Recap 6/23 Sui AMA: Tokenomics with Alonso de Gortari

This recap covers an AMA on Sui tokenomics with Alonso De Gortari.

Recap 6/23 Sui AMA: Tokenomics with Alonso de Gortari

Introduction

Jen: Welcome to our AMA regarding tokenomics. I’ll hand the mic to our guest Alonso De Gortari and let him introduce himself.

Alonso: I’m super excited about this. It’s been almost a month since we released our tokenomics materials on the website. I think this is the first time we’ve had the opportunity to talk about this with the broader community and we’ll surely do this many times in the future.

I’m really excited to hear everyone’s thoughts and make sure that the tokenomics of Sui make the most sense for everyone.

My training was in economics, sort of the traditional economics. I went to grad school, did a PhD, worked on a bunch of supply chain issues during my PhD thesis, then did a post-doc with a professor of economics. At some point, I realized that what was happening out there in the real world is much more interesting, so I decided to join the very early stages of Libra and Facebook. Later, I was designing the Libra stable coin economic incentives around the Libra blockchain; now in many ways, it’s a train wreck compared to what we’re doing with Sui.

So, about six months ago I left Facebook to join Mysten labs, and ever since, I’ve become very much in love with the project. Building at Sui, I can understand how we really have the potential to be a game-changer in terms of blockchain. It actually delivers use cases. I can board the next billion users to web3, and it took a lot of design and engineering to make sure that we can view the Sui blockchain so that it is up to the same mechanics and economic point of view as the engineering of the blockchain.

.   .   .

Question #1: What’s your view on tokenomics and blockchain?

Alonso: In many ways, tokenomics is one of those parts of blockchain ecosystems that describe how different things work. It’s also a part of blockchains that are least developed, at least from a research point of view. So while it’s true that if you look at how different blockchains were in the engineering sense, especially regarding capabilities of L1s, if you look at it from building distributed systems, many of the engineering computer science models are available so you can figure out what you’re doing, what these networks are providing make sense or not. Tokenomics is not there at all if you look at the research, and you can even make very simple issues look not-so-simple, but things that should be well known like proof of work or proof of stake, the economics results are very behind.

So this is sort of both a good thing and a bad thing for tokenomics. It’s good, because it means that there’s a lot of room to improve the way that tokenomics work. A lot of these blockchain ecosystems, in terms of innovation, you can do tricks to solve new problems. For example, think about storage with early blockchains. They were so simple in their functioning that storage wasn’t even something you had to think about, which means that on the tokenomics front, you didn’t really have to think about economically: How you would create the right incentives to price and address those storage capabilities? What would that cost for future participants in those networks?

In Sui, for example, we can handle a large amount of on-chain storage, which means from a tokenomics perspective, we really need to sit down and think very seriously about how we’re going to address that. If we leave these issues unaddressed, that could potentially end up creating really bad dynamics on the network. For example, people might start storing a lot of data on the chain without paying for it, but someone ultimately has to pay for that storage. That creates a bad network externality that can end up killing your network. So tokenomics needs to take this into account. That’s sort of the good thing because there’s a lot of things that we can still think about, and there’s a lot of innovation being done to address these new problems.

The risky part of it is that tokenomics, as a field of research, is so nascent. There’s a lot of risk in what we do. If we change 10 things at the same time, we don’t know what their effects are going to be. Think of a simple analogy like a cake recipe. If you’re baking a cake and change one ingredient, you can likely predict what’s going to happen to the cake. But if you change 10 ingredients, then you might end up with an amazing cake that’s better than anything you’ve ever had or a disaster cake. It’s hard to figure out what’s going to happen. Tokenomics is a bit like that. Once you start opening up the design space and going across all these other verticals, you can dive into: how gas price mechanisms work, how the proof of storage mechanism works, how the storage dynamics work, and so on.

