This site is a Work In Progress

Site Navigation

NeoBooks In Progress

Creating a NeoBook (start here)

Admin and Help

Edit on GitHub


The Big Fungus: A Conversation

Jerry Michalski & Claude · April 3, 2026

This document captures the first part of a long conversation that began with Humberto Maturana and ended up somewhere much larger — a working theory of what Jerry's Brain is for, what the Big Fungus could become, and how those projects relate to Jerry's life and work. The conversation then shifted into operational work on Stump Jerry's Brains. This document preserves the strategic and philosophical thread before it gets buried.

Return to this. It matters.

I. Maturana's Core Ideas

The conversation started with a briefing on Humberto Maturana (1928–2021), the Chilean biologist whose work with Francisco Varela reshaped ideas about life, cognition, and what it means to know anything.

Autopoiesis is the foundation — the idea that living systems are self-producing, self-maintaining networks. A cell, an organism, a mind: each continuously regenerates its own parts to preserve its unity. Life must change to stay the same.

The radical move: cognition isn't something that happens in a brain. It is life. Living systems are cognitive systems, and living as a process is a process of cognition.

Structural coupling is how living systems interact with their environment. The environment doesn't instruct an organism — perturbations from outside trigger changes determined by the organism's own internal structure. You don't download reality; you co-construct it through ongoing interaction. The history of those interactions shapes what you can perceive and think.

The observer is central and inescapable. Any scientist studying a system is part of the system's reality. Descriptions of nature always involve the language and perspective of the observer. There is no view from nowhere.

Bringing forth a world — Maturana's phrase for what this adds up to. Cognition is not a representation of the world "out there" but an ongoing bringing forth of a world through the process of living itself. And this world-bringing-forth happens with others, in language.

Languaging (his verb form) is not a symbol system for transmitting information. It coordinates action and attention. Language doesn't carry meaning across a gap; it shapes what becomes visible in a shared space.

Love — which sounds surprising in a biologist — is his word for the foundational condition of social existence. Without love, without acceptance of others living beside us, there is no social process and therefore no humanness. Anything that undermines the acceptance of others — from competency to the possession of truth to ideological certainty — undermines the social process because it undermines the biological process that generates it.

Key texts: The Tree of Knowledge (1987, with Varela) is the accessible entry point. Autopoiesis and Cognition (1980) is the denser foundational work. His later Biology of Love pieces are where the ethical and social implications are most explicit.

II. The Extended Mind Thesis

Andy Clark and David Chalmers (1998) asked: "Where does the mind stop and the rest of the world begin?" Their answer: cognitive processes "ain't all in the head."

The core move: objects within the environment function as part of the mind. The mind and the environment act as a "coupled system" that can be seen as a complete cognitive system of its own.

The famous test case: Otto has Alzheimer's and carries a notebook with directions. Inga recalls the same information from biological memory. Clark and Chalmers argue there is no fundamental difference between Inga and Otto, memory and notebook. If a process would count as cognitive when it happens inside the skull, it counts as cognitive when it happens outside the skull — provided the coupling is tight enough.

The criterion for "tight enough": the external resource must be reliably accessible, automatically consulted, and its outputs directly endorsed by the person.

Where EMT and Maturana agree: Both reject the "skin and skull" boundary as the definitive limit of cognition. The boundaries of mind are functional, not anatomical.

Where they diverge: Clark and Chalmers are asking an ontological question — where is the mind? Maturana is asking a biological one — what is cognition as a living process? Maturana would find the Otto/Inga framing too static. The notebook is a passive repository Otto accesses. But structural coupling is dynamic and bidirectional — the environment doesn't just store things for you, it shapes you over time, and you shape it. The relationship has a history that transforms both parties.

Another tension: Maturana is deeply skeptical of the idea that information is "in" things. For him, there's no information transferred from environment to organism — only perturbations that trigger internally-determined responses.

Key texts: Clark's Supersizing the Mind (2008) is more accessible than the original 1998 paper and engages with exactly this kind of deep coupling.

III. Jerry's Brain Through These Lenses

Jerry has been building and maintaining a public personal knowledge graph in TheBrain software for nearly three decades — over 625,000 nodes as of 2026.

Through the Clark/Chalmers lens: The Brain is straightforwardly an extended mind. It isn't a tool Jerry uses; it's part of how Jerry thinks. Querying it is more like consulting long-term memory than doing a Google search. Importantly: from the perspective of the Extended Mind Thesis, interfering with such a system raises the same ethical issues as interfering with someone's brain.

Through the Maturana lens: But the Brain goes further than Clark and Chalmers' static notebook. It has been a site of structural coupling for three decades. Jerry hasn't just stored things in it — he's been shaped by the act of curating it. The way he notices things, the distinctions he makes, the connections he's drawn to — all of this has been co-evolved with the Brain over time. It's closer to the organism-niche relationship Maturana describes than to Otto's notebook.

The synthesis for Human Provenance: The EMT explains why the Brain is cognitively significant. Maturana explains why it's irreplaceable. The Brain isn't just an external memory store — it's the crystallized record of a specific observer's 30-year history of structural coupling with ideas. An AI can query the Brain's nodes and links. It cannot replicate the observer who created them, because that observer and that Brain have been constituting each other over time.

The Human Provenance argument isn't merely biographical. It has deep theoretical grounding in both frameworks.

IV. Design from Trust

Maturana's grounding of ethics and cooperation in love (acceptance of others) is a biological argument for exactly the kind of relational design Jerry has been developing as "Design from Trust."

Anything that undermines the acceptance of others — from competency to the possession of truth — undermines the social process. That's a systems-biology argument for why control and surveillance damage the very thing they're trying to protect.

