Distributed journalism — citizen investigation

Source: Citizen Investigation (Google Doc, via MCP).

This note summarizes that document: a design-oriented treatment of peer-to-peer (P2P) citizen investigations as organizational and socio-technical systems—framed as analysis of decentralized sense-making under pressure, not as operational guidance.

Framing

The text argues that decentralized investigations should be modeled as distributed sense-making protocols, not as conventional media production. Centralized “citizen” efforts that hinge on one visible coordinator recreate a Napster-style failure mode (single choke point for pressure, law, and economics); a BitTorrent-style pattern spreads initiation, visibility, and monetization so no one node is required for the process to continue.

Conceptual model (defensive P2P)

• Pattern: Central control and intermediaries become choke points; demand diverges; repression targets visible nodes; protocols can replace platforms. P2P is framed here primarily as defensive (smaller attack surface), not only as emancipation.
• Design goals: Initiation does not imply ownership; visibility does not imply control; contribution is not tied to public attribution by default; leadership is process-shaped, not personality-shaped; value accrual is indirect, delayed, and distributed.
• Infrastructure themes: Federated or protocol-level communication; content-addressed publishing and mirroring; append-only, tamper-evident evidence logs; separation of raw data, analysis, and interpretation; pseudonymous identities with reputation from contribution quality.
• Methodology: Modular tasks (gather, verify, cross-reference, test hypotheses, synthesize); stigmergic coordination (signals on artifacts, not central directives); explicit uncertainty and coexisting interpretations; no forced consensus.
• Governance: No final “truth” authority; validation, challenge, and revision processes; forkability of datasets and narratives as a safety valve; minimal rules (formats, review, dispute signaling).
• Economics: Reward verification, sourcing, and methodology—not headline visibility; indirect monetization for initiators (reputation, future collaboration, voluntary patronage); avoid exclusive ownership of evidence, paywalls on raw material, and platform lock-in (commons-style peer production).
• Culture: Norms favor evidence over authority, process over fixed outcome, forking over faction fights; anti-hero framing—visibility rotated or diffused so the system does not depend on a spokesperson.

Commons-oriented stack (architecture sketch)

The doc specifies layers: ontology (evidence artifacts, claims, analyses, reviews, forkable narratives, contributions); infrastructure (content-addressed redundant storage; separation of storage, indexing, presentation); coordination (federated or relayed messaging; subscriptions to objects/claims); identity (pseudonymous crypto IDs; non-transferable, non-financial-by-default reputation—references to ideas like Flowsta in the original); methodology (artifact fingerprinting/time-stamping; claims that reference artifacts and decay if unmaintained; validation as privileged labor); governance (no editorial sovereign; forking as governance; boundary rules without defining “truth”); economics (monetize labor and infrastructure, not access to evidence/claims; contribution accounting with value signals that can connect to external settlement); culture (anti-celebrity UI, epistemic humility, defensive participant mindset). It explicitly asks whether stacks like Nondominium / ValueFlows could map to this.

Concrete tool landscape (excerpt)

The document surveys social protocols (e.g. Nostr, ActivityPub/Fediverse, AT Protocol, Farcaster, SSB), storage (IPFS, Filecoin, Arweave, blog tooling), and collaboration/leaks (Matrix, SecureDrop, encrypted DMs)—as a composable “off–Web2” stack for independent work.

Failure-mode analysis

Major risks analyzed include: informal re-centralization (power-law attention on a few interpreters); signal flooding (DoS on attention); ideological capture (monoculture, moralized review); economic capture via funders; legal chilling through identification; procedural ossification; fork fatigue without shared evidence pools; reputational laundering; success-induced capture (institutional absorption). The meta-point: dominant failure is often re-institutionalization, not raw shutdown. Design heuristic: mechanisms that maximize efficiency often raise capture risk; resilience favors redundancy, friction, plurality, and procedural humility.

Economics of creators and “independent investigators”

A long tab reviews influencer business models (platform ads, sponsorships, affiliates, products, subscriptions, IP, hybrids) and contrasts them with independent investigator models (direct audience funding, grants, ideological or hybrid outlets, platform monetization, merchandise, consultancy, crowdsourced investigation). It discusses competitive fragmentation in decentralized media (attention markets, tournament dynamics, Nash-style equilibria favoring conflict) and outlines a Decentralized Media Commons (DMC) direction: OVN- and DAO-inspired ideas—shared infrastructure, contribution accounting, reputation, and incentives that make collaboration economically rational (with references to commons-based peer production, Sensorica-style OVNs, and tooling such as ValueFlows / SourceCred-style accounting in the blueprint).

