Purpose
~5 min
Chapter 2 reframes prompting as a professional design literacy. A prompt is not merely a message to an LLM; in prompt cartography, it becomes a design specification, a record of intent, and a reusable artifact that others can critique.
Core ideaStructure beats cleverness. A clever prompt may work once. A structured prompt can be diagnosed, shared, reused, and improved.
Learning Objectives
~3 min
- Distinguish casual prompting from directive prompting in a cartographic workflow.
- Use CRAFT as a framing aid for a map-design prompt.
- Create system, user, and critique prompts that play different roles in one design process.
- Treat prompts as reusable design artifacts with version history and failure notes.
ToolsA web browser, one or more free LLMs, and a place to take notes. No GIS, coding, paid LLM account, or file upload is required.
Setup: Choose a Map Problem
~7 min
Choose one map idea you can use throughout the lab. Professional self-learners should choose a real work-adjacent topic. Students may choose a campus, city, environmental, public health, transportation, or social issue.
- Bike crash risk near campus routes
- Public access to cooling centers during heat waves
- Where invasive species sightings need public reporting
- Walking access to grocery stores for older adults
Setup deliverableWrite a one-sentence map idea and name the audience you care about most.
Part 1 — Casual Prompt Baseline
~12 min
Start with the kind of prompt Chapter 2 warns us not to trust for professional work.
Make a web map about [YOUR MAP IDEA].
Save the response. Then annotate it with these labels:
- Assumed context What background did the model invent?
- Assumed audience Who did it seem to design for?
- Assumed format Did it recommend a map, outline, code, data list, or something else?
- Assumed tone Did it sound public-facing, technical, urgent, neutral, promotional?
Mini-deliverableA 4-row assumption list. Do not fix the prompt yet.
Part 2 — Ask the LLM to Teach the Concept Back
~10 min
Use the LLM as a study partner before using it as a production assistant. This reinforces the chapter’s language while revealing what the model understands generically.
I am studying Chapter 2 of Prompt Cartography, especially the distinction between casual prompting and directive prompting. Explain the difference in the context of designing an interactive web map. Then give me three signs that my prompt is too casual for professional cartographic work.
Compare the answer to the chapter. Mark anything that feels useful, incomplete, or too generic.
ReminderDo not treat the answer as the book. Treat it as a study prompt that helps you test your own understanding.
Part 3 — Build a CRAFT Brief
~18 min
Use CRAFT as a pre-design checklist for your map idea.
| CRAFT Element | Your Draft | Question to Answer |
|---|
| Context | Example: Map for neighborhood comparison | What situation, place, data domain, or design problem does this prompt assume? |
| Role | Example: Map for neighborhood comparison | What kind of reasoning should the LLM perform? |
Starter examples only; expand this in your own notes or submission document.
Part 4 — Write a Directive User Prompt
~15 min
Transform the CRAFT brief into a directive prompt. This prompt should ask for a specific design artifact.
Context: I am planning an interactive web map about [MAP IDEA] in [PLACE OR SCOPE]. The map should help [AUDIENCE] [DECISION, INSIGHT, OR ACTION].
Role: Act as a cartographic design advisor who prioritizes communication, usability, accessibility, and ethical restraint.
Task: Create a map concept brief that includes:
1. communicative purpose
2. intended audience
3. recommended map type and why
4. 3-5 possible data layers, with cautions about over-inclusion
5. interaction ideas that support understanding rather than novelty
6. likely risks, limitations, or misleading interpretations
7. one thing users should remember after using the map
Format: Use a concise table followed by a 150-word design rationale.
Tone: Professional, careful, and plain-language. Do not write code.
Run the prompt. Save the response as Version 1.
Part 5 — Add System and Critique Prompt Roles
~20 min
Chapter 2 distinguishes prompt types by role. Simulate that structure in an ordinary free LLM chat.
System-style framing prompt
For this conversation, behave as a careful prompt cartography assistant. Prioritize purpose, audience, accessibility, uncertainty, and ethical interpretation. If my prompt is vague, identify the missing information before giving recommendations.
Critique prompt
Critique the map concept you just produced against my stated purpose, audience, and CRAFT fields. Identify:
- 2 places where the output follows the prompt well
- 2 places where it drifts or adds assumptions
- 2 revisions I should make to the prompt itself
- 1 risk that remains my responsibility as the cartographer
Mini-deliverableSave the critique and revise your original directive prompt into Version 2.
Part 6 — Treat the Prompt as a Design Artifact
~15 min
Create a version log. This is what makes prompting professional rather than disposable.
| Version | What Changed? | Why? | What Improved? | What Still Failed? |
|---|
| v0 Casual | Example: Short note | Example: Keeps the choice defensible | Example: Short note | Example: Short note |
| v1 Directive | Example: Short note | Example: Keeps the choice defensible | Example: Short note | Example: Short note |
| v2 Critiqued | Example: Short note | Example: Keeps the choice defensible | Example: Short note | Example: Short note |
Starter examples only; expand this in your own notes or submission document.
Submission
~15 min
- Your original casual prompt and assumption list
- Your completed CRAFT brief
- Your Version 2 directive prompt
- Your critique prompt and LLM critique
- Your version log
- A 300-500 word reflection explaining how your prompt became more reusable, traceable, and professional
How to Do Well A friendly self-check: aim to demonstrate clear CRAFT thinking, meaningful revision, attention to audience and constraints, and evidence that the learner can diagnose prompt failure rather than blaming the model first. Think of this as the success path while you work, not a gotcha at the end.
Psst! Before You Turn This In...
~3 minCan you tell Ian thinks this is really, really important yet? Good. Keep going.
Do the tiny-but-mighty judgment check. If these answers are fuzzy, revise the map idea before polishing the prompt.
- Purpose: Can you say what this map helps someone understand or decide?
- Evidence: Can you point to the data, source, or context behind the claim?
- Omission: Did you leave out, downplay, or defer anything that would distract or mislead?
- Risk: What could be overstated, exposed, or misunderstood?
- Human check: What must be verified outside the LLM?
- Stop rule: What would make you redesign, withhold, or simplify this map?
Ian's recurring refrain Prompt frameworks are scaffolds. The LLM can suggest, critique, and surprise you; you still own the cartographic judgment.
Reference
For use with Prompt Cartography: Interactive Web Map Design with LLMs, CRC Press. www.promptcartography.com