Author: Yinghao, Investor at SevenX Ventures, Twitter @linsajiao

"Social media is too important to be controlled by a few corporations. We’re building an open foundation for the social internet so that we can all shape its future."——Bluesky

1. The Fluid Feast — The Eternal Dilemma of Social Media

Eternal Dilemma and Fluid Feast

Over the years of engaging in crypto investments, social media has been a topic of our continuous focus. It is highly enticing because social media is the source of user networks and has strong scale effects and commercial value. We view social media as a form of "non-technical disruptive innovation":

• Technical disruptive innovation: Anchored in technological upgrades, shifts in technological paradigms, etc. For example, smartphones, electric vehicles, artificial intelligence, etc.
• Non-technical disruptive innovation: Anchored in changes in user demands, shifts in content forms, demographic generational shifts, etc. For example, social media, e-commerce, etc.

However, in the process of research and investment practice, we found that social media is difficult to discuss deeply and effectively — social media products are like politics; everyone can comment, but it is hard to articulate clearly. Discussions about it will never cease. The perspectives on social media products are vast and complex, making it difficult to form a unified logical framework. We believe the fundamental reasons are as follows:

• The true boundaries of social media extend far beyond the product itself. It is more like a public utility of the information age, a collection of ideological, value-based, political, cultural, social events, commercial, and technological elements of the era. It often starts with a simple product + a simple user base, but when it grows into a behemoth, it has become the Ship of Theseus;
• Social media serves as a scale for the evolution of human civilization and has a reflexive effect on it. We understand what is happening through it, and it records the real-time standards of human civilization, exerting reverse influence and control over a significant portion of the population. We cannot discuss our reflection from a higher perspective; we can only dance with it.

These reasons make social media products fundamentally different from other products, but it is precisely because of this that social media is a fluid feast: a series of flowing platforms that carry the unchanging demands of users. Users remain skeptical of old products while holding expectations for new ones, always ready to vote for a new king; developers and investors can always look forward to the homerun-style returns that come every few years.

Why Focus on Bluesky

Recently, we participated in the investment of the next-generation social media platform, Bluesky. Its representative vision and philosophy have helped us rethink this issue. We want to share our understanding of the significance and potential of Bluesky.

Bluesky describes its positioning and vision as follows:

“Bluesky is a social application designed to be free from the control of a single company. We are creating a version of social media built by many people, yet still combined to form a cohesive and user-friendly experience.

We hope that modern social media and online public discourse resemble the early days of the web, when anyone could create a blog or use RSS to subscribe to multiple blogs. We believe this will usher in a new era of social media experimentation and innovation. Researchers and communities will have the ability to intervene to help solve the current problems facing social networks, and developers will be able to experiment with many new forms of interaction.

Traditional social networks are typically closed platforms with central authority. A small group of people controls these companies, dictating how users can use the platform and what developers can build. On these platforms, as a user, if you try to leave, you must start from scratch, losing all the connections and content you built there. As a developer, if you try to build a new application, you must overcome network effects and rebuild the social graph from scratch. If you attempt to build based on these companies' APIs, they may cut your connections and kill your company in the blink of an eye. As a creator, you may spend years building an audience, but when the platform changes its rules, you lose access to that audience.”

In an early declaration from Bluesky, we noted that Bluesky was established in 2022 as a PBC (Public Benefit LLC), reaffirming their original intention from when they were founded in 2019: “Our mission is to develop and promote the widespread adoption of open and decentralized public dialogue technologies.”

From a broader temporal perspective, the timing of Bluesky's birth is quite interesting, possessing strong uniqueness and representativeness.

In 2019, the year Bluesky was officially established, global geopolitical and ideological fluctuations were frequent. We can clearly recall that year when the United States withdrew from UNESCO, the INF Treaty, and the Paris Climate Agreement, the UK decided to Brexit, Russia intensified its diplomacy in the Middle East and Africa, Huawei was placed on the entity list, and nationwide riots erupted in Latin American countries like Argentina, Chile, and Bolivia, among a series of black swan events, marking one of the most tumultuous decades, which concentrated ideological conflicts on social media.

Switching the lens, in the subsequent years of 2020-2021, Web3 welcomed the DeFi summer and another bull market, pushing the proposition of blockchain + mass application to practitioners. The earliest batch of decentralized social developers began to take action, deeply contemplating the methods of information distribution. Another one of our portfolio companies, RSS3, is a representative of this, focusing on the “information distribution model” as a primary entry point, paying attention to those “users who still maintain their habits of searching and organizing information — they still actively search, subscribe, unfollow, filter, etc., maintaining their demands for the quality and accuracy of information acquisition,” thus delving into the editor-based distribution and self-based distribution models, which was also the early narrative of RSS3. This direction aligns with Bluesky's later proposal of custom feeds + algorithms marketplace. Clearly, since then, the decentralized concept and technology have seriously impacted developers.

When discussing Bluesky, one cannot overlook its first initiator and funder, X (mainly Jack Dorsey, the founder of X). The relationship between Bluesky and X also serves as a testament to this historically unique period.