The more you start tweaking it, the possibility for creating a richer and more fulfilling ecosystem is very big, but also the risk is higher. From a high level point of view, the people that are designing tokenomics (us included), these are some of the risk/return trade-offs that we’re making. When we’re thinking about how to design a tokenomics framework that makes sense for Sui, we’re building one to the best of our ability that is somewhat stress-tested so we can forecast what it’s going to do.

Question #2: What’s special about Sui Tokenomics?

Alonso: That’s a great question. That’s a great question. I feel like right now crypto is becoming such a large space and it’s hard to keep track of everything that’s happening. There are a lot of things that are randomly emerging and they’re working in very different ways. So what’s special about Sui is that tokenomics are built very much in sync with the innovations and its engineering design.

This basically goes along two main avenues for Sui:

  1. From an engineering perspective, everything that lives on the blockchain is designed as an object. When transactions on Sui are running with objects that are independent of each other, they can be executed in parallel.
  2. Sui has the ability to handle arbitrary amounts of on chain data. We’ve designed Sui to make sure that anyone can use on chain storage. And if you’re going to use that storage, then you should pay for it.

This is different from what happens in more traditional blockchains, where every single transaction needs to be ordered relative to each other, even if the activities are completely independent. The implication for Sui here is that because of its ability to do causal ordering (in which only things that are somewhat related need to be ordered relative to each other but things that are completely independent can be executed in parallel), if demand for network activity increases at any given point in time, the chain can essentially scale horizontally. Each validator can add more computing power to make sure that they can execute these transactions in parallel, providing: high TPS, low latency, and good user experiences.

And here’s where the tokenomics comes in. If you have a chain that can handle a lot more of activity in any given moment in time, you don’t want the pitfalls of more traditional chains. In those instances, network activity increases in the short run, gas fees goes up, and it’s really bad for user experience. If I’m an ordinary user in an emerging market, and I have a lot of things that I wanted to on chain, I don’t want to live in a world where a lot of network activity means I’m gonna have to pay a lot of gas. That’s just bad user experience; it’s bad for my finances and it’s bad for the network.

And so Sui’s gas price mechanism is built in a way such that this horizontal scaling ability, of the chain to scale horizontally and to execute all these transactions at the same moment in time by adding more computing power, is built with a gas mechanism that ensures that gas fees are still low and predictable whenever this happens. That’s one of the big innovations of Sui. It goes hand-in-hand with this ability of making sure that everything that lives on the Sui blockchain is an object and executing these transactions in parallel.

The other big innovation that Sui has is the ability to handle arbitrary amounts of on chain data. For example, this is useful for people that may want to have metadata of their NFT to actually live on chain. From a tokenomics perspective, the main thing we’ve designed is to make sure that anyone should be able to use on chain storage, and if you’re going to use that storage then you should pay for it.

This opens up a lot of questions like:

  • How do you actually make sure that whenever someone is executing a transaction (that’s going to store data on chain) that they pay for it?
  • If they’re going to pay for it, are they going to pay for it for a finite amount of time or for an indefinite amount of time?
  • If it’s going to be an indefinite amount of time, how do you make sure that those fees that they’re paying up front actually end up lasting in a way that is able to pay for those storage fees forever?

So from a high level, the tokenomics has been designed very much in sync with engineering innovations to deliver a tokenomics system that works very nicely with Sui’s infrastructure.

Question #3: How does the gas mechanism work and why is it particular to Sui?

Alonso: Yes, so Sui’s gas mechanism, when you first read about it, it sounds a bit complex; it really is designed to create the right incentives to align the interests of all of these participants. So basically, it delivers a really good user experience for the users, for the validators, and also for people that have an interest in the medium-to-long-term success as the owners of the Sui Token (the native asset of the Sui blockchain).

Essentially the high-level goal was to design a gas mechanism that could always deliver low predictable gas fees for users. One of the objectives of Mysten Labs is delivering a good user experience across the whole tech stack, and for us, the end users of Sui is what’s most important. So how do we do that? How do we create that?

