Staking on the Hedera Network

Oh welcome to gossip about gossip. Powered by Hedera Hashgraph. And each episode will cut through the hype of blockchain promises and explore real world examples of organizations creating the next generation of decentralized applications, which will bring trust back to the internet for us all. OK Lehman we have just closed out what I think is probably one of the most talked about Hedera improvement proposals, and this is around staking on the network. Can you talk a little bit about what that hip process has been like? Yes so the hip process, the Hedera improvement process process is it's redundant, but the hip process has been great. I have loved watching all these hips coming out and the community is proposing hips and in the community discusses these hips and then eventually the council votes on accepting them. The committee, the committee. It has been great, but wow, I have not seen anything like the staking one. The staking one has had an enormous amount of community discussion and proposals and suggestions for changes and then discussions on those. And then people talking about pros and cons and implications and with the long term effects of different things will be this is the way a community should be working. I love it. And I think what we ended up with excuse me, as the consensus of the community is a good system. And so we are going to be doing staking at least if it gets funded, but we're going to be implementing the code for it. And I just can't wait. This is very cool. And so with all of that back and forth, can you walk us through what the final proposal looks like? Yes so the idea is you can have each of the nodes, the Mainnet nodes has weight in the consensus according to how much stake is staked to it, how many bars are officially staked to that node? Proportional to that is, how much weight it has in the consensus. And so if more than 2/3 of the steak is owned by honest nodes, we're all good. But if a full third of the stake can be owned by dishonest nodes and they can attack the network, that's how it works. There's even a cool theorem that says you can't do any better than that. That is the best you can do. And we do that. Well So steak is important. It is critical to the security of the network. And what we have not done in the past is had a program where people can stake and get rewards for staking, but they are directly helping the security of the network. And so it is important that people be rewarded for staking. We want to encourage them to do a good job of choosing who to stake for and for staking. And so the way this will be set up is that we will say each person, if you have an account, you can do a transaction that says, I want to stake my account. I want to stake it to this node here. And then officially your account is staked to that node. And so you can do that. And then you still have Total Access to your account. Your account is totally liquid. There is no lockup, there is no bonding. You simply have access to your account. And once a day at midnight UTC. We take a snapshot of how much is in all the accounts and that affects the weight of the node for the next 24 hours till the next snapshot. So at the beginning of Monday, whatever your balance is, that's how much you are contributing to the weight of whatever node you're staking to throughout the whole day on Monday. And then at the beginning of Tuesday at midnight UTC, we take another snapshot and we reshuffle the weights. You then are rewarded on Tuesday or some time later. You are rewarded for having been staked on Monday and you're rewarded proportional to how much you are staking. And so if you stake more, you get more rewards. Now, if you stick to a node that didn't actually run all day on Monday, they crashed halfway through Monday and didn't run the rest of the day. Then you don't get any rewards for Monday. It's a day by day thing, but on any day when the node is down, you don't get rewarded. So you should pick reliable nodes. This is encouraging people to reward the encouraged the good nodes and to give them more weights and then they're going to be up. So it helps the network keep running, too. So it's not just security, it's also reliability. You're helping with both. So as long as that node was up all day, then you get rewarded. And in the hip it's phrased as a certain fraction of rounds in which that node creates a famous witness, a judge. And that's probably how it ended up being implemented. But that was the example in the hip. So then the question is, what if I am a company, I have a game and I have lots of players in the game and I want them all to have accounts on Hedera so they can use bars for their game. Or I'm an exchange and I want to have one giant account where all my customers put their bars into that account and then that one big account earns rewards. I would like to have the ability and just don't have to but have the ability for them to not have to send all their bars to me, for me to stake for them. I would like them to be able to stake indirectly through me. So this is interesting. Your account, if you want, could be staked to a node or your account if you want, could be staked to another account. And then that other account, if it is linked to a node indirectly, you end up giving that node weight. And if you do this, all of the rewards for your staking go to the account that you staked for. So all those rules I said and how you get rewarded, those rewards accrue to that account. And then presumably whoever has that account is giving you some benefit. They're letting you play their game or there being a DEX for you or I don't know what they're doing if they want to. They could even send you bars if they wanted to. They could just pass on the reward to you. But that's between them and you. So you can stick to an account. The account sticks to a node and then the account gets the rewards. And it's up to you in the account to decide how you want to divide up your rewards. You can even write a wallet. Maybe you write a wallet for people, and you give them the ability to go into the wallet and stick to nodes. But if you have users of your wallet software that don't care enough to even go in and stick to a node, why not have it by default staked to an account owned by you? Because you wrote the wallet software and maybe lots of people will stick to you. You'll actually earn money. Maybe that helps incentivize you to build a good wallet. And so good solid wallet software could be subsidized by this indirect staking model. Kind of a cool concept. So the indirect staking model is something that is, I think, novel to us. And not locking you up is very novel. That's very much what people don't do. And then you can get rewarded. So you get the rewards. The rewards do not come from the Hedera council, that the rewards come directly from