Arbitrum: How do we check and confirm the witch address?

Original title: "All for One and One for All" - An on-chain distribution model for the Arbitrum community

Dark side of the Moon, PANews

TL; DR

•Offchain Labs, the Arbitrum Foundation and Nansen worked together to design a Token allocation model based on community members' participation in the Arbitrum ecosystem

• This distribution pattern relies on Nansen's on-chain data to assign qualifying points based on the wallet's on-chain history of activity. Wallet addresses must meet at least three points to qualify for airdrop.

• The list of eligible wallets covers 625,143 addresses for a statistical period since the inception of Arbitrum One up to 6 February 2023

•拥有社区金库并一直在推动 Arbitrum 生态系统向前发展的协议，也会被邀请参与为自己的社区分配部分 Token

introduction

Arbitrum 基金会与 Offchain Labs 合作宣布为 Layer 2 项目 Arbitrum 进行第一次治理 Token 的空投，并将其分配给 Arbitrum 用户和相关社区。（推荐阅读：《One article to understand Arbitrum airdrop rules")

The Arbitrum Foundation and Nansen worked together to design the airdrop strategy and distribution model to quantitatively ensure that each wallet or on-chain physical address meets the airdrop eligibility criteria.

Figure 1: L2 rise, Arbitrum daily transaction volume surpassed Ethereum for the first time in February 2023

Arbitrum 基金会已授权 Nansen 确保，会考虑到与网络交互的地址数量，以及将 Token 空投到给促进 Arbitrum 生态发展的用户手中。

• The allocation model will be based on on-chain data: past Arbitrum historical data via Nansen Query

• The mechanism design will include multiple parameters and criteria to ensure its security

•Nansen's analysis will follow the vision of Offchain Labs, Arbitrum's parent company

In short, on-chain analysis has an artificial planning component, and even with the use of rigorous analytical standards, the airdrop distribution model cannot be absolutely accurate.

在本文中，我们将介绍 Arbitrum Token 空投的具体流程和分配模型。

Parts 1 and 2 review the specific variables and objectives of the airdrop program. Part 3 explains the specific process between allocation criteria and on-chain analysis. Parts 4 and 5 detail the eligibility criteria and the associated "points" design. The sixth part introduces the idea of anti-witch total. Part 7 summarizes the final statistics for eligible wallet addresses. Part 8 focuses on eligible DApps and protocols.

1. Token 分发模型是空投成功的关键因素

鉴于 Token 空投的复杂操作性，通常需要对用户的参与和治理决策的过程进行优化。因此，协议需要设置具体的空投标准，以支持协议实现长期发展目标（见下文「2. 空投的原则和目标」部分）。

在 Arbitrum Token 空投中，Nansen 专注于分配模型参数的设计，以尽可能帮助 Arbitrum Foundation 制定分配标准并确定 Token 空投的接受者列表。但以下因素与空投分配模型也具有较强关联性：

•需求方因素：主要涉及 Token 在协议治理中的用途，需要在此基础上确定 Token 需求

•供给方因素：供给方由 Token 发行数量（空投与其他渠道发行多少 Token ）、时间（影响未来行为的潜在归属时间表）和分配模型等因素共同决定。

Figure 2: Distribution model variables of the current round of airdrop

2. Principles and objectives of Airdrop

Arbitrum Foundation hopes to optimize the degree of decentralization of Arbitrum protocol governance through air delivery. One approach to consider, therefore, is to try to define organic activities. "Organic" activities include effectively trading on Arbitrum, as well as helping to develop DApps and other agreements on the chain, or making positive contributions to Arbitrum's governance. The above behaviors can be measured in the on-chain data and classified by wallet activity on the Arbitrum chain.

However, some wallet addresses with multiple transactions are not necessarily in real demand, and there is no guarantee that these addresses will remain active after the airdrop. Thus, one goal of on-chain data analysis is to help identify wallet addresses that have the "witch" tint, such as those that are only intended to qualify for this round and future airdrops. The second goal is to introduce "negative criteria" that are highly correlated with witch behavior, which was designed by Offchain Labs anti-witch attack researchers based on data from Nansen Query Arbitrum and Ethereum.

3. Distribution standards and on-chain data analysis: a process of iterative deliberation

The Arbitrum Foundation worked with Nansen to define what constitutes "organic" activity, then converted those behaviors into quantitative criteria and thresholds, and refined those criteria and thresholds through on-chain analysis (see Figure 3).

