5 Domain Architecture

CDI is modular by design. Each domain represents a real-world discipline—such as economics, health informatics, or education—and follows a consistent repository structure scaffolded from a shared CDI template.

5.1 🔧 Template-Based Scaffolding

Each CDI domain repository is scaffolded from a shared, reusable template designed for clarity, reproducibility, and seamless collaboration. This structure ensures that every guide—regardless of domain—follows consistent conventions and is easy to maintain, extend, and publish.

The standard scaffold includes:

  • data/ – Example datasets used in Q&A entries
  • scripts/ – Supporting R, Python, or shell scripts for preprocessing or analysis
  • images/ – Exported plots, diagrams, and workflow graphics
  • library/ – Reference files like .bib (citations) and .csl (citation styles)
  • Domain-specific .Rmd files – Structured Q&A content organized by topic
  • index-*.Rmd – Bookdown entrypoints for building HTML or PDF guides
  • Build scripts – Reproducible automation using combo-build.sh or Makefile
  • GitHub Actions – CI/CD workflows for deploying guides to GitHub Pages

When applicable, a repo may also include:

  • Snakefile – For workflow automation using Snakemake
  • Dockerfile – For containerized, reproducible environments
  • .env, requirements.txt, environment.yml – Environment declarations

5.2 📂 Example Repository Tree

cdi-machine-learning/
├── data/
│   └── titanic.csv
├── scripts/
│   └── clean_titanic.R
├── images/
│   └── ml_workflow.png
├── library/
│   ├── references.bib
│   └── apa.csl
├── 01-0-preface.Rmd
├── 01-1-machine-learning.Rmd
├── index-ml-gitbook.Rmd
├── _bookdown.yml
├── _output.yml
├── combo-build.sh
├── Snakefile              # Optional
├── Dockerfile             # Optional
└── .github/
    └── workflows/
        └── deploy.yml

5.3 🌱 Built for Growth

This structure allows each domain to:

  • Scale horizontally across fields like economics, health, or genomics
  • Scale vertically by adding deeper layers or advanced modules
  • Support reproducibility through Snakemake, Docker, and CI/CD
  • Enable collaboration via consistent layout and templated setup

Whether you’re building a guide for machine learning or microbiome analysis, the CDI scaffold ensures a smooth, professional experience from start to deployment.