In our view, the emergence of blockchain technology, a foundational layer of the Web3 tech stack, has started a seismic shift in the gaming industry. That said, there are 2.7 billion gamers in the world. Meanwhile, DappRadar calculated just over 1 million daily active blockchain game wallets in October, and popular blockchain game Axie Infinity reported 2.7 million DAUs in November. There are clearly barriers that must be overcome to realize mass adoption — let alone a large, real-time, shared, and persistent virtual environment, such as the “metaverse”. We cover these barriers in detail in our previous Insights piece, but they include high and volatile transaction fees, slow transaction speeds, and a complex user experience.

A key challenge to overcome these barriers is improving Web3’s infrastructure without compromising on its benefits (e.g., greater user security and control). In this Insights piece, we’ll explore the potential benefits of Web3 for users and creators, and how decentralized infrastructure may support core parts of the web in the coming years. 

Let’s start at the beginning. Before the web, there was the internet, which was created in the 1960s and 1970s during the peak of the Cold War. The US developed a decentralized network of many computers throughout the country that would allow its defense system to remain operational even after a Soviet nuclear attack. Then, in the 1990s, Tim Berners Lee created the World Wide Web, one of the first applications of the internet. Browsers like Microsoft Internet Explorer allowed users to “surf” the internet and browse content easily. This was Web1 and it was decentralized (powered by regular computers), open-source (anyone could build on it freely), and read-only (very few people had the technical skill to publish on it). 

In the mid-2000s, websites such as Facebook and YouTube emerged, bringing about Web2. These websites allowed anyone to publish content online regardless of technical skill. However, whereas the hallmark trait of the internet and Web1 was its decentralization and democratization of information, Web2 companies sought to build walled gardens to keep users in their ecosystems. Indeed, with Web2, information is increasingly siloed and controlled by a handful of large technology companies. 

The current era of Web2 has largely relied on the centralization of data and, by extension, power. Virtually all web applications — whether it is Facebook, Twitter, Gmail, etc. — are centralized in servers owned by a few large corporations. Companies store and manage users’ data so that users don’t have to and their terms of service dictate how data is treated. For example, a select few companies have an outsized share of the world’s cloud infrastructure market with over 65% controlled by Amazon, Microsoft, and Google. Shifting from data storage to web traffic, Facebook has 2.9 billion monthly active users, equivalent to more than 60% of the world’s 4.7 billion active internet users. Simply put, a few large technology companies effectively control what people see and do on the web. Thus, the internet has evolved into something more closely resembling an oligopoly (where a few corporations own and control user data) as opposed to a democracy (where users own and control their data). 

This centralization results in several issues for users and creators:

• Economic issues. Value accrues to intermediaries and by extension a privileged set of individuals (i.e., founders, employees, and investors) as opposed to users and creators. This creates challenges for creators whose advertising revenue can be significantly impacted by changes to platform advertising policies and algorithms. Many large platforms charge their creators substantial take-rates (20%+), eating into creator profits. 

• Social issues. In order to maintain platform defensibility, Web2 companies typically  retain, harvest, and sell creator and user data. This can lead to data privacy and user sovereignty issues. For example, Facebook has been accused of putting engagement ahead of user well-being and ethical considerations. In October 2018, it suffered a hack that exposed data of over 50 million users. And, according to Pew Research, 91% of American adults agree that consumers have lost control over how personal information is collected and used by companies.

• Distribution issues. In Web2, platforms control distribution and how users connect with their friends. For creators whose businesses sit on top of these platforms, this can have significant implications. For example, mobile video streaming app Meerkat was effectively banned from Twitter, because it competed with Periscope, a similar app that Twitter recently acquired.

• Political issues. Platforms like Facebook and Twitter have become central figures in presidential election results and been accused of perpetuating fake information. These companies have also been forced to take a stance on political debates. For example, Twitter famously suspended President Donald Trump’s account in January 2021. Meanwhile, centralized servers make it easy for governments to block access to them, like when Turkey blocked access to Wikipedia.  

Simply put, there are several problems with today’s web. Web3 seeks to correct these issues, restoring user and creator sovereignty.

The transition to Web3 offers several potential benefits. These include:

• Economic benefits. Web3 fully embraces open economies where everyone shares in the upside. Whereas centralized Web2 platforms did not allow users to participate in growth, Web3 introduces new digital primitives like fungible and non-fungible tokens. These tokens provide upside and ownership for users’ contributions to a project. And as discussed in the prior Insights piece, these are being popularized in collectibles and gaming.

• Social benefits. Web3 projects allow users to decide what data and information they will share. The decentralized web seeks to break away from organizations monopolizing the flow of information on the internet, in which these companies collect, harvest, and sell users’ data.

• Distribution benefits. Web3 allows users to flip the tables on platforms. Web3 composability and data transparency enable applications to be built on top of each other, transcending the current understanding of platforms as walled gardens. This fundamentally changes product and tech distribution and enables exponential and public social graphs. 

• Political benefits. Web2 put control over dissemination in the hands of centralized entities. Web3 seeks to decentralize this decision-making process by putting governance in the hands of a project’s community. These online communities are forming Decentralized Autonomous Organizations (“DAOs”), which are owned and operated via a token. In DAOs, decisions are made by their members. Therefore, decisions about monetization, censorship, and others, that have historically been centralized, are now decentralized among DAO members.

