one is a megabyte that's like a gigabyte well i don't know but that's not going to fit it can be addressed on like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3464s"},{"start_seconds":3472,"text":"preventing query of archival nodes on like historic state for like execution this way you you only","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3472s"},{"start_seconds":3480,"text":"need to care about the staging for cable not so like giving blocks away so if you want to run this","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3480s"},{"start_seconds":3486,"text":"you also have to query indexer which is designed for like tracking some part of the state or like it it can be addressed like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3486s"},{"start_seconds":3494,"text":"transactional state is different from executing something on historical note and demand","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3494s"},{"start_seconds":3502,"text":"and that's why you don't need to like keep it in state and that's why this current model makes sense because you only keep","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3502s"},{"start_seconds":3509,"text":"the current state and state like maybe last and blocks in state until it's finalized and then","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3509s"},{"start_seconds":3517,"text":"you delete it from your uh quick access and only keep it like as a view where you can shift from a past similar","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3517s"},{"start_seconds":3526,"text":"how gas turbo gas optimizes this uh yeah i think we're over time maybe","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3526s"},{"start_seconds":3534,"text":"you can yeah well so like i i do i do want to like what what like i mean get given i think eugene well and bowen potentially","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3534s"},{"start_seconds":3542,"text":"like are against hundred x reduction like what what would be a reasonable like if what do we need to commit","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3542s"},{"start_seconds":3551,"text":"to to make hundred x happen now uh for next whatever six six to twelve months on uh kind of on the back end improving to to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3551s"},{"start_seconds":3560,"text":"make this happen i think solving contract size is one going 10x lower right now like on the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3560s"},{"start_seconds":3569,"text":"next upgrade is also going to help and the third is making a replicated contract","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3569s"},{"start_seconds":3575,"text":"a priority which is like we discuss this and it's not as terribly like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3575s"},{"start_seconds":3584,"text":"implementation-wise so well no i'm asking you if if if we do 100 x","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3584s"},{"start_seconds":3593,"text":"like you're saying we'll need to deal with like one terabyte you know storage syncing stuff like this that's kind of more like what what do we need to do to get","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3593s"},{"start_seconds":3601,"text":"100x to to commit to doing 100x now","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3601s"},{"start_seconds":3609,"text":"well i i just don't see how like going now when some stuff is still unstable gonna help us in the short term like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3609s"},{"start_seconds":3618,"text":"outside of the developer i think we'll have more problems on validators potentially if someone like just deploys four","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3618s"},{"start_seconds":3626,"text":"megabyte contract because it's cheap now oh it's too expensive to run problem","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3626s"},{"start_seconds":3636,"text":"like the megabyte contracts are becoming more expensive to run right so really well we actually have pretty good","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3636s"},{"start_seconds":3645,"text":"like currently we have somewhat linear independence in the contract size and yeah but that moves to deploy","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3645s"},{"start_seconds":3653,"text":"so you only pay once one thing i was going to suggest with the also at the replicating","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3653s"},{"start_seconds":3661,"text":"contract as a priority uh so every mint you do every nft you create you have to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3661s"},{"start_seconds":3668,"text":"add storage right uh what we can do if we want to like fill stadium seats if we do ticketing for a stadium","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3668s"},{"start_seconds":3676,"text":"to deploy the contract if it's really tiny that's really nice but if we wanted to do it for a stadium we could do a thing called lazy minting where","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3676s"},{"start_seconds":3683,"text":"basically uh the customers mint and so they would pay the storage per piece so we don't want the stadium to have to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3683s"},{"start_seconds":3691,"text":"pay hundreds of millions of dollars for the storage of minting all 300 000 tickets right but lazy minting would would kind of solve that that piece but","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3691s"},{"start_seconds":3700,"text":"the store deployment uh being ridiculously cheap would would really help so if that helps with the but you just move","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3700s"},{"start_seconds":3708,"text":"i mean the cost is still the same right you just moved it to page users i think like at least like from perspective of cost","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3708s"},{"start_seconds":3717,"text":"we want to reduce costs for everyone right i mean it's the same as like if like what matters like for eugene what matters doing right with the demo","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3717s"},{"start_seconds":3725,"text":"to get access keys and just accounts work right that actual access key paying for that is still","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3725s"},{"start_seconds":3733,"text":"like it's not zero right and i understand the point is just saying here's a way my brain is thinking about what our custom what customers are going to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3733s"},{"start_seconds":3741,"text":"be going after and so you guys can take that information go well this is one of their ways of actually using this technology","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3741s"},{"start_seconds":3748,"text":"uh how would that help with storage processing so i appreciate it yeah yeah you're