Bluesky was initiated as a project by X in 2019 and funded by X. This attempt by X stemmed from the severe challenges they faced: how to balance survival, protect free speech, and limit abusive behavior. In the United States, India, Turkey, Australia, Russia, and Nigeria, X has engaged in numerous confrontations with governments. Some governments have demanded that X provide personal information about accounts, even threatening to arrest employees. The reality is that the platform operates under the rule of universal free speech, but free speech is defined by local laws in each country. If a country demands that X delete certain accounts, X must comply, as those accounts are classified as illegal. X has also reluctantly introduced the feature to delete accounts and content by country/region (this feature is still widely used today), but this regional governance approach cannot completely resolve the issue. Therefore, X believes that the solution requires a protocol: because protocols cannot be controlled, governments cannot target a single node, and there will not be a centralized leader who can be taken down, just like Bitcoin.

Carrying such a mission, Bluesky has now been in existence for over four years. The representativeness of Bluesky lies in its embodiment of a supreme concept. The core of this concept is: protecting free speech, resisting regulation, resisting censorship, being free from control by a single leader/entity, open protocolization, ecological, and protecting user identity/data ownership, etc. This representativeness comes from both user expectations and the team's proactive attempts, as Bluesky slowly practices X's original will.

Another point is that we position Bluesky as a new generation of "broad social media," rather than a "Web3 social media," which means it possesses characteristics of both Web2 and Web3, absorbing decentralized concepts and technological solutions while aiming for mass adoption as the ultimate goal. This is a relatively unique existence in the current market. We will explore the possibility of this positioning below.

2. Basic Deduction Method — Philosophy, Demand, and Product

Let’s start from first principles and use basic deduction to examine the philosophy, demand, and product corresponding to Bluesky.

The Legend of Two Systems

Years ago, X initiated Bluesky and posed the proposition of "centralized products vs. open protocols." This proposition quickly sparked widespread discussion, with Masnick's "The Technological Approach to Free Speech" being one representative discourse. Let’s revisit the discussion about free speech and how technology delineates protocols, platforms, and the corresponding two systems.

As mentioned above, social media serves as a scale for human civilization, and users generally view it as a tool for accommodating more speech and improving thoughts. However, in recent years, with the expansion of adoption scale, this view has undergone a significant shift: many users have begun to see social media as a cesspool of malicious attacks, paranoia, and hatred. The rapidly fermenting issues include:

• Platform review policies becoming increasingly aggressive, systematically suppressing certain viewpoints
• Platforms collecting user privacy data without users knowing its purpose
• Platforms being used for false propaganda
• Platforms being used for political competition or being targeted by political parties
• Platforms being accused of foreign manipulation of local elections, etc.

To address these challenges surrounding speech content, many solutions have been proposed, including:

• Forming large content review teams; companies like Facebook, YouTube, and X have discussed hiring thousands of reviewers
• Utilizing artificial intelligence technology to attempt to detect controversial content early in the process
• Arguing that no moderation should occur — at least for platforms of a certain scale — so they are viewed as part of the public square, etc.

However, these methods have not yielded good results, leading to the idea of replacing platforms with protocols. Many people mistakenly believe that protocolization is a brand new technological paradigm, but it is actually a development concept that has existed since the early days of the internet. The early internet originated from many different protocols: email used SMTP, chat was done through IRC, Usenet used NNTP as a distributed discussion system, and the World Wide Web itself has its own protocol: the Hypertext Transfer Protocol, or HTTP.

But for a long time, new protocols have become increasingly rare, with more being controlled and operated by private entities, as this not only allows for more efficient rollout of new features and bug fixes but also makes user growth and commercialization easier. In other words, protocols are being deeply utilized by various oligopolistic platforms, establishing walls that lock users in rather than providing interfaces.

Returning to protocolization is a good solution. Instead of leaving the regulation of speech content to a few oligopolistic platforms, it is better to foster widespread competition. Anyone can design information flows and filters, allowing the most effective/popular solutions to emerge. Compared to resorting to a unified censorship system, users will be allowed to determine their own tolerance for speech without completely suppressing anyone or letting the platform itself decide who can speak. Protocolization has other advantages:

• Competition drives innovation, leading to unexpected new features
• Users have better control over their privacy data
• New business models beyond the monetization of user data (traditional advertising models)

Aside from regulation and review, another angle is commercialization, which is a problem that protocols find difficult to solve. There is a case comparing Reddit and Usenet, illustrating the commercialization dilemma of protocols:

• Reddit and Usenet share similar philosophies, both being collective forums organized around specific topics. Different groups on Reddit are called subreddits, while on Usenet they are called newsgroups.
• The difference is that Reddit is a massive publicly traded company, while Usenet is an open protocol. To access Usenet, users need a content reader client (of which there are many options) to connect to Usenet servers. In the early days, this space was dominated by Deja News, which developed one of the earliest web interfaces for Usenet; later, it added a range of additional features, including a search engine. Operating Usenet servers has never been particularly profitable, and Deja News could not find a sustainable business model, ultimately being acquired by Google.
• Another major drawback of Usenet is its extremely low flexibility, especially regarding large-scale iterations. Since it is a set of decentralized protocols, any changes to the protocol require consensus from all parties, and even minor changes often require significant effort. Establishing a new content group is a rather complex process. In contrast, centralized Reddit does not perceive this as an issue; its strong product team can efficiently resolve any problems.