First, let me talk a little bit about how Sui works. Sui basically divides time across epochs. An epoch is probably going to be about 24 hours. The epoch goes through user-submitted transactions, validators process those transactions, and they have the voting power over how the transaction proceeds, which is proportional to the amount of stake they have. At the end of the epoch, the stake rewards are referred to as rewards to the validators and participants in this proof of stake mechanism. And since we’ve run a delegated proof of stake mechanism, most of the stake that validators have actually comes from individual holders of the Sui token that are not running a validator themselves. Most of the reward that the validators make for executing transactions throughout an epoch will be paid out to the Sui token holders that delegated with them in the first place.

Now, how does this relate to the way gas fees work? In Sui, validators are the ones that set the gas free for the epoch. They set what is called a reference gas price. What happens is that validators collectively sit down at the beginning of an epoch and discuss what the reference gas price will be. That gas price is communicated to all the clients and all the participants in the Sui ecosystem so that an end user can know the reference gas price.

So that’s how it works. But then that opens up a bunch of questions. First of all, how do we make sure that validators don’t just set an extremely high gas price? Well, here’s where the gas mechanism is designed to create the right incentives for all participants. Essentially, what happens is that the stake reward distribution rule, which means the share of stake rewards that are distributed to validators at the end of the epoch, gets adjusted depending on what each validator submitted as their bid or quote for the reference gas price. What matters here is that the validators who said at the beginning of the epoch that they wanted the gas price to be relatively lower than others are going to get boosted rewards. This is a way of encouraging validators when they’re collectively setting low reference gas prices. At the end of the epoch, they will be rewarded with more stake awards.

Now, you might be concerned about how this might just be an optimal strategy for validators to get those stake rewards by setting a low reference gas price at the beginning of an epoch, but then start charging high gas prices as the epoch progresses. For that, we have an intricate mechanism in which the validators are in charge of evaluating each other all throughout the epoch. If a validator said, “Well, I want one reference gas price to be very low”, and they submit a quote at that level, but then throughout the epoch they do not execute transactions that come in at those low gas prices, they’re doing something wrong, right? The validators are going to be aggregating information about what each validator is doing. So at the end of the epoch, they submit their best evaluation of each other, and again, the stake reward distribution rule is adjusted given this evaluation.

Essentially, the stake reward distribution rule takes these two elements:

  • It boosts the rewards of the validators that submit low reference gas prices.
  • It penalizes the validators that don’t honor the quotes that they’d submitted originally.

Effectively, this ends up creating a gas mechanism that promotes low reference gas price because the validator is encouraged to set low reference gas prices. But not so low that validators’ business models become unprofitable that they are unable to process these transactions.

This is a win-win mechanism for everyone:

  • End-users win by having low and predictable gas fees throughout the epoch.
  • Validators win because they will only set reference gas prices at which they can operate with a viable business model, but it also encourages them to operate in a somewhat competitive landscape.
  • Ultimately, Sui token holders also win because this promotes the long term growth of the ecosystem.

Question #4: How does the storage fund work and why is it suited to Sui?

Alonso: The gas mechanism is what would be related to more of the traditional workings of the Sui blockchain, except that here, we’ve created a mechanism that delivers low and predictable gas fees in a world where the blockchain is horizontally scalable. The storage fund comes hand in hand with the other part of the Sui blockchain, which is its ability to handle a lot of on chain data.

Now, the problem about the storage is the following:

If I’m a user today that is submitting a transaction and writing a lot of on chain data, that data in the future has to be stored by someone (e.g. validators, full nodes). In the future, once I have conducted my transaction, I can basically disappear off the face of the planet. But validators and full nodes still need to store that data.

What’s going to happen if I don’t pay for that storage today?

If those validators are going to pass those costs of storage to someone, they’ll probably have to pass it on to future users of Sui. And so that creates a negative externality because if I store a lot of data today and don’t pay for it, future users are going to have to pay for my storage and compensate for it. Now, that’s terrible for the network.