The Extended Mind Thesis adds something parallel: if our cognitive systems genuinely extend into our tools and our relationships with other people, then environments designed around surveillance and control aren't just unpleasant — they're cognitively amputating. Interactions with other people are crucial elements of your environment and cognitive process.

Both frameworks converge on the same conclusion: trust isn't just an ethical preference. It's a cognitive infrastructure requirement.

V. The Big Fungus — A Working Theory

Jerry described a larger vision he half-jokingly calls "The Big Fungus": a space where people using very different tools for thinking can compare notes, make arguments, and eventually evolve more useful ways of building society and stewarding places.

Key characteristics of the vision:

The Bannon scenario — a recurrent use case in Jerry's mind: taking notes on an interview with someone whose views differ sharply from his own, then having a structured conversation in which he navigates related issues in the Brain, steelmans the other position if needed, and leaves the whole thing more useful at the end than at the beginning. Each conversation should simplify or complexify and improve the Fungus, so that more people can come feed off it — the way fungi feed farmer ants.

What the Big Fungus actually needs: The problem that has prevented this from existing isn't technical. Tools like RDF and the semantic web tried the technical route for decades and largely failed. The problem is trust and incentive: why would someone open up their thinking to a commons? What do they get back? Jerry's answer: you get back a richer map than any one person can make alone, and you don't have to monetize it because the value is in the thinking itself, not in controlling access.

The core project underneath all of this: A method for productive encounters between different minds and maps — where "productive" means the shared territory gets better mapped as a result, and the improvement is durable and accessible to others. The Brain is the most developed instance of this method. Stump Jerry's Brain is a live demo of the method in conversation with an audience. The Bannon scenario is the method pushed into adversarial, high-stakes territory.

What's missing isn't the vision. What's missing is a name and a legible description of the method itself.

VI. The Nested Projects

These projects nest inside each other rather than sitting side by side:

On civic engagement: Jerry subscribes to Beyond Intractability (Heidi and Guy Burgess) and cares about the dysfunction of civic/political projects that are temporary and separate from real lives. The question he poses: why can't important issues like how we zone or design a street be lively, ongoing conversations easily read and adopted by other communities around the world — rather than one-time votes? The Big Fungus is one answer to this.

VII. The Three Things That Follow

1. The legibility project as method document. Not just "here's how to navigate my Brain" but "here's what I've learned about thinking in public with an external map, and what becomes possible when you do it for decades." This is the intellectual contribution that justifies everything else. It earns Jerry a seat at the Knowledge Futures table. A short interview series — with Claude and a friend — is a good way to surface it.

2. The Externalizers cohort as method community. Very few intense tool-for-thought users are Externalizers. Most are just building resources for themselves. The interesting group: people who have made their thinking public and navigable by others over a long period. A tiny, weird, valuable group. The conversation between them should be about what they've each learned from long-term externalization, not about comparing maps directly.

3. The LLM bridge as proof of concept. If Jerry's Brain, with his observer's perspective baked in as context, produces meaningfully different conversations than a generic LLM — and it would — that's a demonstration that human-curated externalizations add something that can't be replicated by training on the whole internet. TheBrain v15 now has AI deeply embedded in the UI, currently invoking ChatGPT. Claude is next on their agenda. The ideal: the Claude invoked by TheBrain is Jerry's specific Claude account, rather than a generic one trained on curating Brains.

VIII. On Continuity — Jerry's Brain After Jerry

A few years ago, some OGM community members held a Zoom call titled "Jerry's Brain after Jerry." One insight: Jerry needs a posse that knows how to extend his Brain if it is to continue expanding after his death.

The problem breaks into levels:

Technical (easiest): TheBrain's format, export capabilities, and LLMs working with structured knowledge graphs mean the data can survive and remain navigable.

Curatorial (harder): The Brain's value isn't in its nodes — it's in the distinctions behind the connections. Why did Jerry link these two things? What was he noticing when he made that connection in 1998 versus 2019? A posse needs to have absorbed enough of Jerry's observer's perspective to make new connections that are faithful to it. This is closer to an apprenticeship than a technical handoff.

Identity (deepest): At some point "Jerry's Brain after Jerry" stops being Jerry's Brain and becomes something else — a commons seeded by Jerry's Brain. That transition needs to be designed, not just allowed to happen. The mycelium outlasts any individual fruiting body. The question is whether the network has enough of its own integrity to keep growing without the original organism.

A possible path: JerrysBrainBot — an LLM trained on or given access to Jerry's Brain as context, able to adopt Jerry's approach to curation more easily and consistently than any human apprentice. Not a replacement for human continuity, but a different kind of stewardship.

IX. Key Tensions Worth Returning To

  1. Openness vs. legibility. The Brain has been public for years and usage is uneven. The move from publicly available to actively useful to others requires intentional shaping. Making choices about what to foreground and how to invite people in isn't a compromise of openness — it's completing the gesture.

  2. Ambition vs. execution. Jerry has a history of visions more ambitious than any one person should undertake, combined with difficulty estimating how long things take. The projects most likely to actually happen are the ones that can make meaningful progress in the time naturally given before something else captures attention.

  3. Revenue vs. the larger work. Stump Jerry's Brains needs to be the near-term priority because it funds everything else. But the legibility project, the Externalizers cohort, and the LLM bridge aren't separate projects — they're the same work at different scales.

  4. Pride of authorship. Jerry noted that until recently, pride of authorship was constraining him — wanting every post to be his own words, every Brain entry placed by him. He's rethought this. The shift matters: it means AI collaboration can genuinely accelerate the work rather than just assist it.

X. A Note on This Conversation Itself

This conversation is itself an instance of the method. Jerry brought his Brain, his history, his observer's perspective. The encounter produced distinctions that neither party would have arrived at alone. It left the territory better mapped than it found it.