The commercialization issue of platform products is a completely different angle: reliance on advertising. Jack Dorsey pointed out in an interview that one of the root causes of X's predicament is its dependence on a centralized advertising business model:

"I think the core and most serious sin was choosing the advertising model from the start. We needed a business model. We saw that Facebook's model was very good, so we came up with an advertising program and ran it. A year after the IPO, we saw growth decline, which was due to a drop in advertising revenue. When you rely entirely on this, what do you do if brands like Procter & Gamble or Unilever dislike what is happening on the platform and threaten to pull a budget that accounts for about 20% of your revenue? You have no choice. If you take a stand, they cut the budget, and the stock market sees that, and the stock price drops from $70 to $30. Then you will have to let employees go. This is the entire dilemma you need to solve."

When reflecting on this issue, Jack also pointed out a clear direction:

"We will create other revenue streams. We will focus more on commerce. We will pay more attention to payments. We will concentrate on everything they are currently trying to do. All of this was what we were doing before the company was sold. But I would do it faster because, as a private company, you can shut down the advertising business and only engage in commerce. Only applicable to payments. Just do small ads, more like classified ads; I think this is an extraordinary business for a site like X."

Elon Musk is a great reformer, and his will aligns with Jack Dorsey’s. After taking over X, Elon challenged its business model, no longer simply catering to advertisers (remember what he said to Disney CEO Bob Iger? Because Disney was concerned about anti-Semitic speech on the X platform, threatening to pull advertising if X did not take strong measures. Before Disney, companies like Apple, IBM, Coca-Cola, and Microsoft had taken similar actions), laid off a large number of advertising sales staff (who made up more than 50% of X's total employees), while fully promoting subscription services. X's revenue may fluctuate, but it also signals to the market: there is more than one viable advertising model; social media has many more business models to explore.

We can see that the distinction between protocols and platforms is actually a struggle between two systems, which, borrowing from sociology, we can call "elitism" vs. "technocracy." The opposite of elitism is populism, but the evolution of technology stands on the populist side, better resonating with elitism.

The Three-Stage Rocket of Social Media and KSF

Beyond product philosophy, let’s examine Bluesky's opportunities from the perspective of the three-stage rocket theory and KSF (Key Success Factors). We believe the three-stage rocket of social media consists of: necessary factors, boosting factors, and decisive factors, with each stage corresponding to 2-4 key KSFs:

Specifically, in the necessary factors, the two KSFs are:

a. Continuous funding:

• Before achieving scale effects, social media has weak monetization capabilities but requires continuous funding to maintain development, operations, and user growth, making social media a capital-intensive business. X raised over $1 billion before going public and over $1.8 billion through its IPO, while Meta raised over $1.5 billion before going public and over $16 billion through its IPO.
• Bluesky is actively pursuing equity financing after transitioning from a completely open protocol to a tangible company (previously relying solely on grants) and has achieved good results, but it still needs to continuously secure substantial funding, making the team's fundraising capability crucial. Unlike companies like X and Meta, Bluesky also has avenues for fundraising through tokens for private placements and TGE (Token Generation Event) public financing. If the two financing channels of equity and tokens work well together, Bluesky is expected to quickly establish a funding advantage.

b. Growth flywheel:

• This refers to continuously acquiring customers effectively and improving the efficiency of fund utilization. Common methods in traditional social media include: direct user acquisition, which has the advantage of abundant customer acquisition channels and stable costs; building a content ecosystem through high-quality content to attract vertical users and gradually break out, which has the advantage of low cost and rapid penetration into vertical groups, suitable for early growth; acquiring users through incentive mechanisms, such as affiliate and other viral models, which have the advantage of rapid viral growth and high efficiency.
• Bluesky has already established a mechanism for acquiring users through celebrities/KOLs and content in its early stages, achieving good results. Its greater imaginative space lies in launching a points/token system for user acquisition, which is a Web3 native approach, validated by countless protocols, but the challenge lies in distinguishing between real users and incentive-driven users while avoiding the negative impact of incentive-driven users driving out good users from the entire user network.

In the boosting factors, the four KSFs are:

a. Product features:

• For example, the photo wall for Instagram and ephemeral messaging for Snapchat; a truly innovative feature adopted by users can make a social product stand out and even drive an unstoppable trend to become mainstream. Users always have high enthusiasm and expectations for new features, which is also one of the focal points developers love to exploit.
• Bluesky has already established a clear foothold in this regard, including custom feeds, community-driven moderation, etc. Its underlying protocol, AT Protocol, is even more imaginative, gradually establishing a product ecosystem co-created by developers with dozens of products, and this is just the beginning. The feature innovations brought by token economics are also highly anticipated and are one of the areas where decentralized developers excel.

b. Content ecosystem:

• Users' diverse content needs support different social media giants in various niche areas, such as: text + images/videos - X, images - Instagram, videos - TikTok, and some more specialized content, such as: NSFW - OnlyFans, content that cannot be described - deep web/dark web. A single social media platform generally has clear distinctions regarding content forms.
• Bluesky's content forms are not significantly different from X, but the variables in its content ecosystem lie in the decentralized information distribution methods and algorithm-centered content organization brought by custom feeds and community-driven moderation. This change has the potential to break the boundaries of content forms, making it a multi-content type platform. Developers of the AT Protocol are also working on new products based on video and live streaming content.

c. Technological upgrades:

• Part of technological upgrades relates to content, such as the impact of 5G networks on the proliferation of video content, and the emergence of online streaming social platforms like Omegle; another part of technological upgrades relates to user experience, such as recommendation algorithms dominating the current information distribution model, making users increasingly addicted (which is not a good thing). Web3 is also a form of technological upgrade, and SocialFi has gradually become a high-potential niche track.
• Bluesky has a very clear direction for technological exploration, namely the paradigm of protocolization & decentralization, as well as the development philosophy of permissionless & programmable, which will be elaborated on in the third part.

d. Commercialization flywheel:

• The commercialization path of social media can be summarized as first incurring losses and then achieving profitability. User networks rely on dense funding to establish, while the profit model is built on a massive user network. When social media has a vast user base, there are many ways to monetize, commonly including: advertising, subscriptions, value-added services (such as live streaming rewards), e-commerce, etc. The core of commercialization is to gradually improve fund efficiency and eventually turn negative fund efficiency into positive to achieve overall profitability.
• Bluesky has a lot of imaginative potential in terms of commercialization. This includes: an advertising model based on custom feeds, which challenges traditional advertising models; data assetization and trading based on user data ownership, such as reverse sales to platforms and advertisers; and business models based on token economics, such as capturing network value through protocols and using tokens as in-platform payment currency. Innovation in business models will become Bluesky's greatest imaginative potential.

In the decisive factors, the four KSFs are:

a. Content Form Migration:

• The migration process of content forms in social media is very clear: text - images - videos - real-time online content (such as live streaming and online video matching chat).
• The next major content form is still uncertain; it may be mixed reality. We believe that Bluesky will not lead the upgrade of content forms for the time being.

b. User Generational Migration:

• Some users are aging, while others are growing up. This generational flow of users presents significant opportunities for new social media products. The outflow of users from Facebook is a good example, which is also why Meta must maintain high-frequency acquisitions outside of Facebook to retain its users. Over 100 million people globally become Gen Z each year, and products targeting Gen Z users are one of the most valued opportunities for developers.
• Bluesky may become a toy for young people because both developers and users in the younger demographic not only have a better acceptance of new concepts like decentralization and token economics but also hold stronger convictions regarding values such as freedom of speech and ownership. They are highly sensitive to ideological issues and possess a strong rebellious spirit, making them a perfect fit for Bluesky.

c. Geopolitical Arbitrage:

• The time machine theory remains effective. Weibo to X, VK to Facebook, and TikTok to Douyin are all good examples.
• Bluesky also has opportunities for geopolitical arbitrage if it can find regions with lower penetration rates for products like X.

d. X Factors:

• This includes celebrity founders, geopolitical issues, ideological conflicts, and various black swan events. The impact of X factors can be both positive and negative; for example, the banning of Trump's account on X and Pavel Durov's arrest had completely opposite effects on X and Telegram.
• X factors are unpredictable, but the Bluesky team has already proven their ability to navigate them: during the time X was banned in Brazil, Bluesky migrated 10% of X's users in Brazil, which was undoubtedly a successful strike.

The Call of Users and Developers

The struggle between the two systems and the analysis of KSFs lead us to focus on one question: what kind of innovations do users and developers need? What types of opportunities does this create? How can Bluesky accommodate them?

From a meso perspective, we can distill three keywords: new era, new demographics, and new technologies. From a micro perspective, we can identify some more specific pain points:

• From the user side, specific pain points include: centralized review policies, ad-driven information flows, algorithm-driven content, lack of ownership of user data, outdated platforms, etc.
• From the developer side, specific pain points include: closed platforms, cold start dilemmas, high difficulty in getting started with development, and the appropriation of created benefits by the platform, etc.

In response to these pain points, Bluesky has provided a complete set of solutions.

3. The Belly of the Whale — Bluesky's Attempts

From Bluesky to ATmosphere

In their open letter, Bluesky introduced the three stages they have gone through since their inception.

Initially, Jack Dorsey issued a declaration announcing that X intended to fund the development of decentralized social media open protocols. More than a dozen developers gathered in a chat room formed as a result to discuss how to proceed. In the initial Q&A of that room, Jack wrote:
"The biggest long-term goal is to establish a lasting and open protocol for public dialogue. It does not belong to any one organization but is contributed to by as many organizations as possible. It is born and developed on the internet, following the same principles."

Later, Bluesky completed its first fundraising, established a community, and gradually transformed into the company Bluesky PBLLC, operating as a corporate entity to this day. Jack Dorsey's initial statement provided a roadmap for Bluesky, which can be divided into two parts: the Bluesky portal (product) and the AT Protocol (protocol).

