Decentralization vs. Digital Empires: Can We Escape the Algorithmic Gatekeepers?

From its earliest conception, the internet has been imagined as a space which could enable more democratic and open forms of social organization, challenging the concentrated power structures that characterize traditional media and corporate hierarchies. This social imaginary of the internet rests upon a foundational assumption: that distributed architectures and protocols inherently resist centralized control, creating conditions for genuine freedom of expression and association. Contemporary concerns about algorithmic censorship, platform monopolization, and corporate control of information invoke this earlier vision of a more open internet, while proposals for Web3, blockchain-based platforms, and peer-to-peer networks promise a return to decentralized principles through technological innovation. However, as I will demonstrate in this article, the contemporary experience of increasing platform control occurs within an internet infrastructure that was never designed to prevent the emergence of what Anu Bradford terms "digital empires", dominant regulatory and economic powers that shape global digital governance according to their own interests and values. The apparent contradiction between architectural decentralization and the practical reality of platform concentration reflects deeper tensions in how technological form, network topology, and governance structures interact in large-scale internet systems. I argue that the binary framing of centralization versus decentralization obscures more nuanced questions about how power operates in distributed systems, and that effective responses to platform control cannot rely solely on technological solutions. Instead, socially desirable outcomes such as freedom of expression and protection from abuse require careful combinations of technological design, governance innovation, and institutional coordination that acknowledge the inherent political dimensions of internet architecture. To support this argument, I examine the evolution of platform control through what I term the "algorithmic gatekeeper" function: the mechanisms through which platforms shape information flows not through direct censorship, but through algorithmic curation, recommendation systems, and content moderation policies. These gatekeeping functions operate across multiple layers of internet infrastructure, from domain name systems and content delivery networks to application programming interfaces and user recommendation algorithms. The analysis of these layers reveals that decentralization efforts must address not only the technical architecture of alternatives, but also the social, economic, and governance structures that determine how distributed systems evolve over time. Section 1: How Algorithms Quietly Enforce Censorship The mechanisms through which contemporary platforms exercise control over information flows extend far beyond the traditional model of direct content removal or explicit censorship. Instead, platforms deploy what we might call "infrastructural censorship": the shaping of information visibility through algorithmic systems that operate largely outside public scrutiny and democratic accountability. This form of control is particularly insidious because it maintains the appearance of openness while systematically privileging certain voices, perspectives, and forms of content over others. The technological architecture of algorithmic curation systems embeds particular assumptions about relevance, engagement, and social value that inevitably shape the boundaries of public discourse. As recent research into platform algorithms has revealed, the metrics used to determine content reach—engagement rates, click-through rates, time spent viewing—systematically favor sensational, emotionally charged, and polarizing content over more measured forms of public discussion. The apparent neutrality of these algorithmic systems conceals the political and commercial judgments embedded in their design and operation. Corporate platform operators have consistently maintained that algorithmic curation serves user preferences and platform sustainability rather than ideological control. Meta's recent announcement that it would limit political content across its platforms while maintaining that "people have told us they want to see less politics overall" exemplifies this rhetorical strategy. However, the definition of what constitutes "political" content, the mechanisms through which user preferences are determined, and the processes by which these preferences are translated into algorithmic rules remain largely opaque to both users and external researchers. The governance of algorithmic systems reflects the concentration of decision-making authority within a small number of corporate entities whose primary accountability remains to shareholders rather than the broader public affected by their platforms. The criteria used to determine whether content "comes close to violating" platform policies. YouTube's phrase for content that may be algorithmically demoted, illustrate the discretionary power exercised by platform operators in defining the boundaries of acceptable discourse. These boundaries shift in response to advertiser preferences, regulatory pressure, and internal corporate priorities rather than democratic deliberation or transparent policy-making processes. The topological structure of the contemporary internet amplifies the power of these algorithmic gatekeepers through network effects and data advantages that create barriers to entry for alternative platforms. The concentration of user attention and social connections within a small number of platforms creates conditions in which algorithmic visibility becomes essential for effective public communication. As content creators and political organizations have discovered, changes in algorithmic policies can dramatically reduce the reach of particular messages or perspectives, effectively functioning as a form of soft censorship that operates through visibility rather than direct prohibition. The result is a system of distributed censorship in which no single actor explicitly prohibits particular forms of speech, but the combined effect of algorithmic curation, platform policies, and network concentration systematically shapes the contours of public debate. This infrastructure of control operates through what might be called "censorship through curation": the ways in which algorithmic systems determine not what can be said, but what will be heard. Section 2: The Decentralized Alternative The technological response to platform concentration has centered on the development of protocols and systems that promise to return control over information flows to users and communities rather than corporate intermediaries. These decentralized alternatives, including blockchain-based social networks, peer-to-peer storage systems, and federated protocols, represent attempts to encode different values and power relationships into the technological substrate of digital communication. However, the extent to which these technological innovations can address the governance challenges