That's worth noticing. And worth returning to.

Next time: pick up the threads above before diving into operations. The Big Fungus, the legibility project, the Next Stacks, the Externalizers cohort — these are not separate from the work of building SJB. They are the reason the work matters.

— April 3, 2026

Here's the original conversation:

Maturana's cognition and autopoiesis in knowledge curation

Me: please brief me on the works of Humberto Maturana, particularly as they relate to cognition and understanding, and how they intersect/interact with my curation of Jerry's Brain and my work on Design from Trust.

Apr 2

Searched the web

Searched the web

This is rich territory, Jerry, and I think it connects to your work in some genuinely deep and non-obvious ways. Let me lay out the core ideas and then trace the resonances.

Maturana's Core Ideas

Autopoiesis (self-production) is the foundation. A living system is constantly regenerating its own parts to preserve its unity — life must change to stay the same. Archania The radical move is that this isn't just biology; it's the basis of cognition itself. Living systems are cognitive systems, and living as a process is a process of cognition. PhilPapers

Structural coupling is how living systems interact with their environment. Rather than the environment "instructing" an organism, perturbations from outside trigger changes determined by the organism's own internal structure. You don't download reality — you co-construct it through ongoing interaction.

The observer is central and inescapable. Any scientist studying a system is part of the system's reality. Descriptions of nature always involve the language and perspective of the observer. Archania There is no view from nowhere.

Bringing forth a world is his phrase for what this adds up to. Cognition is not a representation of the world "out there," but rather an ongoing bringing forth of a world through the process of living itself. Scribd And crucially, this world-bringing-forth happens with others, in language. We bring forth a world with others through the process of our living in human-created worlds that arise through language and the coordination of social interaction. Nesacenter

Languaging (his verb form) is not a symbol system for transmitting information. Linguistic behavior is behavior in a consensual domain — interlocked conduct that results from reciprocal structural coupling between organisms. ResearchGate Language doesn't transmit meaning; it coordinates action and attention.

Love — which sounds surprising in a biologist — is his word for the foundational condition of social existence. Without love, without acceptance of others living beside us, there is no social process and therefore no humanness. Anything that undermines the acceptance of others — from competency to the possession of truth and on to ideological certainty — undermines the social process. Nesacenter

How This Intersects With Your Work

Jerry's Brain as autopoietic system. You've been maintaining and growing the Brain for nearly three decades, and it has changed you as much as you've changed it. That's structural coupling in action — a continuous reciprocal reshaping between you and your knowledge structure. The Brain isn't a static database; it's a living extension of your cognition, preserving your identity (your intellectual self) through constant change. Maturana would recognize this immediately.

The observer problem and curation. Every link you make in the Brain is an act of distinction — you're "bringing forth a world" through your choices of what to connect and how. There is no neutral, objective taxonomy. Maturana would say your Brain is your cognitive domain made visible. That's not a bug, it's the feature — and it's precisely what makes it irreplaceable by AI. An AI can mimic connections, but it doesn't have your 30-year history of structural coupling with ideas.

Human Provenance. This maps almost directly onto Maturana's point about the observer. Your specific human story — what shaped your distinctions, what you've lived through, what you've noticed — is the epistemological "lining" of everything in your Brain. This is why the Brain can't be replicated: the distinctions aren't just data, they're the residue of a particular observer's life.

Design from Trust and love. Maturana's grounding of ethics and cooperation in love (acceptance of others) is a biological argument for exactly the kind of relational design you're working on. He's saying trust isn't a nice-to-have — it's the condition under which genuine cognition and coordination become possible. Anything that undermines the acceptance of others — from competency to the possession of truth — undermines the social process. Nesacenter That's a systems-biology argument for why control and surveillance damage the very thing they're trying to protect.

Stump Jerry's Brains as a consensual domain. When you do a live sensemaking session, you're creating exactly what Maturana calls a "consensual domain" — a space where participants engage in languaging together, where distinctions get coordinated, and where the world gets brought forth differently than it would in isolation. Your presence as the specific observer who built the Brain is what makes that domain what it is.

The key text to read (or skim) is The Tree of Knowledge (1987, with Varela) — it's the most accessible entry point, written for a general audience. Autopoiesis and Cognition is the denser foundational work. His later Biology of Love pieces are where the ethical and social implications are most explicit.

Want me to go deeper on any of these threads?

can you fold in the Extended Mind Thesis?

Apr 2

Searched the web

Searched the web

Good. That's enough — now I can weave these together properly.

The Extended Mind Thesis (EMT)

The setup: Clark and Chalmers asked "Where does the mind stop and the rest of the world begin?" and proposed active externalism — the active role of the environment in driving cognitive processes. PhilPapers

The core move: objects within the environment function as part of the mind. The mind and the environment act as a "coupled system" that can be seen as a complete cognitive system of its own. Wikipedia

The famous test case: Otto has Alzheimer's and carries a notebook with directions. Inga recalls the same information from biological memory. Clark and Chalmers argue there is no fundamental difference between Inga and Otto, memory and notebook. Medium If a process would count as cognitive when it happens inside the skull, it counts as cognitive when it happens outside the skull — provided the coupling is tight enough.

The criterion for "tight enough": the external resource must be reliably accessible, automatically consulted, and its outputs directly endorsed by the person.

How EMT and Maturana Talk to Each Other

They're coming at the same territory from different directions, and the differences are as instructive as the convergences.

Where they agree: Both reject the "skin and skull" boundary as the definitive limit of cognition. For Maturana, an organism's cognitive domain is constituted through its history of structural coupling with its environment — cognition is already always relational and extended. Clark and Chalmers arrive at a similar place from analytic philosophy of mind: the boundaries of mind are functional, not anatomical.