Bluesky can be understood as the first portal product on the entire AT Protocol, developed by the team itself, akin to a "permissionless & programmable version of X," but with many new features compared to X (as detailed below). The significance of Bluesky is profound; it is not only a benchmark product but also serves the function of accumulating an early user pool for the entire ecosystem. Bluesky has been operating for over a year, surpassing 12 million users, making it one of the fastest-growing social media platforms in history.

The AT Protocol is the protocol layer beneath Bluesky. The team defines the AT Protocol as: a protocol for public dialogue and an open-source framework for building social applications, meaning that people can transparently understand how it is constructed and what content is being developed. It creates standard formats for user identity, following, and data on social applications, allowing applications to interoperate and enabling users to move freely between them. It is a federated network with account portability.

To explain with an analogy: every time a user creates an account on a social platform, it is like moving to a new city. Users make friends and create posts, just like filling their house with furniture they made. But on centralized social platforms, if users leave, it is like leaving all their friends behind without a way to contact them, as if they left their house without taking anything with them. Leaving a centralized site and starting over is very difficult. The AT Protocol essentially allows people to move between cities. Creating standard formats for identity and data is like giving people passports, phones, and property rights. If users do not like the city they initially moved to, they can relocate and take all their belongings (data) with them. Their friends can still find them and keep in touch using the same name and number (identity and following graph).

According to the Bluesky team, the key points of the AT Protocol are:

• Account portability. A person's online identity should not be owned by a company that does not take responsibility for its users. With the AT Protocol, users can transfer their accounts from one provider to another without losing any data or social graph.
• Algorithm choice. Algorithms determine what we see and who we can connect with. If we are to trust our online space, we must control our algorithms. The AT Protocol includes an open algorithm model, allowing users to better control their experiences.
• Interoperability. The world needs a diverse interconnected service market to ensure healthy competition. Interoperability needs to become the second nature of the web. The AT Protocol includes a pattern-based interoperability framework called Lexicon to help address coordination challenges.
• Performance. Many novel protocols have neglected performance, leading to long loading times for users to see their timelines. We do not consider performance optional, so we prioritize building for large-scale fast loading.

The AT Protocol ecosystem is rapidly attracting developers. According to official statistics, over 70 projects have submitted content on GitHub: Project List

Community projects and components developed based on Bluesky itself have also exceeded 100. In the official Bluesky Showcase section, projects and components developed using the Bluesky API can be seen, including clients, custom feeds, social tools, bots, and various functionalities.

For example, deck.blue is a typical third-party client developed by the Bluesky Brazil community. Bluesky users can log in directly using their domain and password, experiencing a different UI design and detailed features: "3rd-party TweetDeck for Bluesky, Featuring columns, multi-account, scheduling, and inline translations." After its launch, deck.blue quickly accumulated users, benefiting from Bluesky's user network.

Bluefeed is a feed discovery and recommendation tool that helps users find, browse, and retrieve various feeds, making it very convenient.

Product Features

Combining the summary of pain points in social media products, Bluesky and the AT Protocol provide a complete solution, with the following four main product features.

First is community-driven moderation services. The drawbacks of centralized moderation policies have been thoroughly explained, and Bluesky's moderation model is:

• Users can delegate moderation rights to moderators they trust.
• Users can customize their moderation preferences, such as hiding rude posts, hiding screenshots, hiding spoilers, and even hiding all images containing spiders. These customization options are continuously being enriched.
• Different moderators are combinable and can be used in conjunction.
• Bluesky's community-driven moderation model also has external interoperability, allowing it to be called by other applications in the AT Protocol ecosystem.

Currently, there are over 100 moderation services in operation.

Secondly, there is user-customized information flow. The information flow on the platform is established by various curators, each of whom has their own algorithm for the feeds they create:

• Some algorithms are quite standard, such as aggregating information from experts or authors in a certain field, automatically fetching the most popular posts from the past 24 hours, etc.
• Other algorithms can be quite peculiar, such as gathering content from those bloggers who do not post very often.

The feeds created by different curators form a marketplace of algorithms, from which users can choose to create their own "algorithm-neutral content." Data validation shows that such customized information flows can enhance user retention by up to 2 times. If Bluesky launches a token economy in the future, curators with excellent curation abilities are likely to become key incentive targets, further increasing the usage rate of custom feeds.

Thirdly, there is complete ownership of identity and data belonging to users. Specifically:

• Users own their identity and content entirely.
• The customized handle represents the user's identity, which can be migrated across applications.
• User data is stored in a database that can be migrated.

The ownership of identity and data has many obvious benefits. In addition to technically achieving true resistance to censorship and regulation, it also allows users the possibility of monetizing based on personal identity and data assets in terms of business models. This business model has been proposed for a long time and is gradually becoming possible with technological advancements.

