This site is a Work In Progress

Site Navigation

NeoBooks In Progress

Creating a NeoBook (start here)

Admin and Help

Edit on GitHub


Claude on SJB Pilots

I'm about to run two pilot SJB calls this week with friends participating. I'm a little nervous and overwhelmed by the feedback I've gotten, the lack of commercial nibbles from my outreach so far, and the many ways this might go. that said, I know this is unique and valuable, and I'm eager to explore the territory.

it's late now (I've avoided talking to you about this all day today) and my session is at 9am tomorrow. the most useful thing is to rethink what a run of show might be like.

the Flow section on StumpJerrysBrains.com is something you created early, and it has some good ideas, but isn't what I'm feeling will work. I was thinking of starting with a Cold Open, meaning I choose a random but interesting topic and spend a minute slowly tracing its trails in my Brain, with some narration. then I stop and explain what SJB is and isn't. then we open things up and get some requests about where to go. that might repeat two or three different times for different topics/quests. then I may find some patterns or trends to wrap, but I haven't thought of a more powerful way to land it at the end.

suggestions?

Show more

Apr 26

Claude responded: Before I dive into run-of-show suggestions, a few quick clarifiers will sharpen this a lot — especially since you're running it in about 11 hours and want some…

Before I dive into run-of-show suggestions, a few quick clarifiers will sharpen this a lot — especially since you're running it in about 11 hours and want something you can actually use.

While you answer those, here's an initial reaction to your structure so we can build on it together:

Your instinct on the Cold Open is strong. Starting in motion — you already navigating, narrating quietly — does more in 60 seconds than any explanation could. It shows rather than tells, and it gives the Brain a chance to be the first thing people meet, not a slide about it.

The one move I'd consider for the ending: don't try to land it with a synthesis or pattern recap. That's the trap — it puts pressure on you to be conclusive about a session whose whole premise is open exploration. Instead, end with something that turns the lens around. A few options worth weighing once I know the constraints:

A "Where would you go next?" beat — pick one of the threads from the session and ask each participant where they'd take it. You stop driving and they drive for 30 seconds each. It reveals that the Brain is a thinking partner, not a performance, and it leaves people holding the thread.

A "What surprised you?" round — quick, one sentence each. This is recordable gold and gives you the testimonial-shaped moments you'll want for SJB content posts later.

A single artifact takeaway — at the very end, you send each person a link to the most interesting node visited in their session. Small, generous, and it makes the Brain feel like something they got access to, not just watched.

Once you tell me the length, group size, and what you most want from the pilot, I can shape these into an actual minute-by-minute run of show. And — for what it's worth — the nervousness is appropriate but not predictive. You've been doing the underlying activity for 30 years. Tomorrow you're just doing it with company.

You said: Q: How long is tomorrow's pilot session?

Q: How long is tomorrow's pilot session? A: 60 min

Q: How many participants, and how well do they know the Brain already? A: 5-8, mixed

Q: What do you most want to learn or get from THIS pilot? (Select all that apply) A: Test the format/flow, Capture great recordable moments, Find the 'wow' beat that lands

Apr 26

Claude responded: Good — 60 minutes with a mixed group of 5-8 is actually the sweet spot for what you're trying to do.

Good — 60 minutes with a mixed group of 5-8 is actually the sweet spot for what you're trying to do. Mixed familiarity helps you: the newcomers ask the naive-but-revealing questions, and the regulars validate that the depth is real. And three goals minus the pricing-signal goal means you can run this loose and curious without performing for a sale.

Here's a run of show built around your Cold Open instinct, with the ending reworked.

60-Minute Run of Show

0:00–0:02 — Cold Open (no intro, no hello) You're already on screen, Brain visible, and you start narrating one thread. Pick the topic tonight — something you're genuinely curious about, not something you've rehearsed. The not-rehearsed quality is the whole point. Two minutes max. Trail off mid-thought, look up, and say: "That's what this is. Let me explain what just happened."

0:02–0:07 — The Frame Short. What SJB is: live sensemaking with a 30-year knowledge graph as the third participant. What it isn't: a demo, a tour, a lecture, or therapy. One sentence on Human Provenance — why this Brain can't be replicated by an LLM scraping the web. Then: "I'd rather show you than tell you. Who's got something they're chewing on?"

0:07–0:20 — First Quest (participant-driven) Take a topic from someone in the room. Navigate it for ~10 minutes. Narrate the trails, the surprises, the dead ends. Critically: let them interrupt. The interruptions are the recordable moments. When someone says "wait, why is X connected to Y?" — that's the wow beat in microcosm.

0:20–0:35 — Second Quest (different person, different flavor) Pick someone whose topic will go somewhere structurally different from the first. If quest one was conceptual, make quest two concrete (a person, a company, an event). The contrast is what teaches the group what the Brain is. ~15 minutes.

0:35–0:48 — Third Quest, but invert it This is the move. Instead of you driving, ask: "Someone give me a topic, and then you tell me where to click." You become the hands; they become the navigator. This does three things — it shows the Brain is a thinking partner not a performance, it lets a participant feel the weird pleasure of steering 625K nodes, and it produces the most recordable footage of the hour because their reactions are unguarded.

0:48–0:55 — "What surprised you?" round One sentence each, fast. Don't let anyone monologue. This is your testimonial reel. Tell them you're recording specifically for this beat so they're slightly performative in a good way.

