How to Digitize a Culture

Building Protocols That Preserve What Makes Us Human

In 1867, farmers in Rochdale, England did something that should have been impossible. Without formal education, without telegraph wires, without any technology more sophisticated than pen and paper, they built an organization that would outlast empires. The Rochdale Pioneers created the modern cooperative movement—not by inventing new ideas, but by encoding existing cultural practices into shareable protocols. They wrote down the unwritten rules of village-scale trust: one member one vote, democratic control, concern for community. Within fifty years, their pattern had replicated across continents.

The Rochdale Pioneers understood something that Silicon Valley has spent thirty years forgetting: culture isn't content. It isn't data. Culture is the way people exchange information with each other—the protocols of trust, the rhythms of feedback, the expectations embedded in every transaction. When you digitize only the content and ignore the protocol, you don't preserve a culture. You hollow it out.

Today, platforms like Instagram flatten the rich visual language of street markets into identical grids of product photos. Amazon reduces generations of shopkeeper-customer relationships into faceless star ratings that tell you nothing useful. Uber erases the reputation systems that taxi drivers built over decades—who drives safe, who knows the shortcuts, who you can trust at 2 AM. The digital transformation of the last thirty years hasn't digitized cultures. It has replaced them with corporate monocultures that extract value while destroying the communicative patterns that made those cultures work.

This paper proposes a different approach. Using two working protocols—Agrinet for agricultural markets and LBTAS for transaction assessment—we demonstrate how to digitize a culture by encoding its communicative patterns into open protocols rather than proprietary platforms.

Culture as Communication Infrastructure

Before examining how to digitize a culture, we must understand what culture actually is. The mistake most technologists make is treating culture as a static thing—a collection of practices, artifacts, and beliefs that can be catalogued and stored. This approach produces digital archives and online museums: useful for historians, useless for living communities.

Anthropologists have understood for over a century that culture is fundamentally dynamic. Culture isn't a noun; it's a verb. It's the ongoing process by which communities create and maintain shared meaning. And the medium of that process is communication—not just the content of messages, but the patterns of who speaks to whom, when, about what, and with what expectations.

Consider a traditional farmers' market. The visible culture is the produce: tomatoes, peppers, fresh bread, honey. But the invisible culture—the culture that makes the market work—is entirely communicative. Farmers recognize repeat customers and adjust prices accordingly. Customers remember which vendor sold them the underripe melon last July. Newcomers observe the flow of transactions and learn appropriate behavior. Reputation travels through gossip networks. Disputes get resolved through social pressure rather than formal contracts. All of this is information flowing through communication channels, building up over time into what we experience as "the culture of the market."

When digital platforms replace this market, they typically preserve the visible culture (pictures of tomatoes) while destroying the invisible culture (the reputation networks, the relationship memory, the informal dispute resolution). The platform substitutes its own communication patterns—centralized, extractive, opaque—for the community's patterns. Users become customers. Vendors become suppliers. The market becomes a vending machine.

Digitizing a culture authentically requires digitizing the communication patterns that generate and sustain it—not just the content those patterns produce.

The Protocol Approach: Learning from TCP/IP

The internet's founding architects faced a problem similar to cultural digitization: how do you enable communication between radically different systems without forcing everyone to adopt a single corporate standard?

Their solution was elegant: instead of building one network, they built a protocol—a set of rules for how different networks could talk to each other. TCP/IP doesn't care whether you're using Ethernet or WiFi, Windows or Linux, a supercomputer or a smartphone. It defines the pattern of communication: how messages get addressed, how they get routed, how receipt gets confirmed. Any device that follows the protocol can participate in the network.

This is precisely the approach needed for cultural digitization. Instead of building platforms that replace cultural practices, we can build protocols that extend them—defining patterns for how agricultural transactions flow, how quality assessments propagate, how reputation accumulates, how disputes get resolved. Any interface that follows the protocol can participate in the network. The protocol preserves the cultural patterns; the interfaces adapt them to different contexts.

The difference is profound. Platforms are owned and controlled—they have terms of service, they can ban you, they extract fees, they change the rules whenever their quarterly earnings require it. Protocols are shared and forkable—if you don't like how one community implements them, you can start your own implementation. Platforms create dependency. Protocols create capability.

Case Study 1: Agrinet and the Culture of Agriculture

Agriculture is one of humanity's oldest cultures—a set of communicative practices for coordinating the transformation of sunlight into food. For most of history, agricultural culture was local and embodied: farmers exchanged seeds and techniques with their neighbors, sold at markets where customers knew their names, passed knowledge to their children through years of demonstration and practice. The Green Revolution of the 1960s began replacing this culture with industrial monoculture, and the digital transformation accelerated the replacement.