right","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3748s"},{"start_seconds":3757,"text":"um okay well i mean i think like we should go with 10x either way it sounds like everybody's on board uh i mean for","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3757s"},{"start_seconds":3765,"text":"specifically for mint base and like multisig and those things uh we should like definitely uh get the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3765s"},{"start_seconds":3772,"text":"no std sdk going uh i i would really like to see like can we bump it yeah","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3772s"},{"start_seconds":3780,"text":"200x reduction like what what would we need to handle like can we commit to like going just like partially stateless for example in a","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3780s"},{"start_seconds":3788,"text":"year or something like this uh to do that um because like i i i mean","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3788s"},{"start_seconds":3796,"text":"as as danny pointed out i think like right now is the time to definitely grow the usage versus like trying to be super","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3796s"},{"start_seconds":3804,"text":"uh you know careful versus like this hundred you can do twitter and","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3804s"},{"start_seconds":3812,"text":"chain and like you'll face problems of like people doing crazy stuff on chain that","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3812s"},{"start_seconds":3818,"text":"shouldn't be unchained so because of how cheap the compute we also like you have to bump the computer like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3818s"},{"start_seconds":3825,"text":"excess of the storage on the gas price and that will affect other contracts because they rely on cheap rights","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3825s"},{"start_seconds":3834,"text":"like our functional token transaction will get more expensive if you bump the price of writing so right now","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3834s"},{"start_seconds":3842,"text":"overwriting is super cheap and this is why lots of transactions are possible but if you go that every operation","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3842s"},{"start_seconds":3852,"text":"increases storage per system storage then we'll have different issues","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3852s"},{"start_seconds":3860,"text":"danny is also asking if we can do something in the middle like 20x well it doesn't affect as much like we","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3860s"},{"start_seconds":3869,"text":"currently trading some stuff that should be addressed differently","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3869s"},{"start_seconds":3881,"text":"then you do you want to like do you have specific reason for 20x um yeah i think just in terms of a short","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3881s"},{"start_seconds":3889,"text":"term patch at least that brings down the acquisition numbers of new users and some of the smart contracts","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3889s"},{"start_seconds":3896,"text":"down to like previous more reasonable levels and it so it gives us a bit more of a patch so like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3896s"},{"start_seconds":3903,"text":"that would bring like minting a contract down to like three and a half dollars getting a new user you know","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3903s"},{"start_seconds":3912,"text":"down to like under a dollar you know just it makes it makes the overall acquisition cost for new applications","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3912s"},{"start_seconds":3920,"text":"more reasonable again i don't i understand the challenges with um having to think about edge cases and attack vectors","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3920s"},{"start_seconds":3929,"text":"and the attack surface that we're opening so like i i respect that you're trying to keep that in mind um i just","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3929s"},{"start_seconds":3936,"text":"want to be an advocate for our developers and our users so that we don't end up keeping everyone out just by","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3936s"},{"start_seconds":3943,"text":"trying to be very vigilant about the bad actors that we also block the good actors from participating so i think that if we did this 20x","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3943s"},{"start_seconds":3952,"text":"reduction it would be a short-term patch for us to think about all the other challenges and all the other ways in which we can address this in a way","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3952s"},{"start_seconds":3960,"text":"that's reasonable and thoughtful and like protects the network long-term but also enables us to move forward and and to keep things manageable for","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3960s"},{"start_seconds":3968,"text":"the for the developers the applications and the users who are playing in the near space right now and who are building and","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3968s"},{"start_seconds":3977,"text":"who need a right now solution yeah i'm pretty sure that like we should","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3977s"},{"start_seconds":3987,"text":"address every problem with a proper solution instead of just like saying let's go cheaper like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3987s"},{"start_seconds":3996,"text":"it's similar argument to gas where the price of gas can be dropped to zero because we have inflation","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=3996s"},{"start_seconds":4002,"text":"and it brings a tax well i mean the gas is still like not like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4002s"},{"start_seconds":4010,"text":"somewhat of an auction right so like as as there's more usage it will like the gas price will increase um","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4010s"},{"start_seconds":4019,"text":"so like it's just we have enough capacity and you know sharding is like allows to increase capacity on","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4019s"},{"start_seconds":4027,"text":"the on the meat from like you know they level to uh just like this okay well let let's rub","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4027s"},{"start_seconds":4034,"text":"this up uh we're definitely over time um i mean i like i'm i'm if we can justify 20x let's do that i","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4034s"},{"start_seconds":4042,"text":"think that's pretty kind of it still was was in reasonable of like calculations that were","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4042s"},{"start_seconds":4050,"text":"originally