Blockchain: Research and Applications 5 (2024) 100173

Contents lists available at ScienceDirect

Blockchain: Research and Applications

journal homepage: www.journals.elsevier.com/blockchain-research-and-applications

Research Article

An overview of Web3 technology: Infrastructure, applications,

and popularity
Renke Huang a , Jiachi Chen a,∗ , Yanlin Wang a , Tingting Bi b , Liming Nie c,d , Zibin Zheng a
a
School of Software Engineering, Sun Yat-Sen University, Guangzhou 510275, China
b
Data61, ACT 2601, Australia
c
Zhejiang Sci-Tech University, Hangzhou, 310018, China
d
Nanyang Technological University, 639798, Singapore

A R T I C L E I N F O A B S T R A C T

Keywords: Web3, the next generation of the Internet, represents a decentralized and democratized web. Although it has
Web3 garnered significant public interest and found numerous real-world applications, there is a limited understanding
Blockchain of people’s perceptions and experiences with Web3. In this study, we conducted an empirical study to investigate
Empirical study
the categories of Web3 applications and their popularity, as well as the potential challenges and opportunities
Survey
within this emerging landscape. Our research was carried out in two phases. In the first phase, we analyzed
200 popular Web3 projects associated with 10 leading Web3 venture capital firms. In the second phase, we
collected and examined code-related data from GitHub and market-related data from blockchain browsers (e.g.,
Etherscan) for these projects. Our analysis revealed that the Web3 ecosystem can be categorized into two groups,
i.e., Web3 infrastructure and Web3 applications, with each consisting of several subcategories or subdomains.
We also gained insights into the popularity of these Web3 projects at both the code and market levels and pointed
out the challenges in the Web3 ecosystem at the system, developer, and user levels, as well as the opportunities
it presents. Our findings contribute to a better understanding of Web3 for researchers and developers, which in
turn promotes further exploration and advancement in this innovative field.

1. Introduction In general, Web3 technology fosters a readable, writable, and ownable

ecosystem, enabling users to manage and maintain control over their
Web3, also known as Web 3.0, represents the third generation of data through public and private keys. Due to the substantial potential of
the World Wide Web, which is usually built upon a decentralized Web3 technology, Web3 applications have become increasingly preva-
blockchain network [1]. Within this network, third parties cannot mod- lent in recent times. As of August 2023, over 14,470 Web3 applications
ify user data, and users maintain control over their data through public have been deployed on various platforms, with more than two million
and private keys. Data changes adhere to predetermined agreements active users [2].
between users and the blockchain, establishing blockchain technology Although Web3 has garnered increased public interest, a clear def-
as the fundamental basis for Web3. Thus, Web3 is also known as the inition remains elusive. Analyzing and understanding popular Web3
“read-write-own” web because it empowers users to retain ownership projects could yield insights into Web3 and its ecosystem. In this pa-
of their data. per, we endeavor to address two research questions. RQ1: What are
Web3 technology offers several key advantages. For example, the the categories of Web3 projects? RQ2: What is the popularity of Web3
Immutability feature ensures the data integrity of the Web3 projects, projects?
and the Trustless feature allows the Web3 projects not to rely on a To answer RQ1, we initially gathered 626 Web3 projects and se-
trusted third party. Web3 also provides the feature to preserve user pri- lected the top 200 based on their investment amounts. We then man-
vacy, and Anonymity empowers users to control their data Ownability. ually analyzed the features and services provided by these projects,

* Corresponding author.
E-mail addresses: [email protected] (R. Huang), [email protected] (J. Chen), [email protected] (Y. Wang),
[email protected] (T. Bi), [email protected] (L. Nie), [email protected] (Z. Zheng).

https://doi.org/10.1016/j.bcra.2023.100173
Received 30 June 2023; Received in revised form 18 October 2023; Accepted 17 November 2023
Available online 28 November 2023
2096-7209/© 2023 THE AUTHORS. Published by Elsevier B.V. on behalf of Zhejiang University Press. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
R. Huang, J. Chen, Y. Wang et al. Blockchain: Research and Applications 5 (2024) 100173

and conducted open card sorting to cluster them into several groups. Each block is characterized by its transaction data, timestamp, and the
Finally, we discovered that the Web3 ecosystem comprises two cat- hash value of the preceding block. The blockchain functions as a public
egories, namely, Web3 infrastructure and Web3 applications with each ledger, with individual blocks housing records of diverse transactions.
further subdivided into several subcategories. The Web3 infrastructure Rather than being stored in a centralized location, the blockchain is
subcategories include DeFi, gaming, layer 2 scaling solutions [3], pri- distributed across a network of nodes, each maintaining a copy. As a re-
vacy, developer tools/services, cross-chain interoperability [4], public sult, records are public and easily verifiable by all nodes, rendering data
chains, decentralized storage, and oracles [5]. The Web3 application modification in the blockchain exceedingly costly. Modifying a block’s
subcategories encompass DeFi [6], NFT [7], Metaverse [8], DAO [9], transactions proves exceptionally challenging without attaining consen-
and Web3 for traditional scenarios. sus among all nodes. Certain blockchains, such as Ethereum, provide
To answer RQ2, we collected code-related and market-related data generic computational capabilities via smart contracts. An Ethereum
for the 200 Web3 projects from GitHub and blockchain browsers (i.e., smart contract represents an account governed by an immutable pro-
Etherscan, Bscscan, and Polygonscan). The analysis results can assist gram (i.e., bytecode). Users can initiate the execution of the bytecode
Web3 practitioners and researchers in understanding the key attributes by transmitting a transaction containing the specified execution param-
of Web3 projects, including dimensions such as code, market, and par- eters to the smart contract account.
ticipant activities. The blockchain and smart contract serve as the foundation for Web3,
In addition, we also point out the challenges and opportunities facilitating the running of applications (i.e., Web3 applications) on a
present within the Web3 ecosystem. Challenges were discussed from decentralized network. Blockchain technology offers several key advan-
the system, developer, and user perspectives. Challenges at the system- tages for Web3, including: 1) Immutability. All the data stored on the
level are interoperability, scalability, and privacy; at the developer-level blockchain cannot be changed. 2) Trustless. The interactions between
are code security and incentive mechanisms; and at the user-level are users in a blockchain do not depend on a trusted third party. 3) Avail-
usability and data recovery. In terms of opportunities, Web3 features ability. The decentralization of blockchain provides high availability,
ensure security and grant users control over their data. Additionally, as the failure of a single point on the network will not affect the normal
Web3 paves the way for new business models, enables novel forms of running of a blockchain. 4) Anonymity. Users do not need to provide
collaboration, and enhances transparency across industries. their real information, and their activities are usually anonymous in
The main contributions of this work are as follows: a blockchain. 5) Ownability. Blockchain allows users to control their
data through private and public keys.
• We conducted a comprehensive empirical study by analyzing 200
popular Web3 projects, from which we derived an understanding of 3. RQ1: The categories of Web3 projects
the Web3 ecosystem.
• We uncovered the popularity of Web3 projects from code and mar- Web3, as a novel concept, currently still lacks a clear definition and
ket perspectives by analyzing code- and market-related data from deep understanding of the ecosystem. In this section, we introduce the
popular Web3 projects. details of how to find out the categories of Web3 projects. The results
• We identified the challenges and opportunities presented in Web3. can help identify popular Web3 projects and provide further insights
This identification serves to inform researchers and developers of into the Web3 ecosystem.
the critical issues and potential prospects within Web3.
• We are open-sourcing a Web3 dataset [10], which includes 200 pop- 3.1. Approach
ular Web3 projects that received funding from venture capital (VC)
firms, as well as their corresponding code- and market-related data. To understand the Web3 ecosystem, we first identified popular
Web3 projects. Currently, the industry has been exploring the Web3
The remainder of the paper is organized as follows. Sections 3 and 4 field for some time and has invested in various Web3 projects. Thus, we
show the answers for RQ1 and RQ2, respectively. Section 5 discusses the collected 200 popular Web3 projects invested by Web3 VC firms. Then,
challenges and opportunities faced in Web3. Section 6 discusses related we conducted an analysis of the selected 200 Web3 projects, reviewing
work, and we conclude our study in Section 7. their documentation to provide a summary for each project. Finally, we
performed open card sorting on these projects. The following provides
2. Background further details on our approach.
Step 1: Web3 project collection. In this step, we first identified
2.1. Web1, Web2 and Web3 200 Web3 projects by examining the portfolios of Web3 VC firms.
We began by using the “10 VC Firms Investing in Web3 Companies”
Web1 [11] is a readable-only network, whereas Web2 [12] is a read- list published by visible.vc1 to identify 10 Web3 VC firms, including
able and writable network, and Web3 [13] is a readable and writable A16Z [14], Sequoia Capital [15], Tiger Global [16], Coinbase Ven-
network that empowers users with data ownership. In Web1, the net- tures [17], Paradigm [18], Pantera Capital [19], Ribbit Capital [20],
work primarily furnishes users with a collection of static pages for Blockchain Capital [21], Digital Currency Group [22], and Slow Ven-
viewing purposes devoid of interactive capabilities. Web2 provides an tures [23]. We then manually examined all the portfolios of these VC
array of dynamic pages, enabling user engagement with the network firms on Crunchbase2 from January 2020 to June 2022, resulting in a
through activities such as content publication, file sharing, and pay- total of 871 projects. For each project, we collected their name, offi-
ment. Nevertheless, Web2 is plagued by issues concerning the collection cial website URL, and the financing amount received. Next, we used
and storage of users’ personal data on third-party platforms, potentially a keyword filtering method to identify Web3 projects. Specifically,
leading to unauthorized sales or illicit usage without the users’ knowl- we checked the official website of each project for the presence of
edge or consent. To address these challenges, Web3 has been proposed, the keyword “Web3” or other blockchain-related keywords, such as
which not only supports readable and writable but also ensures that “blockchain”, “Web3 App”, or “decentralized” [24]. In total, we iden-
users maintain ownership of their data. tified 626 projects (71.87%) as Web3 projects. Finally, we selected the

