# Brain Flywheel and Metapraxis Three-page-equivalent brief · Working proposal · 22 October 2026. The three sections below are intended as separate pages when typeset. The aim is to preserve a simple invitation while explaining the operational and social conditions needed to make it credible. ## Page one: The concept ### Link understanding, not just information Brain Flywheel begins with the idea that everyone has something they explore deeply. Their contribution may be a research collection, a creative technique, a mathematical framework, a repeatable workflow, or practical judgment developed through experience. The proposal is to distill that work into a knowledge brain: a bounded representation of artifacts, concepts, relationships, principles, assumptions, operations, examples, evidence, and limitations. It should be an interface to a person's practice, not a claim to have captured their whole mind. “Link brain” means more than a friend request or a file transfer. It is an invitation to connect the useful structures behind two bodies of knowledge and test what they can make possible together. ### Keep the constructs distinct - **Knowledge brain:** What a person or community makes reusable. - **Link brain:** The voluntary exchange, adaptation, or collaboration. - **Brain Flywheel:** The cycle that makes shared knowledge useful and returns learning and value. - **Metapraxis:** The practical grammar for documenting, standardizing, composing, and revising how work is done. Metapraxis also functions as the user's higher-order “brain brain”: a mesh for understanding how different constructs, principles, logics, and outputs relate and act. Its practical expression is a set of clear descriptions and interfaces, not an abstract vocabulary imposed on every participant. ### A pattern and the playbook within it The user's archetype and hero's-journey analogy distinguishes a recurring pattern from the practical knowledge required to navigate it. “Encounter a challenge and return changed” is the pattern; diagnosis, collaboration, testing, recovery, and documentation form the playbook within it. Metapraxis makes those practical details explicit. It should standardize how a method is described without pretending that every situation calls for the same method. ### Why a connection could create something new Structural analogy offers a useful foundation: map relationships, not just surface similarities. Conceptual blending offers another: selective combination and elaboration can create structure not present in either input alone. ([Northwestern](https://www.qrg.northwestern.edu/ideas/smeidea.htm), [Fauconnier and Turner](https://onlinelibrary.wiley.com/doi/abs/10.1207/s15516709cog2202_1)) For this proposal, the test is practical rather than rhetorical. Does combining two contributions enable a useful result, improve a method, or reveal a previously hidden limitation? --- ## Page two: How the system works ### The six-stage cycle Collect → Curate → Connect → Apply → Learn → Reinvest. Collect real artifacts of work, then curate enough context to make them understandable. Connect complementary modules through explicit mappings, apply them to a real task, evaluate the result, and return useful changes to the shared foundation. The overview diagram compresses these into four operations: Distill knowledge, Link brains, Build and test, and Return value. Metapraxis supports every operation rather than appearing as an additional stage. ### Make the connection a first-class artifact A transfer record should preserve the source and destination, the purpose of the connection, and the roles or relationships that correspond. It should also state what changes, what must remain invariant, what does not transfer, and how the result was tested. For example, a music module and a software-scheduling module might help build an audiovisual sequencer. Beats and bars could inform timing boundaries, but expressive timing and computational correctness should remain distinct, with their relationship explicitly tested. The resulting artifact might improve both source modules. The scheduling module could gain a new timing test, while the music module gains a clearer account of the expressive behavior it needs. ### Preserve existing knowledge as carefully as novelty The codex should distinguish definitions, assumptions, formal consequences, empirical regularities, models, heuristics, procedures, counterexamples, creative twists, and open questions. These are proposed record types, not interchangeable grades of truth. “Memorializing” means attributed, versioned preservation that allows correction. A creative twist should identify what is being adapted or inverted, rather than quietly rewriting the original principle. ### Know whether the wheel is actually turning Track whether another person can reuse a contribution, whether the result passes its intended checks, and whether the next cycle starts from a better position. Capture curation and integration costs as well as benefits. A practical scorecard would include time to validated reuse, output quality, net effort saved, correction propagation, new capabilities, and contributor value. The latter may include learning, friendship, credit, paid work, or agreed revenue sharing. Keep a balancing loop alongside the productive loop: challenge claims, identify conflict, correct or retire weak material, and notify dependent users. More accumulated material is not automatically more useful knowledge. --- ## Page three: Participation, safeguards, and first experiment ### Two problems, not one The representation problem asks whether knowledge can be codified and transferred usefully. The participation problem asks whether people want to offer, request, and receive it, and whether the exchange feels worth the effort. The system should support small exchanges: one question, one method, one worked example, one collaboration. It should not require disclosure of a whole personal archive or private professional context. ### Two central failure modes - **Capture and monopolization:** A platform or participant could control access, discovery, attribution, or the returns from everyone else's knowledge. - **Hollow participation:** Contributions could be fabricated, shallow, outdated, or too poorly contextualized to use. These could reinforce each other: extractive terms discourage substantive sharing, while shallow inputs undermine trust and reuse. This is a risk scenario, not an inevitable prediction. Proposed safeguards include portable formats, nonexclusive contribution, clear licenses, contributor identity choices, provenance, worked examples, test evidence, and visible limitations. Honest incompleteness is acceptable; pretending a partial or invented module is comprehensive expertise is not. Access withdrawal should not promise impossible erasure. The system must distinguish stopping future access from retrieving already exported copies and define those limits before exchange. ### Philosophical inspiration without conflation The user's “rtagama” recollection remains unconfirmed as an exact term, but Vedic ṛta, Old Persian arta, and Avestan Asha correspond to an Indo-Iranian concept of truth/order; Druj, not Asha, is the opposing concept associated with falsehood or deceptive order. ([Iranica on Asha](https://www.iranicaonline.org/articles/ardwahist-av/), [Iranica on Druj](https://www.iranicaonline.org/articles/druj/)) This can inspire a design question: is the apparent structure grounded and useful, or merely persuasive? It should not be turned into a claim that an application's taxonomy embodies cosmic truth or that mistaken contributors are morally corrupt. The turning-wheel metaphor adds a participation principle: the system needs an initial push and continued input. Knowledge cannot compound through a decorative diagram alone. ### A bounded pilot Recruit two willing contributors with different methods and one shared task. Record one usable module from each, agree on scope and rights, propose a mapping, and build a small artifact. Compare against the same task and source materials without the structured modules where feasible. Predeclare success criteria, record limitations, and require a reuse attempt or a documented correction before calling the result a completed turn. The goal is not to prove a universal theory. It is to discover whether this exchange produces a better outcome and whether both parties want another turn.