Closing thread

The piece ends by arguing that much infrastructure already exists; the open question is whether participants adopt coordination and value models that treat investigation as a shared epistemic substrate rather than a sequence of branded media products.

Nondominium in this context

Nondominium is a Holochain application that implements ValueFlows-shaped, agent-centric resource sharing with embedded governance and Private Participation Receipts (PPRs) for participation-backed reputation. It is explicitly aimed at a peer sharing economy and uncapturable, collaboratively governed resources (see README.md, requirements.md, and DOCUMENTATION_INDEX.md). Citizen investigation appears among the project’s adjacent application targets in strategic_development.md.

What is implemented today (relevant capabilities)

At a high level, the codebase delivers a three-zome split—person (identity, roles, capabilities, private-data access), resource (resource specifications and EconomicResource state as the data layer), governance (rule evaluation, commitments, economic events, claims, validation, PPR issuance)—documented in the index and zome docs.

For distributed investigation–style workflows, the following are concrete affordances:

• Agents and membranes: Public profiles and pseudonymous operation, role assignments, capability progression (e.g. Simple → Accountable → Primary Accountable), and time-bounded private data sharing—so participants can separate “public investigator” presence from sensitive fields without a central account database.
• Resources as first-class objects: ResourceSpecification and EconomicResource with lifecycle and governance rules attached to resources—suitable for modeling investigation bundles (e.g. a dataset, a verified extract, a collaborative case file) as resources whose access, transfer, and process use are rule-governed rather than platform-governed.
• Auditability and contribution structure: Commitments, economic events, and claims provide a traceable record of who did what to which resource and when—aligned with the Citizen Investigation doc’s emphasis on provenance, modular tasks, and accounting for verification labor rather than only narrative output.
• Peer validation: Multi-reviewer validation workflows (e.g. N-of-M) and governance-driven transitions support collective checking without a single editorial authority—overlapping the doc’s “validation / challenge / revision” and “review as labor” themes.
• Reputation from participation, not vanity metrics: PPRs (multi-category participation receipts tied to economic activity) are designed so reputation accrues from signed participation in flows, not from follower counts—closer to “review reputation” and contribution quality than to influencer economics.

The Svelte UI and Tryorama test suites exercise these paths end-to-end for the resource sharing domain; applying the same primitives to journalism-specific resource specs and governance patterns is largely a modeling and UX exercise on top of existing APIs.

What is planned or in flight (how the fit improves)

Roadmap and architecture docs describe directions that strengthen the match to commons-based and OVN-style investigation economics:

• hREA integration (integration-strategy.md): richer ValueFlows types and cross-DNA economic operations—useful when investigations need full VF graphs (processes, agreements, fuller commitment/event semantics) while keeping governance and PPR generation atomic with events.
• NDO / generic object model (ndo_prima_materia.md): a more explicit Nondominium Object lifecycle, capability surface, and migration story—relevant if “evidence artifact / claim / narrative fork” become first-class product concepts beyond today’s resource-centric MVP.
• Optional post-MVP bridges: Unyt (economic settlement / proofs) and Flowsta (portable identity across apps)—sketched in post-mvp/—for teams that want external settlement or cross-app identity without re-centralizing the investigation graph.

Requirements also spell out post-MVP agent and organization modeling (collectives, projects, bots, affiliation rules); today’s MVP emphasizes individual agents, so newsroom- or collective-as-agent patterns are design targets, not fully realized in code yet.

How this maps to the Citizen Investigation stack

The Google Doc asks for a protocolized commons: separation of raw data / analysis / interpretation, stigmergic coordination on artifacts, forkability, contribution accounting, and defensive economics. Nondominium does not replace broadcast (Nostr, Fediverse), large-file permanence (IPFS, Arweave), or anonymous drops (SecureDrop)—those remain complementary layers. Where it does plug in is as a coordination and accounting substrate:

In short: Nondominium can serve citizen investigation and distributed journalism as the Holochain-native layer for governed shared resources, auditable contributions, and participation-based reputation—with public narrative and bulk evidence storage still typically composed from federated or content-addressed tools the stack was designed to interoperate with conceptually, not to duplicate wholesale.