The grand opening
This slide wasn't in the workbook — "that's why I ask you to write this down." For Friday, the opening gets maybe three of our fifteen minutes, and Martin calls a good one the grand opening — confetti, ribbon-cutting, the whole image. The reason it matters is contagion: how you enter is how the session enters. "If you start very low energy, very nervous — your whole class is very nervous." The trainer's state in the first minute becomes the room's state for the rest of the hour. The five moves spell GLOSS — grab attention, link to prior knowledge, outcomes, structure, stimulate — a sequence that comes out of the Australian vocational-training tradition, where it has been the standard anatomy of a session introduction for years. "When you use this structure on Friday, you have already a very good three minutes opening. You just fill in the points and how much time you want to spend." A liturgy for the hardest three minutes, so nerves have somewhere to stand.
Five moves, walked through
Grab attention is anything that snaps the room's focus: a powerful picture — teach fire safety and open with a wall of flame, so people know the topic before you say it — a slogan, a motto, a short activity. Singing, Martin conceded, works too, "but if I sing, you all run." His workhorse is the show of hands he calls right hand, left hand, and the third is a joke: "Who of you wants to learn something new? Who of you wants to learn something different? And the rest of you — why are you actually here?" Two questions that get hands moving, one tease that gets a laugh, and the room is his. Link to prior knowledge means docking the topic to something already in their heads. Teaching robotics: "Have you ever seen a KUKA robot?" — everyone has, and nobody needs to know which microprocessor it uses; the components they can name are enough of a foundation. In a multi-day programme the link can simply be continuity: yesterday we covered this, today we build on it. Outcomes has a hard rule attached: the learning outcome must be written on a slide and read out, so everyone knows what to expect. Reaching it by minute two is good timing. Structure means telling the room the shape of what's coming — "today I will use lecture and role play" — or keeping a little mystery: lecture and some other activities. Adults want to know what to expect; Day 2's andragogy, operationalised. And stimulate is the energy push that launches the body of the session: an inspiring image, a strong quote, or pure momentum — "we want to climb Mount Kinabalu — let's go!"
OFF — taking the makeup off
The closing model is OFF, and Martin's mnemonic for the pair was pure cabaret: GLOSS is putting the makeup on; at the end of the session you need to take it off. Same face, reverse order. O is outcomes — review the learning outcomes, and this is structurally where Friday's quiz lives, because the quiz is the review: run it, confirm the outcome landed, congratulate the room. First F is feedback: the Q&A session, invited properly — and in that order, quiz first, then questions, because the quiz is assessment and the Q&A is conversation. The second F is future, and it comes with a prohibition: never close a training with "any questions? No? Thank you very much." "That is not very nice." Instead, end pointing forward — one sentence is enough. Teach fire extinguishers and the future is: next session, the fire hose, for bigger fires. The session ends, the subject continues, and the participants leave mid-story rather than at a dead end. Seeing the two models side by side, the symmetry is the design: outcomes announced in the opening, outcomes reviewed in the close; curiosity opened by the grab, curiosity re-opened by the future. The session is a bracket pair.
Stage-managing the Q&A
Then the most tactically honest advice of the day. On Friday, part of the score is how well we handle questions — but our classmates may know nothing about our topics, and nervous classmates are worse than ignorant ones: "Some of you are nervous, your mind is still on your own presentation, so people cannot answer your question — and people cannot ask." The fix: prepare sample questions and plant them. "You find two people, you give them your two questions: later, you ask me this." Not cheating — staging. A Q&A that exists is worth more than a Q&A that was theoretically possible. The war story made the case better than the rule. At a session for MCMC, most of the room was administrative staff — and one technical officer presented on frequency spectrum. Martin loved it: beautiful animations, genuinely excited. Then came "any questions?" and the entire room turned, in silence, to look at him. Nobody else in the building could ask a spectrum question. In a normal training the audience self-selects — people at a KUKA workshop know robots, people at a 3D-printing course have read about 3D printing. Friday's room is the exception: an audience assembled by administrative accident, which is exactly why the questions need to be catered like the lunch. My margin note: the Q&A is part of the session design, not an act of God. Design it.
Future as a business model
The F of future scales up from a sentence into an industry. In real training, the close is where you promote the next programme: "After the training finishes, we just promote our next training. Next station, next station." Smart training companies, he grinned, always structure offerings as beginner, intermediate, advanced — "so we charge you three times. Just kidding." Nobody in the room believed the just kidding, which was the joke. His own version at EP is the training-to-certification ladder: finish the CATIA course, and the close points at the certification exam. Closings also scale with the relationship. A one-day programme gets a compact OFF and everyone goes home. But a five-day cohort — like ours — has lived something together, so Friday will carry a longer closing, a small ceremony. Team-building events, kids' outdoor camps, long weekends: the deeper the shared experience, the more deliberately it needs to be ended. A course, it turns out, has an offboarding problem exactly like a hotel stay does — the last impression is disproportionately the impression.
Case Study
GLOSS-ing the DJC AI course
I ran tonight's audit on my own opening. The DJC AI course currently begins with what I would charitably call housekeeping: welcome, Wi-Fi password, agenda, my bio. Against GLOSS, that is zero for five — no grab, no link, outcomes buried in slide six, structure disguised as an agenda nobody reads, and the only stimulation is the coffee. The redesign, on today's template: Grab — open with a live sixty-second demonstration, one messy real email transformed into a clean reply by one good prompt, on screen, no commentary until it lands. Link — right hand, left hand, joke, stolen shamelessly: who here has tried ChatGPT? Who has used it for actual work this week? And the rest of you — you're about to. Their WhatsApp-forwarded AI anecdotes are prior knowledge; dock to it. Outcomes — on a slide, read aloud, reached by minute two: by lunch you will rewrite one of your own real work tasks with AI, measurably faster. Structure — honest shape: three hands-on labs, short demos between, no lecture longer than ten minutes. Stimulate — the line I already use but have been saving for the end, moved to the front: today is about you keeping the judgement and delegating the typing. The close gets the OFF treatment too, and the audit stung more there. My current ending is a feedback QR and thanks — feedback without outcomes-review, future without a next station. Fixed: the participants' own before/after task from lab three becomes the outcomes review; the feedback form stays; and the future is finally honest about being a business — the advanced workshop and the in-house version get their one sentence each. Beginner, intermediate, advanced. I laughed at Martin's three-times joke this afternoon and drew a three-step pricing ladder for my own course by dinner. The speed of that conversion is either embarrassing or educational; I have decided it is educational.
Reflections
Openings and closings are where nerves live, and both models are really anti-nerve technology: a fixed liturgy for the minutes when you are most likely to be shaking, so improvisation is only required in the middle, where you are warmed up and among friends. The Day 1 ice-breaker post was about engineering the room's energy; GLOSS is the same engineering, compressed to three minutes and made repeatable. The deeper symmetry is with the forgetting curve from Day 1: primacy and recency mean the opening and closing are the parts a participant will actually remember, and they are also the parts most trainers wing. Structure the two ends, and the middle gets remembered by association. On Friday, I get exactly three minutes of GLOSS and three of OFF around nine minutes of substance — and after today, those six minutes are the ones I am least worried about. That is what a good framework buys: worry, relocated to where it is useful.