Tradeoff between Subjectivity and Exploitability
Public.icon
Objectivity is often celebrated as the primary feature of Bitcoin and indeed has many advantages. However, at the same time, it is also a curse. The fundamental problem is that as soon as one tries to introduce things from outside the cryptoeconomic world, such as real-world currency prices, temperatures, events, reputations, or times, one ends up trying to create connections where there were none before.
The truth is B, and most participants honestly follow the standard protocol that discovers that the truth is B, but 20% are attackers or have taken bribes.
The truth is A, but 80% of the participants are attackers or are pretending to be so by taking bribes.
From the protocol's perspective, these two are indistinguishable. Between truth and lies, the protocol is precisely symmetrical. Therefore, epistemic takeovers (where attackers persuade everyone else to agree to an attack and flip the equilibrium at zero cost), P + ε attacks, beneficial 51% attacks from very wealthy actors, and so on, all start to come into play. While it may seem that an objective system that does not rely on actors who use information other than what is provided through the protocol is easy to analyze, most of the problems with this issue are the opposite. Objective protocols are vulnerable to takeovers and can be taken over at zero cost, making them very poor tools for analyzing the analysis of equilibrium in standard economics and game theory. Therefore, subjectivity is necessary. The power behind subjectivity lies in the fact that concepts such as manipulation, hijacking, and fraud, which cannot be detected by pure cryptographic technology, and in some cases cannot even be defined, can be well understood in the human society surrounding the protocol. Let's look at an example to see how subjectivity works. The example provided here can define a new, third, hypothetical form of blockchain or DAO governance that can be used to complement futarchy and democracy: subjectivism. Pure subjectivism is defined extremely simply. If everyone agrees, decide unanimously.
If there is a difference of opinion between decision A and decision B, split the blockchain/DAO into two forks and implement decision A in one fork and decision B in the other.
All forks are allowed to exist, and which fork to be interested in is left to the surrounding community. Subjectivism is, in a sense, the ultimate non-coercive form of governance. No one is forced to accept a situation where they do not get what they want, and the only catch is that if you have a preference for an unpopular policy, you will end up in a fork where there are few other people to interact with. Perhaps in a future society where almost all resources are digitized and materially useful things are so cheap as to be immeasurable, subjective democracy may become a desirable form of government. But until then, the cryptographic economy seems to be a perfect use case. As another example, let's look at how subjectivism can be applied to SchellingCoin. First, let's define an "objective" version of SchellingCoin for comparison.
The SchellingCoin mechanism has a related sub-currency.
Anyone can "participate" in this mechanism by purchasing a currency unit and depositing it as collateral. The significance of participating is proportional to the size of the collateral, as usual.
Anyone can ask the mechanism a certain fee in the currency of the mechanism.
For a certain question, all eligible voters within the organization vote either A or B.
Those who vote for the majority receive a share of the question fee, while those who vote against the majority receive nothing.
As mentioned in the post on the P + ε attack by Paul Sztorc, there is an improvement where minority voters lose some coins, and the more controversial the question becomes, the more coins the minority voters lose. Note that in a 51/49 split, minority voters come to lose /all coins and become the majority. This significantly raises the hurdle for the P + epsilon attack. However, raising the hurdle alone is not enough. Here, we are interested in the fact that it is not abused at all (again, "vulnerable" is formally defined as "providing an essential opportunity for useful attacks to the protocol"). So let's see how subjectivity can be useful. Details that are not changed are omitted. For a certain question, all eligible voters within the organization vote either A or B.
If everyone agrees, let's reward everyone with a unanimous decision.
If there is disagreement, split the mechanism into two on-chain forks, with one fork behaving as if it had chosen A and rewarding everyone who voted for A, and the other fork behaving as if it had chosen B and rewarding everyone who voted for B.
Each copy of the mechanism has its own sub-currency and can interact separately. It is up to the user to decide which one is worth asking. In theory, if a split occurs, the fork that specifies the correct answer will increase the stake belonging to the truth teller, while the fork that specifies the wrong answer will increase the stake belonging to the liar, so users are expected to prefer asking the fork where the truth teller has a greater influence.
1. Each user answers the question posed (which tag is most likely to apply to the wallet).
2. Each wallet address is given coordinates when more than 80% of the overall answers are in agreement, with adjustments made to the coordinates to account for a tendency of around 40% rather than simply taking a straightforward 51% majority vote. This creates a linear image.
3. This is then mapped to generate a social graph.
4. It exists as an Oracle.
For the decentralized prediction market platform "Gnosis," there is no punishment for entering incorrect answers, so a number of accounts exceeding 51% of total users can enter answers that differ from the "correct answer" (which is, of course, an ideal construct).
This leads to the problem of collusion in entering answers, as discussed in Vitalik's paper.
1. There is a project token for DeCartography.
2. Anyone can participate by purchasing and staking project tokens. The final sum earned is directly proportional to the size of the deposit.
3. Tagging becomes an issue (with each person allowed to vote multiple times).
If everyone agrees, they all earn points. If the answer is divided, then each individual's answer is stored until the final result is determined, at which point they can place their bets. The majority place their bets as follows:
To sound the alarm, enter a bet.
Choosing an alternative method over data science might be beneficial for the world amidst everyone's focus on data.