Today, agricultural data flows through proprietary platforms controlled by a handful of multinational corporations. Four companies control grain price data, selling access back to farmers whose harvest information created the data in the first place. Rural farmers in many countries, including the United States, cannot access internet-based markets—costing an estimated $1 trillion annually in lost value. The communicative patterns of traditional agriculture—direct producer-consumer relationships, local reputation networks, seasonal rhythms of planning and harvest—have been severed or captured.

The Agrinet Protocol

Agrinet represents an attempt to digitize agricultural culture by encoding its communicative patterns into an open protocol. Modeled after TCP/IP, Agrinet defines a standard format for agricultural transmissions—how products get listed, how transactions get recorded, how quality ratings get assigned—while allowing any interface to implement these standards.

The protocol preserves several key cultural patterns:

Geographic proximity as relationship foundation. Traditional agricultural markets are local. Farmers sell to nearby customers because food is heavy, perishable, and seasonal. Agrinet encodes geography at the protocol level: every transmission includes geolocation data, every broadcast is filtered by proximity, every market view prioritizes local production. The protocol doesn't force you to buy local—but it makes local the default, encoding the cultural assumption that food relationships are fundamentally about place.

Producer-consumer dialogue. In traditional markets, transactions are conversations. You ask the farmer about growing conditions. You tell them what you're cooking. You build a relationship over repeated interactions. Agrinet preserves this pattern through its messaging system, which ties all communication to specific posts, transactions, or production updates. Messages aren't abstract—they're always about something, always part of an ongoing relationship.

Temporal rhythms of production. Agriculture operates on seasonal cycles: planting in spring, tending through summer, harvesting in fall. Industrial supply chains abstract away these rhythms; you can buy strawberries in December because they were grown in Chile and shipped across the equator. Agrinet's Plan Post format encodes seasonality: planting date, expected harvest date, production schedule. The protocol makes time visible, preserving the cultural understanding that good food comes from attentive timing, not logistical manipulation.

Production transparency. Traditional farmers' markets involve implicit verification: you can see the soil on the carrots, smell the basil, judge the farmer's hands for signs of real work. Industrial agriculture hides production behind corporate opacity. Agrinet's PING (Periodic Information, News, and Guidance) system requires regular progress updates with photo or video documentation. The protocol creates digital transparency that mirrors the physical transparency of face-to-face markets.

Protocol Architecture as Cultural Encoding

The technical structure of Agrinet reveals how protocols encode culture. Every transmission follows a recursive format:

key1_[key]/key2_[key]/[ui_string]/[user_data]/[transaction_type]/[content]/over

This isn't just technical syntax—it's cultural syntax. The dual-key authentication encodes trust verification. The UI string encodes interface pluralism (many doors into the same protocol). The user data encodes identity and location. The transaction type encodes the kind of communicative act: are you posting a product, sending a message, providing a rating, updating your production status? The recursive structure itself encodes the agricultural cultural assumption that every transaction is part of a larger relationship, every message has a context, nothing happens in isolation.

Consider the Direct Market Post format:

key1_[key]/key2_[key]/[ui]/[user]/dmpst/
t=Fresh_Tomatoes/c=pd-fm/geoLat_38.2527,geoLong_-85.7585/
d=Cherokee_Purples_picked_this_morning/
med:photo1.jpg-photo2.jpg/
term=$4_per_pound_pickup_only/over

Every field encodes a cultural assumption. The title (t=) assumes products have names that matter. The category (c=pd-fm for produce/fruit-melons) assumes products fit into agricultural taxonomies that communities share. The geolocation assumes place matters. The description field assumes farmers tell stories about their products. The media attachments assume visual evidence of quality. The terms field assumes negotiation is normal and conditions vary. The entire format encodes the communicative pattern of a farmers' market stall: here's what I have, here's where I am, here's what I'm asking, come talk to me if you're interested.

Case Study 2: LBTAS and the Culture of Accountability

Every functioning culture has mechanisms for distinguishing good actors from bad ones. In traditional communities, reputation traveled through social networks: your grandmother warned you about the dishonest butcher, the reliable babysitter's name circulated among parents, the skilled carpenter got recommended from job to job. This informal reputation system wasn't just convenient—it was cultural infrastructure. It encoded community standards, provided feedback loops for improvement, and created accountability without centralized authority.

