UbiCity Learning Capture System

Making the city itself a learning environment, one captured experience at a time.

What Is This?

UbiCity (the Ubiquitous City of Learning) is a vision for dissolving the boundaries between formal education and urban exploration. This repository contains practical tools to capture, map, and analyze learning experiences that happen outside traditional institutional boundaries.

Instead of waiting for infrastructure, we’re building tools to document urban learning as it happens - creating a corpus of learning experiences that can reveal patterns invisible within traditional educational frameworks.

The Problem

You have an expansive vision (UbiCity) but need concrete next steps. How do you move from concept to practice without:

1. Infinite scope creep

2. Analysis paralysis

3. Building technology before validating the idea

4. Losing the essence in the implementation

This Solution

Start with data, not infrastructure. Capture learning experiences using a minimal protocol, then use analysis tools to discover what patterns emerge organically.

What You Get

• 📋 JSON Schema - Structured format for learning experiences

• 🎯 Capture Tool - CLI for recording experiences in the moment

• 🗺️ Mapper - Analysis engine finding connections and patterns

• 📊 Minimal Viable Protocol - The simplest possible UbiCity instantiation

• 🔍 Export to Voyant - Text analysis integration for deeper insights

Quick Start

Installation

cd ubicity
npm install  # if needed for dependencies
chmod +x capture.js  # make capture tool executable

Capture Your First Experience

Option 1: Quick capture (30 seconds)

node capture.js quick

Option 2: Full capture (2 minutes)

node capture.js full

Option 3: Generate a template to edit

node capture.js template > my-experience.json
= Edit the JSON file, then manually add to ubicity-data/experiences/

Analyze Your Experiences

Generate a full report:

node mapper.js report

Find learning hotspots:

node mapper.js hotspots

View domain connection network:

node mapper.js network

Track a learner’s journey:

node mapper.js learner <learner-id>

The 4-Week Experiment

Want to test if UbiCity works for you? Follow the protocol in MINIMAL_VIABLE_PROTOCOL.md:

• Week 1: Capture 5 learning experiences in different locations

• Week 2: Run analysis, look for patterns

• Week 3: Decide to stop or add one collaborator

• Week 4: Meet and discuss, capture that collaboration

Success criteria: If you found one unexpected connection and generated one new question, UbiCity works at minimum viable scale.

Data Model

Core Structure (MVP)

Every learning experience needs:

{
  "learner": {
    "id": "who learned"
  },
  "context": {
    "location": {
      "name": "where they learned"
    }
  },
  "experience": {
    "type": "what kind of learning",
    "description": "what they learned"
  }
}

Optional Enrichment

Add any of these to deepen analysis:

• Learner background/interests - For finding similar learners

• GPS coordinates - For spatial mapping

• Situation/connections - Who else was involved

• Domains - Disciplinary areas (enables interdisciplinary analysis)

• Artifacts - Photos, sketches, code created

• Outcome - Success/failure, new questions, unexpected connections

• Tags/privacy - Organization and access control

Full schema: schema/learning-experience.json

Analysis Capabilities

The mapper provides several analytical lenses:

1. Interdisciplinary Connections

Find experiences that span multiple domains:

mapper.findInterdisciplinaryConnections()

Returns learning moments where different disciplines met, often yielding unexpected insights.

2. Learning Hotspots

Identify locations with high disciplinary diversity:

mapper.findLearningHotspots()

These are places where the city itself facilitates cross-domain learning.

3. Domain Networks

See which disciplines frequently co-occur:

mapper.generateDomainNetwork()

Reveals the natural boundaries (or lack thereof) between fields in practice.

4. Learner Journeys

Track how individuals' interests evolve:

mapper.getLearnerJourney(learnerId)

Shows progression, new domains discovered, questions emerged.

5. Location Mapping

Understand the learning characteristics of places:

mapper.mapByLocation()

How many different disciplines meet at each location? How many unique learners?