So then the economics point of view here is how do we solve that and how do we do it? We can solve it in a way that if you store data, you should pay for it. But this opens up a couple of questions:

  • Should we do something simple like paying for storage for a finite amount of time?
  • Or should we do something that has a much better user experience but is much harder to work out from an economic point of view, allowing data to be stored indefinitely?

So how do you make sure that the storage that you’re paying for will last potentially forever? What we’ve done is create a storage fund that recollects all the storage fees you pay when you store data. But the storage fund doesn’t actually pay future entities who are storing the data, it is used to adjust the share of stake rewards that are paid to validators relative to the ones that are paid to Sui token holders. If the validators have to store such a huge amount of upfront data that comes from users who previously stored data, they should get the proportionate amount of stake rewards to compensate for them. That’s how the storage fund works. It’s economically viable because the storage fund never pays out fees out of its principal, but is used to adjust the stake rewards that come from mainly the gas fees.

We wanted to make sure that from a storage perspective, we introduce the right incentive into the workings of Sui. What we’ve enabled is for users to essentially be able to delete their data and get a rebate. And so, in essence, if you keep data on chain today and decide that you no longer want that in six months time or whatever, you can get a pretty big refund of the storage fees that you paid initially. The storage fees are always intended to pay and to compensate for the entities that are storing the data. If you’re not a user and you deleted that data, this storage fee really shouldn’t be there. This is unlike what’s happening in web2. Let’s take DropBox as an example. You might be using DropBox today and paying an annual fee. But if in two years and you decide you no longer want to use DropBox, all that happens is that you close your account, and your data is deleted. No rebates.

Question #5: What will Sui token’s distribution look like?

Alonso: That’s a very important question. There will be 10 billion finite supply of tokens that will exist on the Sui blockchain. In a couple of weeks, we can share the exact percentages but what I can tell you right now is that the distribution of tokens will replicate what is standard in most protocols.

Some of the tokens will be set aside for the Mysten Labs team, Sui contributors, investors, and stake subsidies. There will be a public sale and we’ll deliver this information in the upcoming weeks. In terms of the distribution of tokens, I want to draw a lot of attention to the fact that a huge amount of tokens are being set aside for the Sui Foundation.

The Sui Foundation is an entity that is independent from Mysten Labs. The foundation’s goal is to promote the long-term growth and success of the Sui blockchain. The Sui Foundation will take tokens and use them for projects that support the cause:

  • Fund hackathons, conferences, documentation, applications, and developers teams that are building on Sui, helping them create the right type of products that will bring users to Sui and let them benefit.
  • Promote the strengths of Sui like interoperability, composability, etc.
  • Develop ecosystem enablement. Funding projects like infrastructure, wallets, digital identity, standards type projects that are important for DeFi protocols and general functionality may be challenging to monetize now, but they will be really crucial for the network in the long term.

The real reason for these tokens to be set aside is to make sure that they can be distributed in a smart way across the community to create the projects that will enable the long term success of Sui.

We’ll give out a lot more information in the upcoming weeks specifically about the token allocation but also on the specifics on how the Sui Foundation will work, which as I said, is an independent entity. We’ll start releasing information on how the community members and developer teams can start accessing these grants and these tokens to start funding their own projects. This is so you can also start planning on what you’re going to build and what your business model is going to look like.

Question #6: What will this distribution look like on the mainnet?

Alonso: I think most of the tokens will actually be locked at the mainnet, and will be unlocked over the coming years. The public sale tokens will be liquid immediately and once Sui blockchain becomes live, so will the tokens accruing from stake subsidies that serve as stake rewards for the participants in the delegated proof of stake mechanism. But, the foundation tokens are delivered to the team, depending on the exact framework and the conditions in each specific arrangement. Those will also partially be liquid so that developer teams can actually start funding their projects in a business viable way.

Question #7: There are many different blockchains but in the end, it comes down to EVM. Is it hard to differentiate them? Do you have a perspective on the issue here?