Digital platforms replaced this distributed reputation infrastructure with five-star rating systems—and in doing so, destroyed the culture they claimed to digitize.

The Failure of Five Stars

The five-star system was invented in 1958 by Forbes Travel Guide (then Mobil Travel Guide) to advertise hotel quality along new U.S. interstate highways. It was never designed as a two-way communication system. It was a marketing tool: a way for a corporation to signal quality to transient strangers who would never return to the same hotel twice.

When platforms adopted star ratings for digital commerce, they inherited all the limitations of this one-directional, marketing-oriented design. The scale lacks granularity—what's the difference between four stars and five stars? The ratings are unidirectional—sellers get rated, buyers don't. The meaning is undefined—does three stars mean "acceptable" or "disappointing"? There's no path from rating to improvement—what exactly should the seller do differently?

The result is a system that produces data without producing culture. Ratings accumulate, but communities don't form. Reputation exists as a number, but accountability doesn't flow. Sellers chase ratings through manipulation rather than improvement. Buyers distrust ratings because they've been gamed. The communicative pattern of traditional reputation—specific, bidirectional, embedded in relationships—gets replaced with a pseudo-quantitative abstraction that satisfies no one.

The Leveson Approach

LBTAS (Leveson-Based Trade Assessment Scale) takes a radically different approach. Instead of marketing-derived ratings, it adapts Nancy Leveson's software assessment methodology—developed for aerospace applications where system failures result in loss of life.

The Leveson approach treats assessment as a communicative act with specific information content. Each rating level has a defined meaning:

Rating	Level	Definition
+4	Delight	Interaction anticipates the evolution of user practices and concerns post-transaction
+3	No Negative Consequences	Interaction designed to prevent loss, exceed basic quality
+2	Basic Satisfaction	Interaction meets socially acceptable standards exceeding articulated user demands
+1	Basic Promise	Interaction meets all articulated user demands, no more
0	Cynical Satisfaction	Interaction fulfills a basic promise requiring little to no discipline toward user satisfaction
-1	No Trust	User was harmed, exploited, or received product/service with evidence of no discipline or malicious intent

Notice how this scale encodes cultural assumptions about what good transactions look like. A +4 rating isn't just "better than +3"—it specifically means the other party anticipated your future needs, thought beyond the immediate transaction, treated you as a person with an evolving relationship rather than a one-time customer. The scale distinguishes between "meeting promises" (+1) and "exceeding demands" (+2), between "preventing loss" (+3) and "creating delight" (+4). It encodes a cultural theory of good business: that the best transactions are those that build relationship rather than merely completing exchange.

Bidirectional Assessment as Cultural Pattern

Perhaps more importantly, LBTAS is bidirectional. Both producer and consumer rate each transaction. This encoding preserves the traditional cultural understanding that reputation is mutual: yes, sellers can be unreliable, but buyers can also be exploitative. Markets work when all parties behave well, and accountability must flow in all directions.

In Agrinet, LBTAS ratings are required for every transaction. You cannot complete a purchase without rating the producer; the producer cannot finalize without rating you. The protocol enforces the cultural pattern: no transaction exists in isolation, every exchange creates mutual accountability, reputation is something you build together.

The consequences are proportional. Consistently low ratings affect your visibility in the network. A -1 average results in permanent ban—encoding the cultural pattern of exile for those who harm community trust. The protocol creates the digital equivalent of traditional reputation: distributed, consequential, but not owned by any central authority.

Principles for Cultural Digitization

Drawing from these case studies, we can articulate principles for digitizing culture without destroying it.

Encode Patterns, Not Just Content

When you digitize a farmers' market by putting product photos online, you've digitized content. When you encode geolocation, require progress updates, enforce bidirectional ratings, and structure messages around specific transactions, you've digitized patterns. Content can be extracted and exploited. Patterns must be participated in.

The question to ask: "What do people do in this culture?" The answer is never just "they buy things" or "they share information." The answer involves rhythms, expectations, relationships, accountability structures, ways of building trust over time. Encode those.

Preserve Bidirectionality

Almost every traditional culture involves mutual accountability. Buyers and sellers watch each other. Teachers and students evaluate each other. Neighbors monitor each other. Digital platforms systematically destroy this bidirectionality, replacing it with surveillance from above: the platform watches everyone, everyone watches their rating, no one watches the platform.

Cultural protocols must encode bidirectional assessment, bidirectional communication, bidirectional consequence. The user experience should feel like entering a community, not like being processed by a system.