2.2. Blockchain, smart contract and Web3

1
https://visible.vc/blog/web3-investors/.
A blockchain consists of a sequence of records, referred to as blocks, 2
Crunchbase is a company providing business information about private and
which are interconnected and secured using cryptographic techniques. public companies. The website is: https://www.crunchbase.com.

2
R. Huang, J. Chen, Y. Wang et al. Blockchain: Research and Applications 5 (2024) 100173

Fig. 1. The ecosystem of Web3.

top 200 Web3 projects based on their investment amount from Web3 3.2. Web3 infrastructure
VC firms.
Step 2: Document analysis. We conducted an analysis of the doc- Web3 infrastructure encompasses a variety of applications and solu-
umentation for each Web3 project to understand their main functional- tions designed to support developers and enhance blockchain networks.
ities. Specifically, the first and second authors conducted independent Our analysis identified nine subcategories of the Web3 infrastructure:
reviews of each project’s documentation, identifying texts that could DeFi infrastructure, gaming infrastructure, layer 2 scaling solutions, pri-
provide a preliminary summarization of the project’s functionalities. For vacy infrastructure, developer tools/services, cross-chain interoperabil-
example, Uniswap [25], a decentralized exchange, the summarization ity services, public chains, decentralized storage, and oracle services. In
text is “a peer-to-peer system designed for exchanging ERC-20 Tokens”. the following, we introduce each of these subcategories and the state-
In cases where the documentation did not provide enough informa- of-the-art Web3 projects within these subcategories in more detail.
tion about the project, we manually reviewed additional information,
such as API instructions and use-case examples, to extract relevant de- 3.2.1. DeFi infrastructure
tails regarding the project’s functionality. Based on the information we The DeFi infrastructure is designed to simplify the development of
extracted, we wrote a summarization for each project. If there was a token-swapping functionality by providing a standard set of APIs or
SDKs.
disagreement on the summary of a project between the authors, they
One of the prominent DeFi infrastructure projects is 0x [27], which
discussed and consolidated their results.
has deployed smart contracts that support ERC20 [28] token-swapping
Step 3: Open card sorting. After determining a summarization
on Ethereum, as well as multiple EVM-compatible [29] blockchains.
for each project, we performed card sorting [26] to identify cate-
These smart contracts aggregate liquidity from various decentralized
gories based on these summarizations. Card sorting is a commonly
exchanges (DEXs) [30], such as Uniswap [25]. Then, the 0x proto-
used method for deriving categories from data. There are three main
col provides APIs for token-swapping functionality to developers. By
approaches to card sorting: closed, open, and hybrid. Given that the
utilizing these APIs, developers can easily integrate token-swapping
categories of Web3 projects are relatively unknown, we decided to fol-
functionality into their Web3 applications and share the liquidity from
low an open card sorting approach to analyze these projects. During the
DEXs.
open card sorting process, we created cards for each project that con-
tained its name and summarization. Cards with similar summarizations 3.2.2. Gaming infrastructure
were grouped together to form meaningful groups, each with a specific The gaming infrastructure facilitates the integration of in-game vir-
topic. These groups are equivalent to low-level subcategories and fur- tual assets into the blockchain for developers without blockchain expe-
ther evolve into high-level categories. The resulting hierarchical struc- rience.
ture provided us with a clear categorization of the Web3 ecosystem. The Enjin [31] is a prominent gaming infrastructure that enables devel-
first and second authors participated in the card sorting process and an- opers to design games using common programming languages such as
alyzed and verified each card. Ultimately, we identified two high-level C/C++, Java, and Python. Using the Enjin platform, developers can
categories within the Web3 ecosystem: Web3 infrastructure and Web3 create in-game virtual assets, which are then assigned corresponding
applications. Each category contains several subcategories (as shown in on-chain tokens. By utilizing the API provided by Enjin, developers can
Fig. 1). integrate these on-chain tokens into their games. Enjin is also responsi-
Fig. 1 displays the results of the card sorting, providing an overview ble for monitoring requests from the game, processing requests on the
of the Web3 ecosystem. In the following, we provide an introduction to blockchain, and returning data to the game.
each of the two high-level categories, Web3 infrastructure and Web3
applications, and their subcategories. For each subcategory, we also 3.2.3. Layer 2 scaling solution
highlight the Web3 project that received the highest financing, show- Blockchain networks typically have low throughput, which limits
casing state-of-the-art projects within each subcategory. the widespread adoption of Web3 applications. Layer 2 scaling solutions

