How ossu/data-science works
Project type and purpose
• This is a static documentation/curriculum guide repository for the Open Source Society University Data Science track.
• It’s not a typical full-stack application, but rather a content-centric project that publishes Markdown files (course pages, extras, README) as a browsable website via GitHub Pages (CNAME file present), and supports community contributions via GitHub Issues and Pull Requests.
File structure
• Top-level directories:
• .github/ – CI/workflow definitions (delete-empty-issues.yml), issue templates.
• coursepages/ – per-course Markdown subfolders (intro-cs, intro-programming).
• extras/ – supplemental resources (books.md, courses.md, etc.).
• Root files:
• README.md – project overview, curriculum.
• LICENSE.md – license.
• CNAME – custom domain for GitHub Pages.
• No server or compiled code; content is purely Markdown.
Extract insights from the README
• Purpose: guide for free self-taught Data Science education.
• Workflow: fork → tick checkboxes → use Issues/PRs for contributions.
• Deployment: likely via GitHub Pages (served from master/gh-pages branch).
Main components
a. Content Repository (GitHub)
• Stores all Markdown files, workflows, license, domain config.
b. Continuous Integration / Automation
• GitHub Actions flow: e.g. delete-empty-issues.yml to auto-clean stale issues.
c. Hosting Layer
• GitHub Pages (Jekyll or default static site) + custom domain (CNAME).
d. Users
• Students/Readers: browse curriculum site, view course pages.
• Contributors/Maintainers: fork, open issues, submit PRs, merge changes.
e. GitHub Issues & Pull Requests
• Feedback loop and content updates.
Relationships & interactions
• Students/Readers → HTTP GET → Hosted site (GitHub Pages) ← GitHub Repo
• Contributors ↔ GitHub Repo via fork, pull requests, issues
• PR merge → triggers CI workflows → automatic housekeeping (deleting empty issues)
• Merge → GitHub Pages rebuild & deploy → updated static site
Architectural patterns/design principles
• Content-as-Code: all curriculum content lives in version control.
• GitHub Flow: fork → PR → CI checks → merge → deploy.
• Static Site Delivery: low-overhead hosting, fast page loads.
• Modular Content: each course in its own folder; extras centralized.
Relevant technologies/frameworks/libraries
• GitHub, GitHub Actions, GitHub Pages (Jekyll under the hood)
• Markdown for content authoring
• Custom domain (CNAME)
Detail
• Workflow triggers: “on PR merge → rebuild”.
• Custom domain resolution: CNAME → GitHub Pages endpoint.
• Automatic issue-cleanup action.