Honest Take — Module 9: Storytelling & Speaking #

This module exists over the objection of one of its own parents, and you should know that going in. One earlier draft of this curriculum gave storytelling and stage-speaking two full modules and eighty-plus hours; the other omitted the stage entirely, dismissing the TED-style canon as mistaking "presentation theatre for communication." The merged edition carries the material and the objection together, and I think that's right, because both drafts saw something true. The skeptical draft was right that the theatrics genre is mostly hollow — the rehearsed vulnerability, the walk-and-pause choreography. The maximalist draft was right about the thing underneath: engineers dismiss stories as unserious, as what marketing people use to manipulate non-technical audiences, and that prejudice is wrong and costing them. A story is not a manipulation. It is the cognitive form human brains evolved to absorb — the native format for cause-and-effect over time. When you present information as bullet points, you're sending a payload the receiver must manually convert into narrative before it sticks. When you send the story, it arrives pre-formatted for the runtime. Same content, dramatically different retention. The allergy is the tell: the engineers most resistant to this material are the ones leaving the most leverage on the table.

The single most efficient resource in the module is Kurt Vonnegut's four-minute video on the shapes of stories. Watch it. He draws the arcs on a chalkboard and you'll recognize most of them in talks you've given and posts you've written — most of what you currently produce is "Man in Hole" without your knowing it. Naming the shapes lets you choose them instead of falling into them. Then do the checkpoint's conversion exercise honestly: take an existing list-shaped post — "5 things you should know about X" — and rewrite it as a story, same information, different shape, starting from a specific moment and a specific failure. The story version will outperform the list version on every metric, and you will feel slightly betrayed by the world for caring more about narrative than information. Don't read it as proof that people are shallow. People are exactly as deep as the narrative form is built for. You were the one packaging information for an audience that doesn't exist.

On the speaking half: the body is where the work begins, and most books skip it. In some surveys public speaking polls above death, and the physical response is real — cortisol, cold hands, the working-memory crash that eats your next sentence. You can have the best-structured talk on earth and bomb it because your nervous system believes it's about to be eaten. The free hour that outearns every book is Patrick Winston's MIT lecture How to Speak — watch it three times: once for content, once with a notebook for structure, once watching how Winston himself does everything he recommends, because the meta-layer is the lesson most viewers miss. After Winston, Duarte's one transferable insight: a great talk oscillates between what is and what could be, and the energy lives in the oscillation. Once you hear it, you can't unhear it.

The drill that does most of the work is record-and-watch, and here's how to survive it. You'll record five minutes, watch thirty seconds, cringe, and want to quit. Everyone does. The fix is not pushing through the cringe — it's changing the question. First watch, you ask "do I look stupid?" and the answer is always yes; everyone looks stupid on tape. Second watch, you ask "what are exactly two things I would change?" That question converts cringe into data. Pick two — the "um" rate, the trailing-off sentence endings — fix only those two on the next recording. Two-at-a-time is how the skill compounds; ten-at-a-time is how it stalls. And take the graduated path seriously: the five-minute lightning talk before the twenty-five-minute recorded one. There is no waiting list for new speakers at local meetups — organizers are starved for them. Your reluctance isn't because the slot doesn't exist. It's because it does, and you'd have to use it. The lightning talk teaches you the one thing the big talk can't: the audience does not, in fact, attack you when you finish.

One last asset you may not know you have: if your everyday register is casual and conversational, that's a speaking strength, not a deficiency to polish away. Engineers who talk to a room the way they talk to a colleague — without the conference smarm or the corporate-trainer cadence — are rare and disproportionately memorable. The mistake is treating speaking as performance. Speaking is just talking, with editing, in front of more people than usual.

Two final pieces of equipment for the road. First, Ira Glass's taste-gap segment: when you start performing a craft, your taste is years ahead of your skill, which means you can see exactly how bad your early talks are — and that gap is why most people quit in the first year of any performance discipline. The gap closes only through volume. Knowing the gap is normal is what gets you through the recordings where the gap is all you can see. Second, the Q&A: pre-write the five worst questions someone could ask after your talk and rehearse the answers out loud. The Q&A is a fuzzer — if your model of the topic is shallow, the fuzzing will find it — and the fifth-worst question will, with eerie reliability, actually get asked.

Conclusion #

Storytelling is the conversion layer between information and action; speaking is the non-substitutable channel where careers actually turn — offers, budgets, trust, the relationships that change trajectories all happen in real-time rooms. Neither is given by genetics. Both are built in reps: learn the shapes, convert one post, record weekly, climb the graduated path to the published talk. The recorded talk under your real name is the module's non-negotiable artifact, and it works for you while you sleep.

Predictions #

The Vonnegut video will be the most re-watched four minutes in this curriculum. You'll return to it several times across the year, each time seeing your own current work in one of the shapes.
The republished story-version of your old post will measurably outperform the original — likely 2–3x on time-on-page and replies. You'll find this slightly annoying, in a useful way.
You will feel fake the first few times you deliberately structure something as a story. The fakeness fades around the fourth attempt; it was unfamiliarity, not dishonesty.
The lightning talk will go better than you expect, by a lot. The 25-minute recorded talk will go worse than you expect — it's a different skill with a much higher endurance load. The gap is normal and not regression.
You'll hate watching yourself for the first four or five recordings. Around the sixth, the cringe quietly turns into curiosity. That transition is the skill compounding; it's also permanent.
Recording yourself will improve your 1:1 conversations more than your talks — you'll start hearing tics in everyday speech you've never noticed. This is the module's surprise bonus.
A year from now, someone will quote a specific line from a talk you gave back to you, and you'll be quietly stunned that anyone remembered anything. They did. They do. That's the proof the medium works.