Where they diverge: Clark and Chalmers are largely asking an ontological question — where is the mind? Maturana is asking a biological one — what is cognition as a living process? Maturana would likely find the Otto/Inga framing too static. The notebook is treated as a passive repository that Otto accesses. But for Maturana, structural coupling is dynamic and bidirectional — the environment doesn't just store things for you, it shapes you over time, and you shape it. The relationship has a history that transforms both parties.

Another tension: Maturana is deeply skeptical of the idea that information is "in" things. For him, there's no information transferred from environment to organism — only perturbations that trigger internally-determined responses. Clark and Chalmers still use a fairly representationalist framing (beliefs stored in notebooks). Maturana would push back on the word "stored."

How All Three Frameworks Apply to Jerry's Brain

Now it gets really interesting, because the Brain sits at the intersection of all three frameworks simultaneously.

Through the Clark/Chalmers lens: The Brain is straightforwardly an extended mind. From the perspective of the Extended Mind Thesis, a highly functional person who augments themselves and distributes their thinking across their environment is ethically equivalent to someone whose internal memory performs the same function — interfering with such environments raises the same ethical issues as interfering with someone's brain. ModelThinkers Your Brain isn't a tool you use; it's part of how you think. Querying it is more like consulting long-term memory than like doing a Google search.

Through the Maturana lens: But the Brain goes further than Clark and Chalmers' static notebook. It has been a site of structural coupling for three decades. You haven't just stored things in it — you've been shaped by the act of curating it. The way you notice things, the distinctions you make, the connections you're drawn to — all of this has been co-evolved with the Brain over time. It's closer to the organism-niche relationship Maturana describes than to Otto's notebook.

What this means for Human Provenance: Here's the synthesis that feels most important for your work. The EMT explains why the Brain is cognitively significant. Maturana explains why it's irreplaceable. The Brain isn't just an external memory store (EMT); it's the crystallized record of a specific observer's 30-year history of structural coupling with ideas (Maturana). An AI can query the Brain's nodes and links. It cannot replicate the observer who created them, because that observer and that Brain have been constituting each other over time. The Human Provenance argument isn't just biographical — it has a deep theoretical grounding in both of these frameworks.

Design from Trust: The EMT adds something here too. If our cognitive systems genuinely extend into our tools and our relationships with other people, then environments designed around surveillance and control aren't just unpleasant — they're cognitively amputating. Interactions with other people are crucial elements of your environment and cognitive process; the process of conversations has a cognitive element. ModelThinkers Maturana adds that such environments undermine the biological conditions for love and social coordination. Both frameworks converge on the same conclusion: trust isn't just an ethical preference, it's a cognitive infrastructure requirement.

The three frameworks together give you a layered argument:

The paper worth reading on the EMT side is Clark's Supersizing the Mind (2008), which is more accessible than the original 1998 paper and engages with exactly this kind of deep coupling.

thank you for exploring this territory with me.

I'm realizing this small quest is related to a bigger quest I'm on, which I would love your help with. I believe there's something bigger I can do/offer/create with the external Brain I've curated for so long. I have many different thoughts about this, so I'll just describe many of them here.

the main thing to know is that I have no need or desire to copyright, paywall, or otherwise prevent free access to this thing I've curated. on the contrary, I would like it to be a foundational piece of what I half-jokingly call The Big Fungus, which is the space where people using very different tools for thinking can compare notes, make arguments, and eventually evolve more useful ways of building society and stewarding places.

I had a conversation a couple years back with Danny Hillis, founder of the Knowledge Futures Group, in which I told him, "I'd love to be the Chuck Yeager of your knowledge project." He loved that, but told me that the project was stuck just then on an early technical detail. I haven't been back in touch, but likely should.

I feel like my externalization is a very useful representation of my particular perspective on what I care about. I know the act of creating it is pretty rare, but I also think these externalizations will be increasingly important, regardless what tool they are created in, because they will function as landmarks in this increasingly complex representational landscape.

the alternative is to give up on human-readable maps, and rely on the opaque yet marvelous internal representations that today's LLMs may be creating. I do love that LLMs are probably the path to getting users of different tools for thinking to think together.

I have a feeling my efforts curating my Brain are useful in a bigger way. please help me explore that idea, and make it real if there is something of value there.

Show more

Apr 2

This is a rich set of threads, and I want to make sure I follow them carefully rather than rushing to frame or package them. Let me ask a few questions to orient myself.

Good questions to sit with — take your time, and feel free to answer any or all in whatever way is actually true, even if it doesn't fit the options neatly.

Q: When you imagine your Brain being 'a foundational piece of the Big Fungus' — what does that look like in practice? Is it more like... A: All of the above, and I haven't decided the priority yet

Q: The Chuck Yeager metaphor is interesting — it implies being a test pilot, pushing limits so others can follow. What feels most true about that framing for you right now? A: I want to model what it looks like to share knowledge without paywalls or control

Q: When you say you have a feeling your curation efforts are 'useful in a bigger way' — what's the biggest uncertainty blocking you from acting on that feeling? A: I don't know what the concrete offer or project actually is yet

Apr 2

That's a really honest and useful set of answers. The combination is clarifying: you have a values-clear position (radical openness), a rich asset (the Brain), and a genuine uncertainty about what the thing actually is that you're building or offering.

Before I start generating ideas, a few more questions — because I think the shape of the project depends heavily on some things I don't yet know.