• Composability benefits. In contrast to Web2’s walled gardens, one of the key features of Web3 is composability. Composability is the ability to mix and match software components like lego blocks. The concept is similar to open source, where source code is freely available, but goes a step further in that a system/application can integrate, exchange, and use another system’s functionality and data. It’s why a16z’s Chris Dixon said “Composability is to software as compounding interest is to finance.” Simply put, Web3 more easily allows for software to build on other software.

But it is still early. In order to accomplish a more autonomous and open network for all, systems and protocols need to be re-architected. 

Web3 is an effort to fix the economic, social, and political problems of Web2. In order to realize this vision, Web3 needs infrastructure across storage, computation, indexing, and other services. At the same time, the stack has to 1) allow for users to be onboarded more seamlessly and 2) not sacrifice Web3’s values. Before going further, it may help to visualize the Web3 tech stack:

At BITKRAFT, we are focused on several areas within the Web3 infrastructure stack. Below we discuss two areas of particular interest:

One of Web3’s goals is to move away from centralized servers. Gone are the days of centrally managed repositories storing data. In their place are the decentralized protocols familiar to those in the crypto space. Layer 1 blockchains, like Ethereum, can store a small amount of data across thousands of computers throughout the world. However, one issue is that Layer-1’s are often limited in their amount of on-chain storage capacity. As a result, media attachments and properties associated with NFTs (“NFT metadata”) are not consistently decentralized due to the gas costs associated with storing and modifying this data on-chain. 

There are several decentralized data storage companies working to solve this problem by creating their own blockchains that provide decentralized on-demand or long-term data storage. Some examples include Filecoin, Arweave, and Sia. Additionally, the Web3 benefits of user sovereignty increases the importance of protecting user data. Companies are focusing on decentralized storage solutions as it relates to user identity and privacy. For example, Spruce (a BITKRAFT portfolio company) allows users to store and control their personal data. 

The growth in gaming and collectibles NFT popularity over the past several quarters has created increased demand for decentralized storage. Per The Web3 Index, Filecoin and Arweave are both revenue generating, with Arweave significantly outperforming over the last two months.. The NFT marketplace OpenSea now provides creators with the ability to decentralize their metadata using Filecoin, and allows collectors to view the integrity and immutability of a token’s metadata. Another example is Mirror, the Web3 publishing platform, which stores all data published on its platform on Arweave. 

In summary, storage is a key layer of Web3 infrastructure. As more NFTs and other digital assets are stored on permissionless decentralized blockchains, the storage of the associated metadata will be just as important. 

Compute power is required to support rendering, transcoding, motion capture, and a variety of other applications and services. In Web2, tech giants, like Amazon and Google, bundle these services in addition to providing content to users (e.g., Twitch and Youtube). Web3 companies seek to unbundle many of these services. 

Video transcoding, the process of reformatting raw video files into viewable files based on a user’s device and bandwidth (2G-5G), is one vertical where this unbundling is currently taking place. According to Cisco, video will make up 82% of all internet bandwidth in 2022. Meanwhile, Grand View Research estimates the global video streaming market will reach $224 billion by 2028. This is not surprising given the growth in video streaming across services such as Netflix, Disney+, Fortnite Concerts, Twitch, and TikTok. 

Video transcoding is a key process for streaming video. But, for companies who use Web2 infrastructure, the costs can be expensive and unreliable. For expenses, processing costs are around $3 per stream per hour to a cloud service such as Amazon, up to $4,500 per month for one media server, and up to $1,500 per month before bandwidth for a content delivery network. For reliability, a medium VM on AWS can only transcode 2-3 concurrent streams in a standard bitrate ladder (1080p, 720p, 480p, 360p, 240p). One potential solution is operating media server instances and load balance between them but this can lead to failure if not properly managed.

One company working to address these issues is Livepeer, a marketplace between infrastructure providers and application developers or streaming providers. The Livepeer network takes a distributed architectural design to create a more scalable and cost efficient solution for creators, broadcasters, and developers looking to add live or on-demand video to their projects. Via this decentralized approach, Livepeer’s process can potentially decrease costs 10x+ because it offers pricing based on usage rather than server space. Livepeer doesn’t compete for the same cloud computing supply as AWS or other cloud providers. In addition, the Livepeer network improves reliability by building in layers of server redundancy. If internet congestion happens in a certain region, other transcoding clusters can pick up the work from the ones impacted by the outage. 

Anyone can join Livepeer and help transcode a video by contributing their computer’s resources and earn fees in ether (ETH) or a stablecoin. According to Web3 Index, Livepeer generated $56,519 of protocol fee revenue in October 2021, an increase of 46% month-over-month. Some of Livepeer’s current users include PlayDJ.tv, Korkuma.com, and Vimm.tv.

There are other companies building interesting decentralized networks which provide compute resources. Salad is building a large distributed network that is workload-agnostic. Meanwhile, Akash Network is a permissionless blockchain network that is estimated to allow developers to access cloud computing at up to 3x less than traditional Web2 providers.   

At BITKRAFT, we believe that the transition from Web2 to Web3 is underway. However, this process will take time and require innovative solutions. There are many exciting decentralized gaming applications being developed. But we are just as excited about the efforts to redesign systems powering these applications. And we are spending more and more time digging into the infrastructure layer of Web3. We are trying to understand which approaches are best and where value will accrue. If you are a founder building in this space, please reach out to us.