10 Before the day
Most of what makes a hands-on workshop go smoothly happens before anyone walks in. This chapter is the pre-flight: the timeline you work to, the survey that tells you who is ready, the dry-run that proves the material, and the checks for the morning itself. It pulls together tasks introduced earlier (setup in Part 2, the iteration org and website in Part 3) into the order you actually do them.
10.1 The pre-course timeline
The setup tasks have to land in a sequence that gives learners time to finish the pre-work and gives you time to follow up with anyone who has not. Work backwards from the workshop date so that every due date still leaves a buffer of working days before the day itself. A workable shape, counting back from the workshop:
| When (before the workshop) | Do this |
|---|---|
| ~3 weeks | Generate the cohort website, fill in the survey link and due dates, and publish it (Chapter 9). |
| ~3 weeks | Send the first email: the course website link and a clear due date for completing the pre-work. |
| 2 days before the pre-work due date | Review who has and has not completed the pre-course survey. |
| Pre-work due date | Reach out to explain the next steps (accept the GitHub invitation, clone the setup repository). |
| ~1 week | Give a second due date to finish setup, leaving room before the day. |
| Second due date | Check who has not completed setup. |
| 1 day after the second due date | Reach out to anyone still stuck to offer support; this should still leave at least 2 working days of buffer before the workshop. |
| ~2 days | Run the pre-flight dry-run (below). |
Adjust the exact spacing to your cohort, but keep the principle: every due date is followed by a follow-up window, and the last follow-up still sits clear of the workshop day. The sections below expand the key milestones on that timeline.
10.2 Around three weeks before: launch and invite
10.3 The pre-course survey
The survey does double duty: it tells you who your learners are, and it surfaces who has not finished the setup. The reference run used Google Forms: the lowest friction for learners, and a straight export to the table you feed to the ghclass invitation scripts (Chapter 8). If you would rather keep the data fully under your own control, KoboToolbox (https://www.kobotoolbox.org) is the open-source alternative; the questions are the same either way.
At a minimum, ask for:
Build the survey once and clone it for each cohort. Keep the GitHub-username field as a required short-text question, since it is the one answer the ghclass scripts depend on.
10.4 At each due date: check and follow up
The timeline puts a follow-up window after every due date. Use it to close the gap before the day, not on the day:
10.5 Around two days before: dry-run as the test-learner
This is the highest-value pre-flight step, and it sits near the end of the timeline so any fix you find still lands before the day. Using your test-learner account (Chapter 5), walk the entire workshop as a student would. The example account used throughout this handbook is the test learner rainbow-train (https://github.com/rainbow-train), the test profile of the course author; use your own equivalent. Walk through:
If you can complete the whole workshop from a clean learner account, the material is ready.
10.6 The morning of
10.7 A short triage plan for setup problems
Some learners will still arrive with broken setups. Decide in advance how you will handle it so it does not derail the room:
The real fix for setup problems is before the day, not during it. The timeline above, with its follow-up windows, exists so that the morning-of triage is a safety net for a handful of edge cases, not a plan for the whole room.