3
R. Huang, J. Chen, Y. Wang et al. Blockchain: Research and Applications 5 (2024) 100173

aim to increase network throughput without altering the underlying 3.2.7. Public chains
blockchain protocol. These solutions propose a framework for handling Public blockchains are facing the challenge of the blockchain
transactions off-chain and only reporting little information about the trilemma [42], which means it is difficult to meet the demands of de-
transaction on-chain, thereby achieving higher throughput. centralization, scalability, and security simultaneously.
Optimistic Rollups [32] is a representative layer 2 scaling solu- Many public blockchains are exploring ways to maximize scalability
tion that has been adopted by EVM-compatible layer 2 networks, e.g., while ensuring sufficient security. Near [43] is a representative project
Optimism [33]. This solution improves blockchain scalability [34] by that proposes to improve scalability through sharding technology [44],
bundling multiple transactions on the layer 2 network, compressing named Nightshade [45]. In this sharding technology, data in a block
them, and optimistically assuming their validity before submitting them are divided into multiple chunks based on the number of shards, with
to blockchain layer 1 for verification. During this process, any node each shard responsible for one chunk. The shards are allocated to a
on the layer 2 network can challenge transactions with fraud proofs portion of the validators in the network through a verifiable random
when they observe invalid transactions. The node that submitted in- function [46]. The validator of a shard only needs to store and verify
valid transactions will be punished. This mechanism ensures the va- the chunk corresponding to the shard without storing and validating
lidity of the transactions submitted by the layer 2 network to a great the complete block, thus improving scalability.
extent.
3.2.8. Decentralized storage
3.2.4. Privacy infrastructure Storing data on popular blockchain networks like Ethereum can be
In most blockchain networks, user data are publicly visible, which expensive since the data need to be stored on each node of the net-
poses a challenge to user privacy. Privacy infrastructure is designed to work. To address this issue, a decentralized storage network [47] was
prevent users’ transaction histories from being exposed to the public. designed, which operates essentially like a blockchain. In this network,
One of the representative privacy infrastructures is the Aztec pro- users’ files are encrypted and stored on nodes of the network. The nodes
tocol [35]. It consists of an Ethereum smart contract and a layer 2 only write encrypted metadata of the files (e.g., file location and file
network that adopts ZK-proof [36] technology. Users can transfer to- hash) into blocks of the network without the full files, thus reducing
kens from Ethereum to the Aztec layer 2 network by depositing tokens storage costs.
into the Aztec smart contract. In the layer 2 network, users can send to- Filecoin [48] is a popular decentralized storage network based on
kens to others or interact with some Ethereum layer 1 smart contracts the IPFS protocol [49]. In the Filecoin network, users pay fees based on
that are connected to the Aztec protocol. The transactions made by users file size and storage time, and then their files are stored on nodes. Nodes
in the Aztec layer 2 network are encoded as ZK-SNARK [37] to protect guarantee that the files have been stored and not deleted during storage
by submitting Proof-of-Spacetime [50] and Proof-of-Replication [50]
users’ transaction data from the public. Then, these transactions are
generated from the files to the Filecoin network.
bundled and sent to Ethereum layer 1. Developers can integrate their
smart contracts with the Aztec protocol using the API it provides, en-
3.2.9. Oracle
abling their Web3 applications to support user interaction in a private
An oracle [5] is an interface used to deliver off-chain data to the
manner.
blockchain for smart contracts to consume. However, delivering invalid
or malicious data to the blockchain can potentially put smart contracts
3.2.5. Developer tools/services
and user assets at risk.
Developer tools/services aim to assist developers in developing, test-
To solve this problem, the Razor Network [51] proposed a solution
ing, deploying, and managing smart contracts.
in which data providers are required to stake tokens. Providers of valid
ConsenSys [38] is a popular project that provides development data will receive token rewards, while providers of invalid data will be
tools and services, including Infura [39] and Truffle [40]. Infura is fined. This mechanism enhances the reliability of the data provided to
an Ethereum node provider that allows developers to connect Web3 the blockchain, thus improving the security of smart contracts and user
applications to Ethereum without running their own nodes. It also assets.
provides APIs for Ethereum account management, smart contract de-
ployment, transaction signing, and retrieving on-chain data. Truffle is 3.3. Web3 applications
an Ethereum-based Web3 application development framework that pro-
vides smart contract compilation, testing, deployment, and interactive Web3 applications are a set of applications developed for users. In
console features. It also provides libraries for Web3 application devel- our study, we identified five subcategories of Web3 applications: Decen-
opment. tralized Finance (DeFi), NFT, metaverse, DAO, and Web3 for traditional
scenes. Among them, the subcategories of DeFi, NFT, and Web3 for
3.2.6. Cross-chain interoperability traditional scenes contain several smaller subcategories, which are de-
Communication between different blockchains was difficult. Cross- scribed in detail below.
chain interoperability is a solution that aims to facilitate the flow of
data and value across different blockchains. 3.3.1. Decentralized Finance (DeFi)
Axelar [41] is a popular cross-chain interoperability solution that The term “DeFi” refers to the decentralized financial system that
operates as an independent decentralized network with its validators. runs on top of the blockchain. In our dataset, the DeFi subcategory is
Axelar enables interoperability between blockchains by deploying smart further divided into four smaller subcategories: 1) lending, 2) trading
contracts on these blockchains connected to Axelar. These smart con- & exchange, 3) investment, and 4) wallets. In the following, we will
tracts are controlled by a shared key that utilizes multi-party cryp- describe each of them in detail.
tography, which is divided into multiple parts and held by validators 1) Lending. The lending protocol pools funds from multiple lenders
based on the amount of Axelar tokens they have staked. These valida- to provide loans to borrowers and earns interest from the borrowers.
tors run nodes on blockchains connected to Axelar to monitor on-chain The interest payments are then distributed to the lenders as returns on
activities. They are responsible for approving token cross-chain trans- their lending.
fer requests from users after observing the deposits made by the users Aave [52] is a representative lending protocol that provides over-
to the Axelar smart contract. When more than 50% of the shares of collateralized loans and flash loans [53]. In an over-collateralized loan,
the shared key from validators approve the request, the smart contract the borrower is required to lock up crypto assets with a value exceed-
executes the token cross-chain transfer. ing the loan amount as collateral before the loan is released. In case

4
R. Huang, J. Chen, Y. Wang et al. Blockchain: Research and Applications 5 (2024) 100173

the value of the collateral falls below a predefined threshold, the smart 2) Property rights. NFT has been applied to protect digital/physical
contract initiates an auction to recover the funds lent to the borrower property rights.
and provide returns to the lender. Flash loans are uncollateralized loans In terms of protecting digital property rights, a project called
available to developers with programming experience. In a flash loan, Royal.io [62] provides NFT-based music copyright protection. At
the borrower is only required to repay the entire loan amount and in- Royal.io, artists can mint NFTs for their songs, with the song’s name,
terest before the end of a single transaction without locking any assets. hash value, and royalty share as the NFT’s metadata. When the songs
2) Trading & exchange. Trading & exchange applications are a se- are purchased by third parties, the NFT holders will receive a portion of
ries of DEXs that enable on-chain token swapping without the need the royalties as their return. In terms of protecting physical asset prop-
for a centralized custodian. However, implementing the order book erty rights, a project called Origyn [63] mints high-definition images
model [54] used by centralized exchanges (CEXs) on the blockchain of real-world physical assets such as handbags, watches, and jewelry as
often results in low efficiency. NFTs, which serve as proof of property rights for these physical assets.
In the trading and exchange field, Uniswap is a representative 3) Gaming. NFT technology has also been adopted in the gaming
project that has adopted an automated market maker (AMM) [55] algo- industry, where in-game virtual assets are minted as NFTs, returning
rithm to replace the inefficient order book model. The AMM algorithm ownership of game data to the players. This type of game is also referred
uses a liquidity pool mechanism for swapping, allowing users to have to as GameFi, enabling players to earn profits from the games.
an instant token swap. The swap ratio is determined by the ratio of two Axie Infinity [64] is a prominent GameFi project where players can
tokens in the pool using a specific algorithm. collect, breed, raise, and trade virtual pets, with each pet being stored
3) Investment. The investment protocol acts as a decentralized as an NFT on the blockchain. Players can operate their pets to battle
investment fund that encodes investment strategies into the smart con- with other pets or sell their own pets to earn token rewards.
tract. With this protocol, users can deposit funds into the smart contract, 4) Social. NFT technology has also been applied in the social field,
which then executes predefined rules to transact. After yielding profits, providing a way for users to express themselves freely without the con-
the profits are distributed to the users. trol of tech giants.
TokenSets [56] is a popular protocol in the investment field that One notable project in this field is Mirror [65], which is a decen-
provides users with multiple portfolios. Each portfolio is managed by a tralized blogging platform. On this platform, users can publish their
smart contract, which transacts a specific set of tokens on DEXs based on contents and mint them as NFTs. The content published by the user is
predefined rules. TokenSets also allows investors to write their invest- then stored on the IPFS, ensuring that the user’s data cannot be arbitrar-
ment strategies into the smart contract and provide them to the public. ily altered or deleted. Furthermore, users can purchase NFTs minted by
4) Wallets. Wallets are tools that allow users to manage their crypto creators to express their support.
assets and interact with Web3 applications.
Metamask [57] is a popular wallet that enables users to create or re-
3.3.3. Metaverse
cover their Ethereum accounts on any device by using human-readable
Metaverse [66] refers to a virtual world built by computers that can
phrases without storing complex public and private keys. With Meta-
interweave with the real world.
mask, users can manage their ERC20 tokens and interact with Web3
Many metaverse projects have been built on top of the blockchain,
applications.
characterized by the minting of characters or virtual assets of the virtual
world as NFTs. One popular metaverse built for real estate is “The Sand-
3.3.2. NFT
box” [67]. In The Sandbox, users can purchase land on the game map
NFT refers to a subcategory of Web3 applications that incorporate
and engage in secondary development. The in-game lands are recorded
NFT elements [58]. This subcategory can be further divided into the
on the blockchain as NFTs, and users can also visit lands constructed by
following smaller subcategories: 1) NFT services, 2) property rights, 3)
others.
gaming, and 4) social. The details of each subcategory are described
below.
1) NFT services. NFT services consist of the following three small 3.3.4. DAO
subcategories: i) NFT marketplace, ii) liquidity protocol, and iii) NFT DAO stands for decentralized autonomous organization, which oper-
copyright protection service. ates through smart contracts. The smart contract manages the organiza-
i) NFT marketplace. The NFT market is a platform for NFT trading. tion by executing the rules predefined by the organization’s members.
OpenSea [59] is currently the biggest NFT marketplace, offering In the DAO field, CreatorDAO [68] is a typical example that brings
users the ability to buy and sell various types of NFTs, such as artworks, together investors, creators, and supporters in a community. The DAO
music, game assets, and more. Additionally, OpenSea allows users to provides creators and their works with funding and resources to support
mint NFTs without any prior experience in blockchain technology. their creative endeavors. When creators make profits from their works,
ii) Liquidity protocol. NFTs are often associated with high prices the profits are automatically distributed to community members by the
and low liquidity. smart contract.
To address the liquidity issue, fractional.art [60] has proposed an
NFT liquidity solution that involves dividing the ownership of an NFT. 3.3.5. Web3 for traditional scenes
In this approach, the owner of the NFT locks the NFT in fractional.art’s Web3 technology has also been adopted by many traditional indus-
smart contract. The smart contract then issues multiple ERC20 tokens tries to address challenges related to incentives, collaboration, trans-
based on the locked NFT, with each token representing proportional parency, trust, etc. Based on the functionalities of Web3 in traditional
ownership of the NFT. This method effectively lowers the barriers to industries, three subcategories have been identified: 1) Web3 for incen-
acquiring NFTs and increases their liquidity. tives, 2) Web3 for transparency, and 3) Web3 for eliminating interme-
iii) NFT copyright protection. The NFT market is plagued by coun- diaries.
terfeit NFTs, which are created by malicious users by uploading images 1) Web3 for incentives. Some traditional industries are leveraging
of well-known NFTs and then minting NFTs to list on marketplaces. the incentive mechanism of Web3 by building their businesses on top
To protect the copyright of NFTs, a project called Doppel [61] has of decentralized networks.
provided a solution for detecting fake NFTs. This project uses computer For instance, Helium [69] is a blockchain-based wireless network
vision and AI models to detect and report fake NFTs. Currently, the operator that sells devices for providing wireless network coverage and
company provides detection services for NFT on Ethereum, Solana, and mining. Users can purchase these devices and deploy them to provide
Polygon. wireless networks, thereby earning tokens. This incentive mechanism

5
R. Huang, J. Chen, Y. Wang et al. Blockchain: Research and Applications 5 (2024) 100173

reduces the cost for enterprises to build and run their businesses while Table 1
allowing users to benefit from traditional business scenes. Code-related data of Web3 projects.
2) Web3 for transparency. Web3 has also been adopted by tra- Subcategory Count Code line Commits
ditional industries to address trust issues by recording key business
Public chains 17 3,386,348 124,040
information on the blockchain.
# Cross-Chain 9 3,156,655 39,720
Green Labs [70] and Blocery [71] are two companies that apply Developer tool 8 641,749 11,998
blockchain technology to the field of traceability. They record key in- # Dstorage 7 421,291 17,708
formation about products, such as production time, transfer history, Oracle 3 364,499 872
and sales records, on the blockchain. The information recorded on the # Gaming infra 2 309,826 6,330
blockchain is publicly visible and tamper-proof, which enables the en- Layer 2 scaling 2 2,071,208 15,594
Privacy 2 436,352 7,960
tire process of production and sales of products to be fully traceable,
# DeFi infra 1 105,496 17,137
thereby increasing transparency and trust. Lending 12 595,999 11,205
3) Web3 for eliminating intermediaries. Web3 eliminates the Trading & exchange 7 214,912 2,756
need for intermediaries in traditional business scenes by designing Investment 4 80,193 2,919
smart contracts to run the business. DAO 1 2,793 55
For example, Dtravel [72] is a decentralized travel platform on Wallet 2 162,163 4,533
NFT Marketplaces 2 117,305 2,977
Ethereum. On this platform, travelers can directly book accommoda-
NFT Liquidity 2 56,843 454
tions from property owners by interacting with smart contracts without # NFT CP 1 14,091 3
having to go through online travel agents. # Property rights 2 105,070 806
Gaming 3 28,002 162
4. RQ2: popularity of Web3 projects Metaverse 1 462,462 5,656
Web3 for incentive 2 98,049 3,804
# Eliminating inters 4 20,992 321
In this research question, we conducted an analysis of code-related
and market-related data of Web3 projects to gain insights into their # Cross-Chain denotes the subcategory of Cross-chain interoperability. #
popularity. The findings of this research can assist Web3 practitioners Dstorage denotes the subcategory of decentralized storage. # Gaming infra
and researchers in understanding the key attributes of Web3 projects, and # DeFi infra denote the subcategory of Gaming infrastructure and DeFi
including dimensions such as code, market, and participant activities. infrastructure, respectively. # NFT CP denotes the subcategory of NFT copy-
right protection. # Property rights denotes the subcategory of NFT property
rights. # Eliminating inters denotes the subcategory of Web3 for eliminating
4.1. Approach
intermediaries.