Embed Rather Than Abstract

Five-star ratings abstract reputation into a number. LBTAS embeds reputation in defined categories with specific meanings. Abstract metrics can be gamed and manipulated; embedded meanings create genuine accountability because everyone knows what the rating claims.

When digitizing cultural practices, resist the temptation to abstract. If your culture has specific expectations about quality, name them. If your culture distinguishes between different kinds of good behavior, make those distinctions explicit. The protocol should feel like a formalization of what community members already know—not an imposition of external categories.

Make Time Visible

Traditional cultures operate in real time. Agricultural cultures follow seasons. Market cultures follow business rhythms. Craft cultures follow production cycles. Digital platforms abstract time into "instant availability," destroying the cultural meaning of timing, seasonality, and patience.

Agrinet's Plan Posts, with their explicit planting dates and harvest windows, make agricultural time visible. The PING update system creates expected rhythms of communication. The protocol encodes the cultural understanding that good food takes time, that relationships develop through repeated interaction, that some things shouldn't be instant.

Default to Open Protocols

Platforms are owned. Protocols are shared. If your cultural digitization requires a specific company to continue operating, you haven't digitized a culture—you've merely rented one.

The choice of AGPL-3 licensing for both Agrinet and LBTAS isn't just a technical detail—it's a cultural encoding. AGPL-3 requires that anyone running the protocol as a service must share their source code. This prevents the pattern that destroyed previous open systems: a company takes an open protocol, adds proprietary features, and gradually captures the network into a closed platform. The license encodes the cultural principle that shared infrastructure should remain shared.

The Work Ahead

We have sketched an approach to cultural digitization that preserves communicative patterns rather than replacing them with corporate monocultures. The examples are real and working: Agrinet coordinates agricultural production across multiple user interfaces; LBTAS provides assessment infrastructure for cooperative networks. But the work is far from complete.

Digital infrastructure has systematically replaced cultural patterns with extractive platforms for three decades. The damage is extensive. Farmers who once knew their customers now ship to anonymous warehouses. Craftspeople who built reputations over decades now chase Amazon reviews. Communities that governed themselves through informal accountability now depend on platform moderation teams in distant countries.

Rebuilding will require not just technical protocols but institutional capacity: cooperatives to govern the networks, developers to build the interfaces, communities willing to experiment with alternatives, and sustained funding for infrastructure that produces public value rather than private profit.

Join the Work

The protocols described in this paper exist because people decided to build them. The infrastructure they represent serves communities only if communities adopt and extend it.

If you're a developer who understands DNS infrastructure, agricultural data systems, cooperative governance, or protocol design: join NTARI's Slack workspace and contribute to the technical work. We need people who can write specifications, build interfaces, and solve problems we haven't anticipated yet.

Join the development conversation: https://join.slack.com/t/ntari/shared_invite/zt-39injdzvr-a7jY2FVU00fYPopG7gyP4w

If the vision of community-owned networks and cooperative digital infrastructure resonates with you but your skills lie elsewhere: NTARI operates as a 501(c)(3) nonprofit. Every contribution funds research, development, and advocacy that makes this work possible.

Support the mission: https://ntari.org/#give

The Rochdale Pioneers didn't have the internet. They wrote their principles on paper, shared them through mail, and replicated their pattern across continents over decades. We have faster communication technology, but the fundamental challenge is the same: encoding cultural practices into shareable protocols that communities can adopt, adapt, and extend.

Culture is essentially communicative. Digitizing it means building the communication infrastructure that lets cultural patterns flow, replicate, and evolve—without being captured by systems that extract value while destroying what made the culture work in the first place.

The protocols are open. The community is building. The question is whether you'll participate.

Learn More

Cultural Digitization Theory:

• Elinor Ostrom's commons governance research

• Platform cooperativism

• TCP/IP protocol design

NTARI Projects:

• Agrinet Protocol

• LBTAS Source Code

Licensing and Open Infrastructure:

• AGPL-3 License

• Why AGPL-3 for cooperative infrastructure

Agricultural Systems:

• FAO Digital Agriculture

• Green Revolution

Contact and Participation:

• Collaborate: https://join.slack.com/t/ntari/shared_invite/zt-39injdzvr-a7jY2FVU00fYPopG7gyP4w

• Donate: https://ntari.org/#give

• General Inquiries: info@ntari.org

Document Status: DraftReview Schedule: QuarterlyQuestions: Contact NTARI Communications Team via Slack workspace