Figure 3: Project flow, from first-principles standards to data validity

The study generally followed these steps:

Step 1：根据空投的目标，按照第一性原理去定义分配 Token 的活动

This step defines the prior criteria for on-chain activities to qualify as "organic." At this point, we're trying to answer the following question before looking at the on-chain data: How many transactions should a wallet initiate to qualify for an airdrop? What should organic behavior look like over time? What is the corresponding dollar value of doing the bridge and the cost?

Step 2: Collection of on-chain data and quantification of qualified behaviors

In this phase, we modeled the previously defined hypothesis principle and used Nansen to process Arbitrum One's on-chain data into quantitative criteria, including transactions, transfers, logs, tracking, and Nansen's other labels (see Figure 4).

Figure 4: Nansen query interface: Wallet address transfer activity data on the Arbitrum One chain

Step 3: Sensitivity analysis

We built the interim model by setting up the eligibility criteria in the allocation model and combining them with the address list that had been created previously. The model can attribute a certain number of positive or negative integrals to the criteria.

We reviewed part of the address data to assess whether our criteria and integral design were too strict because they only captured the address data too far in front of the address on the Arbitrum, and vice versa, and also needed to check whether these criteria were too lax. We can illustrate this process by visualizing the distribution criteria (Figures 5 to 8).

This step takes several iterations and eventually leads to reasonable thresholds and standards. For example, the model must capture address data with scores higher than the median threshold shown in Figures 5 to 8: for example, the threshold for the "number of different months in which transactions are conducted" criteria is set to two months or more, and higher than the median of one month.

In addition to sensitivity analysis, cross analysis was conducted to ensure that the eligibility criteria for each category could complement each other and capture the key behaviors of participating Arbitrum users.

Figure 5: Wallet addresses by number of initiated transactions, snapshot on February 6, 2023 (Arbitrum One)

Figure 6: Address data by time interval between first and last transaction, snapshot on February 6, 2023 (Arbitrum One)

Figure 7: Address data by monthly number of transactions, snapshot on February 6, 2023 (Arbitrum One)

Figure 8: Wallet addresses in US dollars, February 6, 2023 snapshot (Arbitrum One)

Step 4: Determine final standards and anti-witch designs

Determine criteria for qualified and disqualified activities and modify the points system based on on-chain analysis results. At this stage, Arbitrum Nova is added as a supplementary standard for Arbitrum One. So re-invoke the model until you get a satisfactory result.

Offchain Labs researchers have also designed the anti-witch attack model by modeling historical transactions and data on Arbitrum and Ethereum. Deleted will part of the address wallet that may involve witches.

Step 5：符合 Token 空投条件的钱包地址预选名单

The model finally gets a valid wallet address. Which are marked as "functional" wallet addresses such as trading platforms, pools of money, addresses for "destruction" and cross-link Bridges. The wallet address list is ready for its final eligibility review.

4. 合规标准和其他 Token 分配标准

The above process resulted in a final concrete standard around the definition of the following broad categories (see Figure 9) :

• Quantified thresholds for the number, value, and time span of transactions or transfers initiated by the wallet

• The behavior of these wallets brings liquidity to Arbitrum

• Experiment with a new Arbitrum chain like Nova

•另外，以下用户需要特别强调： 拥有社区金库的 dApps 也会获得一部分空投，并且将会以符合标准方式分发给他们的社区。通过 Token 分发进行本地化决策一直是 Arbitrum 作为生态系统和社区的重要部分。具体细节参考「8. 分布式治理」。

Figure 9: Airdrop segmentation criteria

The historical data collected by Nansen starts with Arbitrum One's Creation block:

Block number 22,207,817 on August 31, 2022, that is, block number 58,642,080 on February 6, 2023, before Arbitrum's upgrade to Nitro, to account for the increase in Nitro activity after Arbitrum

Supplemental eligibility criteria were added to cover another chain, Arbitrum Nova, to reward early active participation in testing wallet addresses for this new chain. However, the integral will be slightly lower than the standard on Arbitrum One, and the integral obtained in our allocation model will be lower. The snapshot on Arbitrum Nova is:

October 2, 2022 Block number 499,3422023 February 6, 2022 block number 2,108,676

Each "organic" activity earned either a positive (encouraged behavior) or a negative (discouraged behavior) point. The number of tokens a wallet address receives in an airdrop depends on how many credits it collects. The minimum participation standard is that the wallet address must reach three points, the more points obtained, the higher the proportion of participation distribution. All snapshot date credits are combined to determine the final score for each address.

