The current documentation covers data package structure, terminology, and download links, but doesn't yet help users understand how to actually use the data to answer real questions. The goal is to build out documentation across multiple formats to serve different learning styles and skill levels.
Areas of focus:
- Blog posts answering real questions: Write posts that start with a concrete question (e.x, "Which census subdivisions have the highest population growth?") and walk through the process of answering it using D4C data. These serve as both tutorials and demonstrations of what the data can do. We can use the folia spec as a way of creating full examples that combine StatCan data with other datasets for analysis/displaying of data.
- Video screencasts: Short recordings showing workflows in tools like QGIS, DuckDB, Lonboard, and the API bindings — particularly useful for visual learners and for demonstrating map-based exploration that's hard to convey in text.
- Website documentation improvements: Fill gaps in the existing docs: clearer guidance on which dataset to use for a given use case, and improved navigation between related pages.
- AI-friendly documentation: Structure docs so that AI coding agents (ex. Claude) can effectively work with the repository and its associated data. This will include clear instructions on how to build, test, and run the project, and providing complete API references that LLMs can consume to generate correct queries and code.
- Bilingual content: align with the bilingual direction of the project ensuring documentation is accessible to both English and French speaking users
The current documentation covers data package structure, terminology, and download links, but doesn't yet help users understand how to actually use the data to answer real questions. The goal is to build out documentation across multiple formats to serve different learning styles and skill levels.
Areas of focus: