Benefits and Limitations of Web3 (2024)

II-A Web1

In 1989, Tim Burners-Lee sent a proposal to management at CERN where he pointed out that the hierarchical system of keeping records that was in use made it difficult and time-consuming to search for related documents [3]. To solve this problem, Burners-Lee suggested a new paradigm based on hypertext, which links records to each other through hyperlinks. Later, Burners-Lee combined hypertext with the internet to create the first version of the World Wide Web.

This first version of the World Wide Web later became known as Web1. Features first introduced in Web1 have become ubiquitous with the modern web. The defining technology behind Web1 was Hypertext. Developers used HyperText Markup Language (HTML) to create content for web pages; web servers would then host the content created and allow end-users to find the content using Universal Record Locators (URLs); the end-users (will also be referred to as users) could utilize web browsers to access the web pages using the HyperText Transport Protocol (HTTP). While these technologies have been augmented and improved, they remain core to the modern World Wide Web.

In Web1, web pages were static pages that utilized Hyperlinks to connect to other websites. These early web pages did not have the capability for end-users to generate new content. Instead, the early WWW acted as a directory where end-users could find HTML files from worldwide web servers.

This model of a Read-Only web provided end-users with an easy way to traverse the internet. Figure 2 shows a high-level view of Web1. In this paradigm, end-users can read data from web pages that are hyperlinked to each other. Various service providers host these web pages. Because there is no user interactivity, this era of the web was focused on companies providing content to end-users.

While the new technology developed to support Web1 was revolutionary, users quickly realized the limitations of a Read-Only web. This model requires content creators to host their web pages, leading to only those with specialized knowledge being able to create content for the World Wide Web. This meant that creating content for Web1, where businesses created most content, was highly inaccessible. Furthermore, the lack of communication between a content creator and the end-users consuming the content made it difficult for the end-users to engage with the content meaningfully. To address these limitations, Web2 was introduced to make the web more dynamic and interactive.

II-B Web2

Web2 differentiates itself from Web1 by allowing for dynamic user-generated content. By introducing new technologies such as databases, Web2 allows for developing more complex web applications compared to the static web pages from Web1 [10]. These dynamic applications allowed users to interact with content on the web in new ways.

Figure 3 gives a high-level view of the Web2 model. This model builds on Web1, keeping ideas such as hypertext; however, it now allows users to store data in databases controlled by the web applications, allowing users to create their content and interact with the web on a deeper level.

Because Web2 allows for interactivity, this paradigm is called the Read-Write web, where users can read the data just like in Web1 but can also write their data for web applications, a bidirectional data flow; users can send and receive data from applications. To handle the bidirectional data flow, more complex forms of storing data, such as XML and RSS, were created to support Web2. Web2 changes the focus of the web away from companies or site publishers and onto communities. With the rise of Web2, we begin to see new types of applications such as blogs, wikis, and social media.

While Web2 improved the user experience over Web1 [11] [7] [10], it came with new security challenges. Web applications were now storing users’ data, putting these applications at risk of cyberattacks. This also decreased privacy as users now shared personal data with Web2 applications. Users also quickly noticed that the siloed nature of Web2 applications meant that data was often replicated across multiple applications, increasing a user’s attack surface and making updating information across platforms inefficient. While many of these problems were identified in the early days of Web2, they still exist in modern Web2 applications. Researchers began exploring new models to make the web even more versatile and mitigate the limitations presented in the Web2 paradigm.

III-A What is Web3

As early as 2006, researchers noticed the limitations of Web2. To solve these problems, they began to propose a new paradigm of a semantic web[2] [9]. The semantic web expands on Tim Burners-Lee’s original ideas of linking documents to each other through hyperlinks, but rather than linking documents, individual pieces of data are linked to each other.

To highlight the benefits of a semantic web, take an end-user of multiple Web2 social media applications. In each of these applications, the user has a short bio describing their interests. If the user finds a new interest, they must log in to each social media application and update each bio. This is time-consuming and inefficient since users must update the same data across multiple sites. The semantic web allows the user to update the data once and have that change reflected across all platforms on the web.

This new paradigm reduced the data replication in Web2 and gave users more control over their data. However, Web2 applications are currently centralized and siloed off from one another, making it difficult to share data across platforms. Thus, implementing a large-scale semantic web requires decentralization.

