The Future Was Federated

The Promise of a Federation

My journey into decentralized social media began not at the musk of an eel, but in 2009, grounded in my understanding of value-defining network dynamics. My first encounter with the technology that would later be formalized as ActivityPub (the protocol underpinning today’s “Fediverse”) happened with a click on a link to a site called Identi.ca. What captivated me was not merely a new website, but a new paradigm.

The dream: a constellation of independent communities, each living on its own server, yet interconnecting to form a larger social network.

While it is true that my first email was very much federated, here was social media that was open: the software was free and open-source, meaning I could, in theory, run my own node. It presented a vision of the web that resisted the centralizing pull of corporate platforms. Finally a place where users weren’t locked into walled gardens that commodified their data and compromised their privacy. This was the dream: a constellation of independent communities, each living on its own server (or instance), yet interconnecting to form a larger social network.

For years, I witnessed and participated in the unique culture it fostered: a space where thoughtful discourse often trumped algorithmic rage, where one could bridge communities of interest while maintaining a sense of local belonging. The federation wasn't just a technical architecture; it was a social contract, enabling connection without central control. It felt like the future.

My identity was always subject to the will of an admin or, if I ran my own instance, this identity was simply rented from a domain name registrar.

Yet, over the years, flaws in the foundation began to show. The very structure that enabled this federation also imposed annoying limitations. While our content was decentralized our identities were not. My digital self, and all of my social graph were irrevocably tied to a specific URL. Sovereignty was an illusion; my identity was subject to the will of an admin or, if I ran my own instance, this identity was simply rented from a domain name registrar.

What's worse, portability was a myth, locked behind complex technical migrations and the tendency of server software developers to freestyle their own versions of the protocol. The decentralization was selective, it distributed the user-generated value, but funneled users through vulnerable single-points of entry. These pain points, inherent to the protocol's design, shifted my perspective from advocacy to skepticism. The future was federated, but I have come to believe that future was obsoleted by something more robust, more sovereign, and more decentralized.

The Sovereign Protocol: NOSTR and the Architecture of Autonomy

My disenchantment with the federated model did not lead me to abandon the dream of decentralized social networking, but to refine it. In 2022, I found a better realization of that original vision, in a very different paradigm: NOSTR (Notes and Other Stuff Transmitted by Relays). NOSTR addresses the core limitations of the Fediverse by architecting for true user sovereignty, creating a system where identity, social graph, and data are the user's property.

This portability transforms the power dynamic; it means my community and relationships are mine, not artifacts of a particular server's database.

Most notably, NOSTR establishes sovereign identity. Just like old, well tested and popular technologies such as GPG/PGP, my identity is not a URL granted by an admin and rented from a registrar. It is a cryptographic key pair, generated offline on my own device, on my own terms. This self-sovereign identity is completely portable. I can use it to authenticate with relays (NOSTR’s equivalent of a server) in the network, and I can leave any relay at any time without losing my identity or my social connections. Critically, my social graph (the list of people I trust to share important information with me, the people i "follow".) is part of this sovereign identity, stored as a simple, cryptographically signed list that I carry with me. This portability transforms the power dynamic; it means my community and relationships are mine, not artifacts of a particular server's database.

My followers subscribe to me, not to a specific platform or format. This eradicates the fragmentation of the Fediverse, where your digital self is scattered across multiple domain names.

This sovereign identity enables a second advantage: universal content creation under a single identity. In the Fediverse, content format is tied to server software. To microblog, you need a Mastodon account; to write long-form, a WriteFreely account; to share videos, a PeerTube account. Each requires a new identity on a new server. In NOSTR, a "post", a follow, a direct message, is simply a cryptographically signed JSON object. Because the protocol is agnostic about content type, my single cryptographic key can publish microblogs, articles, audio notes, or live streaming metadata. My followers subscribe to me, not to a specific platform or format. This eradicates the fragmentation of the Fediverse, where your digital self is scattered across multiple domain names.

Third, NOSTR achieves a better balance between permissionless networks and moderated spaces. The NOSTR network itself is permissionless; anyone with a key pair can participate. However, individual relays are not. They can be private, curated, or enforce specific rules, much like Fediverse instances. The main difference is that a user is not bound to a single relay. I can publish to (and read from) multiple relays at the same time, participating in different communities without being defined by or subject to the culture of any server or community. I can use a specialized relay for technical discussions and a family-friendly relay for general posts, or why not leverage the local pub's relay while i am there? And all with the same identity.

