In Ostrom’s original formulation, biophysical characteristics describe the nature of the resource system—its boundaries, productivity, predictability, and mobility. For Nostr, the "resource system" comprises the protocol infrastructure that enables information sharing and the information artifacts themselves.

Relays function as facilities in IAD terminology—the physical infrastructure enabling resource access. Each relay is an autonomous WebSocket server that stores and forwards events according to its operator’s policies. Relays exhibit key characteristics: they’re non-excludable in principle (anyone can run a relay), rivalrous in practice (storage and bandwidth cost money, compete for their defined quality user), and independently operated (no requirement to communicate with other relays). The relay network’s topology—hundreds of relays with overlapping user bases—creates natural load balancing and censorship resistance. If one relay rejects content, users can access it via broadcasting to alternative relays.

Events serve as both artifacts and resource units. Every Nostr object—notes, follows, mutes, reports, profiles—is an event with standardized structure: ID, public key, timestamp, kind integer, tags array, content string, and cryptographic signature. Events are signed with Schnorr signatures using the secp256k1 curve, making them tamperproof and verifiable. This cryptographic foundation enables trustless verification: clients validate signatures without trusting relays. Events exhibit non-rivalrous consumption (reading an event doesn’t prevent others from reading it) but rivalrous storage (each relay must allocate resources to store events).

Content and ideas represent the knowledge commons. Information flows through Nostr’s infrastructure create value through network effects: each additional user and relay increases the protocol’s utility. But this commons faces subtractability of a different kind—attention and discourse quality are finite resources that can be degraded by spam, harassment, and low-quality content. Unlike physical commons where the resource (fish, timber, water) is the object of appropriation, in knowledge commons the challenge is often maintaining quality and preventing pollution rather than preventing depletion.

The protocol’s event classification system creates governance structure through technical design. Events fall into four categories based on kind number: regular events (stored indefinitely, depending on operator decisions), replaceable events (typically, only latest version of kind published by pubkey kept), ephemeral events (not stored), and parameterized replaceable events (latest per kind+pubkey+identifier). This classification has governance implications: follow lists (kind 3, replaceable) can be updated without accumulating history, while reports (kind 1984, regular) create long lasting records, which can be broadcasted throughout the open network. The protocol’s constitutional rules are embedded in these technical specifications.

Community attributes in the IAD framework encompass trust levels, reciprocity norms, common understanding, social capital, and cultural repertoire. Nostr’s community characteristics differ markedly from traditional commons due to its global scale, pseudonymous identity, and permissionless participation.

Users identify through public-key cryptography rather than legal identity. Each user controls a keypair (private key and corresponding public key) that functions as sovereign digital identity. This architecture eliminates platform control over identity—users can’t be "banned" globally, only excluded from individual relays. The cryptographic identity system enables censorship resistance but complicates accountability: creating new identities is trivial, making Sybil attacks easy. Social capital must be rebuilt for each new identity, providing some deterrent to bad-faith identity cycling and whitelist exclusive relay membership.

Users are able to dilineate which relays to explicitly connect to, choosing which to read or write to. Users are also not bound to only connect to their explicitly defined relays-clients may also be relay specific, or enable users to ad-hoc insert relays to broadcast their data to. link::https://www.whynostr.org/post/8yjqxm4sky-tauwjoflxs/[The Outbox] model is one such way users can customize the relays they connect to, which is a combination of their own explicit reading and writing relays, while also pulling in notes from relays of their followers, further complicating the legibility of the nostr network.

Relay operators form a distinct community with different incentive structures. Some operate relays altruistically for ideological reasons; others charge subscription fees; some specialize in specific content types or communities. Relay operators make collective-choice decisions—setting content policies, implementing filtering rules, allocating resources—that shape users' operational experiences. The relay operator community exhibits cultural norms around free expression, technical excellence, and resistance to censorship, though significant diversity exists.

Protocol developers constitute a third community focused on evolving the protocol itself. The NIP (Nostr Implementation Possibilities) process governs protocol changes through informal consensus. The official link::https://github.com/nostr-protocol/nips/[NIPs repository] requires working implementations relay before acceptance, ensuring proposals are practical rather than theoretical. Other link::https://nostrhub.io[nostr native solutions] offer opportunities for developers to formally state they are using a specification without needing approval from an authority. In general, developer culture emphasizes simplicity, backwards compatibility, and permissionless innovation. No central authority controls protocol evolution; forks and alternative solutions remain possible if maintainers act against developers' interests.

Trust and reciprocity emerge through repeated interactions and reputation building. Despite pseudonymity, users invest in their identities through consistent participation, quality contributions, and relationship building. Follow graphs create web-of-trust structures that enable reputation to function across the network. Users can weight reports from trusted connections more heavily than reports from strangers, creating personalized moderation without central authority. This distributed trust model replaces platform-based verification with social proof.

Social capital accumulates through visible contributions: insightful posts, helpful reports, maintaining relays, developing clients, documenting the protocol. The public nature of most actions enables reputation assessment. However, Nostr lacks formal global reputation metrics that aggregate these signals, leaving reputation assessment to individual users and client-side algorithms. This represents both a strength (no gameable central score) and weakness (higher cognitive burden on users, no global standardization of quality).

