Prompt CartographyProfessional Self-Learner Path
An untimed two-hours-per-week guide for independent learners
How to Use This Guide
Keep the commitment small and durable: approximately two hours in a normal week, with no penalty for taking longer on a difficult module.
Course Description
This is not a classroom syllabus with deadlines. It is a pacing system for learning Prompt Cartography without turning your evenings into a second job. The only recurring commitment is roughly two hours per week. Some modules may take one session; some may take several. Progress is measured by evidence you create—not by the calendar.
Preparation & Time
Prerequisites / preparation
Curiosity, access to the textbook, a web browser, and an LLM you are comfortable using. Prior cartography/GIS experience helps but is not required.
Time commitment
Keep the weekly ceiling near two hours. A useful rhythm is 25 minutes reading, 70 minutes making, 20 minutes critique/logging, and 5 minutes choosing the next action. If a lab runs long, stop and resume next week; the whole point is sustainable practice.
Learning Outcomes
- Frame a mapping problem in terms of purpose, audience, decisions, uncertainty, and appropriate scope before asking an LLM to generate a map.
- Write and revise structured natural-language directives that function as reusable cartographic design artifacts rather than one-off requests.
- Build a traceable prompt-to-map workflow that separates data, visual, interaction, critique, ethics, publication, and maintenance decisions.
- Evaluate spatial data for fitness, provenance, privacy, bias, representation, and limits on what a map can responsibly claim.
- Direct map form, layout, hierarchy, color, typography, symbology, and thematic representation with explicit visual reasoning.
- Specify animation and interactive behaviors with attention to state, pacing, accessibility, and user tasks.
- Critique AI-assisted map outputs for design drift, misleading defaults, accessibility problems, and mismatches between intent and result.
- Produce a defensible map or map system accompanied by prompts, provenance, critique history, and a plan for publication or maintenance.
Primary Resources
Technology Approach
You do not need to standardize on a single commercial LLM, GIS package, or web-mapping library. Choose tools you can access safely, and keep the emphasis on cartographic reasoning, workflow quality, documentation, and transferability rather than vendor-specific tricks.
Prompt & Provenance Minimum
For any assessed or portfolio map, preserve enough of the production record to explain purpose, data sources, tool/model assistance, important prompts/specifications, human revisions, accessibility/ethics/uncertainty checks, and what should happen when data or technology changes.
Measure Progress, Not Pace
After each module ask: Can I explain the decision? Can I reproduce the workflow? Can I identify what I would critique next? Save the listed evidence in your Prompt Portfolio.
Capstone / Completion Target
Finish one map or small map system you would be comfortable showing a colleague, client, instructor, or future employer—with enough documentation that you could explain how and why it was made.
The Two-Hour Weekly Rhythm
Read a manageable slice. Stop mid-chapter if needed.
Work on the named core lab or your own project.
Save prompts, compare intent to output, and write what you would change.
Decide exactly where you will resume next week.
