Labs / Chapter 6
Chapter 6 argues that a map can work technically and still fail rhetorically, ethically, or experientially. This lab turns critique into a reproducible design dossier: students must preserve prompts, compare critique lenses, detect drift, evaluate metrics as signals, and defend a publish / revise / withhold decision.
Critique is not a complaint box at the end of production. It is a continuous method for asking whether a map's behavior, defaults, tone, evidence, and interaction logic still match its purpose. This lab asks you to critique a map concept as if another person must be able to reproduce your reasoning later.
Use a prior lab map idea or choose one of the scenarios below. You do not need a finished map; you need a plausible map concept detailed enough to critique.
Map concept under critique Topic: Audience: Purpose: Default view: Main layers: Main interaction behaviors: Most important claim: Most sensitive or risky interpretation: Evidence, source, or data limitation I would need to verify:
Ask the LLM to identify what can be technically successful while still needing critique.
Act as a studio cartography critic. Review this map concept. Do not suggest fixes yet. Separate your response into two columns: 1. What appears technically functional 2. What still needs cartographic critique Focus on purpose, audience, default state, interaction behavior, tone, attention, and claims. End with one sentence beginning: "The map works, but it may not yet be right because..." Map concept: [PASTE SETUP NOTES]
Chapter 6 frames critique through description, analysis, interpretation, and evaluation. Run one prompt that keeps those lenses separate.
Critique this map concept using four lenses. Keep them separate. Description: What does the map literally show and do? Analysis: How do data, symbols, defaults, and interactions work together? Interpretation: What does the map seem to argue, imply, or ask users to believe? Evaluation: Do those effects support the stated purpose and audience? For each lens, include one question a human reviewer should still answer.
Do not let the LLM collapse the critique into a checklist. Rewrite the most useful question from each lens in your own words.
Drift is not failure. It is what happens when iteration gradually changes what the map has become.
Audit this map concept for design drift. Use three categories: - aesthetic drift - data drift - ethical drift For each category, identify: 1. the original intent that might be drifting 2. the evidence of drift 3. who would be affected by the drift 4. whether the drift should be accepted, revised, or investigated 5. the prompt or design decision that may have introduced it Do not recommend final fixes until after the audit table.
Command prompts often hide judgment. Critique prompts expose assumptions before deciding what to change.
Convert these command prompts into critique prompts for my map concept. Each critique prompt should reveal assumptions, trade-offs, or risks before prescribing a fix. Command 1: Make the strongest pattern more visually obvious. Command 2: Add a filter so users can explore the data. Command 3: Use a standard metric threshold to decide whether the map passes. Command 4: Make the map feel more neutral. Command 5: Add a short explanation panel. For each converted critique prompt, explain what kind of judgment it forces.
Choose the best converted prompt and run it against your map concept. Save the output and your response to it.
Metrics can point toward trouble, but they do not decide what a map means. Use an LLM to create pressure, then push back.
Act as a quantitative QA reviewer. Identify 4 metrics or standards that someone might use to judge this map, such as contrast, label density, load time, touch target size, projection, aggregation threshold, or classification method. For each metric, explain: - what the metric can reveal - what it cannot reveal - how blindly following it could produce a worse map - what human judgment should decide instead
Close the critique loop with a decision that someone else could audit.
Using my critique dossier, help me draft a publishability verdict with these sections: 1. Decision: publish, revise, withhold, or test with users first 2. Strongest evidence for the decision 3. Most important unresolved concern 4. One design change I accept 5. One suggested change I reject and why 6. What prompt, model response, or human judgment supports this decision 7. What would allow another person to reproduce my critique process Keep the tone professional and accountable.
Yes, we are doing this again. That is very much the point.
Do the tiny-but-mighty judgment check. If these answers are fuzzy, revise the map idea before polishing the prompt.
For use with Prompt Cartography: Interactive Web Map Design with LLMs, CRC Press. www.promptcartography.com
Prompt Cartography