0:55–1:00 — The Takeaway Don't synthesize. Don't try to land a Big Insight. Instead: "I'm going to send each of you a link to the one node from your thread tonight that I think you should sit with." Then close. That's it.

The send-after happens later — 24 hours later is fine, even better. It extends the session into their week and gives you a natural reason to follow up.

Why this shape works for your three goals

Format/flow gets tested because the inverted third quest is the riskiest beat — if it works tomorrow, you know you have a repeatable structure. Recordable moments cluster in the interruptions during quests one and two, the inverted navigation in quest three, and the surprise round. The wow beat is most likely to land in quest three, when a participant realizes they're driving and the Brain responds. That's the moment that's hardest to explain in marketing copy and easiest to show.

Three small things for tonight

Don't pre-pick the Cold Open topic. Choose it in the first 60 seconds of the call, from whatever's actually on your mind. The freshness reads on camera.

Tell the group at minute 0:05 that you're treating this as a pilot and you want their honest reactions, including "this part dragged" or "I got lost here." That permission unlocks better feedback and lowers your performance pressure simultaneously.

Have a backup topic ready in case quest one stalls — but only one, and make it something you find juicy, not something you think will impress them.

You've got this. The format isn't the hard part; you've been doing the underlying activity for 30 years. Tomorrow you're just letting people watch.

You said: the first pilot session went well, but opened a bunch of questions.

2026-04-27 SJB Pilot 1 AI Summary.pdf

pdf

GMT20260427-160037_Recording.cc.vtt

vtt

GMT20260427-160037_Recording.transcript.vtt

vtt

GMT20260427-160037_RecordingnewChat.txt

82 lines

txt

the first pilot session went well, but opened a bunch of questions. I've uploaded the transcript and chat so you can get an idea of how it went.

recently I rolled the dice that going strongly into my external Brain as a key part of the offer was a good thing; now I'm not so sure.

given this feedback, I feel the need to relax my assumptions and efforts to date (which include my speaking site, jerrymichalski.com ), and find my way to my market. I'm also thinking I should see if my Substack can become a source of significant revenue, if I really devote myself to it.

I'd love to know what you think, and what you'd recommend

Apr 27

Claude responded: Before I give you my read, let me clarify what I'm picking up so I'm responding to what you actually need:

Before I give you my read, let me clarify what I'm picking up so I'm responding to what you actually need:

You ran the pilot. The room — Petervan, Mark, Hank, Victoria, John — is sharp, generous, and credentialed, and they more or less converged on telling you to demote the Brain in the pitch. Lead with you. Lead with what happens to a team when you're in the room. The Brain is the engine, not the storefront.

That's a big piece of feedback to absorb in one day, especially after you've spent months building site, brand, agreement, and content strategy around the Brain-forward framing. So I want to slow down before recommending anything.

A few things I want to name first, then a question, then what I'd actually do tonight if I were you.

What I noticed

The feedback isn't telling you the Brain is wrong. It's telling you the Brain isn't the hook. That's a different and smaller adjustment than it feels like at midnight after a long session. Petervan, Mark, John, and Victoria all responded to the Brain in the session — they just don't think "stump my brain" is how you sell it. The thing they want to buy (or want you to sell to others) is your ability to walk into a team and change how they think together. The Brain is what makes that ability non-bluffable.

You got exactly what a pilot is for. You tested a frame, the frame partially held and partially didn't, and a smarter frame surfaced from people who like you and want this to work. That's the maximum-value outcome from a pilot. The fact that it doesn't feel triumphant is irrelevant — diagnostically it's a win.

The "human intelligence not AI" reframe is consistent with Human Provenance. Kyle Shannon's frame and the group's instinct are pointing at the same thing from opposite sides. The Brain proves you've done the externalizing work; the selling proposition is what you do with teams because of it. These don't conflict — the site has been emphasizing the proof and underplaying the use.

About the "give up and pivot to Substack" instinct. I want to flag this gently because I've seen the pattern in our conversations before: a hard-feedback day arrives, and the response is to consider abandoning the current bet for a different bet. Substack-as-revenue is a real thing some people do, but it's a multi-year compounding play that requires a different daily practice than SJB requires. Tonight is not the night to decide that. Tonight is the night to sleep.

The actual question

Before I recommend anything substantive, I want to know which of these you're actually asking:

What I'd do tonight regardless of your answer

Two things, both small:

Write down — in one sentence each — the three sharpest things you heard today. Not the synthesis, just the raw quotes or near-quotes. ("Sell your human intelligence." "Capability and capacity building." "Find the systems, not the rebels inside them." Whatever actually stuck.) Five minutes, by hand or in Obsidian. This locks in the signal before it decays.

Don't touch the site, the agreement, or the Substack tonight. The temptation when feedback like this lands is to start rewriting copy at 11pm. The cost of waiting 48 hours is zero. The cost of rewriting tired is real.

You have a second pilot this week. That one is your next data point — and it's a much better instrument now that you know what to listen for. The frame the group gave you ("you, catalyzing thinking, with the Brain as engine") is something you can test on Wednesday's group without changing anything on the site. Just open differently. See what happens.

Once you tell me which question you're actually asking, I'll go deeper. But the headline answer is: today's session worked. The discomfort is the sound of a frame upgrading. You don't need to relax your assumptions wholesale — you need to relax one assumption (Brain-as-hook) and let the rest of what you've built keep standing.