Alonso: Yeah, and I think we have a surprise for everyone today, which is we brought one of our senior engineers and one of the most knowledgeable engineers that we have at Sui on crypto and blockchain-related issues, Mohan. I think it’s a great moment to bring him in, let him introduce himself and start answering questions too.

Mohan: Hi, this is Mohan! I work on Sui tokenomics as an engineer. I came from the Libra and Facebook team as well, and am now joining Mysten Labs. I have a prediction that in a few years, Sui’s structure and architecture will probably be very popular among other different experiments in this space.

We’re talking about the question of why everything comes down to EVM?

I think the core question that comes to that is, there’s not really much incentive to actually develop a new suitable-for-blockchain, virtual machine execution environment. There isn’t really a great monetization or incentive to actually do anything that is as complicated as a virtual machine in a runtime environment. It’s only made possible because world-class engineers were brought in, like our own Sam and all the other experts to come up with something like Move. I don’t think you can replicate that easily, especially in our space right now, where crypto is so young and propelled by speculation. And that’s probably the reason why everything comes down to VM. But I think that’s not good for the industry and I think pouring enough resources and effort to actually break the game and to actually think and build from first principles perspective, is the maximum value we can bring in Sui tokenomics. We’re the ones who broke the glass and really think from first principles.

I feel like that’s the value that we provide:

  • Right incentive structure and innovation
  • Sustainable growth of the network.

For example, I really love the storage fund design; I think that’s probably one of the most ingenious ones in the current space. All the problems we talked about were EVM compatible chains, they have these stake float issues, and it’s going to come and hit them in the future. But in the case of Sui having a storage account, it gives the user the incentive to free up some of the data in their space. When I think of someone who had been in crypto for almost 10 years and when I think about what’s stored on Ethereum, it sort of scares me. All the useless information that’s accumulated over the past decade, I don’t think storing all that is sustainable and I think a lot of experts in this field already know that.

Question #8: In terms of the overall landscape, what do you think are Sui’s main competitors? And how can we outperform them?

Alonso: Yeah, that’s a great question. We get asked this a lot. I think it’s useful to take a little bit of a step back and reframe this question.

In many ways, web3 is extremely early. If you ask an ordinary person whether they use crypto, they may say something about Bitcoin or their investments, but the truth of the matter is that use cases that web3 has been promising have not been delivered. And that’s both across financial services, digital commerce, NFTs, decentralized social media, anything you can think about. So the question is if web3 makes it, the pie will grow so much that there’s gonna be more than enough to go around. If anything, what’s probably going to happen in the future is multi-chain. Some chains may specialize in specific use cases, perhaps even specific demographics, there might even be really good cross-chain communication protocols.

When we think about Sui, it’s not on our best interest to think that we need to compete with other chains, but rather we need to think about the following:

  • How do we actually onboard those next billion users?
  • How do we create the user experience that is going to make it as simple for my mom that doesn’t know anything about crypto to do something on-chain as it is for her to communicate with her friends on WhatsApp, or to send an email on Gmail?

So, when I think about Sui competitors, we need to zone down and think about what is our product strategy for going to market and then having people build on Sui? And in that sense, we think about optimizing the user experience on all fronts. We’ve talked a lot about infrastructure today about Sui’s ability to have really high TPS and low latency, that’s great for applications. We haven’t talked so much about the Move programming language but, that’s something that’s very important for Sui because it’s really a language that was created to make it very safe for developers to build their products on the Sui blockchain. It’s created to make sure that developers can focus on what they do best, which is building stuff, and delivering new products and new use cases, and not having to spend enormous cycles on auditing their code and looking for bugs that just distracts and makes it very inefficient.

On the product side, we’re making a lot of effort to optimize our user experience. For example, having wallets that are very slick and smooth and let you interact with the blockchain without having to think too deeply about how you’re gonna manage private keys, how you’re going to discover other users on the blockchain, and how you’re going to interact with different applications that are running on the blockchain.

