New crypto mechanism for free usage, or existing economic model?
Every now and then, I go back to the (poorly named) concept of hyperstructures: “crypto protocols that can run for free and forever, without maintenance, interruption or intermediaries.” These so-called structures take the form of protocols that run on blockchains. Something can be considered a hyperstructure if it is: unstoppable, free, valuable, expansive, permissionless, positive sum, and credible neutral.
As defined, hyperstructures seem like impossible things. How can it be both free and valuable? I think maybe if you time-bound some of these requirements, they start to seem a little more realistic. For instance, what if something was free over the long term? Or free under certain conditions?
This seemed like an interesting thread to pull, and while @Andrew Hill and I were chatting about two-sided utility markets and exploring ways of incentivizing both the supply-side and demand-side together, we started to unravel it a bit…
Firstly, there are a number of ways that other projects have tried to incentivize supply-side (e.g. validators) and demand-size (e.g. users) as separate groups, or together. This is particularly important in the early days of a network. In the past, lots of projects have used approaches that required “off-chain” human in the loop mechanisms to achieve this (like airdrops, KYC steps, etc). So as part of our exploration of Filecoin’s hierarchical IPC subnet protocol, we wondered if there were any new opportunities to handle this in a unique and interesting way.
Normal staking is all about supply-side rewards, which are the dominant mechanism for creating incentives for validators/service providers to do work on a network. However, we might also want to include a novel demand-side incentive/rewards mechanism in order to incentivize:
“signaling” to the network that demand is real, ready, and looking for supply,
“curating” network quality by identifying and staking into high-quality subnets,
“demand” by allowing liquid staking, thereby reducing opportunity costs for holders, and
“participation” by providing a mechanism for cheaper transactions in exchange for longer-term staking within a given subnet
Borrowing concepts from liquid staking, Solana’s “set it and forget it” concept, and “key money” or more specifically “jeonsei” from the South Korean rental market, we proposed a type of liquid staking process where users who deposit funds into a subnet are effectively providing “key money” to the subnet. These funds are essentially a “signal” to the network that they intend to use this subnet, with usage roughly equivalent to size of deposit plus some interest.
Jeonsei (or jeonse), meaning "money in full" and often called "key money", is a unique Korean system. Tenants provide a large upfront deposit, usually between 25% and 80% of the property's value, instead of paying monthly rent. They only cover utilities. At the end of the tenancy, the full deposit is returned without interest.
Some portion/percentage of the deposit is “locked” for a period of time, and is slowly unlocked at each block step. After a certain period, the entire deposit can be withdrawn. In the mean time, the deposit is essentially acting like a delegated-stake (hence liquid stake) within the subnet.
Each block the depositor (user) earns some cut of the reward (as in normal protocol staking), with a specific tweak: the earned rewards mustremain within the subnet (or project, protocol, etc). The depositor (user) can spend these earned funds on transactions (or features, utility, etc) within the subnet, effectively leading to near “free” usage over longer periods of time.
Of course, you have to design your staking rewards and transaction fees accordingly so that you incentive the right time-bounded participation levels for your specific protocol.
Elaborating a bit on that tweak: unlike standard staking in other protocols, users/stakers who earn interest/rewards, cannot withdraw any interest/rewards earned from the subnet. Like in a jeonsei market, they may withdraw up to their principal deposit once it has been fully unlocked (this is similar to the rental period in a jeonsei market). As pointed out by our colleague @Sander Pick, in practice this might require the use of colored tokens or other ways to represent tokens limited to a subnet.
We might also want to consider allowing depositors to drawn down their deposit to some specific percentage below their initial principal, to effectively subsidize early usage. However, we would then require them to keep their coins “locked” in the subnet until they have effectively “paid off” their usage from collected interest over time.
The locked deposits act like a signaling mechanism to the network, similar to something like the Graph’s signalers. This incentivizes the supply side to spin up service (e.g. compute, storage, output, etc) relative to the size of the deposit. This also incentivizes depositors to deposit the right amount of funds into the subnet to cover their anticipated costs. They can draw up and down their overall deposit over time, eventually optimizing their “stake” to get to near costless transactions (or features, utility, etc) over very long periods of time, and more costly over the short term.
