Behind the Lab: From Sources to Experience
Speaker notes · Suggested session length: 15 minutes
Launch the presentation · Tutorial home
1. The tutorial is built the way it teaches
The website is not a manually maintained HTML copy of the chapters. It is an output built from the same repository that readers are learning about. Explain the architecture with concrete examples, not a tour of product logos.
Chapter 1: Document-as-Code: Information We Can Work With · Chapter 11: Behind the Lab: How We Built This Tutorial
2. We maintain information before choosing a view
The contract records are maintained as Markdown rather than copied into a dashboard. Presentation YAML contains editorial messages, not complete chapter copies. Keeping these roles separate makes it easier to review the right thing.
Chapter 4: Markdown: Readable Sources · Chapter 9: Information You Can Explore · Chapter 11: Behind the Lab: How We Built This Tutorial
3. Git records changes; GitHub supports collaboration
Distinguish a commit, a PR review, and content approval. A merge does not certify the accuracy of all published information. GitHub supplies an environment for discussion and automation around the maintained files.
Chapter 2: A Shared Workspace for Information · Chapter 3: History Is Useful Information · Chapter 10: Automating Information with GitHub Actions
4. Astro turns sources into a readable website
Important distinction: Astro and Python run during local development or a workflow build, not on the GitHub Pages server for each visitor. GitHub Pages hosts files. The browser executes bundled JavaScript where an interaction is required.
Chapter 7: From Sources to Results · Chapter 10: Automating Information with GitHub Actions · Chapter 11: Behind the Lab: How We Built This Tutorial
5. The browser can explore information without owning it
Do not suggest that the chapter labs write changes back to GitHub. Contract Explorer shows relationships, commitments and possible impacts, but is not a contract management system. Its source records stay authoritative.
Chapter 7: From Sources to Results · Chapter 9: Information You Can Explore · Chapter 11: Behind the Lab: How We Built This Tutorial
6. Presentations reuse knowledge without copying a chapter
Open the presentation collection in the browser. Discuss why an effective slide is a curated interpretation rather than an entire paragraph carried unchanged into another format.
Chapter 9: Information You Can Explore · Chapter 11: Behind the Lab: How We Built This Tutorial
7. GitHub Actions runs the pipeline; Pages serves the result
The repository also uses a scheduled hourly deployment to refresh data snapshots. Actions artifacts and GitHub Pages are separate destinations. A successful build cannot independently prove semantic correctness or content approval.
8. The principles work beyond GitHub
Emphasize equivalent goals, not identical mechanics. Copilot can use permitted documents as knowledge, but a read-only knowledge connection is not an editing tool. Agents need configured actions and permissions to modify files. OneDrive is not Git and SharePoint does not automatically give Git-style PRs or reproducible builds.
9. Build an environment around useful information
End by returning to the tutorial's core distinction: document-as-code is not primarily changing the editor or file format. The objective is working with reliable maintained information across multiple outputs and tasks.
Chapter 1: Document-as-Code: Information We Can Work With · Chapter 11: Behind the Lab: How We Built This Tutorial