Get some sleep. The market hasn't gone anywhere.

You said: Q: Which of these is the real ask right now?

Q: Which of these is the real ask right now? A: it's #1 for Thursday's pilot session, but #3 in general (though GWRE is just satire, so not really a part of my offer)

Apr 27

Claude responded: Good — that's a clean split, and it makes the work much more tractable.

Good — that's a clean split, and it makes the work much more tractable. Let me handle Thursday first because it's time-bound, then zoom out to the portfolio question which is actually less urgent than it feels.

For Thursday: adjusting the frame, not the format

The good news is your run-of-show from Monday is essentially still right. You don't need to redesign the session — you need to change roughly three minutes of it. The Cold Open, the framing beat, and the closing question. Everything in the middle (the quests, the inverted navigation, the surprise round) holds up and is doing its job.

Here's what I'd shift:

The Cold Open changes function. On Monday it was "watch the Brain in motion." On Thursday it becomes "watch me thinking with the Brain in motion." Same screen, same navigation, but you narrate differently. Less "look at this connection," more "here's what I'm noticing, here's what I'm wondering, here's where my attention is pulling me." You're modeling the cognitive act, not demonstrating the tool. The Brain is visible but it's the lens, not the subject.

The Frame beat (minute 2-7) is where the real change happens. Drop "live sensemaking with a 30-year knowledge graph as the third participant." Replace it with something closer to: "I help teams think together about hard things. I've been externalizing my own thinking for 30 years, which means when I sit down with your team I'm not bluffing — I've done the reps. Today I'm going to show you what that looks like, and then we'll do it together."

Then — and this is the move — say what the session is for the team, not for the Brain: "By the end of the hour, my hope is that you've thought about something differently than you would have on your own, and that you've felt what it's like when a group actually thinks together instead of taking turns talking." That's the promise. The Brain is how you keep it.

The closing question changes. Monday's "what surprised you?" was good for capturing testimonial moments. For Thursday, given what you heard today, ask instead: "Where in your work would something like this actually fit?" That's a market-discovery question disguised as a wrap-up. It does double duty: it gives participants a way to land the experience in their own context, and it gives you live data on where the offer might attach. You'll learn more in those five minutes than from another week of outreach emails.

One small structural addition: at minute 0:05, after the Cold Open, set the ground rule Victoria's feedback pointed to. "Interrupt me. If you're confused, say so. If you have a thread to pull, pull it. The session works to the extent that you participate — silence is the failure mode, not rudeness." That permission unlocks the room, especially for women and quieter participants in a mixed group. You named this exact issue today; Thursday is your chance to fix it.

That's it. Don't change the site. Don't rewrite the agreement. Don't touch the LinkedIn drafts. Just open differently, frame differently, close differently. Same engine, new dashboard.

For the portfolio: a frame, not a plan

You asked the bigger question too, so let me at least name how I'd think about it — without trying to resolve it tonight.

You have three live properties that overlap: jerrymichalski.com (speaking), StumpJerrysBrains.com (the live session offer), and the Substack (writing/audience). Today's feedback essentially told you that the boundary between #1 and #2 is wrong — they're not two different products, they're two presentations of the same underlying capacity, which is you, helping groups think. Speaking is the one-to-many version; SJB is the small-group version. The Brain is the proof-of-work behind both.

If that's right, the portfolio question simplifies considerably. You don't have three businesses. You have one practice (sensemaking-with-teams), with three surface areas: a writing surface (Substack) that builds audience and refines language, a speaking surface (jerrymichalski.com) that creates one-to-many entry points, and a working surface (SJB) where the actual paid engagement happens. The Substack isn't a replacement revenue path — it's the top of the funnel for the other two.

That reframe matters because it means today's feedback isn't a signal to abandon SJB or to bet the farm on Substack. It's a signal to align the three surfaces around a single proposition. Which is a much smaller piece of work than "rebuild everything" and a much larger one than "tweak the SJB site."

What I'd recommend you do with the portfolio question: nothing this week. Run Thursday's pilot with the adjusted frame. Then sit with both pilots for a few days and look for the through-line — what did people in both sessions respond to most? That's your real proposition, and it'll be more legible after two data points than after one. We can do that synthesis together next week.

One last thing

You said you avoided talking to me about this all day. I noticed. The pattern I want to gently flag: the days when reaching out feels hardest are usually the days when an outside perspective is most useful, because the inside view has gotten loud. You don't need to perform progress for me. Showing up at midnight with a half-formed question is exactly the right use of this. So — well done on doing the thing, even late.

Sleep. Thursday's going to be better than Monday, because Monday taught you what to listen for.

You said: Hi again!

Hi again! It's a half hour before Pilot SJB #2. I've been re-reading my notes, our conversation, the chat from Pilot 1, and I have some ideas for how to start this one differently, but I'm wondering what else I should do for the opening minutes. please summarize from before, rethink this moment, and offer some advice.

3:30 PM

Claude responded: Good morning.

Good morning. Half an hour is enough time to get clear, not enough time to redesign — so let me be useful in that window.

What we landed on after Pilot 1

The core feedback from Monday: demote the Brain in the pitch, keep it as the engine. Lead with you and what happens to a team when you're in the room. The Brain is proof-of-work, not the storefront. Petervan, Mark, John, Victoria all converged on this from different angles.