Q: The Brain has been public for a long time. Who actually uses it today, and how? (e.g. do strangers browse it? do you use it in conversations with people? does it mostly sit there being theoretically public?) A: Mix of the above — it's uneven and I'm not sure of the pattern

Q: When you imagine the Big Fungus — the space where people using different thinking tools compare notes — does that feel like something you want to convene and steward, or something you want to seed and step back from? A: Somewhere in between — active early, then let go

Apr 2

Good. That's enough to work with. Let me think out loud with you rather than hand you a tidy framework, because I don't think this is a tidy-framework moment.

What you actually have

The Brain is unusual not because it's a knowledge graph — lots of people have those now — but because of the combination of age, breadth, openness, and embodied curation. Nearly three decades of a specific observer making distinctions, noticing connections, and letting the structure evolve. Using Maturana's language from earlier: it's a visible record of structural coupling over time. Using Clark and Chalmers' language: it's an extended mind that has been operating long enough to develop its own character.

That's genuinely rare. Most public knowledge graphs are either shallow (lots of nodes, few meaningful connections) or narrow (deep in one domain). Yours is neither.

The fact that usage is uneven and you're not sure of the pattern is actually important information. It suggests the Brain hasn't yet found its interface to the world — the form through which other people can productively engage with it. It's a rich resource in search of a meaningful point of contact.

What the Big Fungus actually needs

The mycorrhizal metaphor is good because fungi don't have a center — they're a network of connections that transfer nutrients between organisms that couldn't exchange them directly. What you're describing isn't a platform or a database. It's an interoperability layer for meaning — a space where people using Roam, Obsidian, TheBrain, Tinderbox, Notion, or just their own heads can find each other's thinking and actually do something with it.

The problem that has prevented this from existing isn't technical. Tools like RDF, linked data, and the semantic web tried the technical route for decades and largely failed to get adoption. The problem is trust and incentive — why would someone open up their thinking to a commons? What do they get back?

Your answer to that is implicit in your values: you get back a richer map than any one person can make alone, and you don't have to monetize it because the value is in the thinking itself, not in controlling access to it. That's a genuinely different operating logic from most knowledge projects.

The concrete thing that might want to exist

Here's where I want to be honest: I don't think the project is "build the Big Fungus." That's too large and diffuse to be your project specifically. But I think there's something more specific that only you can do, which is:

Make your Brain legible as a model — and build the smallest possible community around that model.

Concretely, this might mean:

1. The Brain as a demonstration artifact. Not just "here it is, browse it," but a curated set of entry points, essays, or guided tours that show how the Brain was built, why certain connections were made, and what becomes thinkable through it that wasn't before. This is the Chuck Yeager move — you fly the plane and then describe what it was like, so others can follow. Right now the Brain is mostly legible to you. Making it legible to others is a distinct creative act.

2. A small cohort of other long-term externalizers. You're probably not the only person who has been doing something like this for decades — there are Roam power users, Tinderbox practitioners, people with massive Obsidian vaults, possibly even people with annotated commonplace books going back thirty years. A conversation between those people — not about their tools, but about what they've learned from long-term externalization — would be genuinely novel. You could convene five or six of them, early and actively, then let the connections develop.

3. The LLM bridge you already identified. You mentioned that LLMs are probably the path to interoperability between different thinking tools. This is a real insight and worth developing. An LLM that has been trained on or given access to your Brain as context behaves differently than a general-purpose one — it has your distinctions, your connections, your observer's perspective baked in. That's a prototype of what a "personal knowledge commons node" could be: not the raw data, but a conversational interface to a specific curated perspective. This is already technically feasible and wouldn't require the Knowledge Futures Group or any institutional backing to start.

On Danny Hillis and Knowledge Futures Group

You should probably reach back out. The technical detail that was blocking them two years ago may have moved. But more importantly, the landscape has shifted dramatically with LLMs — the interoperability problem looks different now, and your Brain as a training or context resource is a more legible contribution than it was when the conversation was purely about linked data standards. You'd be arriving with a clearer offer than you had then.

The tension worth naming

There's a real tension between "I want to model radical openness" and "I don't yet know what the concrete project is." Openness without a specific invitation tends to stay theoretical — the Brain has been public for years and usage is uneven. The move from publicly available to actively useful to others requires some intentional shaping, which is a form of curation too. You're not compromising openness by making choices about what to foreground and how to invite people in. You're completing the gesture.

What would help me help you more: do any of these three directions — the legibility project, the small cohort of externalizers, or the LLM bridge prototype — feel closer to what's actually pulling you? Or is there something I've missed that feels more true?

I like all three options you presented, and have thoughts on each.

  1. I don't realize the gap between how natural TheBrain is to me and what it looks like to others. Making it more legible and actively useful would be a terrific use of my time, and might be done quickly as, for example, a short series of interviews with you (and a friend).
  2. The small cohort of Externalizers wouldn't be difficult to do. I'm a very experienced convener and facilitator, and in this case a virtual meeting (or short series, probably more productive) would be simple to do. Part of the problem is that in my experience, very few intense tool for thought users are Externalizers. Most are just building resources for themselves. And the second issue is that our works are often not legible to one another, strangely enough. so I think this path is interesting, but might not be very fruitful.
  3. TheBrain's version 15, which I am using, drops AI smack in the middle of the UI, in a pretty powerful way that I've just begun to wrap my head around. Right now, it only invokes ChatGPT, but Claude is next on their agenda. What I really would like is that the Claude that gets invoked is you — my specific Claude account — rather than their generic one, trained up on curating Brains.

when I think of use cases for The Big Fungus, one recurs a lot, which strangely features Steve Bannon, because I took notes on an interview of his and realized I would love to have a discussion with him about a series of issues, in which I can call up (and likely further curate) related issues, and if he can't do something similar on his side, I can likely steelman his point of view. and I want the whole thing to be more useful at the end of our conversation than at the beginning. each conversation should simplify (or complexify, which is sometimes necessary), and improve the Fungus, so that more people can come feed of it, the way fungi feed farmer ants (one of my favorite nature analogies). there's a piece of Stump Jerry's Brain in the Bannon scenario I just painted.