To gain insight into the popularity of Web3 projects at both the

code and market levels, we adopted the following methods to collect Web3 projects into two groups: open-source and non-open-source.
and analyze Web3 data. Then, we obtained the medians and maximum values of users and mar-
Step 1: Web3 data collection. In this part, we collected code- ket value for the two groups, which reflect the impact of open-sourcing
related data from Web3 projects through GitHub and market-related of Web3 projects in terms of user numbers and market value. Addi-
data from Web3 projects through Web3 blockchain browsers, i.e., Ether- tionally, for each subcategory of Web3 projects, we calculated the total
scan, Bscscan, and Polygonscan. In terms of code-related data, we number of users and market cap for all applications in each category
first manually checked if the projects had open-source repositories on (shown in Table 2), which were used to quantify the popularity of a par-
GitHub. After manually reviewing the 200 Web3 projects, we identi- ticular subcategory at the market level. Finally, we analyzed the number
fied 96 open-source Web3 projects. Finally, we collected GitHub data of users for each Web3 application, and Fig. 4 depicts the distribution
for all 96 identified projects. In terms of market-related data, we ob- of the user numbers across each Web3 application.
tained transaction and market cap data for each Web3 project between In the following, we present the findings of the popularity of Web3
January 2022 and January 2023 by utilizing the APIs of blockchain projects on two levels: code level and market level.
browsers.
Step 2: Analysis of code-level popularity. For each subcategory
4.2. Code-level findings
of the Web3 infrastructure and Web3 applications, we calculated the
number of open-source projects as well as the total code lines and com-
mits of code repositories for all projects (shown in Table 1), which were The code-level findings reflect the popularity associated with the
used to quantify the popularity of a particular subcategory at the code developer. For example, the preferences for developing specific types
level. A subcategory with more projects, code lines, and commits is of Web3 infrastructure and Web3 applications and the platform for
considered more popular. Additionally, we analyzed the programming deploying Web3 applications. Table 1 shows the code-related data of
languages used by each Web3 application and the blockchain platforms Web3 projects. The first column denotes the subcategories of open-
they were deployed. Then, we summarized the usage frequency of each source Web3 projects. Within them, rows 1–9 denote the subcategories
programming language and the number of Web3 applications deployed of Web3 infrastructures, while rows 10–21 denote the subcategories of
on each blockchain platform to understand the popularity of program- Web3 applications. Besides, Fig. 2 shows the distributions of blockchain
ming languages and blockchain platforms. networks. Following are some observations from the code-level data.
Step 3: Analysis of market-level popularity. We divided Web3 Observation 1: Public chain is the most popular subcategory in
projects into two groups: projects deployed on multiple blockchain plat- the Web3 infrastructure. As shown in Table 1, among the open-source
forms and projects deployed on a single blockchain platform. Each Web3 infrastructure projects, public chain projects have the highest
group includes the names of the Web3 projects, as well as their user number, totaling 17. Moreover, public chain projects have the high-
numbers and market values. We analyzed the median user numbers est total numbers of code lines and commits, which are 3,386,348 and
and market values of the Web3 applications in each group. By compar- 124,040, respectively.
ing the medians and maximum values of the two groups, we assessed Observation 2: Lending is the most active subcategory in Web3
the impact of deploying Web3 applications on multiple blockchains application development. As shown in Table 1, the Lending subcate-
versus a single blockchain, considering both user numbers and mar- gory has the largest number of open-source projects among the Web3
ket value. Following the aforementioned methodology, we categorized application subcategories, with 12 projects. Moreover, the total num-

6
R. Huang, J. Chen, Y. Wang et al. Blockchain: Research and Applications 5 (2024) 100173

Fig. 2. Distributions of blockchain networks.

bers of code lines and commits are higher than in other subcategories,
at 595,999 and 11,205, respectively.
Observation 3: Ethereum is the most popular blockchain net- Fig. 3. Distributions of market cap and user counts. mBCs denotes the ap-
work used to deploy Web3 applications. As shown in Fig. 2, out of plications deployed across multiple blockchains. sBC denotes the applications
the 108 Web3 applications analyzed, 56 are deployed on the Ethereum deployed on a single blockchain. OSAs denote open-source applications, while
network, accounting for 51.85% of the total. By contrast, the number of NOSAs denote non-open-source applications.
Web3 applications deployed on other blockchain networks ranges from
5 to 19, with corresponding percentages ranging from 4.63% to 17.59%.
Observation 4: Solidity is the most popular programming lan-
guage for Web3 application smart contract development. Out of the
41 open-source Web3 applications, 37 of them utilized Solidity as their
programming language for smart contract (back-end) development.

4.3. Market-level findings

The findings at the market level reflect the popularity associated

with user activity. These findings reflect the features of applications
with larger user bases and market caps, indicating the most popular
Web3 activities for users and the user counts for most applications.
Fig. 3 shows the distributions of market cap and user counts, and Fig. 4
illustrates the number of users of a Web3 application between January
2022 and January 2023. Besides, Table 2 introduces market-related
data of Web3 projects. Following are the observations from the market-
level data.
Fig. 4. Number of users of a Web3 application between January 2022 and Jan-
Observation 5: Web3 applications deployed on multiple bloc- uary 2023.
kchains have more users and a higher market cap. In Fig. 3 a), the
first box and the second box represent the distributions of market cap-
italization for Web3 applications deployed across multiple blockchains Table 2
and on a single blockchain, respectively. It can be observed that the Market-related data of Web3 projects.
maximum and median of the market cap for Web3 applications de- Subcategory # Cnt User Market cap
ployed across multiple blockchains are higher than those deployed on
Lending 14 129,188 $1,827,265,578
a single blockchain. In Fig. 3 b), the first and second boxes present the
Trading & exchange 11 843,178 $5,831,446,783
distributions of the numbers of users for Web3 applications deployed Investment 2 22,287 $2,682,758,809
across multiple blockchains and on a single blockchain, respectively. It NFT marketplace 4 2,070,447 $24,687,589,059
is observable that the maximum values and medians of user numbers for # Property rights 4 16,422 $28,864,897
Web3 applications deployed across multiple blockchains surpass those Gaming 17 193,249 $1,249,040,083
deployed on a single blockchain. Social 3 18,780 $8,753,611
Metaverse 1 220,532 $950,498,005
Observation 6: Open-sourcing Web3 applications can increase
DAO 4 49,375 $1,177,325,522
the number of users and market cap. In Fig. 3 a), the third and fourth Web3 for incentive 3 477,043 $522,377,779
boxes display the distributions of the market cap of open-source and # Eliminating inters 2 2,910 $16,049,029
non-open-source Web3 applications, respectively. We can find that the
# Cnt denotes the number of Web3 projects in the subcategory. # property
maximum and median numbers of market caps for open-source Web3
rights denotes the subcategory of NFT property rights. # Eliminating inters
applications are higher than those for non-open-source Web3 applica- denotes the subcategory of Web3 for eliminating intermediaries.
tions. In Fig. 3 b), the third and fourth boxes illustrate the distributions
of the numbers of users of open-source and non-open-source Web3
applications, respectively. We can observe that the median number highest number of users at 2,070,447 and the highest market cap at
of users for non-open-source Web3 applications is lower, while open- $24,687,589,059.
source Web3 applications have a higher maximum number of users. Observation 8: Approximately 65% of popular Web3 applica-
Observation 7: In Web3, NFT trading emerges as the most pop- tions accumulate fewer than 10,000 users in a year. The cumulative
ular activity for users. As shown in Table 2, the NFT market has the distribution function (CDF) of the number of unique users of the popular