done on this uh and i mean yes we should definitely do the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4050s"},{"start_seconds":4058,"text":"kind of longer term stuff that's not a question i think like it's worth thinking again like","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4058s"},{"start_seconds":4066,"text":"my main problem is the current like usage is cumbersome and the the way people use it is more","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4066s"},{"start_seconds":4073,"text":"to store kind of forever um then as a you know store store and remove kind of thing so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4073s"},{"start_seconds":4082,"text":"but yeah let's let's let's go with just uh kind of this fixed reduction for now","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4082s"},{"start_seconds":4090,"text":"and then uh post what post a more longer term uh roadmap","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4090s"},{"start_seconds":4098,"text":"on this going forward people are asking actually what how long would it take to to introduce","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4098s"},{"start_seconds":4105,"text":"this it changed to 10x or like 20x yeah","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4105s"},{"start_seconds":4116,"text":"basically will go um the next uh upgrade that will do if if it's really urgent uh which i don't think so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4116s"},{"start_seconds":4124,"text":"we can push it like you know almost immediately um but but i don't think it's like i i don't think that","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4124s"},{"start_seconds":4132,"text":"this is threatening the uh the house of the entire network at this point so uh yeah i suggest we wait until the next","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4132s"},{"start_seconds":4141,"text":"release so to explain that urgency a little bit uh our big pitch to all of our customers on ethereum was it's too expensive it costs","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4141s"},{"start_seconds":4150,"text":"over 80 to 80 to deploy a store when we go on test net or even mainnet and we deploy and it's the same price","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4150s"},{"start_seconds":4158,"text":"they're kind of kind of look at us and go uh why'd you do all that so so it's","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4158s"},{"start_seconds":4166,"text":"it's going to be a pr nightmare for us a really big one so there's there's urgency","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4166s"},{"start_seconds":4173,"text":"i mean does that urgency dictate that we should deploy tomorrow for example uh all right what's wrong like what is","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4173s"},{"start_seconds":4181,"text":"the like yeah what is the timeline you're looking at is it like are looking a week or like a day or like two weeks or what is so","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4181s"},{"start_seconds":4189,"text":"we're trying to release on test net on uh for east denver when we do our thing on saturday if everything goes the plan","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4189s"},{"start_seconds":4197,"text":"that's test nets that gives us a month of basically a couple weeks of testing the heck out of it then the","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4197s"},{"start_seconds":4204,"text":"audits will start two weeks after that and then may net month month and a half from now um but there's definitely going to be the pr piece of","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4204s"},{"start_seconds":4212,"text":"wow it actually deploys this it costs uh 40 near to deploy on testnet is not going to be the same on on main net and we'll have to say yes","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4212s"},{"start_seconds":4220,"text":"why did you guys move everything to near and we'll have to say well there's a plan to make it cheaper that's the okay so basically i'm just","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4220s"},{"start_seconds":4228,"text":"i'm hearing that uh uh we should deploy this within a month is that correct east denver is in four days that's when he's going to","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4228s"},{"start_seconds":4236,"text":"pr oh i see uh well then okay i'll have a discussion with eugene later today to see but but","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4236s"},{"start_seconds":4244,"text":"if i hear it right we need to deploy on test net within a day and then on the main net within the month right give or take yeah yeah i mean that's probably a month and a half","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4244s"},{"start_seconds":4253,"text":"uh as long as we can go well as long as yeah okay well we can definitely deploy this on tesla within four days","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4253s"},{"start_seconds":4260,"text":"uh or just oh thank god yes that'd be great because then we can see it in action and it won't be oh we think it might happen that would be okay okay yeah we can we can definitely","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4260s"},{"start_seconds":4269,"text":"yeah we can i i guarantee you we can deploy this within four days on tesla yeah you know how things always take longer than they should so a month month","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4269s"},{"start_seconds":4277,"text":"and a half to you know as it goes okay sounds cool yeah i mean the the main net is like you","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4277s"},{"start_seconds":4285,"text":"know atomic swaps and multi-sticks that's where that's who's waiting on it on main net kind of or i mean i don't know about atomic spells but multi-sticks","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4285s"},{"start_seconds":4293,"text":"are waiting on this right now so yeah we have atomic swaps coming in about four four to five weeks so it would be","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4293s"},{"start_seconds":4301,"text":"helpful to have that yeah i think i think a mouse is like quite reasonable uh we're going to have another release","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4301s"},{"start_seconds":4308,"text":"anyways so yeah okay sounds good all right thanks everyone uh thanks nate","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4308s"},{"start_seconds":4316,"text":"and thor for joining um and providing feedback yeah thanks for having us very clear thank you for having this minute","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4316s"},{"start_seconds":4324,"text":"awesome","url":"https://www.youtube.com/watch?v=Ge1Cfq-0y5w&t=4324s"}],"youtube_url":"https://youtube.com/watch?v=Ge1Cfq-0y5w","url":"https://near-pedia.xyz/videos/Ge1Cfq-0y5w"}