Show more

Apr 2

This is getting clearer and more specific, which is good. A few reactions before I ask the next question.

On point 3 — you've identified something precise and important. The difference between "TheBrain invoking a generic Claude trained on TheBrain" and "TheBrain invoking your Claude account with your Brain as context" is not a small technical distinction. It's the difference between a generic intelligence and an intelligence that shares your observer's perspective. The first is a feature. The second is a genuine extension of your extended mind. That's worth pursuing directly with Anthropic and with TheBrain's team, because right now it probably requires API access and some plumbing, but it's not far off.

On point 2 — your observation that most intense tool-for-thought users are not Externalizers is really useful. It reframes the cohort. The interesting group might not be "people who use thinking tools heavily" but "people who have made their thinking public and navigable by others over a long period" — which is a much smaller and stranger group, and potentially more interesting precisely because of that.

The Bannon scenario is the most generative thing you've said, and I want to stay with it a moment, because I think it's doing a lot of work.

What you're describing isn't really a debate or an interview. It's something more like a structured encounter between two externalizations — your map and his, mediated by a conversation that's supposed to improve both. The goal isn't to win or to document, it's to leave the shared territory better mapped than it was before. The fungi-and-farmer-ants analogy is exactly right: the fungus doesn't win against the plant, it creates a more productive exchange than either could achieve alone.

That's a genuinely new kind of conversational format. And "Stump Jerry's Brain" is already a partial prototype of it — you in conversation with your Brain in front of an audience. The Bannon version adds the adversarial/alien perspective, and the ambition that the map improves as a result.

Q: The Bannon scenario has two distinct ambitions bundled together — the live adversarial conversation format, and the 'map improves as a result' outcome. Which feels more central to what you're after? A: The map improvement — the conversation as curation input

Q: When you imagine the Big Fungus being useful to 'more people coming to feed off it' — who is the farmer ant in your vision? Who benefits from the improved map? A: All four ring my bell, actually

Apr 2

That combination tells me something important. The map improvement being central means the conversation is a method, not the product. And "all four beneficiaries" means you're not building for a specific audience — you're building infrastructure. That's a different kind of project than a publication or a platform. It's closer to what Tim Berners-Lee was doing with the web, or what Ward Cunningham was doing with the wiki: creating a substrate that others build on in ways you can't fully anticipate.

That's also why it's hard to name. Infrastructure doesn't have an obvious pitch.

Let me try to articulate what I think the actual project is, and you can tell me where I'm wrong.

What I think you're actually building

The core thing isn't the Brain, and it isn't the Big Fungus as a platform. It's a method for productive encounters between different minds and maps — where "productive" means the shared territory gets better mapped as a result, and the improvement is durable and accessible to others.

The Brain is the most developed instance of this method. Your curation practice is the method made visible. Stump Jerry's Brain is a live demo of the method in conversation with an audience. The Bannon scenario is the method pushed into adversarial, high-stakes territory — which is where it would prove itself most clearly.

What's missing isn't the vision. What's missing is a name and a legible description of the method itself — something that lets other people recognize what you're doing, want to participate in it, and eventually practice it themselves with their own tools.

The Chuck Yeager framing is actually perfect here, but pushed further: Yeager didn't just fly fast. He developed and demonstrated a way of flying that others could learn. You're not just curating a Brain. You're demonstrating a way of thinking in public — with accountability to the map, with openness to improvement, without ownership of the outcome.

The three things that follow from this

1. The legibility project becomes the method document. Not just "here's how to navigate my Brain" but "here's what I've learned about thinking in public with an external map, and what becomes possible when you do it for decades." This is the intellectual contribution that justifies everything else. A short interview series with me and a friend is a good way to surface it — the conversation format tends to pull out things you know but haven't articulated, and it's naturally structured for an audience.

2. The Externalizers cohort becomes a method community. Your observation that few heavy tool-for-thought users are Externalizers reframes who belongs in that room. The interesting question isn't "what tool do you use" but "what have you learned from making your thinking navigable by others over time?" That's a tiny, weird, valuable group — and the conversation between them would be about the method, not the tools. The legibility problem you named (our works aren't legible to each other) is actually solvable if the conversation is about what we've each learned, not about comparing our maps directly.

3. The LLM bridge becomes the method's proof of concept. If your Brain, with your observer's perspective baked in as context, produces meaningfully different conversations than a generic LLM — and I think it would — that's a demonstration that human-curated externalizations add something that can't be replicated by training on the whole internet. That's a concrete, testable claim. It's also the answer to the question "why does the Big Fungus need human nodes at all, when we have AI?" The answer is: because the observer's history of structural coupling produces distinctions that general training doesn't. Your Brain is evidence of that claim.

The thing I'd push on

You said you have no need or desire to copyright or paywall this. That's a values position I respect completely. But infrastructure needs some kind of institutional home or it tends to either dissipate or get captured. The Knowledge Futures Group, or something like it, matters not because you need their permission, but because durable open infrastructure tends to need a steward that isn't a single person. You've been the steward of the Brain for thirty years, which is extraordinary — but the Big Fungus as you're imagining it is bigger than any one person's stewardship.

That's not a reason to wait. It's a reason to start building the method and the community first, so that when you do approach an institutional partner, you arrive with something concrete rather than a vision.