Use Cases

For Individuals

• Document self-directed learning

• Find patterns in your interests

• Justify learning time to yourself/others

• Create a learning portfolio

For Small Groups

• Track collaborative learning

• Find intersection of interests

• Identify productive meeting locations

• Build shared knowledge maps

For Researchers

• Study informal learning patterns

• Analyze urban learning environments

• Generate papers from captured data

• Validate educational theories in practice

For Institutions

• Understand learning outside classroom

• Find community partners

• Document experiential learning

• Support lifelong learners

Integration with Existing Tools

Zotero Connection

This repo started as a Zotero → Voyant exporter. The connection:

• Zotero: Captures research reading and sources

• UbiCity: Captures experiential learning and practice

• Voyant: Analyzes text patterns in both

You can export UbiCity experiences to Voyant:

const mapper = new UrbanKnowledgeMapper();
mapper.loadAll();
const corpus = mapper.exportToVoyant();
// Upload corpus to Voyant for text analysis

Other Integrations

• Echomesh: Context preservation across time

• JanusKey: Reversible operations for experimental learning

• Oblíbený: Type-safe learning protocols

• MAA Framework: Accountability in public learning spaces

Scaling Considerations

Don’t Scale Until You Have 50+ Experiences

Seriously. Capture data first. Then:

1. Add visualization - Map experiences geographically

2. Add recommendations - "Learners like you also explored…​"

3. Add mobile capture - Quick smartphone entry

4. Add social features - Share discoveries

5. Add institutional integration - Course credit, portfolios

But each of these is optional. The core works with just the CLI tools.

Design Principles

1. Capture First, Analyze Later

Minimize friction in the moment. Rich analysis can wait.

2. Privacy by Default

Pseudonymous learner IDs. Privacy levels. Opt-in sharing.

3. Failures Are Data

Experiments that fail still generate valuable learning experiences.

4. Location Matters

Where you learn shapes what you learn. The city is co-teacher.

5. Interdisciplinary by Nature

Cities don’t respect academic boundaries. Neither should learning.

6. Incomplete Is Better Than Perfect

A sketchy capture is better than waiting for perfect data.

7. Tools, Not Platforms

Composable tools you control, not a locked-in platform.

Project Philosophy

This is not a startup. Not a platform. Not a product.

This is infrastructure for a practice - like a yoga mat or a notebook.

UbiCity isn’t something to complete. It’s a continuous process:

• You capture experiences

• Patterns emerge

• You explore those patterns

• New questions arise

• You capture more experiences

• Patterns evolve

The tools here support that cycle. They don’t try to optimize it, gamify it, or monetize it.

Connection to Your Other Work

Formal Methods

• Learning experiences as verifiable events

• Safety properties: learner wellbeing, privacy

• Verification conditions for "successful learning"

Solidarity Economics

• Learning as commons, not commodity

• Mutual accountability in public learning

• Skills-sharing over credentials

Phase-Separated Development

• MVP protocol provides clear phase boundaries

• Each week of experiment is a phase

• Natural pause points prevent scope creep

Seven Tentacles Framework

• Character-driven narrative of learning journeys

• Urban spaces as characters in learning story

• Multi-threaded exploration of ideas

Next Steps

If This Resonates (Immediate)

1. Read MINIMAL_VIABLE_PROTOCOL.md

2. Capture one experience right now

3. Decide: 4-week experiment or not?

If You Want to Extend (Month 2+)

1. Add visualization (D3.js maps of learning)

2. Create web interface (form + map view)

3. Integrate with Zotero (reading → practice)

4. Write paper: "Formal Verification of Urban Learning Systems"

If This Becomes a Life Project

1. Establish quarterly 4-week cycles

2. Each cycle explores new aspect

3. Publish findings after each cycle

4. Tools evolve based on real usage

5. Community emerges organically

Contributing

This is a personal exploration toolkit, but if UbiCity resonates with you:

• Capture your own experiences

• Share patterns you discover

• Extend the tools for your context

• Write about your experiments

The GPL3 license means you can fork, modify, and share freely.

Credits

• Concept: UbiCity vision from 2017+ work with UAL Creative Communities

• Implementation: Built on learnings from zotero-voyant-export project

• Inspiration: Voyant (text analysis), Zotero (research management), and years of trying to formalize the informal

License

GPL3 - Same as the original zotero-voyant-export project.

Learning should be free as in freedom.

---

Get Started

node capture.js quick

Then:

node mapper.js report

That’s it. You’re doing UbiCity.

Welcome to the Ubiquitous City of Learning. 🏙️