7
R. Huang, J. Chen, Y. Wang et al. Blockchain: Research and Applications 5 (2024) 100173

Web3 applications between January 2022 and January 2023 is shown if he or she cannot understand the transaction details, resulting in the
in Fig. 4. transfer of all his or her assets to another party. Currently, many phish-
ing scams exploit this vulnerability, putting users’ assets at risk.
5. Challenges and opportunities
5.4. Opportunities in Web3
In this section, we point out the challenges faced by Web3 on three
levels: system level, developer level, and user level. Then, we discuss We present multiple opportunities that Web3 brings in the follow-
the opportunities in the Web3 ecosystem. ing:
O1) Business model. Web3 has motivated the emergence of new
5.1. Challenge at the system level business models through its decentralized design and smart contracts.
For example, in the finance industry, DeFi provides users with new ways
As the underlying infrastructure that supports Web3 applications, to store, trade, and invest their assets. In the mobile communication in-
problems in the blockchain system could potentially limit the devel- dustry, Helium [69] explores the deployment of network infrastructure
opment of Web3. We identified three key aspects that may impact the in a distributed manner, allowing people to participate and profit from
Web3 ecosystem. network deployment and operations. These examples demonstrate how
S1) Scalability. Blockchain’s transaction throughput limits hinder Web3 enables the creation of innovative business models through de-
Web3 applications from fulfilling real-time processing needs for mil-
centralization.
lions of transactions. This problem also results in users paying high gas
O2) Collaboration. DAOs facilitate new collaboration models. They
fees during network congestion. While some layer 2 scaling solutions
transform the traditional organizational model through decentralized
have been proposed, they risk reduced decentralization and potential
governance and transparent operations, providing a significant oppor-
security issues.
tunity to create a more efficient and democratic form of organization.
S2) Interoperability. Different blockchains are not able to commu-
MakerDAO [73] is a DeFi protocol built on Ethereum that allows users
nicate with each other directly, which restricts the flow of data and
to obtain stablecoins by staking crypto tokens. In MakerDAO, impor-
tokens across them. Even though some cross-chain interoperability so-
tant decisions such as adjustments to fees and the addition of staking
lutions have been proposed, they only support the transfer of tokens but
token types are decided through a voting process by MakerDAO token
not communication between applications on different blockchains.
holders.
S3) Privacy. The data stored on a blockchain are publicly visible,
O3) Transparency. Web3 introduces a new method of informa-
which makes user privacy more vulnerable to threats, especially when
tion recording that enhances transparency in various business scenarios.
the relationship between a wallet and a physical identity is revealed.
Recording information on a blockchain enables it to be publicly visible
While some privacy protection solutions have been proposed, they are
and immutable, thereby addressing issues related to trust and trans-
still unable to fully hide users’ account balances and transaction histo-
parency. For instance, supply chain management utilizes blockchain to
ries.
record relevant information, resulting in increased transparency, trust-
worthiness, and management efficiency.
5.2. Challenge at the developer level
O4) Ownership of data. Web3 restores data ownership to users. In
Web3, personal user data are stored on the blockchain and controlled by
In developing Web3 applications, developers may face two chal-
the user who holds the private key. This eliminates the control of cen-
lenges. The details are as follows:
tralized institutions over user data. For example, when a game adopts
D1) Code security. The code of Web3 applications deployed on the
NFT technology, the game operator cannot modify the user’s data, en-
blockchain is publicly accessible, which makes it easier for hackers to
suring the security of user data.
exploit and launch attacks. Currently, there is a lack of mature code
auditing tools, which puts pressure on developers to be cautious when
writing application logic. Moreover, there have been instances where 6. Related work
even audited code has been attacked, resulting in significant losses of
crypto assets. There have been several previous studies on the topic of Web3. The
D2) Incentive mechanism design. Poorly designed economic sys- details are as follows:
tems in Web3 projects can lead to failure. Many Web3 projects use Wang et al. [24] conducted the first empirical investigation of exist-
token-based incentive mechanisms, but if the reward and consumption ing Web3 projects, extracting a concise backbone model that delineates
rules of the token are not well thought out, the token’s value may plum- the participating roles and operational workflows of Web3 projects.
met, causing the project to collapse. There have been several cases of Subsequently, they proposed twelve distinct Web3 architectural de-
Web3 projects failing due to issues with their economic systems. signs, capturing the operational mechanisms of typical Web3-based ap-
plications. In their study, they deconstructed a complete Web3 service
5.3. Challenge at the user level into three-tiered components based on data workflow considerations.
The data within each component can be managed through on-chain, off-
The following describes two challenges that users face while inter- chain, or hybrid methodologies. The identified design types effectively
acting with Web3. represent the full spectrum of potential Web3 applications and ser-
U1) Data recovery. The loss of a private key can result in the loss vice combinations. To evaluate the merits of each architectural design,
of personal data and assets in Web3. Private keys serve as credentials the authors meticulously assessed each configuration using a variety of
for users to access and manage personal data in Web3. Losing them can property metrics derived from classic blockchain systems. Additionally,
lead to the crypto assets being locked and unrecoverable. Therefore, it they explored which participating entities stand to gain the most advan-
is necessary to design a mechanism for recovering user data. tages under varying design types. Expanding the scope of their research
U2) Usability. The user interface of Web3 applications is more com- beyond architectural design, they examined the broader implications of
plex and challenging to operate than that of Web2 applications. When Web3, discussing its impacts, opportunities, and challenges.
interacting with a Web3 application, users are typically presented with Sheridan et al. [1] delineated the fundamental components of Web3
a wallet interface containing complex strings, which can be confusing implementation, encompassing blockchain networks, Web3 program-
and result in potential asset loss. For example, a user may inadvertently ming languages, Web3 libraries, smart contracts, and wallets. Further-
sign a “setApprovalForAll” transaction launched by a Web3 application more, they presented an overview of Web3 developers as well as the

