Prompt Cartography: Interactive Map Design with LLMs is a practical, research-informed guide to directing large language models and AI agents to design modern interactive maps across web, mobile, and emerging natural-language GIS platforms. It highlights how cartographic principles such as design, ethics, and spatial communication remain vital despite technological shifts and emphasizes the importance of understanding coding and data literacy to responsibly integrate AI tools like LLMs into workflows. It reframes cartographers as media directors who articulate intent, ethics, and design decisions in structured natural language that LLMs then execute, critique, and iterate at scale. Features Demonstrates how to design complete prompt-to-map pipelines for web cartography, from semantic data discovery and cleaning through visual form, symbology, interactivity, and publication decisions. Connects classic cartographic principles like visual hierarchy, visual literacy, thematic map types, color, and typography rules directly to prompt structures and reusable natural-language patterns. Shows how to manage LLM defaults, guard against hallucinations, and encode ethical constraints around provenance, privacy, bias, and civic accountability directly into prompts and agent workflows. Explains techniques for orchestrating multiple LLM agents, such as data finders, critics, visualizers, and publishers, into robust map design pipelines that make expert judgment visible instead of implicit. Provides end-to-end seminar-style case studies, from exploratory analysis to civic-facing web maps, that expose real missteps, revisions, and decision points in prompt cartography practice. Includes conceptual exercises, applied labs, and companion website materials with evolving prompt templates, schemas, and implementation examples for a variety of web mapping platforms. Acts as a stand-alone text, but it can also serve as a bridge for readers of Web Cartography to bring their existing cartogr
...
閱讀全文