So I think this is really the way Sui should look at it. This is probably how all blockchain should look at it. We shouldn’t really be competing with each other, we should all be thinking about how we are actually going to bring those billion users that know nothing about crypto into web3, and actually build use cases that are going to serve their purposes. Because the vast majority of the world don’t really care about the tech that they’re using, they just care about having products that work and serve their interests. I think this is how we should think about it.

Question #9: How does Sui plan to expand users and markets?

Alonso: Yeah, for us in particular, we’re trying to address this across various verticals. We want to work very closely with the community but also, the Sui Foundation as an independent entity will work very closely with developer teams, to create all these use cases.

In addition to that, we’re working with external partners that will be able to either bring their existing web2 type applications to Sui or develop independent web3 native applications.

Here are the verticals that we particularly are most interested in pushing right now:

  • The gaming front, where we think there’s a lot more terrain to be covered than has been so far.
  • The finance front, this includes both traditional finance and DeFi, the use cases in DeFi that haven’t been successful yet (e.g. uncollateralized lending) are going to be very important for people that don’t have access to not only on-chain assets that they can use as collateral but just assets in general. So, we need to think about what is the complimentary type of infrastructure that we need, maybe it has something to do with digital identity, and maybe it has to do with reputation-based systems.
  • Digital commerce, that has to do everything from payments to peer-to-peer transfers to letting big merchants distribute coupons on the blockchain and letting users redeem them, and even ticketing for concerts.
  • Social media including everything related to it. Communication and content creation.

Our hope really is to make success on this promise of really onboarding the next billion users to web3. We’re really thinking about the applications, developing the new use cases, and documents to serve that purpose.

Question #10: How does the roadmap look and when will Sui launch it to the world?

Alonso: That’s a good question, I’m going to let Mohan tell a little bit about the technical stuff related to what’s already running. Just to give you a very high level of the roadmap, right now, the devnet is already live. If you haven’t tried it out, please do, you can go to Sui.io and all the information is there. Hopefully, we’ll give a precise date soon but in the coming weeks, we will start releasing the test net and the incentivized test net. The goal really is to make sure how we can be hard in the protocol and make sure that validators and network operations are operating smoothly.

Whilst we’ll be testing some of the tokenomics elements during that period, after we’re done with the incentivized tests that will go to the mainnet, hopefully, by the end of this year, the Sui token will be live. It will be a fully permissionless, fully decentralized blockchain and you’ll be essentially able to do everything you do with other decentralized and permissionless blockchains. Except that here, you will have all the power of the Sui’s infrastructure, you’ll have the Move programming language, and you’ll have hopefully, the good user experience of the Sui tokenomics will give you.

So that’s sort of a very high-level overview of our roadmap. Mohan, would you like to say anything from an engineering point of view in terms of how the devnet, testnet, and mainnet progression are going?

Mohan: Yeah, I think you pretty much covered most of it, and I’ll go to the basic stuff running for the devnet over the staking delegation, and the basic structure for the gas incentives has been implemented and some of them are on-chain using written loops on the off-chain written in Rust code. For the incentives, we’ll start testing the structure of some of the rules, where we try to make sure to produce a stable gas reference price. We can ensure our gas price is changing around that reference gas price. In the incentive test we’re also planning to get a storage fund up and running, to work on the refund of the storage rebate, and to monitor and collect some data on that as well. We’re gonna go with most of the future we described in the paper. That’s the plan right now.

Question #11: In terms of early contributors, they’re an important asset to the protocol. How do you reward them as the protocol grows?

Alonso: Yeah, that’s a great question. I want to underline one thing. Sui is nothing without its community. This is a community project, and early contributors are key elements of that. A lot of the things that the Sui Foundation does is making sure that the early contributors get incentives that will let them thrive both in the short term but also in the long run. We made a token allocation to incentivize good projects. These will essentially deliver the Sui token holder the ability to thrive and benefit from the network’s success.

