The 'judge' behind the Celestia airdrop, how does Trusta Labs perform witch detection?

2023-09-27 18:30

Read this article in 10 Minutes

Score all addresses and let the project team set the 'passing score'.

Original title: "The "Judge" behind the Celestia airdrop, how does Trusta Labs perform witch detection?"

Original source: Odaily Star Daily

On the evening of September 26th, Celestia, the leading project in the modular track, suddenly announced the launch of a Genesis airdrop, which will distribute a total of 60 million TIA tokens (6% of the total token supply) to the community.

This Celestia airdrop is quite generous. In addition to airdropping to early participants in the modular ecosystem, developers and researchers of public affairs, early users of Rollups, IBC relay node operators, and Cosmos staking users from different groups are also included in the airdrop range.

However, in addition to "sunshine", Celestia is also working hard to mitigate the increasingly rampant witch behavior through various measures. The specific measures adopted include directly using the witch address blacklist of Hop Protocol and Optimism during the airdrop years, and using Trusta Labs' witch screening mechanism to analyze specific addresses to determine whether they are suspected of witch behavior.

How does Trusta Labs conduct airdrop detection? Let's take a closer look.

In August of this year, Trusta Labs published an article detailing its on-chain behavior scoring standard MEDIA Score. MEDIA Score has two layers of utility. First, it provides project parties with an objective, fair, and quantifiable indicator to comprehensively evaluate the participation and value level of user addresses, accurately positioning those users who have truly contributed to the project. Second, it helps users themselves to have a clearer understanding of their address behavior, evaluate their potential community rewards (i.e. airdrops) opportunities, and thus conduct on-chain interactions more reasonably and efficiently.

From the specific scoring criteria, the five letters M, E, D, I, and A in MEDIA Score represent five dimensions - Monetary, Engagement, Diversity, Identity, and Age. The total score of the five dimensions is 0-100 points, but the corresponding weightings are different.

M: Monetary

"Currency" dimension aims to evaluate the financial value of a specific address. Trusta Labs will convert the token balance and the record amount that has been interacted with in the address into a value denominated in US dollars and compare and analyze it with other addresses. The "currency" dimension accounts for 25% of the total weighted score.

From the perspective of inspection content, the "currency" dimension will focus on checking three records:

Address balance: the total value of all tokens in the address.

Total interactive amount: the cumulative amount of all agreements that have interacted.

Official Bridge Amount: The total amount of cross-chain transactions conducted through the official bridge.

E: Engagement

"Participation" dimension aims to evaluate whether a specific address has deeply participated in the interaction of the on-chain ecosystem. This dimension not only requires users to have multiple interaction times, but also requires that their interactions cannot be concentrated in a single time period. The "participation" dimension accounts for 30% of the total score weight.

From the perspective of checking content, the "participation" dimension will focus on checking five records.

Active days: At least one active interaction within a calendar day counts as one active day;

Active weeks: At least one active interaction within a calendar week counts as one active week.

Active months: At least one active interaction within a calendar month counts as one active month;

Total interaction times: the total number of interactions with all on-chain projects.

Interaction time span: The time span from the first interaction to the most recent interaction.

D: Diversity

"Diversity" dimension aims to evaluate the breadth of the specific address's interacting projects (contracts, protocols, categories). The "diversity" dimension accounts for 15% of the total weighted score.

From the perspective of inspection content, the "diversity" dimension will focus on inspecting three records:

Interactive contract count: the number of independent contracts that have been interacted with;

Number of interaction protocols: the number of independent protocols that have been interacted with;

Number of interactive protocol categories: the number of interactive protocol categories (DeFi, NFT, Web3 Game, Infra, etc.) that have been interacted with.

I: Identity

"Identity" dimension aims to evaluate the specific identity roles and qualifications of the given address in the L1/L2 ecosystem. The "Identity" dimension accounts for 10% of the total weighted score.

From the perspective of inspection content, the "identity" dimension will focus on checking multiple records including:

Multi-signature members: for example, as a member of a well-known DAO;

Historical airdrop recipients: such as Arbitrum airdrop users, or Optimism airdrop users;

From the perspective of inspection content, the "age" dimension will focus on checking the following multiple records:

Calculation Method for Total Score:

Trusta Labs has concluded that the total score of address 64 ranks in the top 26% after comparing it with all zkSync Era user addresses. However, this cannot be used as a basis for whether the address can receive potential zkSync Era airdrops, as the right to set the "cutoff line" ultimately lies with the project team, such as the standard adopted by Celestia for early Rollups users, which is the top 50%.

Summary:

However, for more real users (of course, it is also difficult to avoid the wool party operating "boutique accounts"), this is a good opportunity for them to win the "scientists" in the bend, after all, considering the current popular projects with millions of interactive addresses, if more strict screening is not carried out, the airdrop benefits that real users can get will probably be diluted into "mosquito legs".

For project parties, the game situation with the witch may also undergo some changes. Prior to this, the game between the project party and the witch was mostly conducted positively. The project party needed to screen airdrop targets based on user interaction portraits, such as Hop, Optimism, Paraswap, and so on. Now, with the emergence of third-party witch screening mechanisms such as Trusta Labs and Gitcoin Passport, the project party can more easily delegate this part of the work to third parties, thereby focusing their energy on their own business.

Finally, let me share the rating page of Trusta Labs here. You can check your address's score and ranking on the page.