Investigations of Top-Level Domain Name Collisions in Blockchain Naming Services

Daiki Ito, Yuta Takata, Hiroshi Kumagai, Masaki KamizonoWWW 2024 · OpenReview

TL;DR

As blockchain naming services (BNS) like Handshake and Decentraweb allow users to register any string as a Top-Level Domain (TLD), they create a "Wild West" of naming that conflicts with both each other and the traditional ICANN-managed Internet. This paper quantifies these "name collisions," identifying thousands of overlaps that pose significant security risks, such as man-in-the-middle attacks and misdirected cryptocurrency transfers.

What Problem Does This Paper Address?

The traditional Internet relies on a centralized authority, ICANN, to ensure that every Top-Level Domain (like .com or .org) is unique. Blockchain Naming Services (BNS) disrupt this by allowing permissionless registration of TLDs on a blockchain. However, because there is no central coordinator across different BNS platforms, two different people can register the same TLD (e.g., .meta) on two different blockchains.

This creates a name collision. If your browser or crypto wallet doesn't know which "root" to trust, a TLD might resolve to different IP addresses or wallet addresses depending on the software you use. This isn't just a technical glitch; it’s a security nightmare that enables credential theft, information leaks, and financial loss.

Methods

The researchers performed a large-scale empirical study focusing on two major BNS platforms that allow TLD management: Handshake (HNS) and Decentraweb (DWEB).

  1. Data Collection: They synchronized a Handshake full node (hsd) to extract TLD strings and metadata from block height 0 to 184,000. For Decentraweb, they used the NFTPort API and SDKs to collect TLD data from the Ethereum and Polygon blockchains.
  2. Collision Analysis: They cross-referenced the TLD lists to find overlaps between the two BNSs and compared them against ICANN’s official list of operational and candidate TLDs.
  3. Squatting & Brand Analysis: To identify domain squatting, they compared BNS TLDs against the Tranco Top 1M list and a database of companies with over 1,000 employees (D&B Hoovers).
  4. Technical Audit: They analyzed DNS Resource Records (RRs) to see how these TLDs were actually being used (e.g., pointing to IP addresses vs. just being held as assets).

Datasets

The study utilized several key datasets:

  • Live Blockchain Data: 11,595,406 Handshake TLDs and 11,889 Decentraweb TLDs.
  • ICANN TLD List: The official list of operational TLDs and the "New gTLD Program" candidate list.
  • Tranco Top 1M: Used to identify famous web service strings.
  • D&B Hoovers: Used to identify major corporate brand names.
  • Zonefiles.io: Used to count the number of domain names registered under specific ICANN TLDs for comparison.

Key Insights and Ideas

  • The Irony of "De-centralization": While Handshake is built to be decentralized, the study found that 99.9% of Handshake TLDs with IP records pointed to the same two IP addresses owned by Namebase, a popular marketplace. In practice, the ecosystem has reverted to a centralized model managed by a few dominant players.
  • Digital Identity vs. Web Infrastructure: A massive number of collisions occurred in "lastname" TLDs (e.g., .gonzalez, .perez). This suggests users view BNS primarily as a way to establish a permanent digital identity (e.g., firstname.lastname) rather than building new web infrastructure.
  • The "Buzzword" Collision: Squatting is rampant. Nearly 45.6% of colliding TLDs matched famous brands or organizations. Users are speculative, registering TLDs they hope to sell back to companies later.
  • Complexity of Mitigation: Traditional ICANN mitigation (like "Controlled Interruption," where colliding names resolve to a specific loopback IP) is hard to enforce in a permissionless BNS environment where no one has the authority to "shut down" a record.

Key Results

  • BNS-to-BNS Collisions: The researchers identified 6,973 TLDs that were registered on both Handshake and Decentraweb.
  • BNS-to-ICANN Collisions: They found 10 collisions with currently operational ICANN TLDs (including .music and .kids) and identified 153 unique TLDs that conflict with ICANN’s upcoming "New gTLD" candidates.
  • Usage Gap: Most TLDs are "parked." While Handshake had a high rate of DNS record settings (due to automatic marketplace defaults), only 0.04% of Decentraweb TLDs had DNS records configured, suggesting they are being held purely as speculative assets.
  • Character Trends: 3-character TLDs were the most common site of collisions. Interestingly, combining alphanumeric characters with emojis significantly reduced the likelihood of a collision.

Limitations and Future Work

The authors acknowledge that their study is a snapshot in time (as of August 2023). Because BNS is rapidly evolving, these numbers change daily. A major limitation is the lack of public zone files; while the TLDs themselves are on the blockchain, the subdomains (SLDs) registered under them are often kept in private name servers, making it difficult to see the full scale of the ecosystem.

Future work should involve collaborating with the IETF and W3C to create standards that allow BNS and the traditional DNS to coexist. The authors also suggest that BNS could find a legitimate niche as a "backup" or "mirror" for retired ICANN TLDs, rather than just competing for the same namespace.

Summary updated

Abstract

Traditionally, top-level domains (TLDs) are managed by the Internet corporation for assigned names and numbers (ICANN), and the domain names under them are managed by registrars. Against such centralized management, a blockchain naming service (BNS) has been proposed to manage TLDs on blockchains without authority intervention. BNS users can register TLD strings as non-fungible tokens and manage the TLD root zone. However, such decentralized management results in the introduction of a new security issue, BNS TLD name collision, wherein the same TLD is registered in several different BNSs. In this study, we investigated BNS TLD name collisions by analyzing TLDs registered on two BNSs: Handshake and Decentraweb. Specifically, we collected TLDs registered in Handshake and Decentraweb and the associated data, and analyzed the data registration status of BNS TLDs and BNS TLD name collisions. The analysis of 11,595,406 Handshake and 11,889 Decentraweb TLDs revealed 6,973 BNS TLD name collisions. In particular, lastname TLDs, which are intended for use as person names, yielded a large number of registered domain names. In addition, the analysis identified 10 name collisions between the BNS and operational ICANN TLDs. Further, the ICANN TLD candidates under review also had name collisions against the BNS TLDs. Consequently, based on the characteristics of these name collisions and discussions in BNS communities, we considered countermeasures against BNS TLD name collisions. For the further development of BNSs, we believe that it is essential to discuss with the existing Internet communities and coexist with the existing Internet.
BibTeX
@inproceedings{ito2024investigations,
  author = {Daiki Ito and Yuta Takata and Hiroshi Kumagai and Masaki Kamizono},
  title = {{Investigations of Top-Level Domain Name Collisions in Blockchain Naming Services}},
  booktitle = {WWW},
  year = {2024},
  url = {https://openreview.net/forum?id=5nINTZKe4d},
}

Loading related content...

Figure