5. Point system and partial wallet address data that meet the conditions

Figure 10: Arbitrum One and Nova criteria and scores

To summarize the following, we get various behavioral data for eligible wallet addresses, such as:

• Bridge connected to Arbitrum One's wallet address, made more than a dozen transactions in two different months, and had a non-zero balance at the time of the airdrop snapshot

• The bridge to the wallet address of Arbitrum One and Arbitrum Nova requires at least 25 transactions through Arbitrum One in two different months, and more than a dozen transactions for the wallet on Arbitrum Nova

• A wallet address that wired assets worth more than $250,000 into Arbitrum One made more than 100 different transactions over nine different months, and on Arbitrum Nova, the same wallet initiated more than a dozen transactions

6. Identify the witch address

在最初有资格接收 Arbitrum 空投 Token 的地址中，其中部分地址不是出于实用目的，而是聚焦于获取未来可能的空投（见图 11 和 12），这些地址被认为不太可能一直在使用 Arbitrum，也不太可能积极参与治理。

Witch addresses are often hard to distinguish from actual behavioral addresses, as in the old days with airdrop data. Two different types of addresses can exhibit exactly the same pattern on the chain, but they have different goals, one for real use and the other for potential airdrops.

Therefore, tracking witch addresses requires a qualitative evaluation step, that is, a manual evaluation of whether the model used to detect the witch will violate the airdrop target.

Offchain Labs 研究人员通过在来自 Nansen Query 的 from_address / to_address 交易数据使用聚类算法，并在 Arbitrum 和 Ethereum 上整合跟踪和 Token 转移，致力于识别可能的女巫钱包。并且对可能误报的数据进行人工「检查」。有关女巫地址的具体信息，请参阅 Arbitrum Foundation Github。

Figure 11: Suspected witch addresses Example: Two addresses in a set of about 400 have highly similar activity (sending funds to the same centralized trading platform deposit address)

图 12 疑似女巫行为的地址示例，由来自同一地址的 Token 进行链上转移

7. List of eligible wallets and summary of statistics

在 2023 年 2 月 6 日之前桥接到 Arbitrum One 的约 230 万个钱包中，有 625,143 个或约 28% 获得了超过 3 个以上的积分，并有资格获得新发行的 Arbitrum Token 。这个资格号码是 ex-sybil ex-functional-entity。大约 3.7 万个地址与功能类的钱包相关联，例如桥接智能合约、中心化交易平台钱包或销毁地址，这些会被排除。大约 13.5 万个地址被确定为女巫地址，也被排除在外。

我们可视化了每个地址的 Token 分配，符合条件的地址总数中只有约 8.6% 获得了 4,000 个以上的 Token ：

图 13：按可得到的 Token 数量划分的合格钱包8. 去中心化治理

The above criteria do not cover all positive contributions to the Arbitrum ecosystem. In addition to the technology stack part, Arbitrum is also a community of developers, builders, and users with a primary goal of evolving into a decentralized governance community. Thus, a portion of the airdrop is allocated to the community Treasury on the Arbitrum so that it is distributed according to the value of each subcommunity. These entities cover the various DApps and protocols that run on Arbitrum.

• Based on time deployed to Arbitrum

• Does not differentiate between Arbitrum native, Arbitrum + Ethereum, or other multi-chain protocols

• Based on total volume and daily volume data

• The economic value of these trading activities

• The demand is in direct proportion to the activity statistics of each dApps (Figure 14 shows the contract log and transaction share within each entity since the establishment of Arbitrum One)

将 dApps 作为空投分发的一部分，目标是建立更广泛的利益相关群体，他们有资格为 Arbitrum 的未来发展做出贡献。这与 Arbitrum Token 分配的核心目标一致：过渡到更去中心化的 Arbitrum 决策模型。

Figure 14: Total transactions and smart contract logs within entities as of February 6, 2023, an Arbitrum entity conclusion marked by Nansen

通过利用链上分析提取出有资格接收 Arbitrum 空投 Token 的钱包地址，Arbitrum 基金会和 Nansen 试图回答这个问题：「如何公平地办事处所有帮助过并可能继续帮助的人 Arbitrum 生态系统的努力？」。

Under the final allocation model, one-third of the Arbitrum One's addresses and the Dapps on Arbitrum meet the conditions as of the snapshot date.

Original link

欢迎加入律动 BlockBeats 官方社群：

Telegram 订阅群：https://t.me/theblockbeats

Telegram 交流群：https://t.me/BlockBeats_App

Twitter 官方账号：https://twitter.com/BlockBeatsAsia