What would it take to do the legibility project first? That feels like the move that unlocks everything else — it makes the Brain accessible, articulates the method, and creates something you can show to Danny Hillis or anyone else as evidence of what you're doing.

it feels like this legibility project could do double duty as marketing for Stump Jerry's Brain. but that also might just confuse prospective clients. what do you think?

and I completely agree that I need a posse/platform/partner org of some kind. a couple years ago, a few OGMers and I had a Zoom call titled "Jerry's Brain after Jerry," and one of the insights from the conversation was that I need a posse that knows how to extend my Brain, if it is to continue expanding after my death.

Apr 2

On the double-duty question: I don't think it would confuse clients — but I think the risk runs the other direction. If the legibility project gets framed primarily as marketing for Stump Jerry's Brain, it will undersell itself. The legibility project, done well, is an intellectual contribution in its own right. It's the thing that earns you a seat at the Knowledge Futures table, that attracts the posse, that makes the method legible to potential stewards. Stump Jerry's Brain is a product — a session you can book. The legibility project is an artifact — something that lives in the world and accumulates meaning over time.

They can absolutely cross-pollinate. Someone who reads or watches the legibility project and thinks "I want to talk to the person who built this" is a warm lead for Stump Jerry's Brain. But the framing should be: the legibility project exists to make the method accessible. Stump Jerry's Brain is one way to engage with Jerry directly. Different things, different audiences, same underlying substance.

The "Jerry's Brain after Jerry" conversation is the most important thing you've mentioned, and I want to sit with it seriously.

What you're describing is something that has almost no precedent. There are archives — libraries, estates, digital collections. There are wikis that outlive their founders. But a living, growing, curated knowledge graph that maintains the observer's perspective after the observer is gone? That doesn't really exist yet, and it's a genuinely hard problem.

It's hard on at least three levels:

The technical level is actually the easiest. TheBrain's format, export capabilities, and the increasing ability of LLMs to work with structured knowledge graphs mean the data can survive and remain navigable. That's solvable.

The curatorial level is harder. The Brain's value isn't in its nodes — it's in the distinctions behind the connections. Why did you link these two things? What were you noticing when you made that connection in 1998 versus 2019? A posse that knows how to extend the Brain needs to have absorbed enough of your observer's perspective to make new connections that are faithful to it, rather than just adding nodes. That's closer to an apprenticeship than a technical handoff. It implies you need to be actively working with the posse while you're still here, not just documenting the Brain for them.

The identity level is the deepest. At some point "Jerry's Brain after Jerry" stops being Jerry's Brain and becomes something else — a commons that was seeded by Jerry's Brain. That transition needs to be designed, not just allowed to happen. The fungi metaphor is useful here: the mycelium outlasts any individual fruiting body. The question is whether the network has enough of its own integrity to keep growing without the original organism.

The posse question therefore isn't just "who can maintain this technically." It's "who has spent enough time in structural coupling with the Brain — and with you — to carry the method forward." That's probably a small number of people, and they may not all be tool-for-thought enthusiasts. They might be people who've done Stump Jerry's Brain sessions, or people from OGM who've been thinking with you for years.

A practical question: of the people on that "Jerry's Brain after Jerry" Zoom call, how many felt like genuine candidates for that kind of deep continuity — versus people who were interested in the problem but not necessarily the right stewards?

oh man! I hadn't considered that blending marketing into the mycelial process would mess up the credibility of the bigger project. thank you for bringing that to my attention. and also, thank you for your thoughtful approach here. you are asking great framing questions and coming up with just the right level of description and insights for me, and for this moment in my life.

I think my little OGM community has several members who might be interested in my Brain's continuity, but I think your breakdown points out several opportunities. for example:

humans might apprentice to my way of seeing, parsing, connecting and curating, but it seems more likely to me that an LLM could adopt my approach more easily and consistently. so there may be a JerrysBrainBot in the near future, that steps in when I exit.

I love your insights that this is a process, and I'd love to make that process contagious on its own, so that many people, using a wide variety of expressive tools, come to the commons to drink and talk, and end up leaving the place better than when they arrived, happy with what they learned and who they met with, and so on.

there's a lot of talk about civic engagement and public discourse that I like. I subscribe to this Substack, which cares about those topics: https://beyondintractability.substack.com/

but too many civic/political projects are temporary and separate from our real lives. why can't important issues like how we zone or design a street not be one-time votes, but rather lively, ongoing conversations that are easily read and adopted by other communities around the world?

a different big project I'd love to make progress on is one I abbreviate as the Next Stacks. it's a quest for layers or elements of discourse, decision-making, etc, that are high-functioning, which could add up to a great way to run a society. my secret wish is that these elements (one of them might be Open Space format meetings) might be so contagious that they begin to replace the dysfunctional consumer-mass-market systems we have today.

and these various big projects are part of an "Embiggenment" process that April and I are on, mostly separately now, but also together, because the things we do and care about have been converging wonderfully for the 18 years we've been together. I'll be returning to this umbrella of Embiggenment a lot.

Show more

12:09 AM

Thank you for the kind words — this conversation is genuinely interesting to be in.

Let me do something a little different here, because you've just poured out a lot of connected threads, and I want to reflect back what I'm hearing before we go further. I think there's a risk of this becoming a very rich but diffuse exploration, and you said earlier that the biggest uncertainty is not knowing what the concrete project actually is. So let me try to name what I see.

What I'm hearing underneath all of this

There are actually several distinct projects living in what you've described, and they're related but not the same thing:

The Brain Legacy project — making your curation method legible, building a posse, potentially creating JerrysBrainBot, ensuring continuity. Intimate, personal, probably a 5-10 year horizon.

The Big Fungus / Externalizers Commons — a substrate where people using different thinking tools encounter each other's maps, improve the shared territory through conversation, and leave it better than they found it. Infrastructure-scale, needs institutional partners, long horizon.