Fourthly, there is a protocol-based development platform that addresses the issues of closed ecosystems, cold start dilemmas, and high difficulty in getting started with development. The problems of social media platforms can be summarized as: users are locked inside, and developers are locked outside. For example, in the case of X's third-party clients, there were once a large number of quality developers who created third-party clients based on X's API, such as Tweetdeck. In China, there is also a similarly large ecosystem of third-party Weibo clients, such as Weico. These clients brought a large number of users to X and Weibo with their innovative features, more attractive UIs, and better user experiences (such as ad-free), but ultimately X and Weibo chose to shut down their APIs, causing significant losses for third-party client developers.

Bluesky's AT Protocol advocates for completely permissionless development, allowing developers to freely access any resources within the protocol to develop products with different functionalities and become members of the ATmosphere (AT Protocol ecosystem):

• The user network of the AT Protocol (including Bluesky) is shared, so developers do not have to undergo time-consuming and costly cold starts.
• Commonly used functions are abstracted and packaged into a complete toolkit, allowing developers to easily make composable calls.
• Developers can also engage in composable development based on a single platform, adding rich plugins and potentially adding support for tokens.

Bluesky collaborated with user @davis.social to create a comic that explains what makes Bluesky unique.

In addition to Bluesky, several other protocol-based social products have made attempts, such as Farcaster, Lens, and Nostr, each with its unique philosophy.

• Farcaster was established in 2020 by executives from Coinbase and is currently valued at over $1 billion, with investments from Paradigm, A16Z, and others. Farcaster has chosen a balanced approach to on-chain/off-chain issues, aiming to enhance performance while ensuring user data ownership. Farcaster launched its flagship application Warpcast in February 2024, which has been well-received by Web3 degens. The subsequent launch of the Frames development framework has also attracted a large number of developers. Meanwhile, the success of the ecosystem meme coin $DEGEN has drawn many users to Warpcast.
• Lens was founded in 2022 by the founder of AAVE and is currently valued at over $500 million, with investments from IDEO, Variant, and others. Lens has achieved a high degree of modularity in its technical architecture, making it very developer-friendly, and has introduced the concept of assetizing social elements centered around NFTs. Its open actions development framework also supports the implementation of more personalized features. The Lens ecosystem currently includes applications like Phaver, Hey, and Orbs.
• Nostr was established in 2018 and has also received funding from Jack Dorsey. Unlike Farcaster and Lens, Nostr does not support a token economy and does not plan to launch a token. It is a minimalist P2P messaging system composed of clients and relays. Nostr is characterized by simplicity, neutrality, and robustness, allowing users to run their own relays. Its ecosystem application Damus launched in February 2023 and sparked a wave of interest, with many crypto users experiencing it right away.

Before these protocols, believers in decentralized social media began actively experimenting since Steemit in 2016. These products and protocols have explored various aspects such as technical architecture, token economy, and developer engagement for over 8 years, achieving remarkable results after overcoming many challenges. However, they still face challenges in user acquisition and building user networks, with daily active users not exceeding 100,000. But this is just the beginning, and we look forward to decentralized social products exploring different paths and ultimately gaining user favor.

Phase Achievements

Bluesky's data is impressive, representing a high-scoring phase report:

• In terms of user data, Bluesky had only 30,000 users in April 2023, but by October 2024, its user count had surpassed 12 million, with a staggering monthly growth rate of 38%. This almost replicates the rapid growth curve of major social media giants in their early stages. Monthly active users and daily active users have also exceeded 3.5 million and 2 million, respectively, with a DAU/MAU ratio exceeding 45%, reflecting a quite active user ecosystem.
• In terms of user behavior, the proportion of "lurker" users is gradually decreasing, with the proportion of posting users approaching half of that of "lurker" users; the total of 12 million users has generated 440 million posts and 1.72 billion likes; 33% of users logged in more than 5 days in a week, indicating healthy user behavior data.
• In terms of user distribution, users from the United States account for over 31%, users from Brazil account for over 25%, users from Spain account for about 8%, users from Canada account for about 4.5%, and users from the UK account for about 3.6%. These are all regions with high user quality, and the distribution is gradually becoming more diverse, with users from over 20 countries currently using Bluesky.
• In terms of content and developers, Bluesky has generated over 50,000 custom feeds, over 100 moderation services, and more than 40 various types of clients, including representative ones like Skyfeeds, Deck.blue, Whitewind, Bluecast, and Fresky.tv. Users and developers are actively building the ecosystem.
• To some extent, the data achieved by Bluesky proves that it has found the right early growth flywheel, thanks to its clear GTM strategy. Bluesky has adopted a strategy similar to Reddit and Discord, leveraging celebrities/KOLs to attract high-quality users from different niche topics and using them as the main contributors of content to build an early high-quality content ecosystem for the platform: "We already have a large amount of information flow and a vast amount of content flowing through the network. Unlike many decentralized social or crypto platforms that typically cater to a tech-savvy, crypto-focused audience, we have thriving communities across various disciplines, including scientists, writers, and artists, as well as fan groups like BTS." For example, the author of the popular chiikawa series, which has exploded in popularity over the past two years, was brought onto the platform and began updating their comics, quickly becoming one of the most followed accounts in Japan, similar to the case of the Korean group BTS. Bluesky allows content on the platform to be less serious, featuring many chaotic and humorous memes and images (similar to the early days of X), but the quality of KOLs and active users must be guaranteed to allow the platform to ferment effectively during its early growth process.