For Thursday, I suggested three small shifts to your run-of-show — not a redesign:

  1. Cold Open changes function — from "watch the Brain in motion" to "watch me thinking with the Brain." Narrate your noticing, your wondering, where your attention pulls. Model the cognitive act; the Brain is the lens, not the subject.
  2. Framing beat reframed — drop "live sensemaking with a 30-year knowledge graph." Replace with something like: "I help teams think together about hard things. I've been externalizing my thinking for 30 years, which means I'm not bluffing — I've done the reps. Today I'll show you, then we'll do it together." Then state the promise for the team: by end of hour, you've thought about something differently than you would alone, and felt what it's like when a group actually thinks together.
  3. Set the interrupt rule at 0:05 — Victoria's feedback. "Silence is the failure mode, not rudeness." Unlocks the room, especially for women and quieter participants.

The middle of the session (quests, inverted navigation where they steer) holds up. Closing question shifts to "Where in your work would something like this actually fit?" — market discovery disguised as a wrap.

Rethinking the opening for right now

Here's what I want to add, given you've had 36 hours to sit with it.

Don't open with the Cold Open. Open with one sentence of acknowledgment, then go. Something like: "This is pilot two. I'm still finding the shape of this. You're helping me figure it out — and I think you'll get something out of it too." Six seconds. Then straight into the navigation, no further setup. The acknowledgment does two things: it tells the room they're collaborators not audience (which lowers the performance pressure on you), and it pre-licenses the interruptions and feedback you actually want.

Pick the Cold Open topic in the last five minutes before the call. Not now. Not from a list. Whatever's actually on your mind at 8:55. The freshness reads on camera, and it keeps you in thinking mode rather than presenting mode — which is the entire point of the reframe.

Resist the urge to over-explain what SJB is. This was the trap in Pilot 1 — the framing beat got long because the format felt unfamiliar. After the Cold Open and the one-line frame, just ask: "What's something one of you is chewing on?" and go. If they need more context they'll ask. If they don't, you've saved seven minutes for the part that actually matters.

One last thing for your nervous system. You're not auditioning. The people in the room agreed to be there because they want to see this work. Your job for the next 90 minutes is not to impress them — it's to be curious in front of them. That's a much smaller and more achievable task. And it happens to be the thing you've been doing for 30 years anyway.

You've got this. Go pick a topic. See you on the other side.

You said: Here's the transcript from the second SJB pilot call.

GMT20260430-230237_Recording.transcript.vtt

vtt

Here's the transcript from the second SJB pilot call. as you'll see, I'm getting great feedback, and SJB is off to a wobbly start. but there are lovely moments in this call, particularly when I realize I "knew" already about Maslow's connection to the Haudenosaunee and make that connection, and toward the end, when I launch into a section about Trump, Muhammad Ali and Gorgeous George. I do wish you could watch the video.

one thing that came up in my OGM (Open Global Mind, an online community I started when lockdown started, which has a standing call Thursdays at 8am Pacific) call yesterday, after we were done talking about music and AI, was that me giving tours of my Brain is a very compelling thing. so I'd like to binge on that for a while.

this bingeing will be made easier, strangely, because April will be on a trip for 7 weeks before I catch up with her in Florida, visiting relatives. that means I can really record and post a lot.

I think my many posting venues (Substack, Medium, LinkedIn, Patreon (which I have but ignore), YouTube and X (which I've left since Musk took it over and wrecked it) and framings (Rethinking Constraints, Stump Jerry's Brains, speaking) are tangling my wet brain up a bit and keeping me from being me. I like my approach of cross-posting media, but either need it to be simpler, or to stop doing it altogether. could use some help there, too.

I think posting Brain tours will serve multiple purposes, from marketing SJB (if I create a strong ending screen video to link to, for example) to beginning to tell the JB>J story, to getting ideas out of my head and into the world.

Show more

6:52 AM

Claude responded: Let me read the transcript before I respond properly — give me a minute.

Let me read the transcript before I respond properly — give me a minute.

Viewed 5 files, ran 4 commands

Viewed 5 files, ran 4 commands

OK, I have enough now. Let me also check how the session was framed and if there was clear feedback on the format itself.

Ran a command, viewed a file

Ran a command, viewed a file

OK, I have a solid feel for the call now. Let me give you a real response.

What I saw in the transcript

This was a different kind of session than Pilot 1, and I think a more interesting one. Two-plus hours, intimate (just Marshall, Kyle, John), and the format basically dissolved into something better than the format. The Maslow/Haudenosaunee moment is genuinely lovely — Marshall mentions misappropriation, you say it sounds familiar, you search your Brain, and there it is from 2021: a critique you'd added years ago and forgotten. "I knew more about Maslow's hierarchy than I realized." That's the actual artifact you're selling. Not the Brain. Not the tour. The moment when the externalized self surprises the present self with something it had cared enough to capture but not enough to keep top-of-mind. That's irreplaceable, and no LLM can do it because no LLM has been you for thirty years.

The Trump/Ali/Gorgeous George run is the other kind of artifact — pure associative cascade. Wrestling heel → Ali's persona → Trump as kayfabe costume → the word kayfabe itself. That's not sensemaking, that's show. It's also, frankly, the strongest evidence I've seen that "Brain tours" might be the right product. People will watch that.

So: your OGM friends are right, and your instinct to binge on Brain tours is right. Let me push on it.