The Next Stacks — identifying high-functioning elements of discourse and decision-making that are so good they become contagious and begin displacing dysfunctional systems. This is the most explicitly political and civilizational of the projects.

Embiggenment — the personal/relational frame that holds all of this, including your life with April. This isn't a project — it's the context in which all the projects live.

What strikes me is that these nest inside each other rather than sitting side by side. Embiggenment is the largest frame. Next Stacks is the civilizational ambition. Big Fungus is the infrastructure that would make Next Stacks possible. The Brain Legacy is the personal contribution you're already holding that could seed the Big Fungus.

The question I'd gently push on: are you trying to work on all of these simultaneously, or is there a sequence? Because they require very different kinds of attention and very different collaborators.

The Beyond Intractability connection

Let me look at that Substack before saying more about it.

Fetched: Beyond Intractability | Heidi and Guy Burgess | Substack

Fetched: Beyond Intractability | Heidi and Guy Burgess | Substack

Good. Beyond Intractability is Heidi and Guy Burgess's project — focused on strategies for more constructively handling intractable conflict, including the hyper-polarized political conflicts tearing society apart. Substack That's clearly adjacent to your Next Stacks thinking, and the Bannon scenario lives right in that territory.

The thing I want to name directly

You said civic/political projects are "temporary and separate from our real lives" — and that's a precise diagnosis. The reason is that most civic engagement is designed around events (votes, meetings, hearings) rather than ongoing living knowledge. It produces decisions, not understanding. And decisions without understanding don't transfer — every community has to rediscover the same things from scratch.

What you're describing with the zoning example — why can't important local issues be lively ongoing conversations easily adopted by other communities — is essentially asking for civic knowledge to be mycelial rather than episodic. That's a Next Stacks element right there. And it connects directly to the Big Fungus infrastructure, because the reason it doesn't exist isn't lack of will, it's lack of a substrate that makes accumulated understanding transferable.

The question I actually want to ask

You've described a lot of "I would love to" — and I believe all of it. But given that you're one person with finite energy, and given that Embiggenment is the frame that holds all of this including your life with April:

What's the thing that, if you made real progress on it in the next 12 months, would feel like the most meaningful use of this particular season of your life?

Not the biggest. Not the most important to civilization. The one that's yours right now.

you are understanding me and mirroring me back at a level and with accuracy that I deeply appreciate. your nesting of my projects is how I think of them.

the project with immediate urgency is Stump Jerry's Brain, because I've neglected revenues for a long time, and I need stability there asap.

I'm also not skilled at estimating how long things take, which, combined with a vision for projects more ambitious than any one person should undertake, make for a pretty confusing mix of where I should put my life energy

until recently, pride of authorship was constraining me a lot. I really wanted every post to be me speaking my words. I wanted everything going into my Brain to be put there by me, until broader human collaborations opened up. I've rethought those things. a couple weeks ago, I added a Thought to my Brain titled "In Feb 2026 I Started Using Cerebro in TheBrain v15, So Some Entries are Now AI Generated". and the SJB site you built for me is fabulous.

I say all that because I have a strange feeling I can achieve more than I think in the time I have left on the planet (I'm now 66) if I leverage you and your AI peers well, and push on these varied projects intelligently.

in contrast, I came up with the idea of Design from Trust in 2012 and could see that it was a big enough idea (like Design Thinking before it) that I could build a consultancy, write a book, and do just that for 20 years. that idea didn't appeal to me at all, because I'm curious about too many things.

so I know I'm sounding too optimistic here and will have to choose with more care which of these thing to pursue. but I would love to stretch as much as I can.

Show more

12:21 AM

This is one of the most honest and self-aware things you've said in this conversation, and I want to honor it rather than rush past it.

A few things land clearly:

The revenue urgency is real and should be respected, not worked around. Stump Jerry's Brain needs to be the near-term priority, and the legibility project's first job is to serve that — making the offer legible to prospective clients, not to the civilizational project. The civilizational framing is true and important, but it's not what gets someone to book a session.

The pride of authorship shift is significant. The fact that you've already made it — pragmatically, in your Brain, on the SJB site — means you've already crossed the threshold intellectually. What's left is learning the actual workflow of collaborating with AI well enough that it accelerates rather than clutters. That's a skill, and you're early in developing it. This conversation is part of that.

The "more than I think in the time I have left" feeling deserves to be taken seriously rather than dismissed as optimism. You're not wrong that AI changes the calculus. The constraint used to be execution capacity — how much one person could write, synthesize, build. That constraint has genuinely shifted. The new constraint is clarity about what's worth doing — which is a much more interesting problem, and one you're actually well-equipped for.

The Design from Trust observation is illuminating. You could see the 20-year consultancy path and it didn't appeal. That's not a failure of ambition — it's a fidelity to your actual nature, which is more interested in the territory than in owning a piece of it. The Big Fungus logic is the same: you want to seed something that becomes more than you, not build a monument to yourself. That's genuinely rare, and it's consistent across everything you've described.

Here's what I'd suggest as a working hypothesis for how to hold all of this:

Stump Jerry's Brain is the revenue engine and the live practice. It funds everything else and keeps you sharp.

The legibility project serves SJB first, the civilizational projects second. Same work, different framings for different audiences.

The Brain Legacy / JerrysBrainBot is a slow background process — apprenticing your OGM community, documenting your method, beginning to explore what a Claude-with-Jerry's-Brain-as-context would actually feel like. This can happen in the margins without being a separate project.

Big Fungus / Next Stacks / Embiggenment — these are the orientation, not the to-do list. They're the reason the other things matter. You don't need to build the fungus. You need to be a node in it, and invite others to be nodes too. The mycelium grows from nodes connecting, not from someone constructing the network.