The Different Types of Token Curated Registries
This First Half of this Post is Completed. The Second Half of this Post is in Beta.
Classifying TCRs into categories.
Finite Element TCRs: FE TCRs are lists with a finite number of elements that can be included in them.
Ideas: Coffee Shops in San Francisco TCR, Theme Parks in USA TCR
Ideas for Token Economics: Fixed token supply from Genesis. ‘Pay-to-Apply’. MIN_DEPOSIT is the fee to apply. Use Reparameterization via the On-Chain Governance property of the TCR token to Reparameterize MIN_DEPOSIT as it gets too expensive. It may start out as 100 COFFEE tokens to lockup to pay for the equivalent 100 DAI listing fee. In the future it may only be 20 COFFEE tokens to lockup and pay for the equivalent 100 DAI listing fee. MIN_DEPOSIT is required for spam resistance and to reward voters for voting. 50% of MIN_DEPOSIT is given to voters for voting. Other 50% of MIN_DEPOSIT is locked up for the duration of being on the list. Monthly/Quarterly/Yearly Staking fees of MIN_DEPOSIT are required after that, where SUBSCRIPTION_TIME (the time you have to wait before you have to stake another MIN_DEPOSIT) can also be Reparameterized. Since the price for the token is proportional to the demand to be listed on the list, there needs to be continuous buying demand even as the list is near/at completion. Recurring Staking Fees solves this problem.
Also, utilize Messari’s system of liquid democracy to allow Voters who work a 9-to-5 to delegate their vote to a Validator who votes smart (after a lot of due diligence) & takes care of the TCR as their 9-to-5. Validators would earn a portion of their delegated voter’s fee income, controlled by the parameter FEE_SHARE. This would be subject to Reparameterization & could start at some fixed number like 5%. This means Delegating Voters would get 95% of application fees awarded to them, and Validators would get 5% of Delegating Voters’ fees awarded to them. There should also be some Parameter called DELEGATE_PCT that limits the amount of votes an address can have delegated to it. This could initially be set at 7% and then be changed via Reparameterization in the future.
Continuously Growing TCRs: CG TCRs derive their continuous growth from the fact that they start out with a large pool of possible Initial Listees, and the rate of ‘possible listees being created’ is greater than ‘possible listees being destroyed’. This allows an initial list to have value to viewers even as only a small % of the Initial Listees are added. Then the list picks up more value over time as a larger portion of the Initial Listees are added, and then all the possible Initial Listees have been added- but there have also been new listees created during the time the TCR began until now, so we have to add them too! And then maybe the growth rate of possible listees is greater than the rate at which they can be added to a TCR- in which case you would have an Infinitely Growing TCR!
(Ex: Human Population. Large number of listees if you started a TCR today, and rate of birth is higher than rate of death, meaning there will be some meaningful rate of growth going into the future).
Ideas: Phonebook TCR, KYC/AML Whitelist for Accredited Investors TCR, Messari TCR with token
Ideas for Token Economics: Same exact Token Economics as FE TCRs, but no Recurring Staking Fees. Continuous growth in the number of listees into the future will make the TCR token have demand well into the future, with no need for Monthly/Quarterly/Yearly Recurring Staking Fees.
Or, follow the Messari model. Use a stable coin to pay the application fees. Use the TCR token for Voting, On-Chain Governance, and a ‘Claim on Cash Flows’. I buy Messari’s argument that using a stable coin to pay application fees will “facilitate the bootstrapping phase of the TCR as initially there may not be a liquid token market for Applicants”.
Additionally (aka a variation of Messari’s model), require a voter to stake/lockup their coins for a period of time to take advantage of the proxy voting of the liquid democracy and ‘claim on cash flow’ property, which really drives down V (Velocity) and increases H (Holding Time) [they are basically the same property], which will prop up the Price of the TCR token. But it is true that this is an artificial price boost for early adopters, while also making it artificially more expensive for newcomers to join the system.
Still in beta….
Data Accrual TCRs: DA TCRs seek to accrue data with 0 rent-seeking. They can be either FE TCRs or CG TCRs, but with a little bit different Token Economics because we don’t want to have people paying to do the work to add data to the lists. There will probably not be recurring Monetary or Social Rewards relating to the data you submit to be added to these types of TCRs (but you will be paid a one-time ‘Data Collection’ fee [called ACCEPTANCE_AWARD] for having those Data Points [DPs]successfully added to the TCR). Current ‘Free & Open Education/Data Platforms’ are good examples of these.
Ideas: ChainShot TCR, Wikipedia TCR, KhanAcademy TCR, Tokenless-Messari TCR, Events from Every Year in the Entire History of the World TCR
Three Ideas for Token Economics:
1. Use MIN_DEPOSIT to prevent spam.
2. Or have NO MIN_DEPOSIT and no spam prevention, allowing even people with no money to contribute to the list.
3. Or use MIN_DEPOSIT+Pending Pool to prevent spam and still allow anyone in the world (even those with no money) to submit data to the list.
Digital currencies use transaction fees to prevent spam. In DA TCRs, ‘spam’ can be categorized as ‘poor data applications’. We have to prevent too much spam accumulating to be voted on, because we have a finite amount of Voters/Validators and those Voters/Validators do not have infinite time.
I already thought through using Idea 1 and 2, and they were not feasible. But I found a way to make Idea 3 feasible, and prevent the ‘Good Enough’ Subjectivity Problem from occurring.
