A Map is a curated index - a page that tells you where to read, in what order, and why. Maps do not carry new knowledge; they organize existing Articles into a useful traversal. Some span more than one Path. Open one when you want a guided reading order through a whole subject in a single session.
Four Maps in the library right now - two broad stacks for Web and AI, a narrower cut on the developer-facing AI surface, and the engineering platform that runs underneath modern software.
The narrower cut of the AI library - just the tools that show up in a developer's day-to-day. Coding with Copilot or Claude Code, designing with Claude, generating images, plus the model and protocol underneath.
Open the map →Every AI Path article sorted by layer - from foundation models up to the developer tools - with three reading orders for developers new to AI, agent builders, and depth-first readers.
Open the map →The platform craft underneath modern software - orchestration with Aspire, observability across services, and API documentation on top of OpenAPI. Three layers, every Engineering Path article in scope.
Open the map →Every Web Path article sorted by layer - language, build tools, frameworks, API docs, orchestration - with a one-line reason and three reading orders for different readers. Pulls in API doc pieces from Engineering.
Open the map →A Map is a peer of an Article, not a parent or child. Where an Article gives you a self-contained piece of knowledge ("What is Vite?"), a Map gives you a curated path through several Articles ("here is the Web stack in 2026 - read these eight in this order, organized by layer").
Maps can cross Paths. The Web stack Map sits under the Web Path conceptually but pulls in Engineering Articles where it makes sense. That is the whole point of having Maps as a distinct content type.
For the full model, read How StackNova is organized.