The Bluesky team is also very adept at seizing sudden opportunities. On August 30, 2024, the Brazilian Federal Supreme Court issued an order to immediately suspend X's operations in Brazil due to X's repeated failure to remove content deemed in violation of Brazilian law and its refusal to provide user data required for an ongoing criminal investigation. Bluesky quickly seized this opportunity, and as users flocked in, it became the hottest app in Brazil, with a migration rate from the X platform reaching 10%. Brazilian President Luiz Inácio Lula da Silva also joined Bluesky, and as one of the country's most influential political figures, Lula's presence not only enhanced the platform's credibility but also suggested that Brazilian politicians and government officials might communicate with the public through Bluesky.

Many novel operational strategies are also brewing: "We envision exciting opportunities for a bidirectional embedded gaming platform where users can participate through games, especially during events like elections, where they can make friendly predictions with each other. We are actually pushing quite a few flywheels."

Although Bluesky's data is still far from social media giants like X and Facebook, it has already gained a significant lead compared to other Web3 native social protocols that also focus on protocols. The total user count of several leading Web3 protocols is around 500,000 to 600,000, with monthly active users in the tens of thousands. The most protocol-oriented, Nostr, has about 30,000 weekly active users and is still in the early stages of growth.

It can be seen that Bluesky is the decentralized social project with data metrics closest to centralized social platforms. Bluesky has also clearly stated that their benchmarks are not Web3 social platforms; they hope to use products like X and Instagram Threads as benchmarks, aiming to become a new generation of global mass social media.

In Joseph Campbell's "The Hero with a Thousand Faces," the author poses the question—how does a hero become a hero? After distilling and summarizing mythological stories from around the world, Campbell outlines the entire process a protagonist undergoes from departure to ultimately becoming a hero, known as the hero's journey.

One of the stages is referred to as the "belly of the whale." The belly of the whale symbolizes a dark space filled with uncertainty and is a threshold for inner transformation. After enduring prior trials, the hero inadvertently enters the belly of the whale, where they must confront their fears and anxieties, grow their inner strength, and continuously reflect in a state of absolute solitude, reinforcing the goal of their journey and preparing for the upcoming adventure. Ultimately, the hero undergoes a transformation akin to rebirth, emerging from the belly of the whale as a more resolute warrior.

Bluesky has undergone countless trials and transformations, achieving remarkable milestones such as establishing its values, iterating personnel, corporate operations, building protocols, and cold-starting products. Although greater challenges lie ahead, Bluesky has completed its rebirth from the belly of the whale.

4. There Are Always Young People

A Young Team with Strong Value Propositions

There is a famous internet meme whose origin is hard to trace—The Three Laws of Technology & The Three Laws of Society:

The Three Laws of Technology:

• Any technology that existed when I was born is a part of the ordinary world order.
• Any technology that was born between the ages of 15 and 35 is a revolutionary product that will change the world.
• Any technology that is born after I turn 35 is a violation of natural laws and will be punished by heaven.

The Three Laws of Society:

• Anyone who was born 10 years or more before me is a stubborn old fogey who is stuck in their ways.
• Anyone born within 10 years of me is the elite and backbone of society.
• Anyone who is born 10 years or more after me is a hopelessly collapsed generation.

Social media is a flowing feast; although there is skepticism and mockery regarding this "Three Laws of Society" among young people, we will always believe in the new generation of teams. Bluesky is led by a core team consisting of Jay Graber, Rose Wang, Paul Frazee, and Dan Holmgren. From multiple conversations with the team, we can feel that this is a young team with a strong style and value proposition, possessing a strong long-termism value perspective.

Jay graduated from the University of Pennsylvania and has been the head of Bluesky since 2019. She has years of experience as a software engineer and entrepreneur. Jay also has another identity as a seasoned digital rights activist, focusing on privacy and net neutrality movements since the early stages of her career. Meanwhile, Jay has worked at a cryptocurrency mining company and several blockchain startups, contributing to the launch of Zcash.

When Jack Dorsey was planning Bluesky, he spent two years selecting the right leader and ultimately found Jay. "We spent about two years interviewing 'the people who would build the protocol.' We finally found Jay Graber. She looked great, and we decided to fund her." Jay raised $14 million in funding from Jack Dorsey.

Jay is at the center of the storm, but she embodies great resilience. As Bluesky developed, continuous financing, development, and growth became urgent, and she had to return to a corporate model, which was also the most important reason for Jack Dorsey's exit from Bluesky. Although Jack Dorsey expressed different views when Bluesky transitioned from a fully open protocol to a physical company, he still maintained recognition of Jay, stating, "I really respect Jay. She is under a lot of pressure to survive and do what she does," making her a CEO worthy of respect.

Rose's experience is quite different. After graduating from Harvard University, she founded a food company called Chirps and ran it for nearly six years before joining the AI startup Forethought to manage customer experience. During this time, she met Jay in the San Francisco entrepreneurial community and decided to join Bluesky to leverage her offline community-building experience. Rose currently oversees partner management and user growth at Bluesky. As a relatively rare female entrepreneurial partner, we also feel the different gentle yet firm strength of Jay and Rose, and we are honored to support such a team.