account 800. So account 800 is an account that the council can't touch. The council cannot pull bars out of account 800 and once a day, account 800 rewards or figures out how much it's going to reward everybody. And it keeps track of how much it's rewarding everybody. Interesting fact. If you do stake to a node or have somebody staked to you as you stick to the node, you get rewarded for each day, 24 hour periods, but you don't get the reward immediately. You get the reward the next time you touch that account. So the next time you do a transaction and you pay for it from this account, this account gets all the rewards it has earned up to that point. Or the next time someone transfers bars to you, this account will get the rewards. Or the next time something indirectly changes a setting on that account. This account gets the rewards. So any time we touch this account, it gets all the rewards it has built up. And with our expiration we finally have entity expiration working. Entities have to renew every three months anyway. The account will be automatically renewing itself every three months. Even that will trigger your rewards. So if you do absolutely nothing and no one in the world touches your account, you'll still get your reward quarterly every quarter. You'll get all the rewards you earn during that quarter. So you don't have to do anything. But if you use your account normally, then you get your rewards slightly faster, which is kind of cool. And of course, they compound. I've had some people have said, oh, no, if I only get rewarded quarterly, that's going to be vastly worse than compounding every day. It turns out if you look at the formula for compounding, it's almost the same. It's just a couple of different percentage difference. It's not like you earn 7% instead of 6% It's like you earn 6.1% instead of 6% That's what I mean. A small percentage of the 6% difference percentages are weird. Anyway, if you are doing stuff with your account, you compound your rewards every day. But if you do it quarterly, you compound almost as fast. Quarterly interest is almost as good as daily interest. But it's up to you. You do whatever you want and and hopefully you're actually using your account. And so because it's not or transactions aren't expensive, this isn't designed for you to just put your stuff there. Let it sit there forever. You're designed to be able to do things with it. And as long as you're doing things, you get rewarded. Now, you could also ask what happens if everyone in the world all stick to the same node? That node could end up with a third or more of all the stake or 100% of the steak. Couldn't that node then destroy the network? Yes, that's right. So we have a cap. Each node has a maximum amount that can be stick to it. And if you stake over that, you can do so. But you don't get rewarded for that extra and it doesn't have any effect on the voting. And so the nodes have weights proportional to what steak to them, but anything over the max doesn't count. Anything under the Min doesn't count. If a node can't get the certain minimum amount, then the node just goes to 0 weight and has no effect on consensus. And so when we do sharding, we'll break things up to make sure that every shard is balanced. Before we have sharding, we at least when our whole network to be pretty balanced, we don't want one whale to be able to destroy the whole network. New at least have to have the nail whale running several nodes, malicious nodes to run through in the network. So it raises the bar on how hard it is to attack the network. So we have the max, we have the Min. If you're below the Min for that note, if the total to that note is below the Min, then the node has none of its stakes count. No one gets rewarded for sticking to it. And if you're over the max, then by over whatever amount you're over the max, that little bit of what you're staking doesn't count and doesn't get rewarded and it doesn't affect the weight. Other than that you get proportional to what you staked if you were directly staking, and if other people are indirectly staking through you, you get rewarded according to them as well, which is kind of nice. That is staking. It will be rolling out soon in the software and it also has a threshold that says it will start actually rewarding the stakers until account 800 reaches a certain threshold. Now once we reach that threshold, once staking continues for eternity. It's only when it starts. It only controls when it starts. And so we will roll out the software to the network. But staking will not start staking rewards will not start until we also get the threshold met. And so we're looking forward to that and hopefully people will fund that account. We'll see if that happens. If not, then the account will be funded from transactions and so a certain fraction of transaction fees goes into that account and a certain fraction goes into Hedera to support Hedera and a certain fraction even goes to the node that you submitted through. So you get to choose which node you submit through, and that node gets paid a very tiny fraction of the transaction fee for having done the work of submitting it for you, which is nice of them. That is staking. I haven't even covered all the little details we have to round off to the nearest bar just to stop rounding errors from killing us. So if you stake a million and a half bars, you only get credit for a million bars. Sorry what else? I think that pretty much is it. It, will you be able to so each account can only stake to one node? Or can you divide your account balance to stick to multiple nodes? One account can stick to only one node or can stake to one account that stakes to one node. Either way, if you want to stake to multiple nodes, then break it up into multiple accounts. Lots of people break up their accounts into multiple accounts. Then if you want to stick to multiple nodes, you can do that. Hedera itself is going to do that. It has a big Treasury that it wants to stake. It's going to break it up into multiple accounts and each account will just stick to one node. So if you stake to another account and you've talked about the fact that you have to the account has to have some activity for the rewards to be delivered. Does that mean that your end account has to have some activity or the account that you have staked has to have some activity to earn the rewards? If you stick to an account that is then sticking to a node, you get no rewards. That account gets a reward every time its balance changes. And every time your balance changes. And every time the owner of that account changes one of the parameters on the account, lots of things trigger a reward. Got it. Well, I think the community is very excited for staking and like you said, lots of engagement on that hip. So we are glad to see all of that input that has been put into