identified in the previous section remains an open empirical question. Web3 platforms such as Farcaster and DeSo exemplify efforts to rebuild social media infrastructure using blockchain technologies that eliminate single points of control over user data and content distribution. Farcaster's protocol separates identity management from content hosting, allowing users to maintain consistent identities across multiple applications while storing their data on a decentralized network. This architectural separation addresses some of the lock-in effects that characterize centralized platforms by ensuring that users retain control over their social graphs and content archives even if particular applications cease operation. The InterPlanetary File System (IPFS) represents a complementary approach to decentralization through content-addressed storage that distributes files across multiple nodes rather than relying on centralized servers. By using cryptographic hashes to identify content, IPFS enables censorship-resistant publishing and reduces the ability of any single actor to remove or modify information once it has been distributed across the network. The integration of IPFS with blockchain-based naming systems creates the technical foundation for what proponents describe as a "user-owned web" that operates independently of traditional internet service providers and content delivery networks. Federated protocols such as Mastodon offer a middle path between full decentralization and centralized control by enabling independent servers to interoperate while maintaining local governance autonomy. The ActivityPub protocol underlying Mastodon allows users on different servers to follow and interact with each other while enabling server administrators to set community-specific rules and moderation policies. This federation model distributes both the technical infrastructure and governance responsibilities across multiple actors, reducing the systemic risks associated with single-platform dominance. However, these decentralized alternatives face significant challenges in scaling beyond early adopter communities and achieving the network effects necessary to compete with established platforms. The technical complexity of managing cryptographic keys, understanding protocol interoperability, and navigating multiple applications creates substantial barriers to adoption for users accustomed to the simplified interfaces of centralized platforms. As the DeSo documentation acknowledges, existing blockchains were not designed to handle the storage requirements of social media applications, necessitating fundamental innovations in blockchain architecture to support "infinite-state" applications that generate new data with every user interaction. The governance challenges facing decentralized platforms mirror those of the broader internet: technical standards must be maintained, disputes must be resolved, and resources must be allocated in ways that maintain network stability and growth. The apparent technical neutrality of protocol design conceals political choices about privacy, scalability, and user experience that ultimately determine which communities and use cases these systems can support. The evolution of decentralized alternatives thus depends not only on technological innovation, but on the development of governance institutions capable of coordinating collective action across distributed networks of users and developers. Section 3: From Empires to Federations The transition from centralized platforms to federated and decentralized systems introduces new topological configurations that redistribute power rather than eliminating it entirely. While federation promises to democratize control over digital infrastructure by enabling multiple actors to participate in governance and operation, the practical implementation of federated systems creates new centers of power that may reproduce some of the concentration dynamics they were designed to address. The examination of these emergent power structures reveals the limitations of purely technological approaches to decentralization and illuminates the institutional challenges that must be addressed to realize the democratic potential of distributed systems. Federated networks such as the Fediverse depend on relay operators and instance administrators who exercise significant control over user experience and content moderation within their jurisdictions. These intermediaries occupy positions in the network topology that allow them to shape information flows, determine community standards, and facilitate or impede cross-instance communication. As research into Mastodon's governance structure has documented, instance administrators must navigate complex decisions about content moderation, resource allocation, and community management that mirror the challenges facing centralized platforms. The distributed nature of these decisions does not eliminate their political character; instead, it multiplies the sites at which power is exercised while reducing the visibility and accountability of these governance processes. The economic sustainability of federated infrastructure creates additional pressures toward concentration and hierarchy within ostensibly decentralized systems. The costs of operating servers, moderating content, and maintaining technical infrastructure must be supported through volunteer labor, community donations, or commercial arrangements that may compromise the independence these systems seek to preserve. Large instances often emerge as dominant nodes within federated networks, attracting users through superior resources and technical capabilities while smaller instances struggle with sustainability and technical challenges. This process of organic centralization suggests that topological decentralization alone cannot prevent the emergence of hierarchical power structures. Protocol governance represents another site where power concentrates within decentralized systems through the control of technical standards and network evolution. The development and maintenance of protocols such as ActivityPub, Nostr, and blockchain consensus mechanisms require coordination among developers, node operators, and user communities whose interests may diverge over questions of scalability, security, and functionality. The informal authority exercised by protocol developers and early adopters shapes the trajectory of decentralized systems in ways that may favor particular technical approaches or use cases over others. The apparent openness of protocol development processes conceals the practical barriers to participation and the ways in which technical expertise translates into governance authority. The interdependence between technical infrastructure and social organization in federated systems creates what might be termed "confederal governance": distributed authority structures that maintain coordination through negotiated standards and voluntary compliance rather than hierarchical control. However, the stability of these confederal arrangements depends on ongoing consensus among participating actors whose interests and capabilities may change over time. The fragility of this consensus becomes