Use MIN_DEPOSIT to prevent spam, but also have a ‘Pending Pool’ where Data Contributers who do not have the means to reach MIN_DEPOSIT (but do have a TCR token address) can submit their Data Points: You believe that content on the TCR would be better if nobody needs any money to be able to contribute to the list, but you also want to prevent Spam. People who do not have enough TCR tokens to reach MIN_DEPOSIT can submit their data to the Pending Pool. This does not cost anything. We call these people ‘Fishermen’. They ‘fish’ for Data Points from their brain or from books/internet, and then they release those Data Points into the Pending Pool. People who have MIN_DEPOSIT can then take the Data Points of the Pending Pool, and then submit those to the DA TCR. We call these people ‘Trawlers’. Like the big fishing trawlers, they cast a wide net and accumulate a lot of Data Points from the Pending Pool and then submit the good Data Points and then forget about the bad Data Points. For doing this work, Trawlers receive some share of the ACCEPTANCE_AWARD, determined by POOL_SHARE. POOL_SHARE can be Reparameterized using the On-Chain Governance Property of the TCR’s Native Token. It should be initialized fairly high to account for a possible 100% loss of MIN_DEPOSIT if a bad/duplicate Data Point is submitted. I think a POOL_SHARE of 50% makes sense.
The way this prevents spam is: The Pending Pool prevents the time of Voters/Validators from being used on bogus applications. Trawlers are the only ones who can interact with the Pending Pool- and they don’t actually have to look at the Pending Pool at all! They can use their MIN_DEPOSIT to submit Data Points on their own (in which case they are the Default: Data Contributor). This means even if an adversarial TCR wanted to overload our TCR, they would do it through the section that has no MIN_DEPOSIT requirement. In that case, it is only the Fishermen who get the short end of the stick, but the TCR can still function just fine.
And now for the Token Economics:
There are many Permutations of these Token Economics that were not written in this Post. They would already thought through and written about here. Below is v7 of these Token Economics and I think it’s the best one.
How do these TCR tokens retain value as t → ∞? Achill Rudolph and I posit in this thread (read the comments below the initial post as well) that there would always be buyers of a TCR Token to have the power to decide what goes on the list. This would be because the list is Societally Valuable, therefore there is some non-zero value in owning the TCR Tokens and having the power to decide what goes on the list. This could also apply to FE TCRs as well, but we’ve already solved the Valued Leakage problem there by having Recurring Staking Fees. In DA TCRs, there is no Social or Monetary Benefit to having your data on the list, so Recurring Staking Fees would not work.
The most valuable DA TCRs will be ones where Fisherman are creating DPs as a natural byproduct of something they are already doing in life, and not doing it only to get paid in the TCRs token. Example: If you had a machine that was powered by CO2 in the air, you could have ‘infinite’ fuel because everyone around the world is constantly creating CO2 as a byproduct of their daily lives. Nobody’s behavior has to be changed for you to get fuel. I think people learning new languages would naturally be creating many valuable DPs as a byproduct of their daily life and would be a great resource for a Language Learning Database TCR.
So, inflation funding via ACCEPTANCE_AWARD for these DA TCRs should be set up in a way such that
It may mean a lot of these DA TCRs start out in low-standard-of-living countries where earning xx TCR Tokens (equivalent to $0.01) per completed data point (because of the 0.0006% inflation for ACCEPTANCE_AWARD) is a meaningful amount for the time put in to think about, research, and submit that data point. Then, as the value of the DA TCR grows, that 0.0006% inflation per each ACCEPTANCE_AWARD is worth more people’s time in developed countries. I think the most valuable DA TCRs will follow this model: Be a list of something that people of low-standard-of-living countries have in their brain, and that people of middle-standard-of-living countries have in their brain, that people of high-standard-of-living countries have in their brain. This is why I like the Language Database TCR. Language is something that is in everybody’s brain.
TCR Platform Protocols: Seek to use 1 token across all TCRs built on the Platform AND/OR Seek to use 1 Web GUI across all TCRs built on the Platform. 1 Token may help building a stronger TCR community. 1 standard Web GUI would make it easy for all players (Consumers, Candidates, Token Holders) to view/apply for listing/open challenges/vote in challenges/vote on the Parameterization of the Registry for all TCRs built on the platform.
The only way DA TCRs would work is if only the token holders were the ones contributing the data to increase the value of their own TCR Platform Token
Reputation System- This is a positive for that specific TCR Ecosystem.
Examples: DIRT Protocol
Ideas for Token Economics: One Token or Many Token
Comments:
[5/20/18]
I buy Simon de la Rouviere’s idea that the friction introduced by bundling a a token with a Dapp is good for coordination. (Here is his second post about it). Think: Would people be shilling 0x as hard if it didn’t have a token?
I also buy Mike Goldin’s idea that each TCR should have its own token so that the token holders can fully realize the positive or negative impact they are having on that one specific Registry.
The intrinsic tokens of token-curated registries are necessary elements of self-sustaining systems which are public utilities. Token-curated registries are peak predators of capitalism that perform a useful function at the lowest possible marginal cost.
A token is a necessary element of a system if the use of any other in its place would damage the system’s normal functioning. Token-curated registries require intrinsic tokens because token holders must realize both the upside and downside of their good or bad work in order to be motivated to perform their essential curation task. The price of Bitcoin will not be responsive to reduced demand for it in the application of registry listings, meaning token holders’ only incentives would be to take all the Bitcoin they could from candidates by issuing spurious challenges and colluding on votes against the interest of the registry’s curation criteria. A token whose only fundamental utility is its necessity for making applications to some registry will see its price fluctuate on the basis of demand for those listings, which is determined by the quality of curation done by token holders. Token-curated registries satisfy token-necessity.
And even with those 2 data points, I still think Yin Wu’s argument that “the people taking care of the 500 TCRs built on DIRT will come and help the 1 TCR that is being poisoned even though it won’t really affect the value of their DIRT tokens” is plausible.