Ostrom distinguished three nested levels of rules. Constitutional rules define who can participate in collective choice and under what procedures. Collective-choice rules specify how operational rules are made and modified. Operational rules govern day-to-day appropriation, provision, monitoring, and sanctioning activities. This hierarchy appears clearly in Nostr’s governance architecture.

Constitutional-choice rules are embedded in the protocol specification itself. These rules define what counts as a valid event (cryptographic signature requirements), how events are structured (standardized JSON format), what kinds exist and their semantics, and how clients and relays communicate (WebSocket message types). Constitutional rules change through the NIP process, which requires community consensus, multiple implementations, and demonstrated utility. The high barrier to constitutional change (backward compatibility requirements, need for broad adoption) provides stability while permitting gradual evolution.

Key constitutional rules include: (1) events must be signed with valid Schnorr signatures; (2) event IDs are SHA256 hashes of canonical serialization; (3) clients validate signatures; (4) relays may reject events but cannot modify them; (5) users control their private keys and thus their identity (with managed keys being a lesser, but possible function). These rules cannot be violated without breaking compatibility with the protocol. They constrain all collective-choice and operational rules, creating the constitutional foundation for decentralized governance.

Collective-choice rules are set by relay operators and govern which content their relays accept, store, and serve. Each relay operator makes policy decisions that structure users' operational experiences: which event kinds to store, retention policies for different kinds, rate limiting rules, content filtering criteria, and sanctions for policy violations. These collective-choice rules are sovereign within each relay’s domain but don’t bind other relays. Users experiencing undesirable collective-choice rules can shift to relays with preferred policies, creating competitive pressure that shapes relay governance.

Relay information documents (NIP-11) make collective-choice rules visible. Relays can publish metadata specifying supported NIPs, retention policies (which event kinds stored and for how long), content policies, contact information, payment requirements and access limitations (e.g. rate limiting). This transparency enables informed relay selection. However, many relays don’t publish comprehensive policies, and actual enforcement may differ from stated policy. The protocol provides mechanisms for making collective-choice rules observable but doesn’t mandate their use.

Operational rules govern day-to-day user behavior and emerge from both explicit relay policies and implicit social norms. Explicit operational rules include: posting limits, prohibited content categories, required proof-of-work for events, payment requirements, and use restrictions. Implicit operational norms develop through community practice: norms around content warnings for sensitive material, courtesy in discussions, reporting standards, and appropriate use of tags. These norms aren’t encoded in protocol or enforced by relays but shape behavior through social pressure of users, developers and reputation effects.

The interplay between rule levels creates governance dynamics. Users dissatisfied with collective-choice rules (relay policies) can establish their own relays, essentially exercising constitutional rights to self-organize and exit. Relay operators facing competitive pressure may adjust policies, demonstrating how operational outcomes feed back into collective choice. Protocol developers proposing constitutional changes must consider whether clients and relays will implement them, creating bottom-up constraints on constitutional evolution. This nested structure enables both stability and adaptation.

The action arena in IAD framework analysis represents where participants interact, make decisions, and generate outcomes. For Nostr, the primary action arena is feed curation and content moderation decision-making distributed across users, clients, and relays.

Positions in this action arena include: (1) Content creators who publish events; (2) Content consumers who request and view events; (3) Relay operators who accept, store, and serve events; (4) Client developers who implement filtering and display logic; (5) Reporters who file kind 1984 events; (6) Moderators (emerging role) who aggregate signals and provide curation services. Each position has distinct capabilities, information access, and incentives. Intermediated activity, through clients and relay curation offer novel opportunities beyond individual actors, ranging from explicit relay defined feeds to any number of algorithmically generated processes based on user interactions.

Actions available to actors vary by position. Content creators can publish events to selected relays, tag content with warnings or categories, and respond to reports or feedback. Content consumers can follow users (kind 3), mute users/content (kind 10000), file reports (kind 1984), choose which relays to query, and select clients with preferred filtering logic. Relay operators can accept or reject events, apply retention policies, filter queries, and block specific public keys. Client developers can implement recommendation algorithms, integrate reputation systems, and provide moderation tools. These actions combine to create content governance without central authority.

Information asymmetries shape action arena dynamics. Users see only events from relays they connect to, creating filter bubbles determined by relay selection. Relay operators see all events submitted to their relay but lack visibility into other relays' content. Client developers can observe patterns across users but may not access all relays. Reports (kind 1984) partially address this asymmetry by creating public signals about content quality that propagate across the network. Users can leverage reports from trusted sources to filter content they haven’t personally evaluated.

Control and autonomy differ markedly from centralized platforms. No actor has unilateral control over the network. Users control their identity and content production but not how relays handle their events. Relay operators control their servers' behavior but not user actions or other relays' policies. Protocol developers control neither users nor relays, only their ability to propose standards. This distribution of control creates checks and balances: no single actor can capture the system, but coordination requires alignment of interests rather than command.