apparent during periods of technical crisis, resource scarcity, or political disagreement when the costs of coordination rise and the benefits of defection increase. The result is a complex governance landscape in which power operates through multiple, interconnected mechanisms: technical control over infrastructure, social authority within communities, economic resources for sustainability, and institutional capacity for coordination. The effectiveness of federated systems in democratizing digital governance depends not only on their technological architecture, but on the development of institutions and practices that can manage these different forms of power in ways that preserve both system functionality and democratic accountability. Section 4: Freedom or Chaos? The fundamental tension underlying debates about decentralized alternatives concerns whether radical expansion of speech freedoms can coexist with the social stability and trust necessary for productive public discourse. Proponents of decentralization argue that removing corporate and state gatekeepers will enable more authentic forms of democratic participation by eliminating the systematic biases embedded in centralized moderation systems. Critics contend that the absence of coordinated content governance will create conditions in which disinformation, harassment, and extremist organizing flourish at the expense of marginalized communities and democratic institutions. The empirical evidence from existing decentralized platforms provides partial support for both positions, suggesting that the relationship between technological architecture and social outcomes remains contingent on institutional design choices that extend beyond technical specifications. Recent survey research across ten countries has revealed that substantial majorities of social media users support some form of content moderation, particularly for content involving physical threats, defamation, and targeted harassment. This preference for "online safety" over "unlimited freedom of expression" persists even in contexts such as the United States where free speech protections are constitutionally enshrined and culturally prominent. However, the mechanisms through which users would prefer to see moderation implemented—whether through centralized corporate policies, community self-governance, or algorithmic filtering—remain underexplored in both research and public policy discussions. The practical implementation of content moderation in decentralized systems raises complex questions about the relationship between local autonomy and network-wide coordination. Platforms such as Nostr have experimented with "bottom-up content moderation" in which individual users and communities maintain their own filtering preferences rather than accepting system-wide policies imposed by platform operators. This approach distributes both the authority and responsibility for content decisions across multiple actors, potentially reducing the systematic biases that characterize centralized moderation while preserving user agency over information consumption. However, the effectiveness of this model depends on the development of shared standards and technical tools that enable users to express and implement their preferences without requiring extensive technical expertise. The governance challenges become more complex when considering forms of harmful content that have network-wide effects rather than localized impacts on particular communities. Coordinated disinformation campaigns, financial fraud, and technical attacks on infrastructure require responses that extend beyond individual user preferences or community-level policies. The distributed architecture of decentralized systems may actually facilitate certain forms of harmful activity by making it more difficult to coordinate detection and response efforts across multiple platforms and jurisdictions. The absence of centralized chokepoints that can be used to interrupt harmful activity represents both a feature and a vulnerability of decentralized architectures. The economic and political incentives that drive content creation and distribution in decentralized systems may also shape the forms of expression that flourish in these environments. The removal of advertising-based business models and algorithmic engagement optimization could reduce some of the systematic biases toward sensational and polarizing content that characterize centralized platforms. However, alternative funding mechanisms such as cryptocurrency rewards, subscription models, and peer-to-peer payments may introduce different incentive structures that privilege particular forms of content or communication styles. The long-term effects of these economic changes on the quality and diversity of public discourse remain largely unknown. The sustainability of decentralized content moderation approaches depends partly on the capacity of user communities to develop and maintain social norms that balance expressive freedom with collective well-being. Historical examples of successful self-governing communities—from Wikipedia's editing processes to open-source software development—suggest that voluntary coordination around shared standards is possible under certain institutional conditions. However, these successful cases typically involve relatively homogeneous communities organized around specific technical or intellectual projects rather than the diverse, contentious environment that characterizes general-purpose social media platforms. The tension between freedom and chaos in decentralized systems thus cannot be resolved through technological design alone, but requires ongoing experimentation with governance institutions that can adapt to changing social conditions and user needs. The question is not whether decentralized alternatives will eliminate the challenges associated with content moderation, but whether they can create more democratic and responsive mechanisms for addressing these challenges than the centralized systems they seek to replace. Section 5: The Next Internet We Build The development of alternatives to platform-controlled internet infrastructure requires institutional innovations that extend beyond technical architecture to encompass governance structures, economic models, and social practices that can sustain democratic participation in digital systems. The experience of internet governance over the past three decades from the multistakeholder processes that coordinate domain name administration to the distributed standardization work of the Internet Engineering Task Force provides instructive examples of how collective action can be organized across networks of independent actors with diverse interests and capabilities. However, the scaling of these governance models to encompass the broader social and economic functions currently performed by centralized platforms presents challenges that cannot be addressed through existing institutional frameworks alone. I propose that effective alternatives to digital empires must address three interconnected dimensions of power distribution: technological architecture, economic sustainability, and democratic accountability. The