This architecture disempowers bad actors, both users and operators. A relay operator has no power over my identity. They cannot deplatform me; they can only refuse to host my data on their specific relay. This disincentivizes the admin "power trips" seen in some Fediverse instances, as operators are encourage to provide stable service in good faith, or the users (and their valuable content) migrate elsewhere. Conversely, I am also incentivized towards good behavior, as persistent abuse could lead to my exclusion from the relays. Reputation becomes attached to the immutable cryptographic identity, creating a more durable and user-responsible social contract.

Finally, this cryptographic foundation makes privacy a protocol feature, not an add-on. Because my identity is a cryptographic key pair, end-to-end encrypted private messaging is native to NOSTR. Of course, true privacy needs more than that. On the modern internet, privacy is more of a personal hygiene than a choice in software. But it's a good foundation to build on. A relay simply passes along the encrypted data. It has no ability to read my private communications, offering a level of privacy that federated servers, which always work with plaintext data, cannot match without complex external tools.

NOSTR replaces the fragile, admin-dependent federation of domains with a resilient, user-centric network of cryptographic identities.

In summary, NOSTR inverts the Fediverse's model. Instead of federating servers and tethering users to them, it federates sovereign users who temporarily utilize servers for the purpose they see fit. It replaces the fragile, admin-dependent federation of domains with a resilient, user-centric network of cryptographic identities.

The Burden of Sovereignty: Challenges in the NOSTR Paradigm

My goal here is not to declare NOSTR a perfect utopia. Its radical architecture may give unparalleled sovereignty, and enable a web of trust dearly needed in these times of disinformation, it also imposes burdens and presents problems that do not exist in the federated model. These are not mere technical bugs to be fixed, but intrinsic features of a system where the responsibility shifts from server administrators to the individual user.

The biggest challenge is the difficulty of key management. Sovereign identity means the user is their own root certificate authority. There is no "Forgot Password?" link that emails you a reset. The private key is the singular, irrevocable proof of self. Losing it means the permanent loss of that identity and every connection built with it. Compromising it means handing over the entire digital life created around an identity, past and future, to an attacker. This places a huge cognitive and security burden on users, most of whom struggle with basic password hygiene. Imagine trusting them with lifelong guardianship of a cryptographic key. Furthermore, the act of signing is a final, legal-grade attestation. Every post, every like, is cryptographically undeniable. This demands a level of intentionality and caution in communication that is alien to the more ephemeral, deletable culture of mainstream or even federated social media. It forces a coming to terms with a hard truth: privacy, and the permanence of our digital footprints, is not a feature an app can provide. It is a discipline of personal hygiene that NOSTR makes explicit and non-negotiable.

This creates a single point of failure for one's social presence, a problem the federated model largely avoids through account recovery mechanisms controlled by instance admins.

Closely related is the immaturity of key lifecycle management. Robust methods for key rotation, delegation, or inheritance (common in enterprise cryptography) are still nascent in NOSTR clients. This lack compounds the risk of loss or compromise. In the PGP "Web of Trust" model, keys can be revoked and new ones signed. In NOSTR's current state, your original key is your eternal, un-rotatable self. This creates a single point of failure for one's social presence, a problem the federated model largely avoids through account recovery mechanisms controlled by instance admins. This is also an issue for organizations: if their social properties are managed by a key, how do we allow a social manager to write, without surrendering the entire account and it's future? While multiple solutions for multi-signature are being advanced, the protocol still lack meaningful tools for an organization to leverage it.

Beyond key management, there is a significant conceptual hurdle in understanding the network's topology. The decoupling of identity (keys), data (events), and access points (relays) is disorienting. Understanding that "where" you post is subjective and fluid, while "who" you are is absolute, requires a mental model shift. Finding content, and shaping one's experience are functions of the client software. The separation of substance from form is powerful but abstract. It creates a steep learning curve that will make mainstream adoption complicated.

Finally, there is the reality of the current network culture and scale. NOSTR remains a small space. Its early adopters are heavily skewed towards cryptocurrency advocacy, with Bitcoin culture undeniably prevalent. For those opposed to this world, it can feel like an unwelcoming niche. Yet, this very origin also creates a unique strength. The embedded capability for micropayments via the Lightning Network fosters a nascent gift economy (often termed "Value 4 Value"). This allows direct, frictionless support between creators and audiences, a feature that is protocol-native rather than a corporate monetization add-on. While the community is actively building vibrant spaces for art, technology, and philosophy beyond finance, the network's size and its economic layer remain barriers for a broader public seeking a simple social media alternative.

This is not a system designed for passive consumption; it is a toolkit for active, conscious digital citizenship, with all the power and danger that entails.