Net costs and benefits create incentive structures. Publishing content requires minimal cost (signing events, bandwidth), but quality content production requires significant effort. Relay operation costs money (servers, storage, bandwidth) but can be offset by fees, donations, or altruistic satisfaction. Filing reports costs attention and reputation if reports prove inaccurate, while effective filtering saves attention and improves experience. These cost-benefit structures shape participation patterns and governance outcomes.

Patterns of interaction emerge from actors making choices within the action arena’s constraints. For Nostr, key patterns include network effects from social graph construction, feedback loops from reporting and reputation, and relay specialization through competitive differentiation.

Follow networks (kind 3 events) create the fundamental social graph. Each user publishes a replaceable event listing followed public keys with optional relay hints and petnames. These follow lists are public by default, enabling social graph analysis and web-of-trust algorithms. Clients can implement features like "friends of friends" discovery, relay suggestions based on follows, and content recommendations from your network. The follow graph creates positive network effects: following valuable contributors increases your feed quality, which demonstrates their value to others, increasing their audience and influence.

Engagement events are any form of interaction from one user to an event, where the data formally lives within the tags field (e.g. referencing an event through an a or e tag, engaging another user through a p tag) through commenting (kind:1/kind:1111), boost/repost (kind:6), reacting (kind:7) or sending a monetary transaction (kind:1935). These are only the main types of measurable engagement established by conventions, and more may be created.

Mute lists (kind 10000 events) enable individual filtering without public visibility. Users can mute specific public keys, event IDs, or content categories. Mute lists can be public (tags array) or private (encrypted content field), giving users control over visibility of their filtering decisions. While mutes are private by default, aggregate mute patterns could reveal consensus about problematic actors if users chose to make mutes public.

Reports (kind 1984 events) create public, semi-costly signals about content quality. Users file reports by publishing events that reference the reported content (e` tag) or user (p tag) and specify a report type: nudity, malware, profanity, illegal, spam, impersonation, or other. Reports are permanent, public, and cryptographically signed, making them observable governance traces. Filing inaccurate reports risks reputation damage, but reports also a personal choice for feed filtering with their own personal definitions. Users and relays can weight reports from trusted sources, creating distributed content moderation. The public nature of reports enables research on governance dynamics in ways impossible on centralized platforms.

Relay filtering creates content governance through inclusion/exclusion decisions. Relays may implement policies like accepting only certain event kinds, requiring proof-of-work for events, blocking specific public keys, filtering content based on analysis (NSFW detection, spam filtering), or rate-limiting high-volume publishers. These filtering decisions are observable through relay responses to submission attempts but opaque in their criteria. Users experiencing censorship on one relay can shift to others, but relay filtering still shapes content visibility across the network.

Network exclusion—a theoretical, but most severe sanction—occurs when sufficient relays reject a user’s content that network participation becomes impractical. While no central authority can ban users globally, if major relays coordinate exclusion (through shared blocklists or similar policies), effective network exclusion results. This represents emergent governance through relay coordination rather than protocol-level enforcement. The boundary between acceptable decentralized moderation and problematic censorship is a negotiation between a relay and the community members connecting to it.

Outcomes in the IAD framework represent the results of interactions within the action arena, evaluated against multiple criteria. For Nostr, key outcomes include discourse quality, censorship resistance, network growth, and sustainability of relay infrastructure.

Discourse quality emerges from the aggregated effects of individual filtering decisions, relay policies, and client algorithms. Early evidence suggests variable outcomes: some communities achieve high-quality discussion through careful curation and reporting, while others struggle with spam, harassment, or low-engagement. The distribution of discourse quality across relay-client combinations represents an institutional experiment in decentralized content governance.

Censorship resistance—a primary design goal—appears robust against state-level and platform-level censorship. Users censored on one relay can access the network through others. Cryptographic identity prevents account seizure. Protocol simplicity enables rapid implementation in new jurisdictions. However, coordinated relay-level filtering can still achieve effective censorship, and reliance on cloud infrastructure creates centralization risks. True censorship resistance requires geographic and institutional diversity in relay operators.

Network growth and adoption serve as revealed preference indicators of institutional success. Nostr has grown to hundreds of public relays and about 25 daily active users as of link::https://nostrarchives.com/analytics[April 2026], demonstrating that decentralized social protocols can achieve meaningful recurrent adoption.

Sustainability challenges include relay operator incentive alignment and spam management without central control. Operating relays costs money but generates limited revenue; many relays operate altruistically, which may not scale. Paid relays create barriers to participation. Finding sustainable relay economics remains an open question. Similarly, spam prevention without central authority requires distributed solutions like proof-of-work requirements, reputation systems, or payment barriers, each with trade-offs.

Evaluative criteria from Ostrom’s framework can assess these outcomes. Economic efficiency: Does distributed moderation achieve comparable content quality at lower social cost than centralized alternatives?

Redistributional equity: Does governance concentrate power or distribute it broadly?

Accountability: Are relay operators and influential users accountable to community?

Conformance to values: Does the system align with free expression, privacy, and censorship resistance values?

Sustainability: Can governance institutions persist and adapt long-term?

These questions remain partially answered as Nostr’s natural experiment continues and new communitys emerge.