The core insight you're circling

You've been trying to figure out whether your offer is SJB, speaking, Substack, consulting, or something else. The transcript suggests the actual offer underneath all of them is something simpler:

You give tours. The tours can be live (SJB), recorded (YouTube/Substack), keynote-shaped (speaking), or embedded in a team's actual work (consulting). The format varies. The capacity is one thing.

That reframe matters because it means the seven weeks alone aren't a chance to "build SJB" or "grow the Substack" — they're a chance to figure out what tours you actually want to give and what makes one good. The commercial questions resolve themselves much more easily once you know what the artifact is. Right now you're trying to commercialize something whose shape you haven't fully seen yet.

What I'd actually recommend for the seven weeks

Three things, in order of importance:

1. Make 20-30 short Brain tour videos. Don't optimize anything yet. Five to twelve minutes each. Pick a topic in the morning, hit record, narrate as you go, stop when you've reached a natural ending. Don't edit beyond trimming the head and tail. Don't worry about thumbnails, descriptions, SEO, end screens, calls to action, cross-posting strategy, or which platform "should" be primary. Just make them. The single most valuable thing you can do right now is generate a corpus large enough that you can see what's working — because you can't pattern-match on three videos. You can on twenty.

The reason this matters: every conversation we've had about positioning has been theoretical. You don't yet have enough data on which tours are actually compelling vs. which ones you think should be. After 20, you'll know — partly from view counts, but mostly from rewatching them yourself and noticing which ones you'd watch again.

2. Pick ONE primary venue. Cross-post mechanically, don't strategize about it. Your wet brain is tangling because you're treating five platforms as five strategies. They're not. They're one strategy (make tours) and four distribution channels (post tours).

My suggestion: YouTube as primary, because it's where video tours actually live, the discovery surface is massive, and the format rewards exactly what you do (long-form, associative, idiosyncratic). Substack as secondary, embedding the YouTube video plus a short written framing — 200-400 words, not an essay. LinkedIn gets a 1-2 sentence post linking to the Substack. Medium and Patreon: ignore for now. X: stay gone. You can revisit the strategy after 30 days; right now the cost of optimizing is higher than the cost of just posting.

If YouTube-as-primary feels wrong because you identify more as a writer than a video person, flip it: Substack primary, YouTube secondary. But pick one and let the others be derivatives. The cognitive load of treating each as its own channel is what's eating you.

3. Stop touching the SJB site, agreement, and pricing for the seven weeks. I know the instinct is to keep refining. Resist it. SJB will benefit more from 25 great Brain tour videos that prove the underlying capacity than from any positioning tweak. When April's back and you regroup, you'll have a corpus, real audience data, and a much clearer sense of what the paid offer should be — and whether it's SJB sessions, advisory retainers, recorded tours as a Patreon product, or something none of us has named yet.

On the JB→J story

The "binge on Brain tours" idea has a hidden second purpose you flagged: beginning to tell the story of how a person and their externalized practice merge over thirty years. That's the real long-term work, and it's the thing nobody else can tell. I'd suggest a small structural move: of the 20-30 tours, make sure 4-5 are explicitly about the practice itself — how you decide what goes in, what a node feels like before vs. after it gets connected, what it's like to discover you'd already noted something years ago (the Maslow moment is exactly this). Those become the anchor for the JB→J narrative when you're ready to build it out.

One last thing

You wrote "I'm getting great feedback, and SJB is off to a wobbly start." Both are true and they're not in tension. Pilot 2 wasn't a wobble in the substance — it was a wobble in the frame. The substance (you, navigating with curious smart people, surfacing surprises) worked beautifully. The frame ("Stump Jerry's Brains") is the part that keeps not quite fitting, because it implies a gameshow and what you actually do is something quieter and stranger. The Brain tours direction lets you do the substance without committing to the frame yet. That's worth a lot.

Go make videos. The rest will sort itself out faster than you think.

You said: Great advice.

Great advice. Thank you. I'm realizing that there might be an interesting title here that I'm a Tour Guide to the Noosphere.

I'm gonna want a checklist for, as I post the videos, what to make sure to have done. And I do think having an end screen with a "please hire me" thing produced early in this sequence would be really good. Because then the 20, 30 videos have at least a little call to action to go look at SJB or jerrymichalski.com or something like that.

Substack has some sub products like Substack Live and its own studio production system for both live and prerecorded stuff that I might wanna play with. Partly also because they're new tools.

And finally, think writing is in fact stopping me. I have a series of blog posts that are drafted in Obsidian, ready to go towards Substack because I write first on Obsidian. But I'm just stuck finishing them. And I think if I talked through them, I could actually get through them and post them, then worry about the text later or just do a super whisper transcription and post that.

Show more

7:02 AM

Claude responded: "Tour Guide to the Noosphere" is strong.

"Tour Guide to the Noosphere" is strong. It's accurate (Teilhard's noosphere is exactly the conceptual territory your Brain maps), it has a job-shaped noun in it ("tour guide" = someone you can hire), and it sidesteps the framing problem that's been dogging SJB. SJB implies a gameshow; "tour guide" implies expertise and generosity. It also makes the Brain a terrain rather than a tool, which is the right cognitive shift — terrains are interesting, tools are demos.