8
R. Huang, J. Chen, Y. Wang et al. Blockchain: Research and Applications 5 (2024) 100173

impacts and risks associated with Web3. Finally, they projected possi- CRediT authorship contribution statement
ble future integrations of Web3 with other emerging technologies.
Yu et al. [74] delved into the essence of Web3 by examining an The contributions of each author to the paper are summarized as
extensive range of real-world Web3 projects. They conducted a com- follows.
parative analysis between existing Web2 solutions and self-proclaimed Renke Huang’s contributions include:
Web3 projects to identify the fundamental characteristics of Web3 ap- 1. Collecting data which include Web3 projects, and their corre-
plications and their dependencies, as well as to distinguish their differ- sponding market-related and code-related data.
ences from traditional Web2 applications. Through a thorough investi- 2. Conducting an in-depth analysis of the gathered data on Web3
gation, the authors proposed a seamless transition framework, dubbed projects.
WebttCom, aimed at facilitating the transition from Web2 to Web3. 3. Writing paper.
The innovative WebttCom framework offers efficient and reliable access
Jiachi Chen’s contributions are characterized by:
control and user management across both decentralized Web3 and cen-
1. Proposing the idea.
tralized Web2 environments. Furthermore, it presents a viable method
2. Providing crucial guidance during the data collection, analysis,
for implementing the transition with widely-used software as a ser-
and writing of the paper.
vice platform, and showcases a practical use case implemented by this
3. Engaging in meaningful discussions with the first author, which
framework. In order to assess the framework’s effectiveness from the
further refined the research.
developers’ perspective, they conducted interviews with five proficient
developers. The feedback obtained from these interviews demonstrated Yanlin Wang’s contributions can be encapsulated as:
that the research question was adequately addressed by the proposed Partaking in substantial discussions with the first author and propos-
WebttCom framework. Subsequently, the authors proposed several rec- ing vital suggestions for enhancing the quality of the paper.
ommendations for enhancing the WebttCom framework, which include
extending its compatibility to incorporate additional blockchain plat- Tingting Bi has contributed by:
forms and exploring further potential business cases. Proposing significant suggestions aimed at improving the overall
Liu et al. [75] introduced the first comprehensive and quantifiable clarity of paper writing.
metric, referred to as verifiability, for delineating the Web3 era, which Liming Nie has contributed by:
has been grounded in empirical observations and rational analysis of Proposing significant suggestions aimed at improving the overall
the evolution of blockchain infrastructure in recent years. In light of clarity of paper writing.
this characterization, the authors identified three fundamental infras-
tructural enablers for Web3: individual blockchains with smart contract Zibin Zheng’s role is critical in:
capabilities, centralized or federated state publishers, and interoper- Proposing the idea and guiding the research direction.
ability platforms designed to bridge the gaps between these disparate
systems. Subsequently, the authors offered an in-depth exploration of Declaration of competing interest
one of these three core enablers: HyperService, the pioneering interop-
erability platform that seamlessly connects heterogeneous blockchains The authors declare that they have no known competing financial
and federated or centralized state publishers to establish a unified, interests or personal relationships that could have appeared to influence
cohesive computing platform for Web3 applications. The researchers the work reported in this paper.
implemented a prototype of HyperService, consisting of approximately
62,000 lines of code, and assessed its performance using three distinct Funding
categories of cross-chain decentralized applications. The experimental
results demonstrate that HyperService imposes an end-to-end dApp ex- This work is partially supported by fundings from the National Key
ecution latency on the order of seconds while also exhibiting horizontal R&D Program of China (2022YFB2702203), the National Natural Sci-
scalability on the platform. ence Foundation of China (61972359).
In contrast to previous studies, our research focuses on delineating
the Web3 ecosystem through popular projects, analyzing overviews and References
representative applications for each field, and evaluating the popularity
[1] D. Sheridan, J. Harris, F. Wear, et al., Web3 challenges and opportunities for the
of Web3 projects. Additionally, we underscored the opportunities pre- market, arXiv, 2022, preprint, arXiv:2209.02446.
sented by Web3 and emphasized the challenges it poses for blockchain [2] Dappradar industry overview, https://dappradar.com/industry-overview. (Ac-
systems, developers, and users. cessed 29 June 2023).
[3] C. Sguanci, R. Spatafora, A.M. Vergani, Layer 2 blockchain scaling: a survey, arXiv,
2021, preprint, arXiv:2107.10881.
7. Conclusion
[4] B. Pillai, K. Biswas, V. Muthukkumarasamy, Cross-chain interoperability among
blockchain-based systems using transactions, Knowl. Eng. Rev. 35 (2020) e23,
In this study, we conducted a comprehensive empirical study to gain https://doi.org/10.1017/S0269888920000314.
insights into the Web3 ecosystem. We first selected the top 200 Web3 [5] A. Beniiche, A study of blockchain oracles, arXiv, 2020, preprint, arXiv:2004.07140.
projects and performed open card sorting to investigate the categories [6] P. Schueffel, Defi: decentralized finance - an introduction and overview, J. Innov.
Manag. 9 (3) (2021) I–XI, https://doi.org/10.24840/2183-0606_009.003_0001.
of Web3 applications, which comprise two main categories: Web3 in- [7] Q. Wang, R. Li, Q. Wang, et al., Non-fungible token (NFT): overview, evaluation,
frastructure and Web3 applications, each containing several subcate- opportunities and challenges, arXiv, 2021, preprint, arXiv:2105.07447.
gories. We introduced each subcategory and its representative project [8] M. Sparkes, What is a metaverse, New Sci. 251 (3348) (2021) 18, https://doi.org/
to provide a comprehensive overview. Then, we collected code-related 10.1016/S0262-4079(21)01450-0.
[9] Y. El Faqir, J. Arroyo, S. Hassan, An overview of decentralized autonomous organi-
and market-related data for the 200 Web3 projects from GitHub and
zations on the blockchain, in: Proceedings of the 16th International Symposium
blockchain browsers. This information helps developers better under- on Open Collaboration, ACM, 2020, pp. 1–8, https://doi.org/10.1145/3412569.
stand and navigate the Web3 landscape. In addition, we discussed the 3412579.
challenges facing the Web3 ecosystem at the system, developer, and [10] Web3 dataset, https://github.com/popularWeb3projects/Web3-dataset. (Ac-
user levels, while also highlighting the opportunities within Web3. Our cessed 29 June 2023).
[11] Web 1.0, https://www.w3.org/2010/Talks/0119-next-web-plh/web10.html. (Ac-
findings can help researchers, developers, and other stakeholders en- cessed 29 June 2023).
hance their understanding and contribute effectively to the evolving [12] Web 2.0, https://en.wikipedia.org/w/index.php?title=Web_2.0&oldid=
Web3 ecosystem. 1083050427. (Accessed 29 June 2023).