Bluesky has a strong technical foundation, especially with a strong open protocol gene. Chief Technology Officer Paul Frazee has five years of blockchain startup experience, having founded Blue Link Labs, where he built Patchwork (the first application using the SSB distributed social protocol) and Beaker Browser (the first web browser supporting the Hypercore distributed web protocol). He later joined Bluesky as a protocol engineer and was promoted to CTO. Protocol lead Daniel Holmgren is one of the founding members of Bluesky, with IPFS development experience at Fission. Daniel previously co-founded a Consensys-supported startup focused on peer-to-peer data transfer of scientific datasets.

Bluesky's technical advisory team is equally strong, including: Jeremie Miller, the inventor of XMPP; Martin Kleppmann, author of "Designing Data-Intensive Applications" and a researcher at Cambridge University, who has collaborated with the Ink & Switch team to publish cutting-edge research on distributed protocols; Jeromy Johnson, the first employee of Protocol Labs and chief engineer for IPFS and Filecoin, as well as other projects including Coinlist and Estuary; and experts from X and Meta dedicated to trust, security, and developer experience, such as Amir Shevat (former head of the developer platform at X), Dan Abramov (former product lead at Facebook), and Yoei Roth (trust and safety).

The Potential of Bluesky

From an investment perspective, we believe social media represents a very special investment opportunity: they possess the most efficient network effects, the largest scale of coverage, and the broadest social impact. A small spark can ignite a prairie fire; social media is one of the few high-quality non-technical trend investment themes.

Social media is a chaotic model, and no one can make effective predictions. We speculate that Bluesky's potential may lie in the following areas:

a. A truly efficiently developing open protocol ecosystem. Operating an open protocol in a quasi-corporate manner to expand developers in the most efficient way and allow developers to compete widely, thereby spawning a large number of high-quality innovative projects (this is similar to the operational methods of many Web3 native social protocols, which have already attempted a path in ecosystem building).

b. The next generation of super applications. This could be Bluesky itself (currently far ahead) or other applications in the ATmosphere, continuously iterating on core features like custom feeds and moderation services and finding product-market fit while achieving the right growth flywheel.

c. A brand new business model:

• First, an effective incentive model. One of the main reasons early internet protocols gradually declined was due to business model issues; establishing and maintaining protocols has long been a struggle. Most of the work is usually done by volunteers, and over time, if not carefully managed, the protocol can wither.
• Second, a challenge to traditional advertising models, perhaps a more limited data-based advertising model that focuses more on matching intent. For example, Google's initial advertising model did not rely heavily on knowing everything about the user but rather on understanding the user's internet searches at specific moments. Alternatively, it could hark back to the more traditional world of brand advertising, where popular advertisers seek suitable communities. For instance, a card company may want to advertise within a micro-community on a platform that has an interest in cards.
• Third, a business model based on user data ownership, such as reverse auctions: users assetize their data and provide it to platforms, advertisers, or other users in exchange for ad access and facilitating transactions.
• Fourth, a business model based on token economics. The token system for Bluesky and the AT Protocol is still in design, but two points can be confirmed: first, the token's value reflects the network effects of the protocol; as a protocol with over 10 million users, its token demand and value already have a solid foundation. Second, as a protocol token, it can effectively capture some of the commercial value generated by ecosystem applications. On top of this, Bluesky's token may also be used for creator payments/subscriptions and more imaginative in-platform financial payments, among other uses. Web3 native social protocols like Farcaster have already made creative attempts in this area.

Can protocol-based social media succeed? We do not know. To conclude, I would like to quote a passage from Masnick's "A Technical Approach to Free Speech—Changing the Economic and Digital Infrastructure of the Internet to Promote Free Speech": “For the past half-century of network computing, the pendulum has swung between client-side computing and server-side computing. We have moved from mainframes and dumb terminals to powerful desktop computers, to web applications and the cloud. Perhaps we will begin to see a similar pendulum in this area. We have transitioned from a world dominated by a single protocol to a world where centralized platforms control everything. Returning to a protocol-dominated platform world could bring tremendous benefits to online free speech and innovation. Such initiatives have the potential to bring us back to the early promises of the web: to create a place where like-minded people can communicate on various topics from around the globe, where anyone can discover useful information on a variety of different subjects without being polluted by abuse and misinformation. At the same time, it can foster greater competition and innovation on the internet while enabling end users to have better control over their data.”

On March 22, 2006, Jack Dorsey posted the first tweet in history, “just setting up my twttr,” marking the beginning of a social media empire. Thirteen years later, on December 11, 2019, Jack posted another tweet: “X is funding a small independent team of up to five open source architects, engineers, and designers to develop an open and decentralized standard for social media,” initiating the self-iteration chapter of replacing platforms with protocols.

Four years have passed, and one of the relays has been passed to Bluesky, with everyone looking forward to it. Our predictions are quite limited, but that is precisely the charm of social media products. We see that Bluesky is doing the right things, and thus we look forward to Bluesky taking us to unknown and further places.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。