NOSTR does not eliminate the problems of online interaction. It replaces the problems of centralized control and platform/server dependency with the problems of absolute personal responsibility and conceptual complexity. It trades the risk of an admin's capriciousness for the risk of one's own human error. This is not a system designed for passive consumption; it is a toolkit for active and conscious digital citizenship, with all the power and danger that entails.

Forging a Sovereign Future

My journey from the federated promise of ActivityPub to the sovereign architecture of NOSTR is not a rejection, but an evolution. It is the recognition that the fundamental flaw of the federated model; tethering identity to a domain, was a limiting compromise that could not be patched, only replaced. My thesis extolled NOSTR's design for autonomy; my antithesis laid bare its demanding realities. Synthesizing these reveals that NOSTR is not merely a better protocol, but a necessary step toward a user-empowered internet, one that demands and rewards greater personal responsibility.

1. Sovereignty is a discipline, not just a feature. The antithesis identifies key management and the permanence of the signed event as NOSTR's greatest burdens. Yet, these are not bugs to be wished away. They are the direct, logical consequences of the sovereignty promised in the thesis. True ownership means true responsibility. The difficulty of this model is its entire point. It forces a societal shift from viewing digital interactions as disposable, platform-managed chatter to recognizing them as durable, personal attestations. The "hygiene" it demands is the foundation of a digital society where individuals have authentic agency, not just user profiles. The technical challenges of key rotation and recovery are significant, but they are engineering problems to be solved around the sovereign core, not by dismantling it.

2. Complexity is the price of true choice. The conceptual difficulty of understanding relays and clients is real. However, this complexity is the result of decoupling form from substance, which the thesis celebrated as an advantage. In the Fediverse, the instance (form) dictates the community and rules (substance). In NOSTR, I choose my experience (client, relay filters) independently of my immutable identity and social graph. This abstraction is powerful precisely because it is complex. It mirrors the real world, where our identity is constant, but we choose which forums, gatherings, and media to engage with. Making this model accessible is the great design challenge for NOSTR clients, not a refutation of the underlying principle.

3. A network's culture is its adolescence, not its destiny. The current size and Bitcoin-centric culture of NOSTR are functions of its early stage, not its end state. The federated Fediverse, too, was once a small niche of FOSS idealists. The embedded gift economy, born from its cryptocurrency roots, is not a cultural baggage but a revolutionary core—a built-in, protocol-level alternative to surveillance-based advertising. As the tools mature and the burdens of sovereignty are mitigated by better software, the network can and will diversify. The vibrant communities I cherished in the Fediverse can be rebuilt on NOSTR, but this time on a foundation of portable identity and unbreakable social bonds, free from the threat of arbitrary exile.

The future was federated, a necessary experiment that revealed the true goal: not federation, but sovereignty.

In conclusion, I stopped trying to grow ActivityPub not because its dream of a decentralized social web was wrong, or because federation is an obsolete concept, but because its architectural blueprint was too limited. It federated servers but left users as tenants. The future was federated, a necessary experiment that revealed the true goal: not federation, but sovereignty. This doesn't mean that federation is an obsolete model, it is still very relevant in systems where a trusted 3rd party is required. It only means that, for online socialization NOSTR, and all its current difficulties, is better adapted to the modern web. It accepts that a free, resilient, and human-centric internet cannot be made convenient at the expense of user agency. It replaces the fragility of rented domains with the resilience of cryptographic keys. It is a harder path, requiring different tools and more thoughtful users. But it is currently the only known path that leads to a future where our digital selves are irrevocably our own. The work now is not to patch the old federation, but to build the accessible bridges that will allow others to cross into this sovereign territory. And then again, everything we know and think about social media may soon be obsoleted by an entirely different form of interaction. Who knows what the future holds?

Tune in to the discussion 💬

(These services are bridged: join your favorite and reach them all)

🗨️ IRC
🗨️ Matrix
🗨️ Telegram
🗨️ Discord

Support Dyne 🫱🏿‍🫲🏾

Help dyne.org stay focused on hacking the planet!

🪙 Bitcoins: bc1qz9wz2f9swcefra2tfrhk4fx49evqsv03m9nx4l
☕ Ko-Fi
🍴 Github.com
🧁 LiberaPay
🍥 Patreon.com

Follow Dyne.org 🗞️

Social Media everywhere!

🐭 Lemmy
🐘 Mastodon
🎬 Peertube
📸 Instagram
🐦 Xitter
👔 Linkedin
🪞 Facebook
🧵 Threads
✍️ Medium