As cryptocurrency gained popularity and evolved, its underlying blockchain technology evolved in such a way that blockchain could provide a decentralized backbone for a large-scale semantic web [13] [8]. Cryptocurrency, through the use of blockchain, allows for a trustless, decentralized exchange of data. The same principle required by semantic web, making blockchain a viable candidate to support the semantic web.

Like the web, blockchain has evolved, adding more dynamic features [6]. Early blockchain projects such as Bitcoin were only focused on allowing users to transact cryptocurrencies. The next generation of blockchain, such as Ethereum, allowed users to create smart contracts, code that could be executed on a distributed state machine. With the dawn of generation 2 blockchains, developers began to combine semantic web ideas with decentralized smart contracts. Gavin Wood, one of the co-founders of Ethereum, first referred to this new paradigm as Web3 in 2014.

In Web 2, users relied on various third-party service providers to store and maintain their data. For example, each platform must keep a copy of the user’s bio in the social media use case. With Web3, users can store their bio on a public blockchain, such as Ethereum, via a smart contract, allowing each social media application to access this data. Thus, the user only needs to update their bio on the blockchain once and have that change reflected across all the social media platforms. In this Read-Write-Own model, the user controls their data and stores it on a blockchain, unlike Web2, where the platforms own the user’s data and are responsible for storage. Figure 4 highlights the Read-Write-Own model.

Table I provides a summary comparison of the various generations of the web; more detailed comparisons can be found in [11] [7] [4].

III-B Benefits of Web3

One of the immediate benefits of Web3’s Read-Write-Own paradigm is that users are in control of their data. Thus, users can restrict who has access to their data and when access is granted. This is in contrast to Web2, where once users give their data to a platform, the platform is free to use that data in any manner, including selling the user’s data to data brokers.

Likewise, in Web2, if a user shares their data with an application, there is no guarantee that the user will be able to remove their data from the application. Users must trust that third-party application providers securely handle their data and will remove it when requested. In contrast, in Web3, the user controls the data and can grant or revoke an application’s access to their data anytime. Furthermore, users can utilize blockchain to record which applications have access to their data and what data these applications have access to, allowing them to understand their attack surface and risk better.

In Web3, users no longer need to trust third-party applications to keep their data secure. In Web2, a security-conscious user may review SOC audits or other security reports to ensure that an application is taking steps to keep data secure. However, in Web3, since the user owns their data, they can implement their controls to ensure that the data is secured. While this may be daunting for individual users, security-conscious businesses can greatly benefit from this paradigm.

Another advantage of the Web3 paradigm is that users no longer need to replicate data across multiple applications. In Web2, a user may need to enter the same data into multiple applications, for example, updating their bio in multiple social media applications. With Web3, the user only needs to update this data one time. This makes interacting with multiple platforms on the web more streamlined, as data no longer needs to be replicated. For enterprises that need to manage complex relationships between applications, simplifying how data is entered can reduce costs and increase efficiencies.

Likewise, since users no longer need to replicate their data in multiple places, this reduces a user’s attack surface. In Web2, attackers targeting a user must find one weak application where a user has stored sensitive information to successfully steal that user’s data. In Web3, the user only needs to worry about securing one data source.

Since applications no longer store data, the user and the provider’s security risk is reduced. In Web2, an application accumulates data from many users, centralizing this data into one storage system, such as a SQL database. Since this storage system contains the data for many users of a centralized application, it makes it a high-value target for attackers. They only need to breach the storage system once to get the data for many users. In contrast, in Web3, the data is decentralized; thus, attackers no longer have a single point of attack, making it more difficult to breach all of the users’ data. Application providers can reduce threat levels and liability by switching to Web3 and decentralizing users’ data. It is critical to note that Web3 must be implemented correctly and securely for applications to experience a risk reduction; if implemented incorrectly, applications may expose themselves to higher levels of risk.

Likewise, in a Web3 model, service providers no longer need to store users’ data, which allows applications to spend less on storage. This cost reduction allows developers to focus their resources elsewhere to improve the application. In addition, since the storage is passed off to users, Web3 smart applications no longer need to worry about data storage when scaling the application.

While Web3 offers many benefits, it has struggled to become mainstream due to several drawbacks of implementing the technology.