Being an early contributor to Sui will give the contributors two things:

  • The advantage of being on the protocol and knowing how to develop a smart contract on it. If you’re successful, because of the beautiful things that blockchains have like composability and interoperability, other blocks building on the Sui blockchain can interact with yours.
  • The token allocations will end up being more valuable in the future, given as more activity runs on the blockchain.

If anyone has any more ideas, we have an active Discord and members that can answer questions on anything related to engineering, tokenomics, or testnet. If you have any feedback or anything you’d like to see, please do let us know.

Question #12: What happens when the big staking firms set their fees to zero to attract delegated stake, and smaller validators can’t afford to get started?

Alonso: Yeah, that’s a great question. I want to state again that Sui is a delegated proof of stake system. So for the vast majority of validators, most of the tokens that they get will not actually be made by them, but will actually be delegated by the people who own the Sui token and want to participate in the community. Having said that, there’s of course, the risk that big players can amass a large amount of tokens and in a sense “control the network”. That would be very detrimental to Sui’s long-term success. So the Sui Foundation has an enormous amount of tokens that they’ll use to fund projects throughout the lifetime of the Sui blockchain. In the beginning, most of the tokens will probably not be given out yet because the Sui foundation is not going to fund all projects on day one. Rather, it will distribute these tokens over several years.

In the meantime, most of those tokens are going to be used for staking and that’s also what’s great about Sui Foundation because that means you can get some of the rewards that are associated with that token. That will fund even more trajectory in the ecosystem.

What that means for validators, is that since the Sui Foundation’s interests are aligned regarding views of decentralization and distributed networks, Sui will see to find validators that are perhaps even owned by the community that has the resources and expertise to run a validator. The Sui foundation will lend them tokens and delegate tokens so that these small validators can compete with larger rigs. Because it’s a delegated proof of stake mechanism, those small validators getting large validator tokens from the Sui Foundation will be able to increase their own capital and fund their business model. And so while these concerns are very clear, multiple entities on the Sui ecosystem that exist to help community members that are running validators be successful.

Jen: So couldn’t the storage fund model be too costly for the use cases of short term data storage? Let’s take the DropBox analogy that was shared. You pay for a year upfront, then renew if you want to keep storing. If you had to pay for perpetual storage upfront, it really would lock up a lot of capital inefficiently. Is that the case?

Alonso: For the vast majority of use cases on this Sui blockchain, storage fees will be negligible. So everything that has to do with, for example, peer-to-peer transfers, payments, things that have to do with digital commerce, probably most of the stuff related to financial services, almost everything there will have negligible storage fees. So for the vast majority of people, this is not something they will have to worry about.

In some ways, the model is a little bit more capital heavy than what you would have in a world where you are paying for that storage on an ongoing basis, for example a rent model.

In a rent model, we’d say we’re going to pay for storage every year and whenever you stop paying that then the storage gets deleted. That’s why we went for our storage model, we think it just delivers a much better user experience in the sense that it’s much better to just rent and pay the storage fee, forget about it, and if you ever want to delete, you can delete and get the rebate. But the rent model in which you have to continuously think about how much data you’re storing, I pay a storage fee this year, maybe next year I think about it again, but for the vast majority of users this just introduces friction and makes things much more complicated. This sort of friction scales when you start having millions or even billions of users and transactions every year.

You might say fine, but some users just don’t have the capital to do that. Right? Some users might just not have to have the amount to pay the perpetual service fee. They might get a rebate in the future, but they just don’t have that liquidity.

I think what will happen is that you will see some DeFi-type protocols that will essentially lend to you what is needed to pay for that storage, they charge a small premium, and this contract might be such that the moment they stop paying that premium, the data gets deleted. So in many ways, once the financial ecosystem becomes more complete we’ll create solutions that will somewhat replicate what the rent model is, which is more capital efficient, but has worse user experience for the people who are interested in it.

.   .   .