that. Lehman I know there have been some security concerns around staking on other networks. Any key features that you want to address there? No, I'm just going to say that we're doing this carefully. Things like the Min and the max are to make sure that you don't have one computer and one node being a whale. That's not a panacea. Nothing in this industry is a panacea, but it's helpful. We've done lots of little things that make things helpful. The distinctions of us, I would say, that we are not locking. There is no bonding, there's also no slashing. We do not steal coins from you for punishing you for doing something bad. There's no bonding or slashing and and it's very liquid. So you can use your coins at any time you want and you can do this indirect, which is a possible business model, could be part of a possible business model for someone monetizing a game or wallet software or an exchange or a million other things or examples where they might want to have you indirect stake through them. And I really like that because otherwise they would be incentivized to not let you hold your own coins. And I would say it is better if all of these different people would allow their customers to have their own account and the customer has the key to their own account, rather than putting it all in one big pot owned by the central player. And they often do that because they can make money on DeFi and things, but now they can make money on staking without having to take your key. And I would just. Anyway, I feel much better if everyone has their own key and we don't just all give our key to a central pot. OK Lehman last year, we rolled out scheduled transactions, but we are now talking about longer term scheduled transactions. Can you talk about the differences there and what that will mean and what it will allow applications to do? I love this feature. I hope every Leger will someday implement this feature. So we have multisig. You can have a transaction with multiple signatures on it that you submit to the network and you have to have multiple people sign it in order for it to go through. We actually have hierarchical, multi layer, multi sig, which is no one else does that. And we can talk about that if you want. But we have multisig. Every ledger has multi sig and the hard part of multi sig is you have to go collect signatures from all the people and we do that. We have a way of doing that. But wouldn't it be cool if you could just submit your transaction to the network and then all those individual people could submit their signatures to the network? And when it collects enough signatures, then the transaction happens and we implemented that for a half hour. You can submit a transaction without all the signatures and then it can just sit there and collect signatures for a half hour and all the people that need to sign it can log in and sign it over that half hour and then it executes. That's cool. That's a nice way to do multi sig. We don't have to be emailing each other back and forth or having our computers connect with each other. It can be in some sense more decentralized because it's the decentralized network itself. It's collecting the signatures for you, but that's only a half hour. What we would really like is that you can submit a scheduled transaction to the network scheduled to actually execute in the far distant future, could execute a month from now. And then all those people that need to sign it have a whole month to submit a transaction that adds their signature. And then you can set it up either when it gets enough signatures, it executes immediately or the reason it's called scheduled transactions. You can say I am scheduling this for a certain time and date a month from now. When we get to that point, if it has received enough signatures, it executes then or shortly thereafter, and if it hasn't got enough signatures, then it disappears and it failed. And so we don't fail after a half hour. We fail after whatever date you scheduled. This is fun. There are lots of use cases for this, but the one that I think is just fun, which is isn't the one the business guys always talk about the biz dev guys talk about. But it's one that I think is fun is that you can actually now use a scheduled transaction. Where these transaction that you have scheduled calls a smart contract and executes it. And while that smart contract is running, it can do all sorts of activities. And, you know, one of the activities it can do is to create a schedule transaction for next month. And then it stops. It ends. It ends its call and it goes quiet. But then next month, the scheduled transaction comes to life and calls the smart contract, which runs and then schedules the one for next month and then goes to sleep. Smart contracts can be alive. They can call themselves you can write a SMART contract where you have, say, a bond that you have to pay off the bond holders once a month or something like that. And you can have a smart contract programmed to pay everybody once a month, and no human being has to be involved in triggering it every month. It triggers itself. It comes alive and does something every month on its own. In fact, if it had a way of earning money, it could earn enough money to keep itself alive forever. It could keep doing this for eternity. It could always be earning enough that it can afford to do the transaction, the scheduled transaction, which isn't all that expensive. And then once a month it would do that and it would run itself once a month and it would pay for its own transaction fee to run once a month. This is cool. Now people have done this as L to's. People have built other networks whose sole purpose is to call, say, Ethereum periodically and call it smart contract. But isn't it cool to have it in the ledger itself? In the 01 itself, the smart contract can actually be a living thing that has its own resources, can earn its own resources, can manipulate its own resources, and can call itself once a month. I love that idea. You can imagine. For a DAO you want certain things to happen periodically. You don't have to rely on any human being to call it. It just happens. You could owe mortgage payments to the bank and over time and you just make it automatically pay them for you. There's just all sorts of things that you can talk about when you have the ability for turning complete smart contracts to now be alive, where they can call themselves and can come to life at scheduled times on their own. This is just immensely fun. So I'm hyped about this and we'll have this rolling out and I can't wait to see the creative ways that people will use this. Oc we've given you smart contracts that can be alive. Now what are you going to do with it? I can't wait. All right. Thank you, Lehman, for joining us today. And Thank you to our audience for joining us on our latest episode.