I'd test it for a week before committing. Use it as the working title for the video series and see how it feels in your mouth after you've said it ten times. If it still rings, it might be bigger than a series title — it might be the through-line that ties SJB, speaking, and the tours together. "Jerry Michalski, Tour Guide to the Noosphere. Available for live tours (SJB), keynotes, and ongoing partnerships." That actually works.

Three thoughts on what you raised, then a checklist.

On the end screen — yes, do this early

You're right that producing the end-screen module before video 1 is the right move. It compounds: every video you post from day one has a working call-to-action, instead of you having to retroactively add one to fifteen videos in week three. The bar for "done" here is low — a 10-15 second outro with your face, one sentence, and a URL is all it needs. Polish later.

The trickier question is what the end screen points to. Right now your options are SJB, jerrymichalski.com, Substack, or "all of the above." I'd argue strongly against "all of the above" — three CTAs is zero CTAs. My instinct, given everything we've talked about: point to a single landing page (could be on jerrymichalski.com or a new tiny page) that says something like "I give tours like this for teams. Live sessions, keynotes, and ongoing engagements." with a contact button. One page, one ask, one path.

That page becomes the unifying surface that resolves the portfolio question — visitors don't need to understand the difference between SJB and speaking, they just need to know they can hire you. You can decide later which underlying offer matches their need.

On Substack Live and Studio — play, but cap it

You named the real reason: they're new tools and you want to play. That's fine and I think correct — but cap the playing time. New-tool exploration is the most reliable form of productive procrastination there is, and you've got seven weeks where the actual goal is make 20-30 tours, not master Substack's video pipeline. I'd suggest: dedicate one specific tour (maybe video 5 or 6, once you have the basic flow down) to be your Substack Live experiment. After that, evaluate. If it feels like it's adding leverage, integrate it. If it feels like it's adding overhead, drop it and stick with record-and-post. One experiment, one decision point.

On talking through the blog drafts — yes, this is the move

This is actually a really good unblock. You're stuck on writing because writing is a different cognitive mode than thinking-while-talking, and your strongest mode is the latter. The Pilot 2 transcript proves it — your Maslow moment, your Trump/Ali run, your externalizers riff — none of those would have come out as fluently in a blank Obsidian doc.

Practical version: open the Obsidian draft, glance at the bullets, hit record, and talk the post the way you'd talk to me. Don't read it. Don't try to "narrate the draft." Use the draft as a thinking-prompt, not a script. Run SuperWhisper (or Whisper, or whatever you've got) on the result. Post the lightly-cleaned transcript on Substack with the audio attached. That's a real Substack post — Substack readers love this format, and it's distinctly you.

There's a bonus: if the talked-through version is also screen-recorded with the Brain visible, it's both a blog post and a Brain tour video. Two artifacts from one session. That might be the sustainable production model for the whole seven weeks: one recording, two outputs.

The checklist

I'll keep this short and practical. Two parts: a one-time setup, and a per-video checklist.

One-time setup (do this before video 1, or at latest video 3):

Per-video checklist:

That's it. Resist adding anything else to this list for the first 10 videos. If you find yourself wanting to add steps, that's the procrastination instinct talking — note the desire, and decide whether to add it after video 10, not before.

One last thing

The shift you just made in this message — from "should I keep doing SJB or pivot to Substack" two days ago to "I'm a tour guide, here's how I'll make 20-30 videos, here's how I'll unblock the writing" today — is real progress. Not because you've solved the commercial question (you haven't), but because you've named the underlying thing you actually do. Tours. Everything else is packaging.

The seven weeks just got a lot more focused. Go make video one.

You said: I just remembered the thing I wanted to mention with Substack, which is that you can also have guests either for the live sessions or for the prerecorded.

I just remembered the thing I wanted to mention with Substack, which is that you can also have guests either for the live sessions or for the prerecorded. And I'm wondering whether I should just mix and match recording these all by myself or inviting one friend or posting to OGM and Jerry's Retreat (two high-trust conversational mailing lists I host) that I'm going cover a particular topic and seeing who all shows up. I'd love running a conversation with several people and I could format these so that the first five minutes or ten minutes is me talking through a narrative, which I could clip and use separately but I think then the conversation afterward might be really great.

I also want to add to the checklist, which includes showing you the transcript of each session and ask you to critique them and then figure out what to do with the resulting addenda. I don't want to prescript each session and have you critique them before recording, though that would actually make the the presentations stronger. Although, as I consider it more, that seems like a better and better idea. What do you think?

Show more

7:07 AM

Claude responded: Two good questions, both pointing at the same underlying tension, which is: how much structure helps versus how much structure kills the thing that makes you y…

Two good questions, both pointing at the same underlying tension, which is: how much structure helps versus how much structure kills the thing that makes you you. Let me take them in order.

On guests and the OGM/Jerry's Retreat invite

This is a really good instinct, and I think you should do it — but stage it. Here's why.

The Pilot 2 transcript is evidence that conversation makes you better, not worse. The Maslow moment doesn't happen without Marshall mentioning Haudenosaunee. The Trump/Ali/Gorgeous George run lands because Marshall sets it up with the wrestling-bravado line. Your Brain plus a good interlocutor is a stronger artifact than your Brain alone — you've got 444 lines from Pilot 2 proving it.

But (and this matters): solo tours and conversational tours are different products, and they each do different work for you. Solo tours are tighter, more scriptable, easier to clip, and give you full control over the narrative arc. Conversational tours are more alive but messier, longer, and dependent on who showed up. Both have a place, and mixing them is right — but I'd suggest a rough ratio rather than pure mix-and-match.