technological dimension encompasses not only protocol design and infrastructure operation, but also the development of user-facing tools that enable meaningful participation in system governance without requiring extensive technical expertise. The economic dimension involves creating funding mechanisms that can support infrastructure operation and protocol development while avoiding the capture by commercial interests that characterizes advertising-based platforms. The democratic dimension requires institutional innovations that can coordinate collective decision-making across diverse communities while maintaining the local autonomy that decentralized systems are designed to preserve. The "distributed internet governance ecosystem" proposed by scholars such as Stefaan Verhulst and Beth Noveck offers a framework for addressing these challenges through what they term "open governance": institutional arrangements that enable broader participation in policy-making while maintaining the technical expertise and operational efficiency necessary for complex system management. This approach builds upon the internet's existing governance architecture while extending its principles to encompass the social and economic functions that have been captured by centralized platforms. The key insight from this work is that sustainable alternatives to platform control require not just technological decentralization, but institutional innovation that can manage the complex interdependencies between technical standards, economic resources, and democratic participation. The practical implementation of distributed governance faces significant coordination challenges that mirror those encountered in other forms of collective action. The development of shared technical standards, the allocation of infrastructure costs, and the resolution of conflicts between different communities or use cases all require mechanisms for coordinated decision-making that can operate across organizational and jurisdictional boundaries. The Internet Society's research into multistakeholder governance identifies four key attributes that characterize successful distributed coordination: inclusiveness and transparency in decision-making processes; collective responsibility for system outcomes; effective implementation mechanisms; and collaboration through distributed and interoperable governance structures. However, the application of these principles to the governance of social media and content distribution platforms introduces complexities that extend beyond the technical coordination challenges that characterize traditional internet governance. The regulation of expression, the management of community conflicts, and the allocation of algorithmic visibility involve inherently political decisions that cannot be resolved through technical expertise alone. The legitimacy of distributed governance arrangements depends partly on their capacity to represent the diverse interests of users and communities affected by platform policies, rather than simply coordinating among technical operators and protocol developers. The development of economic models that can sustain decentralized alternatives without reproducing the concentration dynamics of advertising-based platforms represents another critical challenge for the next generation of internet infrastructure. The integration of cryptocurrency-based payment systems, community-funded infrastructure, and cooperative ownership models offers potential mechanisms for distributing economic control more broadly while maintaining the resources necessary for technical operation and continued development. However, the long-term sustainability of these alternative economic arrangements remains uncertain, particularly as they interact with existing financial systems and regulatory frameworks that may favor centralized coordination. As I have shown throughout this article, the escape from algorithmic gatekeepers cannot be achieved through technological innovation alone, but requires coordinated institutional change that addresses the political, economic, and social dimensions of digital governance. The internet we build next will reflect the governance institutions we create today to manage the transition from digital empires to more distributed forms of power. The challenge is not simply to decentralize existing systems, but to develop institutions capable of coordinating collective action in ways that preserve both individual autonomy and collective well-being in an increasingly interconnected digital world. Conclusion The contemporary debate over decentralization versus digital empires reflects deeper questions about how power should be organized in systems that span multiple jurisdictions, communities, and technical domains. The analysis presented in this article shows that the binary opposition between centralized and decentralized systems obscures the more complex ways in which power operates through technological architecture, network topology, and governance institutions. The algorithmic gatekeepers that characterize contemporary digital empires exercise control not through direct censorship, but through the more subtle mechanisms of visibility, recommendation, and platform design that shape the contours of public discourse while maintaining the appearance of user choice and market competition. The decentralized alternatives examined in this article, including blockchain-based social networks, federated protocols, and peer-to-peer storage systems, offer important innovations in technological architecture that can reduce some forms of platform dependency and enable new models of user control over data and digital identity. However, the governance challenges facing these alternatives mirror those of the broader internet: the need to coordinate collective action across diverse actors while maintaining system functionality and democratic accountability. The transition from digital empires to more distributed forms of power thus requires not only technological innovation, but institutional experimentation that can address the political and economic dimensions of digital governance. The tension between freedom and chaos in decentralized systems cannot be resolved through technological design alone, but depends on the development of governance institutions that can balance expressive freedom with collective well-being while adapting to changing social conditions and user needs. The sustainability of alternatives to platform control depends ultimately on their capacity to create more democratic and responsive mechanisms for addressing the complex challenges associated with content moderation, resource allocation, and community coordination than the centralized systems they seek to replace. The internet we build next will reflect the governance institutions we create today to manage the distribution of power in digital systems. The escape from algorithmic gatekeepers requires not simply the decentralization of existing platforms, but the development of institutional innovations that can coordinate collective action in ways that preserve both individual autonomy and democratic participation in an increasingly interconnected digital world.