Self-Paced Modules
| Module | Focus | Read / Practice | Minimum Evidence Before Moving On |
|---|---|---|---|
| Start here | Build your learning system | Set up a Prompt Portfolio (one folder or notebook for prompts, outputs, critique notes, source links, and decisions). Bookmark the book site, labs, WebMapGPT, and the design-schema tools. | Evidence: a one-page map idea list and a blank provenance template. |
| 1 | Director mindset + prompt literacy | Ch. 1: From Web Cartographer to Prompt Cartography Director; Ch. 2: Learning to Speak via Prompt. Complete Lab 1: The Prompt Director's Chair; Lab 2: From Casual Prompt to Cartographic Design Artifact. | Evidence: one before/after prompt pair showing how purpose, audience, and constraints changed the map request. |
| 2 | Memory + repeatable workflow | Ch. 3: The Cartography Process as Knowledge Recall; Ch. 4: The Prompt Cartography Workflow and Pipelines. Complete Lab 3: Designing Cartographic Memory; Lab 4: Build a Prompt-to-Map Pipeline. | Evidence: a small knowledge document and a diagram of your prompt-to-map pipeline. |
| 3 | Ethics + critique | Ch. 5: Provenance, Ethics, and Privacy; Ch. 6: The (Constant) Critique Pipeline. Complete Lab 5: Ethical Provenance and Privacy Packet; Lab 6: The Critique Pipeline Dossier. Optional deeper reflection: read the “Disputation on the Power and Efficacy of Human Cartographic Expertise.” | Evidence: an ethics/provenance packet and one critique that causes you to revise or reject something. |
| 4 | Data engineering | Ch. 7: The Data Engineering Pipeline. Complete Lab 7: Build a Task-Specific Data Pipeline. | Evidence: a data brief that states what the dataset can support, what it cannot support, and how you know. |
| 5 | Form + layout | Ch. 8: Learning the Language of Map Form and Style; Ch. 9: Map Elements, Layouts, and the Curse of LLM Defaults. Complete Lab 8: Prompting Map Form and Visual Dialect; Lab 9: The Map Elements Specification. Use the Map Design Lab / schema library to remix a style and edit the resulting design language. | Evidence: a design schema plus a “default autopsy” noting at least three choices you deliberately overrode. |
| 6 | Color + typography + symbology | Ch. 10: Color Theory and Palette Design; Ch. 11: Prompting for Typography; Ch. 12: Map Symbology. Work through Lab 10: Palette as Argument; Lab 11: The Typographic Voice System; Lab 12: The Symbology Specification across as many two-hour sessions as you need. | Evidence: one coordinated visual-system specification with accessibility notes. |
| 7 | Thematic representation | Ch. 13: Thematic Representation. Complete Lab 13: The Thematic Method Trial. | Evidence: a chosen method, one rejected method, and a short explanation of what each would imply. |
| 8 | Time + interactivity | Ch. 14: Change over Time; Ch. 15: Bringing Your Visual Plan to Life via Interactivity. Complete Lab 14: The Animation Worthiness Test; Lab 15: The Interactive Behavior Blueprint. | Evidence: an animation decision and an interaction blueprint with explicit state, disclosure, and persistence choices. |
| 9 | Direct the whole map | Ch. 16: From Prompt Cartographer to Interactive Web Map Director. Complete Lab 16: Direct an Earlier Map into Excellence. | Evidence: revisit an earlier map and make at least three deliberate revisions, including one subtraction. |
| 10 | Plan for the map’s second life | Ch. 17: Sustaining a Map That Has Left the Station: The Maintenance Pipeline. Complete Lab 17: The Map Maintenance Dossier. | Evidence: a maintenance dossier with update triggers, ownership, QA checks, and a graceful-retirement condition. |
| 11 | Accountability + uncertainty seminar | Ch. 18: Mapping Community Data with Accountability; Ch. 19: Mapping Presence, Absence, and Uncertainty. Apply both chapters to a real public-facing or community-relevant map you can inspect. | Evidence: a one-page “publish / revise / do not publish” decision with rationale. |
| 12 | Capstone + handoff | Ch. 20: The Main Feature Presentation: Visualizing Complex Connections. Build or finish one defensible map or small map system. | Evidence: final map/prototype + prompt stack + provenance + critique history + accessibility/ethics note + maintenance plan. |
Course / Practice Policies
Responsible AI use
Use any suitable LLM or agentic tool, but keep a decision record. Treat generated code, data transformations, claims, and cartographic choices as drafts that require human review.
Confidential and client data
Do not place proprietary, regulated, client-confidential, or personally identifying information into an external AI service without authorization. Substitute public or synthetic data for training exercises when necessary.
Tool neutrality
The course is workflow-first. Specific LLMs and mapping libraries may change during the course; participants are assessed on transferability, judgment, reproducibility, and communication rather than platform-specific tricks.
Accessibility and professional standards
Treat accessibility, provenance, uncertainty, and ethical review as production requirements rather than end-of-project cleanup.