9
R. Huang, J. Chen, Y. Wang et al. Blockchain: Research and Applications 5 (2024) 100173

[13] Web 3.0, https://en.wikipedia.org/w/index.php?title=Web3&oldid=1083301005. [44] G. Wang, Z.J. Shi, M. Nixon, et al., Sharding on blockchain, in: Proceedings of the
(Accessed 29 June 2023). 1st ACM Conference on Advances in Financial Technologies, ACM, 2019, pp. 41–61,
[14] A16z, https://a16z.com/, 2022. https://doi.org/10.1145/3318041.3355457.
[15] Sequoia Capital, https://sequoiacap.cn/, 2022. [45] A. Skidanov, I. Polosukhin, Nightshade: near protocol sharding design, https://
[16] Tiger Global, https://www.tigerglobal.com/, 2022. pages.near.org/downloads/Nightshade.pdf.
[17] Coinbase Venture, https://ventures.coinbase.com/, 2022. [46] S. Micali, M. Rabin, S. Vadhan, Verifiable random functions, in: 40th Annual Sym-
[18] Paradigm, https://www.paradigm.xyz/, 2022. posium on Foundations of Computer Science (Cat. No. 99CB37039), IEEE, 1999,
[19] Pantera Capital, https://panteracapital.com/, 2022. pp. 120–130, https://doi.org/10.1109/SFFCS.1999.814584.
[20] Ribbit Capital, https://ribbitcap.com/, 2022. [47] N.Z. Benisi, M. Aminian, B. Javadi, Blockchain-based decentralized storage net-
[21] Blockchain Capital, https://blockchain.capital/, 2022. works: a survey, J. Netw. Comput. Appl. 162 (2020) 102656, https://doi.org/10.
[22] Digital Capital Group, https://dcg.co/, 2022. 1016/j.jnca.2020.102656.
[23] Slow Ventures, http://main.slow.co/about/, 2022. [48] Filecoin, https://filecoin.io/. (Accessed 29 June 2023).
[24] Q. Wang, R. Li, Q. Wang, et al., Exploring web3 from the view of blockchain, arXiv, [49] J. Benet, Ipfs-content addressed, versioned, p2p file system (draft 3), arXiv, 2014,
2022, preprint, arXiv:2206.08821. preprint, arXiv:1407.3561.
[25] Uniswap protocol, https://uniswap.org/. (Accessed 29 June 2023). [50] J. Benet, D. Dalrymple, N. Greco, Proof of replication, Protocol Labs, July 27, 2017,
[26] D. Spencer, Card Sorting: Designing Usable Categories, Rosenfeld Media, 2009, 20, https://filecoin.io/proof-of-replication.pdf.
https://rosenfeldmedia.com/wp-content/uploads/2014/11/Card-Sorting-Excerpt. [51] Razor network, https://razor.network/. (Accessed 29 June 2023).
pdf. [52] Aave, https://aave.com/. (Accessed 29 June 2023).
[27] 0x, https://0x.org/. (Accessed 29 June 2023). [53] D. Wang, S. Wu, Z. Lin, et al., Towards understanding flash loan and its applications
[28] Erc-20 token standard, https://ethereum.org/en/developers/docs/standards/ in DeFi ecosystem, arXiv, 2020, preprint, arXiv:2010.12252.
tokens/erc-20/. (Accessed 29 June 2023). [54] M.M. Abudy, A. Wohl, Corporate bond trading on a limit order book exchange, Rev.
[29] R. Jia, S. Yin, To EVM or not to EVM: blockchain compatibility and network ef- Finance 22 (4) (2018) 1413–1440, https://doi.org/10.1093/rof/rfx054.
fects, in: Proceedings of the 2022 ACM CCS Workshop on Decentralized Finance [55] J. Xu, K. Paruch, S. Cousaert, et al., Sok: decentralized exchanges (DEX) with auto-
and Security, ACM, 2022, pp. 23–29, https://doi.org/10.1145/3560832.3563442. mated market maker (AMM) protocols, arXiv, 2021, preprint, arXiv:2103.12732.
[30] S. Malamud, M. Rostek, Decentralized exchange, Am. Econ. Rev. 107 (11) (2017) [56] Tokensets, https://www.tokensets.com/. (Accessed 29 June 2023).
3320–3362, https://doi.org/10.1257/aer.20140759. [57] Metamask, https://metamask.io/. (Accessed 29 June 2023).
[31] Enjin, https://enjin.io/. (Accessed 29 June 2023). [58] Q. Wang, R. Li, Q. Wang, et al., Non-fungible token (nft): overview, evaluation,
[32] T. Schaffner, Scaling public blockchains. A comprehensive analysis of opportunities and challenges, arXiv, 2021, preprint, arXiv:2105.07447.
optimistic and zero-knowledge rollups, University of Basel, https:// [59] Opensea, https://OpenSea.io. (Accessed 29 June 2023).
www.unibas.ch/fileadmin/user_upload/wwz/00_Professuren/Schaer_DLTFintech/ [60] fractional, https://fractional.art/. (Accessed 29 June 2023).
Lehre/Tobias_Schaffner_Masterthesis.pdf. [61] doppel, https://www.doppel.com/. (Accessed 29 June 2023).
[33] Optimism, https://www.optimism.io/. (Accessed 29 June 2023). [62] Royal, https://royal.io/. (Accessed 29 June 2023).
[34] A. Chauhan, O.P. Malviya, M. Verma, et al., Blockchain and scalability, in: 2018 [63] origyn, https://www.origyn.com/. (Accessed 29 June 2023).
IEEE International Conference on Software Quality, Reliability and Security Com- [64] Axie Infinity, https://axieinfinity.com/. (Accessed 29 June 2023).
panion (QRS-C), IEEE, 2018, pp. 122–128, https://doi.org/10.1109/QRS-C.2018. [65] Mirror, https://mirror.xyz/. (Accessed 29 June 2023).
00034. [66] T.R. Gadekallu, T. Huynh-The, W. Wang, et al., Blockchain for the metaverse: a re-
[35] Aztec, https://aztec.network/. (Accessed 29 June 2023). view, arXiv, 2022, preprint, arXiv:2203.09738.
[36] X. Sun, F.R. Yu, P. Zhang, et al., A survey on zero-knowledge proof in blockchain, [67] The sanbox, https://www.sandbox.game/. (Accessed 29 June 2023).
IEEE Netw. 35 (4) (2021) 198–205, https://doi.org/10.1109/MNET.011.2000473. [68] Creatordao, https://creatordao.com/. (Accessed 29 June 2023).
[37] M. Petkus, Why and how zk-snark works, arXiv, 2019, preprint, arXiv:1906.07221. [69] Helium, https://www.helium.com/. (Accessed 29 June 2023).
[38] Consensys, https://consensys.net/. (Accessed 29 June 2023). [70] Greenlabs, https://greenlabs.co.kr/. (Accessed 29 June 2023).
[39] Infura, https://www.infura.io/. (Accessed 29 June 2023). [71] Blocery, https://blocery.io/. (Accessed 29 June 2023).
[40] Truffle, https://trufflesuite.com/. (Accessed 29 June 2023). [72] Dtravel, https://dtravel.com/. (Accessed 29 June 2023).
[41] Axelar, https://axelar.network/. (Accessed 29 June 2023). [73] Makerdao, https://makerdao.com/. (Accessed 29 June 2023).
[42] G.D. Monte, D. Pennino, M. Pizzonia, Scaling blockchains without giving up decen- [74] G. Yu, X. Wang, Q. Wang, et al., Towards web3 applications: easing the access and
tralization and security: a solution to the blockchain scalability trilemma, in: Pro- transition, arXiv, 2022, preprint, arXiv:2210.05903.
ceedings of the 3rd Workshop on Cryptocurrencies and Blockchains for Distributed [75] Z. Liu, Y. Xiang, J. Shi, et al., Make web3.0 connected, IEEE Trans. Dependable
Systems, ACM, 2020, pp. 71–76, https://doi.org/10.1145/3410699.3413800. Secure Comput. 19 (5) (2022) 2965–2981, https://doi.org/10.1109/TDSC.2021.
[43] Near, https://near.org/. (Accessed 29 June 2023). 3079315.

10