Instructor Notes
Timing and Schedule
This lesson is designed for a single 60-minute slot,
taught after the collaborative-git section (so learners already know
branches, commits, and @-mentions).
| Time | Episode | Notes |
|---|---|---|
| ~8 min | Introduction | Benefits table, the StarSort scenario setup, and the first scavenger hunt on spack. |
| ~22 min | Basic Issue Tracking | The new “What makes a good issue?” content + the roleplay bug report. The heart of the lesson — protect this. |
| ~14 min | Labelling Issues | Triage roleplay + the filtering scavenger hunt. |
| ~16 min | Issue Templates | Custom template + external contact link, reskinned into the scenario. |
Total ~60 min.
The StarSort scenario
To make it more fun and interactive, the lesson is wrapped in scenario: learners are new contributors to StarSort, a fictional open-source tool that sorts/catalogs telescope images. All the hands-on work still happens in each learner’s own practice repository — StarSort is just the framing that makes it feel like real teamwork.
The story threads through the episodes:
- Basic Issues — file a good bug report for a StarSort crash; loop in a “maintainer” (you).
- Labels — tidy StarSort’s label set; triage a new feature idea.
- Templates — the maintainers request a Design Discussion template and a community link.
Lean into it: play the “maintainer” when learners
@-mention you, react to their bug reports, etc. The more
you ham it up, the better it lands.
Gamification (keep it light)
Two optional bits of friendly competition. They’re fun but skippable — don’t let them eat the core exercises:
- Scavenger hunts (Intro + Labels): first to call out the correct spack numbers / fastest to build the filter wins (brownie points)
- Best Bug Report (Basic Issues): spotlight 2–3 of the clearest reports and ask the room why they’re easy to act on
GenAI thread
GenAI is woven through rather than tacked on:
- Introduction — seeds “AI drafts, you review.”
- Basic Issues — turn messy notes into a structured bug report (then verify).
- Labels — LLM as a triage/label-suggestion assistant.
- Templates — generate a template from a prose description, then trim.
Reinforce the through-line: AI drafts; you own the review.