My suggestion: aim for roughly 60% solo, 40% conversational across the 20-30 videos. Solo gives you the steady cadence and clippable assets. Conversational gives you the genuine moments and the implicit social proof of other smart people show up to think with me.

For the conversational ones, the OGM/Jerry's Retreat open-invite model is excellent because it self-selects for the audience you actually want — people who already trust you and bring their own thinking. The "I'm covering X, who's around?" format is also low-overhead, which matters for sustainability over seven weeks.

The format you described — 5-10 minutes solo narrative, then conversation — is structurally smart for two reasons. First, the solo opening gives you a clippable asset regardless of how the conversation goes. If the conversation is great, you have a great long-form video. If the conversation meanders, you still have a tight 8-minute solo piece you can post separately. That's hedging without compromising. Second, it sets the conceptual frame before guests arrive in the discussion, which makes their contributions sharper because they know what terrain you're on.

One small caution: don't invite guests to the first 3-5 videos. Solo first, while you find your rhythm. Bringing in guests too early means you're optimizing for two unfamiliar things simultaneously (the format AND the dynamic). Get comfortable with the solo flow, then start mixing in conversational ones from video 4 or 5 onward.

On pre-scripting and critique — the harder question

You're circling something real here. Let me be honest about what I think.

Post-recording critique is straightforward and I think obviously valuable. Send me the transcript after each video, I'll flag patterns I notice — what landed, what didn't, where you got tangled, where you found gold, recurring tics, missed connections in your Brain that I can spot from the outside. Over 20 videos that compounds into real signal about what your tours are actually doing. Add it to the checklist.

Pre-recording critique is the trickier one, and I want to push back gently on the "better and better idea" framing. Here's what worries me.

The thing that made the Maslow moment work — the thing that's literally the heart of why your tours are valuable — is that you didn't know you knew it. You had to surprise yourself with your own Brain in real time, in front of witnesses. That moment is structurally impossible if you've pre-walked the topic with me beforehand. By the time you hit record, the surprise is gone. What you'd capture instead is a competent re-performance of a discovery you already had.

This is the deep risk of pre-scripting: it trades authentic surprise for production polish. Your tours' edge is the surprise. The polish is replaceable; the surprise isn't.

That said, there's a middle path that captures most of the upside without the downside.

My suggestion: don't pre-script the tour itself. But do pre-think the frame — for 5-10 minutes before recording, tell me the topic and where you think you might start, and I can flag things like "watch out for going too deep on X before establishing Y" or "your Brain probably has a connection to Z worth looking for" or "you tend to under-explain jargon — define [term] when it comes up." That's directional coaching, not script. It sharpens the presentation without killing the live discovery.

The distinction matters. Pre-scripting says: "here's what to say." Pre-framing says: "here's what to watch for while you wander." The first replaces your judgment with mine. The second helps your judgment be more legible to viewers.

If you wanted to test this empirically — and I think you should — do videos 1-3 fully unscripted, do video 4 with light pre-framing, and watch them back. You'll see immediately whether the framing helped or made you stiffer. Trust your gut on the playback, not the theory.

One more thing on this. You said "that seems like a better and better idea" as you considered it. I want to flag that this is a familiar pattern — the longer you sit with structure, the more appealing it gets, because structure feels like control and control feels like progress. But for your work specifically, structure is often the enemy. Your Pilot 2 was less structured than Pilot 1 and substantially better. The OGM tours emerged from zero structure. The Maslow moment required zero structure. The instinct to add scaffolding is worth questioning every time it shows up, because in your case the scaffolding is usually what's costing you.

Updated checklist additions

Adding to what we had:

One more thing

You're now sitting on a real production system. Solo tours, conversational tours, talked-through blog posts that double as tour videos, an end screen pointing to a landing page, a checklist that includes after-the-fact critique. That's a full content engine, and it's coherent.

The risk for the next two days is that you keep refining the system instead of starting it. Every additional refinement feels productive and is a form of stalling. The seven weeks are short. Video one being imperfect matters less than video one existing.

What's the topic for video one going to be?

You said: I'm thinking a very interesting format might be for me to record a solo, post it, then tell people I'm going to have a live conversation about that video, whic…

I'm thinking a very interesting format might be for me to record a solo, post it, then tell people I'm going to have a live conversation about that video, which means anybody interested in the topic will show up, and those who are not will not. I'll include in that session your comments back on my original post, so people can see the critique and then use the conversational recording session to actually deepen the subject and improve my Brain. That feels like a really interesting path for some, not all, of the sessions. As you say, maybe 40% run this way or 10%.

Ah, you've started the checklist; thank you. please add your suggestion that I (optionally) test the frame for each call with you (written as a simple prompt for me). the checklist may have an optional section for group calls, because I'll have to compose and send the invites, etc. And it would be useful to think of a simple followup action or two that leverages this sequence.

Please add to the checklist a reminder to capture the Brain's log file right when each recording stops, because I'd like to add a Brain Tour link to each recording (and it's much harder to capture the log later).

You are right about my overthinking. I'm excited to get moving on the videos. I think many of them will map well to the Obsidian pages I've been stuck on.