This describes the ultimate pay as you go model, with built-in scaling. If a user comes in with serious demand, they have a mechanism to signal to the network they need resources. The network has a mechanism to ensure those resources are well spent, and not just a flash in the pan grab for one-off compute (or resources, etc). It also provides a way for users to potentially get really cheap services while maintaining their initial investment. Aaaand, it also helps with security on the network (”staking”) while maintaining liquidity and actual utility of the deposited tokens (”key money”).
So I’m calling this something like Jeonse staking, but I’m guessing something like this already exists… so what is this called? Do other networks have similar concepts? Will it work? What do you think?
New crypto mechanism for free usage, or existing economic model?
Every now and then, I go back to the (poorly named) concept of hyperstructures: “crypto protocols that can run for free and forever, without maintenance, interruption or intermediaries.” These so-called structures take the form of protocols that run on blockchains. Something can be considered a hyperstructure if it is: unstoppable, free, valuable, expansive, permissionless, positive sum, and credible neutral.
As defined, hyperstructures seem like impossible things. How can it be both free and valuable? I think maybe if you time-bound some of these requirements, they start to seem a little more realistic. For instance, what if something was free over the long term? Or free under certain conditions?
This seemed like an interesting thread to pull, and while @Andrew Hill and I were chatting about two-sided utility markets and exploring ways of incentivizing both the supply-side and demand-side together, we started to unravel it a bit…
Firstly, there are a number of ways that other projects have tried to incentivize supply-side (e.g. validators) and demand-size (e.g. users) as separate groups, or together. This is particularly important in the early days of a network. In the past, lots of projects have used approaches that required “off-chain” human in the loop mechanisms to achieve this (like airdrops, KYC steps, etc). So as part of our exploration of Filecoin’s hierarchical IPC subnet protocol, we wondered if there were any new opportunities to handle this in a unique and interesting way.
Normal staking is all about supply-side rewards, which are the dominant mechanism for creating incentives for validators/service providers to do work on a network. However, we might also want to include a novel demand-side incentive/rewards mechanism in order to incentivize:
Borrowing concepts from liquid staking, Solana’s “set it and forget it” concept, and “key money” or more specifically “jeonsei” from the South Korean rental market, we proposed a type of liquid staking process where users who deposit funds into a subnet are effectively providing “key money” to the subnet. These funds are essentially a “signal” to the network that they intend to use this subnet, with usage roughly equivalent to size of deposit plus some interest.
Some portion/percentage of the deposit is “locked” for a period of time, and is slowly unlocked at each block step. After a certain period, the entire deposit can be withdrawn. In the mean time, the deposit is essentially acting like a delegated-stake (hence liquid stake) within the subnet.
Each block the depositor (user) earns some cut of the reward (as in normal protocol staking), with a specific tweak: the earned rewards must remain within the subnet (or project, protocol, etc). The depositor (user) can spend these earned funds on transactions (or features, utility, etc) within the subnet, effectively leading to near “free” usage over longer periods of time.
Elaborating a bit on that tweak: unlike standard staking in other protocols, users/stakers who earn interest/rewards, cannot withdraw any interest/rewards earned from the subnet. Like in a jeonsei market, they may withdraw up to their principal deposit once it has been fully unlocked (this is similar to the rental period in a jeonsei market). As pointed out by our colleague @Sander Pick, in practice this might require the use of colored tokens or other ways to represent tokens limited to a subnet.
The locked deposits act like a signaling mechanism to the network, similar to something like the Graph’s signalers. This incentivizes the supply side to spin up service (e.g. compute, storage, output, etc) relative to the size of the deposit. This also incentivizes depositors to deposit the right amount of funds into the subnet to cover their anticipated costs. They can draw up and down their overall deposit over time, eventually optimizing their “stake” to get to near costless transactions (or features, utility, etc) over very long periods of time, and more costly over the short term.
This describes the ultimate pay as you go model, with built-in scaling. If a user comes in with serious demand, they have a mechanism to signal to the network they need resources. The network has a mechanism to ensure those resources are well spent, and not just a flash in the pan grab for one-off compute (or resources, etc). It also provides a way for users to potentially get really cheap services while maintaining their initial investment. Aaaand, it also helps with security on the network (”staking”) while maintaining liquidity and actual utility of the deposited tokens (”key money”).
So I’m calling this something like Jeonse staking, but I’m guessing something like this already exists… so what is this called? Do other networks have similar concepts? Will it work? What do you think?
From SyncLinear.com | NOT-26