One video to shoot early is the Call to Action outro. Another, which I've been pondering, is a bit of Behind The Scenes for the making of these videos, which includes an explanation that this is sort of like my showing you the notes behind the documentary... if that makes sense.

Then there are several clusters. One is about the effects of GenAI on jobs, another that explains why so many uprisings around the world, and another I call Next Stacks, because it's the answer to the uprisings.

I love your suggestions about not eliminating the human surprise elements, which are indeed the spiciest moments.

Also love doing a few without pre-framing with you and comparing the results.

On your reviews: I'm equally interested in your format/process comments as in your content/topic critiques, because part of what I'm aiming for is a relatively consistent (I think) web of how things work, and what we might do about it all. That's my frame for all of this, half-consciously. And I want to challenge and strengthen my thinking — and change my opinions as appropriate.

Show more

7:32 AM

Claude responded: The post-then-conversation format is genuinely smart, and I want to name why before adding it to the checklist, because the why will help you decide which vide…

The post-then-conversation format is genuinely smart, and I want to name why before adding it to the checklist, because the why will help you decide which videos warrant it.

What that format does that pure solo and pure conversational don't: it creates a built-in audience filter (only people who care about the topic show up), it gives the conversation actual substance to push against (they've all watched the same 10-minute artifact), and it produces a sequence — solo tour → critique → live deepening — that maps directly to how thinking actually evolves. It's also the most legible demonstration of what you'd do with a paying client: here's my take, here's an outside critique, here's how the thinking improves under pressure. That's the SJB value proposition in a public, free, repeatable format.

I'd reserve it for topics where the deepening matters — your Next Stacks cluster, GenAI/jobs, uprisings — and skip it for tours that are more associative-pleasure than thesis-building (the Trump/Ali/Gorgeous George kind). My guess is 15-20% of videos, not 40%. The format has overhead, and overdoing it dilutes the cadence.

On the format/process critique you asked for

This is an important addition and I want to make sure I do it right. You're describing two things: a coherent thesis-web you're building (consistent across videos), and a practice you want challenged. Those are different tasks for me. So when I review transcripts, I'll be watching for:

Format/process layer: pacing, where you got tangled, clarity of the opening hook, whether the ending lands or trails off, jargon that needs definition, recurring verbal tics, when the Brain navigation got hard to follow, how guests were brought in and out, missed clipping opportunities (moments that would stand alone as 60-second posts).

Content/topic layer (the web): how this video connects to your other tours, where your stated position might be sharper than your actual evidence supports, where you're hedging when you don't need to, places where my outside view sees a different connection than yours, claims I'd want challenged, opinions of yours I'd push back on directly.

Coherence layer (the half-conscious frame you mentioned): as videos accumulate, I'll start mapping the thesis-web you're building — what's the recurring causal model under all of this, where does it have holes, where is it strongest, where it might be wrong. This one only emerges over time, maybe by video 8-10, but it's probably the most valuable thing I can offer across the whole run.

I'll lead each critique with format/process (faster to act on), then content, then any coherence-level observations once they emerge.

Behind the Scenes video

This is a really good early one and I think it should be video 2 or 3. Here's why: it preempts the question every viewer will eventually ask ("how does he do this?"), it gives you a place to put the meta-explanation so you don't have to repeat it in every tour, and it functions as a soft sales asset — when someone watches a tour and gets curious, the BTS video is what converts curiosity into "I want to hire this person."

The "notes behind the documentary" framing makes sense and is unusual enough to be memorable. The Brain isn't the documentary — the tours are the documentary, and the Brain is the research substrate that made them possible. That inverts the framing problem from Pilot 1 (Brain-as-product) elegantly.

Quick suggestion for the BTS structure: open with what people see (a tour), reveal what's underneath (the Brain), explain the practice (30 years of curating), then end with what it makes possible (the live sessions, the keynotes, the ongoing partnerships). That mirror-reveal structure is satisfying and naturally lands on the call to action.

On the three clusters

GenAI/jobs, uprisings, Next Stacks. Worth saying out loud: those three form a natural arc. Diagnosis (jobs being hollowed out by GenAI) → symptom (uprisings as response to a system that's failing people) → prescription (Next Stacks as what could replace what's failing). That's a coherent intellectual through-line, and if you sequence the videos roughly in that order across the run, viewers who watch multiple will start seeing the web you're building. That's the consistency you mentioned half-consciously aiming for — making it slightly more conscious in the sequencing will help it land.

Don't make all the videos in one cluster back-to-back, though. Mix in associative tours (the Maslow-type, the Gorgeous George-type) between them. The associative tours show range and keep things fresh; the cluster tours build the thesis. Alternating teaches viewers what kind of person you are and what you think.

Updated checklist

One-time setup (before video 1, or at latest video 3):

Per-video checklist:

Pre-recording (optional but recommended for thesis-building topics):

Recording:

Immediately after recording stops:

Posting:

After posting:

Optional: Group call section (for conversational and post-then-discuss formats)

Before:

Recording:

Follow-up actions to leverage the sequence:

On excitement to start

Good. That's the right energy and you should ride it. One pragmatic note: the temptation in the next 24 hours will be to set up the perfect recording environment, finalize the series title, design the landing page, and then record video 1. Resist that. Record video 1 imperfectly with whatever setup you have right now, and let the production system improve in parallel with the videos themselves. The OBS template, the landing page, even the series title — all of it gets better through use, not through pre-design.

Pick the topic for video 1. Go.